183

INDOOR ENVIRONMENTAL QUALITY

This chapter deals the indoor environmental quality assessment in the following parts

with respective literature survey included:

Introduction to Indoor Environmental Quality(IEQ)

Methodology of IEQ optimization in Office Buildings with parameter and algorithm

wise results and discussions

Methodology of IEQ optimization in Resident Buildings with parameter and

algorithm wise results and discussions.

Bibliography

4.1. INTRODUCTION

There is a large untapped opportunity for economic benefits, resulting from

improvements in indoor environmental quality (IEQ) in non-industrial work places and

houses. (WilliamFisk & OlliSeppanen, 2007). The most clearly established sources of

economic benefits include improved work performance, e.g., work speed or quality, reduced

absence, and reduced health care costs. There is also evidence that providing better IEQ can

improve student learning which, in turn, should lead to more effective future workforces.

(WilliamFisk & OlliSeppanen, 2007). At the societal level, economic value can also be

assigned to the reduced suffering of ill health and to extended average lifetimes expected

when IEQ is improved.

The effects of indoor air quality on productivity became an issue only in the

last decade, as a result of extensive research and an understanding of the strong connections

between factors such as ventilation, air-conditioning, indoor pollutants and the adverse

effects on health and comfort. The complexity of a real environment makes it very difficult to

evaluate the impact of a single parameter on human performance, mostly because many of

them are present at the same time and as a consequence, act together on each individual. One

way of evaluating the performance is the use of self-reported performance.

184

It is clear that the indoor environment was evaluated to have the biggest

influence on performance, providing better job satisfaction and reducing job stress. A

common approach, to evaluate the influence of climatic factors on human performance could

be to measure the extent to which the Sick Building Syndrome (SBS) symptoms occur, as

these symptoms are known to cause distraction from work or even short-term absenteeism.

However, this link is not well established yet and must be better understood and recognized.

Possible reasons may be cited as follows: (1) inadequate ventilation or superfluous

emissions from different sources increase the concentration of pollutants, which negatively

affect perceived air quality; (2) reduced air quality negatively affects the central nervous

system, increasing SBS symptoms such as headache, difficulty in concentration, tiredness;

(3) these symptoms will cause distraction from work and decreased work ability, i.e.

productivity loss. (Olesen & Bjarne, 2009), indoor pollution may also exacerbate the

sensation of dryness and irritation of eyes. As a consequence, a higher blinking rate and

watery eyes will negatively affect visual skills and decrease the performance of visually

demanding work.

The indoor environmental quality (IEQ) in offices and residences are

examined from the prospect of an occupants‘ acceptance in four aspects: thermal comfort,

indoor air quality, noise level and illumination level. Based on the evaluations made by the

occupants of the IEQ, empirical expressions have been proposed to approximate an overall

IEQ acceptance of an office environment at certain operative temperature (To), carbon

dioxide concentration (CO2), equivalent noise level (Leq) and illumination level (lux). The

overall IEQ acceptance is calculated from a multivariate logistic regression model.

4.1.1. Literature survey

Physical environmental parameters such as air temperature, relative humidity,

acoustics, air quality, lighting, ventilation and air distribution are all interrelated, and the

feeling of comfort is a composite state of an occupant‘s mind, responding to the senses to

these factors (Goldman, 1999) (Haghighat & Donnini, 1999) (Mendell M. , 2003) (Naganoa

& Horikoshib, 2005). This state of mind is an intricate response to the indoor environmental

factor groups, including physical environment sustained by the building and its service

system, and individual physiological conditions such as health, social relations, financial

state, etc. Studies showed that an occupant‘s acceptance of an environment depended on a

185

number of environmental parameters. Four basic components, namely thermal comfort,

indoor air quality, aural and visual comfort were identified for determining an acceptable

IEQ. Conventional studies on indoor environment addressed each of them separately. They

are still addressed independently by designers for many office designs. More recently, the

equivalence of the discomfort caused by different physical qualities has been considered

(Clausen, Carrick, Fanger, KIm, & Rindel, 1993) ; (Fanger, Olf, & Decipol, 1988) (Pellerin

& Candas, 2004). Discomforts caused by indoor air pollution, thermal load and noise were

investigated. It was reported that at an operative temperature between 23 and 29 1C, each

degree Celsius change would associate the same effect on human comfort with a change in

perceived air quality of 2.4 decipol, or a change in noise level of 3.9 dB (Fanger, Olf, &

Decipol, 1988) . For levels of perceived air quality up to 10 decipol, a unit change had the

same effect on human comfort as a change in noise level of 1.2 dB (Clausen, Carrick, Fanger,

KIm, & Rindel, 1993).

The equivalence of acoustic sensation to the thermal one was proposed for

short-term exposure. Each degree Celsius change in temperature had the same effect of 2.6

dBA (Pellerin & Candas, 2004). Workplace variables inducing the largest number of health

symptoms, comfort or odour concerns were investigated by multivariate regression analysis

(Bulysssen & Cox, 2002); (Dan, 1993); (Toftem, 2002); (Sofuoglu & Moschandreas, 2003);

(Wallace, Nelson, & Dunteman, 1991). It was realised that a successful control of the indoor

environment required an understanding of the integral indoor environmental parameters. An

overall IEQ index would be derived to describe the state of mind of a user in balance with the

indoor environment. Unacceptable indoor environments are often manifest in some forms and

symptoms of sick building syndrome (SBS) prevail in many office buildings (Sofuoglu &

Moschandreas, 2003).

This study argues that the subjective evaluation of an indoor environment

being perceived by an occupant can be used to assess the acceptance of the IEQ. In particular,

occupants‘ acceptance of the four basic components of IEQ was evaluated and correlated

with the overall IEQ acceptance of an office environment. The occupants‘ attitudes towards

the operative temperature, CO2 concentration, equivalent noise level and illumination level

and the overall IEQ acceptance recorded by a dichotomous scale were studied (Houser &

Tiller, 2003); (InternationalStandardISO7730, 1994); (Wong & Leung, 2005).

186

Mathematical expressions were proposed for the overall IEQ acceptance using

a multivariate logistic regression model with the former four parameters recorded. The

proposed overall IEQ acceptance can be used as a quantitative assessment criterion for an

office environment and similar environment where an occupant‘s evaluation is expected. The

occupant‘s acceptance of the perceived indoor environment given by four aspects, namely

thermal environment, indoor air quality, equivalent noise level and illumination level, was

studied with a dichotomous assessment scale (Portney & Watkins, 2000).

Indoor environmental quality (IEQ) and occupant comfort are closely related.

Current indoor environmental assessment includes four aspects, namely thermal comfort

(TC), indoor air quality (IAQ), visual comfort (VC) and aural comfort (AC) (Clausen,

Carrick, Fanger, KIm, & Rindel, 1993); (Wong, Mui, & Hui, 2008). Maintaining satisfactory

thermal comfort conditions for the occupants by an adjustable indoor temperature set point of

the air conditioning system is one of the primary concerns in many air conditioned office

buildings. Thermal comfort relates human sensation and perception with a number of

environmental and physical parameters (Faanger, 1970). It is by definition the perception of

satisfaction, a subject experiences, in a given thermal environment (ANSI/ASHRAE55-2004,

2004). Extensive studies resulted in a number of thermal comfort equations as proposed in

some widely used design guides and standards (ANSI/ASHRAE55-2004, 2004);

(InternationalStandardISO7730, 1994). Three indices were derived based on Fanger‘s

comfort equation: Predicted Mean Vote (PMV), Predicted Percentage Dissatisfied (PPD) and

Lowest Possible Percentage Dissatisfied (LPPD). The PMV and PPD indices are relatively

common in practical applications (Fanger.P.O., 2002); (Han, Yang, Zhou, Zhang, Zhang, &

Moschandreas, 2009).

`The former predicts the mean thermal comfort votes among a large group of people;

the latter is a quantitative measure of the number of thermally dissatisfied persons in a group

under particular thermal conditions. Field studies on the thermal comfort of occupants

working in an air-conditioned environment can be used to examine the neutral temperature—

a temperature associated with a neutral thermal sensation (Oseland.N.A, 1995); (Fanger,

1995); (Mui,K.M & Wong,L.T, 2007). This temperature is a key factor for selecting an

appropriate air temperature set point for an indoor thermal environment (Wong,L.T,

Mui,K.W, Fong,N.K, & Hui,P.S, 2007). As climate, besides occupant factors, including

lifestyle, economic status and adaptive behaviour, plays an important role in affecting the

187

indoor thermal environment (Yoshino, et al., 2006); (Brager,G.S & Dedear,R.J, 1998),

neutral temperatures for different climatic zones have been studied (Clausen, Carrick, Fanger,

KIm, & Rindel, 1993); (Mui,K.M & Wong,L.T, 2007); (Wang, 2005). IAQ, as the nature of

air in an indoor environment in relation to occupant health and comfort, is neither a simple

nor an easily defined concept. In a broad context, it is the result of the complex interactions

among buildings, building systems and people. Comparative risk studies performed by the

United States Environmental Protection Agency (USEPA) ranked IAQ, as one of the top five

environmental risks to public health (Wang, 2005).

Over the past decades, exposure to indoor air pollutants is believed to have

increased due to a variety of factors, including the construction of more tightly sealed

buildings, the reduction of ventilation rates (for energy saving), and the use of synthetic

building materials and furnishings as well as chemically formulated personal care products,

pesticides and household cleaners. As investigating all types of indoor air pollutants for

general air quality monitoring and assessment is a complicated matter (Wong, Mui, & Hui,

2006); (Hui, Wong, & Mui, Feasibility study of an express assesment protocol for the indoor

air quality of air-conditioned offices, 2006); (Mui K. , Wong, HUI, & Law, 2008), it was

suggested that the measurement and analysis of indoor carbon dioxide (CO2) concentration

could be useful for understanding IAQ and ventilation effectiveness (Persily, 1997); (ASTM,

2003).

Although healthy people can tolerate a CO2 level up to 10,000 ppm without

serious health effects, an acceptable indoor CO2 level should be kept below 1000 ppm or 650

ppm above the ambient level in order to prevent any accumulation of associated human body

odour (ANSI/ASHRAEstandard62-2007, 2007); (Mui.K.W & Wong,L.T, 2007). In terms of

occupant satisfaction, acceptable IAQ means room air in which no known contaminants are at

harmful concentration levels and at least 80% of the people exposed to it do not express any

dissatisfaction (ANSI/ASHRAEstandard62-2007, 2007); (Mui.K.W & Wong,L.T, 2007);

(Hui, Wong, & Mui, 2008). Light enables humans to see. A number of available guides and

codes of practice provide recommendations on adequate indoor lighting designs. For

example, an illumination level of 2000 lx with a colour rendering index not less than 90 is

required for a fabric inspection factory while 500 lx with a colour rendering index between

60 and 80 should be maintained in a general office. Lighting quality attributed to the quantity

and colour spectrum of light can be expressed by a comprehensive comfort, satisfaction, and

performance (CSP) index (Bean & Bell, 1992).

188

Since indoor visual comfort is closely related to the horizontal and cylindrical

illumination levels, adjustment of the illumination level is essential to improve visual comfort

and occupant acceptance (Mui & Wong, 2006). All sounds that are distracting, annoying, or

harmful to everyday activities such as work, rest, study and entertainment are regarded as

noises. In fact, any sound judged undesirable by the recipient can be considered a noise.

Noise can be continuous or impulsive and both types can cause adverse effects on physical,

mental and social well-being. A number of measures were proposed for indoor aural comfort

evaluation, for instance, the equivalent sound pressure level (SPL), the noise criterion (NC)

curves, the balanced noise criterion (BNC), the noise rating (NR), the preferred noise

criterion (PNC), the room criterion (RC), and the loudness level. Given the nature of noises

generally encountered in offices, a previous research indicated the A-weighting equivalent

sound pressure level that has been widely adopted in the studies of noise level within

buildings would be the best and most convenient measure, although the loudness level would

also be a very good alternative (Ayr, Cirillo, Fato, & Martellotta, 2003).

In Hong Kong, surveys showed that the equivalent continuous noise level

correlated well with the occupant acceptance reported in air-conditioned offices (Mui &

Wong, 2006), and in construction site offices where the background noise was not dominated

by air-conditioning but by outdoor noise sources (Mui, Wong, & Wong, 2009).The neutral

sound pressure level found for aural comfort in some typical air-conditioned offices was

between 45 and 70 dBA, with a mean of 57.5 dBA. Studies showed that an occupant‘s

acceptance of an acceptable IEQ environment would be closely related to aforesaid

environmental parameters. More recently, the equivalence of the discomfort caused by

different physical qualities for indoor air quality, thermal sensation and noise has been

investigated (Fanger, Olf, & Decipol, 1988); (Clausen, Carrick, Fanger, KIm, & Rindel,

1993) ; (Pellerin & Candas, 2004); (Hikmat, Hind, & Muna, 2009). It was reported that every

1oC change of operative temperature in range between 23 and 29

oC would produce the same

feeling on human comfort due to a deviation 2.4 decipol regarding the perceived air quality

(Fanger, Olf, & Decipol, 1988), or due to a change in noise level of 3.9 dB. For levels of

perceived air quality up to 10 decipol, a 1 decipol change in perceived air quality had the

same effect on human comfort as a change in noise level of 1.2 dB (Clausen & Wyon, 2008);

(Clausen, Carrick, Fanger, Kim, Poulson, & Rindel, 1993).

189

The equivalence between acoustic and thermal sensation was proposed for

short-term exposure as a 1 8C change in temperature had the same effect on 2.6 dBA

(Pellerin & Candas, 2004). Workplace variables inducing the largest number of health

symptoms, comfort or odour concerns were investigated by multivariate regression analysis

(Hikmat, Hind, & Muna, 2009); (Mendell M. , 2003); (Bulysssen & Cox, 2002); (Dan, 1993);

(Toftem, 2002).. It was realised that successful control of the indoor environment required an

understanding of the integral indoor environmental parameters. The occupants‘ acceptance of

the four basic IEQ components was evaluated and correlated with the overall IEQ acceptance

of an office environment (Wong, Mui, & Hui, 2008).

Mathematical expressions were proposed for the overall IEQ acceptance,

using a multivariate logistic regression model and it can be used as a quantitative measure for

an office environment design. IEQ parameters are interdependent and must be considered

interactively. However, conflicts were reported in the above cases where these parameters

were treated discretely, e.g. maximization of openings for natural ventilation and daylight

resulted in poor acoustic performance and thermal discomfort in an indoor environment in

subtropical climates (Koenigsberger, Ingersoll, Mayhew, & Szokolay, 1974). Laboratory

experiments of controlled auditory and visual stimuli also showed that the visual parameter

was predominant in audio–visual interactions and the visual information would affect the

auditory judgment (Viollin, 2003). Reportedly, attention to a visual form would reduce the

conscious perception of sound, and vice versa when the sound was related to a scene (Yang

& Kang, 2005). In fact, interactions between visual and auditory perceptions gave people a

sense of involvement. This study investigated the occupant acceptance of residential IEQ

through physical measurements and subjective surveys. Mathematical expressions were

proposed for the overall IEQ acceptance using a multivariate logistic regression model with

the four environmental parameters discussed above. The study provides useful information

for developing quantitative assessments for residential environments where an occupant‘s

evaluation is expected.

4.2. Components of IEQ

IEQ depends on the following four components

1. Thermal comfort

2. Carbon dioxide

190

3. Sound

4. Illumination

4.2.1. Thermal comfort

Thermal comfort models that take into account human adaptability have been

developed over the years (Fanger.P.O, 1970); (ANSI/ASHRAE55-2004, 2004). The concept

of adaptive thermal comfort can be described as (Auliciems.A, 1983) ‗When a change occurs

causing thermal discomfort, people react in such a way that their thermal comfort is re-

established.‘ This description refers to behavioural adaptation that can be discerned in

personal, technical, environmental, cultural and organizational adaptation. Physiological

adaptation or acclimatization does not seem to affect peoples‘ neutralities, but there is some

evidence that the acceptability is altered (Nicol & Humphreys.A', 2001). Psychological

adaptation implies a changed perception of, or response to, sensory information. Thermal

sensations are influenced by an individual‘s experiences and expectations in a direct manner.

When applying models of adaptive thermal comfort, one should distinguish between different

types of buildings, usage and climatic circumstances. Occupants in naturally ventilated

buildings have possibilities for increasing the air velocity in the room by operating windows.

By doing so, they can still create a comfortable environment in higher indoor temperatures.

Additionally, it turns out that psychological adaptation plays an important part especially in

this type of buildings: because of the more direct contact to the weather outside, higher

temperatures are also expected for the indoor climate. Fanger‘s PMV-model can only take the

effects of behavioural adaptation into account: the adjustment of clothing and the level of

activity, and the increase of the air velocity.

4.2.2. Carbon di-oxide (CO2)

CO2

has long been used as a basis for ventilation (providing fresh outdoor air

to indoor spaces) design and control. CO2

is a natural product of human respiration whose

rate can be predicted based on an occupant‘s age and activity level. Beginning as early as

1916, CO2 of 800 to 1,000 ppm and 1,000 ppm respectively were recommended. However,

the key point is that CO2

levels are good predictors or surrogates for human emitted bio

effluents (i.e., odours) that are considered undesirable for the overall human comfort inside

conditioned spaces. Thus CO2

is a surrogate for levels of other bio effluents that cause odours

191

that are likely to be viewed as unacceptable by others in the space, not because of their

presence as a direct health hazard.

Since people exhale CO2 as a consequence of their normal metabolic

processes, the concentrations of carbon dioxide inside occupied spaces are higher than the

concentrations of CO2 in the outdoor air. In general, a larger peak difference between indoor

and outdoor CO2 concentration indicates a smaller ventilation rate per person. The ventilation

rate per person can be estimated with reasonable accuracy from the difference between the

maximum steady-state (equilibrium) indoor CO2 concentration and the outdoor CO2

concentration, if several critical assumptions are met, including: the occupied space has

nearly constant occupancy and physical activity level for several hours, the ventilation rate is

nearly constant, and the measured indoor CO2 concentration is representative of the average

indoor or exhaust airstream concentration in the space . For example, in an office space under

these conditions, if the equilibrium indoor CO2 concentration is 650 parts per million (ppm)

above the outdoor concentration, the ventilation rate is approximately 15 cubic feet per meter

(cfm) per person. In many real buildings, occupancy and ventilation rates are not stable for

sufficient periods and other critical assumptions may not be met to enable an accurate

determination of ventilation rate from CO2 data. The American Society for Testing and

Materials (ASTM) states that this technique has been misused, when the necessary

assumptions have not been verified and the results have been misinterpreted. Nevertheless,

CO2 concentrations remain a rough and easily measured surrogate for ventilation rate. In

addition, many studies have found that occupants of buildings with higher indoor CO2

concentrations have an increased prevalence of sick building syndrome symptoms. However,

indoor CO2 concentrations may be poor indicators of health risks in buildings and spaces with

strong pollutant emissions from the building or building furnishings, particularly when

occupant densities are low.

4.2.3. Sound

Noise has several adverse effects on human beings. On the physiological side,

these effects include hearing damage and hearing loss. On the psychological side, they

include interference with speech communication, impairment of performance, and

annoyance. Noise can be very distracting and prevent concentrated mental work. In extreme

192

cases, it can also result in physical disorders. Noise can be characterized in two ways, - direct

and indirect. A direct noise is determined by the intensity of the source and the distance from

the ears. Reflected noise is dependent on the reflection factors of the floor, walls, ceiling, etc.,

and on the position of these surfaces. Direct noise should be suppressed by placing covers

over or by isolating sources of noise from the rest of the work area. A distinction should also

be made between meaningful noise and general background noise. Most working

environments will have some background noise. However, this noise can become

uncomfortable if an irregularity, such as a malfunctioning machine, develops. Reflected noise

can be reduced by introducing sound absorbing materials into the environment. Acoustical

noise is considered a human factor because it affects such factors as a workers‘ comfort, job

satisfaction and performance. Fortunately, however, the noise levels of modern workstations

designed for office or laboratory environments are relatively low, much lower than those of

typical data processing equipment found in computer room installations.

The noise is generated primarily by the single small fan in the system unit,

used to cool the electronics or by the spinning hard disk drive. Displays are usually cooled by

convection and are very quiet. Thus, the primary concern from a human factor point of view

is that, the noise from a workstation may be disrupting and annoying. The noise is not very

loud, but it may be reported as objectionable by the user simply because the office

environment itself is very quiet. Annoyance is a subjective response and difficult to quantify,

but it should not be treated lightly. From an employer's point of view, an annoyed employee

can present morale problems which may affect performance and reliability. Noise control

engineers are striving to lower noise levels of workstations, while at the same time studying

and identifying the psycho acoustical aspects of particular noises that most contribute to

annoyance.

4.2.4. Illumination

Today there is great value in the task/ambient approach to lighting. This

method at first provides general room illumination and then specific, brighter illumination -

only where needed. In this respect, ambient lighting levels may be reduced to save energy

and task area lighting may be increased for optimum human performance.

Poor lighting can be a safety hazard misjudgement of the position, shape or

speed of an object can lead to accidents and injury. Poor lighting can affect the quality of

193

work, specifically in situation where precision is required, and overall productivity. Poor

lighting can be a health hazard too much or too little light strains eyes and may cause eye

discomfort (burning, etc.) and headaches. The amount of light we need varies and depends

on:

The type of task being done (such as demands for speed and accuracy),

Type of surfaces (does it reflect or absorb light),

The general work area, and

The individual's vision.

The amount of light falling on a surface is measured in units called lux.

Lux = Lumens (quantity of light) per square metre.

Illuminance is the amount of light falling on a surface. The unit of

measurement is lux (or lumens per square metre = 10.76 foot candles (fc)). A light meter is

used to measure it. Readings are taken from several angles and positions.

4.2.4.1. IES - RECOMMENDATIONS

Since 1958 the Illuminating Engineering Society has published illuminance

recommendations in table form. These tables cover both generic tasks (reading, writing etc),

and 100's of very specific tasks and activities (such as drafting, parking, milking cows,

blowing glass and baking bread). All tasks fall into 1 of 9 illuminance categories, covering

from 20 to 20,000 lux.

To reach proper light levels, many light fixtures are designed to reflect light

off walls, ceilings and objects. The amount of light reflected off a surface can be measured.

Canadian centre for occupational health and safety suggest the percent of light reflected off

surfaces in a typical office include:

Window blinds (40-50%),

Walls (50% maximum),

Business machines (50% maximum),

Ceiling (70-80%),

Floor (20-40%), and

194

Furniture (25-45%).

The percent value refers to the amount of light that a surface reflects, relative

to the amount that falls on the surface. In addition, light fixtures that are too widely spaced or

wrongly positioned can create shadows. Objects between the light fixture and work being

done can block the light and cast shadows. Likewise, workers sitting with their backs to

windows, with light fixtures directly overhead or to the rear, cast shadows on their own work

surfaces. The immediate work area should be brighter than surrounding areas. If the

surrounding area is brighter than the work area, your attention is distracted away from the

work area. The contrast between colours of objects, such as between the print itself and paper

or text and background on computer screens, can also cause problems. Too little contrast

between print and the paper or little contrast between characters on a video display terminal

screen and the background makes reading tasks difficult. In an industrial setting, moving and

stationary machine parts are hard to distinguish if they are of the same colour.

4.2.4.2. Luminance

Luminance is the amount of light reflected from a surface. The unit of

measurement is candela per square metre (equals 0.29 foot-lamberts). An Illuminance meter

is used to measure it. Several measurements are made and averaged. Luminance tables are

consulted for reference values. Illuminance is a measure of the amount of light falling on a

surface. It is defined as: 'the density of the luminous flux incident on a surface'.

One footcandle is the Illuminance at a point on a surface which is one foot

from, and perpendicular to, a uniform point source of one candela. One Lux is the

Illuminance at the same point at a distance of 1 meter from the source. One lumen uniformly

distributed over one square foot of surface provides an illumination of 1 footcandle.

If you work in feet, your results will be in footcandles - (1 footcandle = 1 lumen/square ft.)

If you work in meters, your results will be in Lux - (1 Lux = 1 lumen/square meter)

Formerly the term 'ILLUMINATION', was used for Illuminance.

195

4.2.4.3 Illumination Levels

The Illuminating Engineering Society (IES) measures light in foot candles, or

"lux," which translates in scientific terms to one lumen per square foot. Full daylight is

characterized as approximately 10,752 lux, while an overcast day measures only 1,075. Lux

is measured in terms of the amount of illumination, or light, covered per square foot.

Recommended lux for the workplace varies by field. A classroom environment has a

recommended lux standard of 250. Workers who do detailed drawing work should have

lighting illumination at 1,500 to 2,000 lux. Specialized visual tasks can require upward of

10,000 lux depending on the profession. It is important today that the lighting designer

provide appropriate lighting levels for the required task(s). It is also equally important not to

diminish light in a task. There is generally little value in reducing lighting in a task where

human performance is concerned. The electrical energy saved is often offset by a far greater

loss in human performance or productivity.

As the eye ages, it requires more light to see the same detail with the same

speed and accuracy. For this reason, lighting systems must be designed with specific human

needs in mind.

Energy restrictions and building codes often tend to limit lighting to 'x'

number of watts per square feet (or meter) in new constructions. It must be remembered that

these are usually 'average' figures in that, a storage room might require lower lighting levels

and an office area might require higher lighting levels than average. These average levels can

and should be increased with the object of providing sufficient lighting for effective human

performance.

4.2.4.4 . Leveraging Daylight

The Illuminating Engineering Society (IES) recommends employers to

leverage available daylight in order to cut down on energy costs while improving lighting for

workers. Sunlight provides a higher lux level than indirect lighting and is more cost effective.

Glazing windows is one approach the IES recommends to prevent glare while still allowing

daylight into the office. Placement of offices, such as east-facing or west-facing, should also

be considered carefully to make the most of the available sun light.

196

4.2.4.5 . Indirect Lighting

Providing the proper levels of illumination to workers prevents safety hazards

and increases overall productivity. The IES recommends use of indirect lighting at

consecutive intervals. Measurements should be obtained to ensure workers have the

recommended light levels. The use of task lights is recommended for certain areas.

Employers must consider the nature of the work an employee is doing while establishing an

illumination target. Greater density of lighting fixtures should be used in offices where detail-

oriented work is performed. In this problem, to find the optimum indoor environmental

quality, ten solvers are used. Each solver has its own characteristics. The characteristics lead

to different solutions and run times. The results are examined based on various criteria.

4.3. IEQ OFFICE

4.3.1. Introduction

Today, the concept of an acceptable indoor environmental quality (IEQ) as an

integral part of the total building performance approach is still not fully appreciated. This

state of mind is an intricate response to the indoor environmental factor groups, including

physical environment sustained by the building and its service system, and individual

physiological conditions such as health, social relations, financial state, etc.

Four basic components, namely thermal comfort, indoor air quality, aural and

visual comfort were identified for determining an acceptable IEQ. Conventional studies on

indoor environment address each of them separately. They are still addressed independently

by designers for many office designs. It was realised that successful control of the indoor

environment required an understanding of the indoor environmental parameters. An overall

IEQ index would be derived to describe the state of mind of a user in balance with the indoor

environment.

Subjective evaluation of an indoor environment being perceived by an

occupant can be used to assess the acceptance of the IEQ. In particular, occupants‘

acceptance of the four basic components of IEQ was evaluated and correlated with the overall

IEQ acceptance of an office environment. The occupants‘ attitudes towards the operative

temperature, CO2 concentration, equivalent noise level and illumination level and the overall

IEQ acceptance recorded by a dichotomous scale were studied .Mathematical expressions

were proposed for the overall IEQ acceptance, using a multivariate logistic regression model

197

with the former four parameters recorded. The proposed overall IEQ acceptance can be used

as a quantitative assessment criterion for an office environment and similar environment

where an occupant‘s evaluation is expected.

Note: We could not find any standards for India, and so we have used ASHRAE,

since they seemed reasonable and applicable in Indian context also. (living-smartly.com)

4.3.2. Methodology

Subjective evaluations made by 220 occupants of indoor environmental

conditions in natural ventilated Faculty rooms in Karunya University were studied. The

sample offices had floor areas ranging from 233.3 m2 to 937.77 m

2. The offices were

spacious with ordinary design and with good quality finishes; flexible layout, average-sized

floor plates, adequate lobbies; good lift services zone and parking facilities were available.

The occupant‘s acceptance of the perceived indoor environment given by four

aspects, namely thermal environment, indoor air quality, equivalent noise level and

illumination level, was studied with a dichotomous assessment scale . This scale was used for

a direct feedback of acceptability with the question ‗Is the thermal environment/indoor air

quality/noise level/illumination level being perceived in the office environment acceptable to

you?‘ being asked. The ranks ‗(1) Yes, acceptable‘ and ‗(0) No, not acceptable‘ were self-

explanatory. In order to confirm the validity of their responses, each respondent had to use a

semantic differential evaluation scale for the subjective assessment of the first two aspects,

and a visual analogue assessment scale for the evaluation of the aural and visual comfort . At

the end of the survey, an overall acceptance of the IEQ was determined.

A total of 220 occupants were interviewed and their evaluations of the IEQ and the

four parameters were obtained. The results are summarized in Table 4.1. The correlation

between subjective response to each parameter and the overall IEQ acceptance was evaluated

by a statistic χ2-test. Results showed that all the four parameters contributed to the overall

IEQ; and the significance between the acceptance votes on the latter and those on the former,

ranking from the most important to the least, was as follows: p = for thermal

environment, for air quality, for noise level and

for illumination level.

198

Table 4.1 Occupant’s votes on acceptance of a perceiving indoor environmental quality

U - Unacceptable; A -Acceptable.

The overall IEQ acceptance θ for an office environment perceived by an occupant

expressed by a multivariate logistic regression model is proposed

Where the regression constants determined from the 220 occupant evaluations are

k0 = -14.98; k1 = 6.04; k2 = 4.92; k3 = -4.70; k4 = 3.74;

Values of k1, k2, k3, k4 confirm the relative importance of the four contributors to θ, the larger

the value, the greater the importance and it is seen that the occupants were very sensitive to

the operative temperature as compared with the other three parameters.

Various combinations of contributors i=1, 2, 3, 4 and the corresponding

overall IEQ acceptance were considered. A total of 24 possibilities were found. Taking the

binary notation for the acceptance i.e., 0 for ‗unacceptable‘ and 1 for ‗acceptable‘ the

predicted acceptance of IEQ (θ) is calculated.

φ1=1 - (PPD/100).

Where, PPD = 100 – 95 x e-(0.03353 (PMV^4) + 0.2179 (PMV^2));

-2 PMV

φ2=1-

(

Overall

acceptance

Θ

Votes Thermal

environment

φ1

Air quality

φ 2

Noise level

φ 3

Illumination

level

φ 4

U

A

98

122

U A U A U A U A

47

3

51

119

52

12

46

110

30

7

68

115

26

9

72

113

TOTAL 220 50 170 64 156 37 183 35 185

199

φ3=1 -

; 67 ≤ ≤ 78,

φ4=1 -

187 1522.

Table 4.2. Overall IEQ acceptance

4.3.3. Algorithms

4.3.3.1. Genetic algorithm

4.3.3.1.1. Options Set for the Algorithm:

Initial population: 20.

Elite count: 2.

Cross over fraction as 0.8.

Max Time Limit: ∞.

Max Generations: 100.

Fitness Limit: -∞.

Selection: Stochastic.

4.3.3.1.2. Stopping Criteria:

If the maximum generations is reached (100).

If maximum time is reached (∞).

Case

No.

Survey

Sample

Contributors Predicted

acceptance

of IEQ θ

Φ1 Φ2 Φ3 Φ4

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

0

5

1

14

3

3

5

20

1

8

9

27

3

16

13

92

0

0

0

0

0

0

0

0

1

1

1

1

1

1

1

1

0

0

0

0

1

1

1

1

0

0

0

0

1

1

1

1

0

0

1

1

0

0

1

1

0

0

1

1

0

0

1

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

3.1208 10-7

1.313 10-5

2.84 10-9

1.1949 10-7

4.275 10-5

1.7967 10-3

3.88 10-7

1.637 10-5

5.486 10-3

1.31 10-4

1.1918 10-6

5.017 10-5

0.017636

0.43045

1.6326 10-4

6.827 10-3

200

If average change in function value < 10¯⁶.

TABLE.4.3. Results of GA in 20 trails for office IEQ

Trails

no. Thermal CO2 Sound Illum IEQ Time Iterations

1 2 1799.9 59.6 1553.7 1 1.208193 51

2 2 1740.6 62.8 1370.9 1 1.133192 51

3 2 1799.7 71.7 1569.3 1 0.775846 51

4 2 1070.5 64.3 1196.7 1 1.089181 51

5 1.9 1662.3 57 1513.8 1 1.059298 51

6 2 1713.5 55.7 1520.9 1 1.140209 51

7 2 1217.2 60.2 1524.2 1 0.768988 51

8 2 1775.7 69.7 1395.2 1 0.91929 51

9 2 1724.6 57.4 1503.8 1 1.172938 51

10 2 1771 63.8 1018.5 1 0.917832 51

11 2 1295.4 71.3 1440.9 1 0.913941 51

12 2 1668.8 60.2 1434 1 1.169327 51

13 2 1549.2 71.4 1232.4 1 1.059507 51

14 2 1510.7 52.7 792.1 1 0.887249 51

15 1.9 1799.9 69 1538.8 1 1.115581 51

16 2 1722.7 65.6 1495 1 1.195485 51

17 2 1611.4 63.8 1265.3 1 1.032374 51

18 1.9 1756.2 54.5 1234.4 1 0.932092 51

19 -2 1789.5 71.9 1580.4 1 1.162234 51

20 2 1295.4 71.3 1440.9 1 0.913941 51

Avg 1.785 1613.71 63.695 1381.06 1 1.028335 51

201

Fig.4.1.Convergence of GA

4.3.3.2. Simulated annealing

4.3.3.2.1. Options Set:

Initial Temperature: 100.

Annealing Function: Fast Annealing.

Reannealing interval: 100.

Time Limit: ∞.

Max.function evaluation: 3000* No. of variables.

Max. Iterations: ∞.

Function Tolerance: 10¯⁶.

Objective Limit: 10¯⁶

0 10 20 30 40 50 60 70 80 90 100-1

-0.5

0

0.5

1

1.5

Generation

Fitness v

alu

e

Best: -0.99999 Mean: -0.9526

202

4.3.3.2.2. Stopping Criteria:

Max. Time reached.

The average change in value of the objective function is < 10¯⁶.

Max. Iterations are reached.

If the number of function evaluations reached.

If the best objective function value is less than or equal to the value of Objective

Limit.

TABLE.4.4. Results of SA in 20 trails for office IEQ

Trails

no. Thermal CO2 Sound Illum IEQ Time Iterations

1 2 1379.6 68.4 1014.9 1 2.582472 2004

2 -2 1213.4 70.6 1125.6 1 2.227751 2003

3 2 1513.7 50.5 970.4 1 2.322216 2001

4 2 1160.6 64.2 846.1 1 2.482702 2006

5 -2 1341.8 72 746.3 1 1.90792 2005

6 2 1467.3 53.6 1100.5 1 3.029657 2017

7 -2 1318.7 62.4 1029.3 1 2.866003 2031

8 -2 1219.8 58.4 878.4 1 2.279764 2005

9 2 1127 71.3 959.8 1 2.089766 2007

10 -2 1486.7 56.2 1042.4 1 2.134044 2006

11 -2 982.3 64.3 1116.7 1 2.91151 2011

12 2 1330.5 45.3 789.4 1 3.014331 2002

13 2 1041.9 61.3 854.1 1 2.283927 2013

14 -2 1284.2 71.5 964.4 1 2.42402 2007

15 -2 1303.7 45.5 791.1 1 3.054213 2007

16 2 1174.6 46.5 961.3 1 2.085067 2015

17 2 1410.5 52.1 1145.8 1 2.568922 2016

18 -2 1313 68.8 1081.6 1 2.904454 2004

19 2 1228.1 71.2 828.1 1 2.865376 2016

20 2 1157.9 60.2 994.8 1 1.904684 2025

Avg 0.2 1272.765 60.715 962.05 1 2.49694 2010.05

203

Fig.4.2.Convergence of SA

4.3.3.3. Pattern search

4.3.3.3.1. Options Set:

Poll Method: GPS positive Basis 2N.

Initial Mesh size: 1.

Expansion Factor: 2.

Contraction Factor: 0.5.

Mesh Tolerance: 10¯⁶.

Max. Iteration: 100* No. of Variables.

Max. Function Evaluation: 2000* No. of Variables.

Max. Time Limit: Inf.

Function Tolerance: 10¯⁶

4.3.3.3.2. Stopping Criteria:

Mesh Tolerance: 10¯⁶.

Max. Iteration: 100* No. of Variables.

Max. Function Evaluation: 2000* No. of Variables.

Max. Time Limit: ∞.

Function Tolerance: 10¯⁶.

0 500 1000 1500 2000 2500-1

-0.5

0

0.5

1

Iteration

Function v

alu

eBest Function Value: -1

1 2 3 4-200

0

200

400

600

800

1000

1200

1400Best point

Number of variables (4)

Best

poin

t

0 10 20 30 40 50 60 70 80 90 100

Time

Iteration

f-count

% of criteria met

Stopping Criteria

0 500 1000 1500 2000 2500-1

-0.5

0

0.5

1

Iteration

Function v

alu

e

Current Function Value: -0.99999

204

TABLE.4.5. Results of PS in 20 trails for office IEQ.

Trails

no. Thermal CO2 Sound Illum IEQ Time Iterations

1 2 1800 70.5 1600 1 0.156075 60

2 2 1800 70.5 1600 1 0.148124 60

3 2 1800 70.5 1600 1 0.148314 60

4 2 1800 70.5 1600 1 0.14655 60

5 2 1800 70.5 1600 1 0.146723 60

6 2 1800 70.5 1600 1 0.150499 60

7 2 1800 70.5 1600 1 0.148128 60

8 2 1800 70.5 1600 1 0.147931 60

9 2 1800 70.5 1600 1 0.148575 60

10 2 1800 70.5 1600 1 0.147136 60

11 2 1800 70.5 1600 1 0.149625 60

12 2 1800 70.5 1600 1 0.146856 60

13 2 1800 70.5 1600 1 0.146429 60

14 2 1800 70.5 1600 1 0.146925 60

15 2 1800 70.5 1600 1 0.147402 60

16 2 1800 70.5 1600 1 0.147624 60

17 2 1800 70.5 1600 1 0.146341 60

18 2 1800 70.5 1600 1 0.14618 60

19 2 1800 70.5 1600 1 0.149296 60

20 2 1800 70.5 1600 1 0.149342 60

Avg 2 1800 70.5 1600 1 0.148204 60

Fig.4.3.Convergence of PS

0 10 20 30 40 50 60-1

-0.8

-0.6

-0.4

-0.2

0

0.2

0.4

0.6

0.8

Iteration

Func

tion

valu

e

Best Function Value: -1

205

4.3.3.4. Particle swarm optimization

4.3.3.4.1. Options Set:

Max.Generation = 200.

Max. Time Limit=∞.

Average change in fitness value= 10-6

.

Time Limit = ∞.

Function Tolerance= 10-6.

Cognitive Attraction = 0.5.

Population Size = 40.

Social Attraction = 1.25.

4.3.3.4.2. Stopping Criteria:

Max.Generation = 200.

Max. Time Limit=∞.

Average change in fitness value= 10-6

Time Limit = ∞.

Function Tolerance= 10-6

0 10 20 30 40 50 600

200

400

600

800

1000

1200

Iteration

Mes

h si

ze

Current Mesh Size: 9.5367e-007

206

TABLE.4.6. Results of PSO in 20 trails for office IEQ

Trails

no. Thermal CO2 Sound Illum IEQ Time Iterations

1 2 1528.9 61.2 1302.9 1 0.14490

4 57

2 2 1436.2 53.6 991.6 1 0.11841

2 73

3 -2 1769.3 45.2 945.7 1 0.08162

9 53

4 -2 1687.9 54.7 1191.7 1 0.09000

9 56

5 2 1335.3 60.5 1222.3 1 0.08634

6 53

6 2 1758.5 50.7 1192.7 1 0.08790

1 55

7 -1.9997 774.4371 49.8432 966.685

2 1 0.09591

1 59

8 2 1073.6 49.9 1359.2 1 0.08510

6 51

9 -2 1785.2 60.3 1181.9 1 0.08805

8 54

10 -2 1718.7 60.1 1148.9 1 0.07966

8 52

11 -2 1770.7 55.8 1139.2 1 0.10200

6 66

12 2 1753.2 65.1 1049.7 1 0.09481

4 62

13 -2 1777.2 61 1218.2 1 0.09750

2 63

14 -2 1585.4 66.4 1365.9 1 0.09300

3 53

15 2 1712.2 49.8 1128.5 1 0.09859

6 59

16 -2 1675.8 50 1152.2 1 0.08977 57

17 -2 1641.3 67.9 1459.9 1 0.08328 53

18 -2 1455.5 70.2 1466.1 1 0.08996

2 51

19 2 1706.3 58 1308 1 0.09390

9 58

20 2 1678 71.5 981.6 1 0.07920

3 51

avg -0.19999 1581.18185

5

58.0871

6

1188.64

4 1 0.09399

9 56.8

4.3.3.5. GODLIKE

4.3.3.5.1. Options Set & Stopping Criteria:

Max.FunEvals = 10-5

.

Max. Iterations= 20.

Min. Iterations = 2.

Total. Iterations = 15.

Function Tolerance = 10-4

207

TABLE.4.7. Results of GL in 20 trails for office IEQ

Trails

no. Thermal CO2 Sound Illum IEQ Time Iterations

1 -2 1791.1 69.7 1565.3 1 1.546532 4

2 2 1799.9 58.7 1548.5 1 1.885316 4

3 -2 1799.9 70.6 1565.7 1 1.012803 4

4 2 1799.6 54 1560.7 1 1.505692 4

5 -2 1756.2 56.8 1380.8 1 1.191908 4

6 -2 1799.8 71.7 1580.9 1 1.099644 4

7 2 1799.5 58.7 1557.4 1 1.044684 4

8 2 1798.7 66.2 1592.6 1 1.15372 4

9 -2 1799.5 61.4 1491.2 1 1.076416 4

10 -2 1795.6 63 1541 1 1.219463 4

11 2 1799.5 65.8 1589 1 1.07033 4

12 -2 1799.1 53.2 1435.2 1 0.935505 4

13 -2 1799.1 55.6 1599 1 1.032483 4

14 2 1799.9 66.7 1599.2 1 1.285674 4

15 -2 1776.1 58.6 1473.2 1 0.933089 4

16 -2 1794.9 59.7 1164.1 1 0.953743 4

17 -2 1800 71.1 1556.6 1 1.379827 4

18 2 1769.3 48.3 1451.6 1 1.021947 4

19 2 1787.9 62.7 1446.4 1 1.07574 4

20 2 1780.6 70.9 1483.2 1 1.512711 4

avg -0.2 1792.31 62.17 1509.08 1 1.196861 4

4.3.3.6. Fmincon.

4.3.3.6.1. Options Set for ‘Fmincon’:

Max.Iterations:400.

Max.function Evaluations: 100* No. of Variables.

Max.Time: ∞.

Max. Function Tolerance: 10-6

.

4.3.3.6.2. Stopping Criteria for Global Search:

Max.Time: ∞.

Max Wait cycle: 20

208

4.3.3.6.3. Stopping Criteria for Fmincon:

Max.Iterations > 400.

Function Tolerance: 10-6

TABLE.4.8. Results of Fmincon in 20 trails for office IEQ.

Trials

no.

Therma

l CO2 Soun

d

Illu

m IEQ Time Func.Coun No.Local.Solver

s 1 2 180

0 72 1600 1 8.86601

9 5002 198

2 2 180

0 72 1600 1 13.4064

1 7277 334

3 2 180

0 72 1600 1 14.7796 7042 362

4 2 180

0 72 1600 1 12.6790

6 6897 324

5 2 180

0 72 1600 1 7.11649

2 4067 144

6 2 180

0 72 1600 1 11.6128

6 5507 266

7 2 180

0 72 1600 1 7.61832 5312 160

8 2 180

0 72 1600 1 6.78551

4 4672 184

9 2 180

0 72 1600 1 8.38266

8 4847 204

10 2 180

0 72 1600 1 13.3524

9 6477 373

11 2 180

0 72 1600 1 9.97782

8 4662 270

12 2 180

0 72 1600 1 11.5853

1 6032 253

13 2 180

0 72 1600 1 9.27154

5 5712 199

14 2 180

0 72 1600 1 8.60228

6 4612 199

15 2 180

0 72 1600 1 12.3196

6 6592 344

16 2 180

0 72 1600 1 14.4853

1 5987 344

17 2 180

0 72 1600 1 10.1845 4567 213

18 2 180

0 72 1600 1 7.90361

3 4912 167

19 2 180

0 72 1600 1 13.6878

9 6077 396

20 2 180

0 72 1600 1 12.0780

2 6227 236

Avg 2 180

0 72 1600 1 10.7347

7 5624 258.5

4.3.3.7. Direct evolution

4.3.3.7.1. Options Set:

Min. Value to Reach = 10-6

.

Population Size = 10*D.

Max. Iterations = 200.

Step Size F = 0.8.

Cross Over Probability = 0.5.

Strategy= 7 (DE/rand/1/bin)

209

DE/x/y/z, where DE stands for DE, x represents a string denoting the vector to be

perturbed, y is the number of difference vectors considered for perturbation of x, and

z stands for the type of crossover being used (exp: exponential; bin: binomial).

4.3.3.7.2. Stopping Criteria:

Max.Value of function reached= 10-6

.

Max.Iterations=200

TABLE.4.9. Results of DE in 20 trails for office IEQ

Trails

no. Thermal CO2 Sound Illum IEQ Time Iterations

1 1.8902 858.2017 45.1112 555.2472 1 0.00303 40

2 2 1060.1 60.6 865.3 1 0.003735 40

3 -2 1729.5 48.6 1302.6 1 0.002558 40

4 -1.9906 897.3013 53.8414 790.6261 1 0.01054 40

5 -2 1583 58.2 877.1 1 0.003255 40

6 1.9 1234.7 54.9 550.2 1 0.004085 40

7 -2 1305.5 66.3 1110.2 1 0.00335 40

8 1.9833 690.9403 48.5607 478.5444 1 0.001541 40

9 2 1652.5 59.7 1587.9 1 0.001507 40

10 1.9283 725.1315 46.8846 491.9793 1 0.003296 40

11 -1.9913 649.2797 63.5242 584.1255 1 0.002578 40

12 2 1251.8 64.7 1380.1 1 0.002499 40

13 1.9 1020.9 71.7 1450.5 1 0.00378 40

14 1.9933 970.858 47.7356 803.1949 1 0.003476 40

15 -2 1697.5 52.6 1197.2 1 0.00366 40

16 -1.9362 708.0032 65.3522 799.9138 1 0.00321 40

17 -2 1610.8 60.2 1566.2 1 0.002003 40

18 2 1770.6 59.1 957.6 1 0.003846 40

19 2 1225.4 45.9 1537 1 0.003841 40

20 2 1152.7 65.8 1581.2 1 0.001516 40

Avg 0.38385 1189.735785 56.9655 1023.337 1 0.003365 40

4.3.3.8. LGO

4.3.3.8.1. Stopping Criteria:

If the current best solution did not improve for

Program execution time limits > 600 seconds.

210

4.3.3.8.2. Local search termination criterion parameter:

First local search phase ends, if the function difference is less than

If max. constrain violation exceeds

TABLE.4.10. Results of LGO in 20 trails for office IEQ

Trials

no.

Therma

l CO2 Sound Illum IEQ Time Func.Coun

1 2 1759.80654

8

68.7651

6

1561.97

1 1 0.91222

6 2699

2 2 1759.80654

8

68.7651

6

1561.97

1 1 0.73261

5 2699

3 2 1759.80654

8

68.7651

6

1561.97

1 1 0.73121

9 2699

4 2 1759.80654

8

68.7651

6

1561.97

1 1 0.72443

2 2699

5 2 1759.80654

8

68.7651

6

1561.97

1 1 0.71583

8 2699

6 2 1759.80654

8

68.7651

6

1561.97

1 1 0.75345

1 2699

7 2 1759.80654

8

68.7651

6

1561.97

1 1 0.73637

8 2699

8 2 1759.80654

8

68.7651

6

1561.97

1 1 0.74391

7 2699

9 2 1759.80654

8

68.7651

6

1561.97

1 1 0.74018

2 2699

10 2 1759.80654

8

68.7651

6

1561.97

1 1 0.96158

5 2699

11 2 1759.80654

8

68.7651

6

1561.97

1 1 0.71396

3 2699

12 2 1759.80654

8

68.7651

6

1561.97

1 1 0.94339

1 2699

13 2 1759.80654

8

68.7651

6

1561.97

1 1 0.97151

4 2699

14 2 1759.80654

8

68.7651

6

1561.97

1 1 0.96947

1 2699

15 2 1759.80654

8

68.7651

6

1561.97

1 1 1.00635 2699

16 2 1759.80654

8

68.7651

6

1561.97

1 1 0.96290

8 2699

17 2 1759.80654

8

68.7651

6

1561.97

1 1 0.74592 2699

18 2 1759.80654

8

68.7651

6

1561.97

1 1 0.983 2699

19 2 1759.80654

8

68.7651

6

1561.97

1 1 0.811 2699

20 2 1759.80654

8

68.7651

6

1561.97

1 1 0.70414

6 2699

avg 2 1759.80654

8

68.7651

6

1561.97

1 1 0.82817

5 2699

4.3.3.9. glcCluster

4.3.3.9.1. Stopping Criteria:

Maximum Iterations = 10000;

Maximum Function count = 10000;

Tolerance of Variables = 10-5

Function Tolerance =10-7

211

TABLE.4.11. Results of glcCluster in 20 trails for office IEQ

Trials

No. Thermal CO2 Soun

d Illum IEQ Time Iteration

s

Func.Coun

t 1 1.987562 1150.00001

8 49.5 433.33

36 1 0.59777

8 1 1515

2 1.987562 1150.00001

8 49.5 433.33

36 1 0.71545

5 1 1511

3 1.987562 1150.00001

8 49.5 433.33

36 1 0.61269

5 1 1514

4 1.987562 1150.00001

8 49.5 433.33

36 1 0.55204

4 1 1512

5 1.987562 1150.00001

8 49.5 433.33

36 1 0.72806

2 1 1513

6 1.987562 1150.00001

8 49.5 433.33

36 1 0.63516

5 1 1511

7 1.987562 1150.00001

8 49.5 433.33

36 1 0.62625 1 1513

8 1.987562 1150.00001

8 49.5 433.33

36 1 0.56273

4 1 1511

9 1.987562 1150.00001

8 49.5 433.33

36 1 0.59834

3 1 1513

10 1.987562 1150.00001

8 49.5 433.33

36 1 0.73717

1 1 1511

11 1.987562 1150.00001

8 49.5 433.33

36 1 0.58418

5 1 1516

12 1.987562 1150.00001

8 49.5 433.33

36 1 0.61178

3 1 1513

13 1.987562 1150.00001

8 49.5 433.33

36 1 0.56361

1 1 1513

14 1.987562 1150.00001

8 49.5 433.33

36 1 0.58698

7 1 1513

15 1.987562 1150.00001

8 49.5 433.33

36 1 0.57058

9 1 1511

16 1.987562 1150.00001

8 49.5 433.33

36 1 0.66160

9 1 1513

17 1.987562 1150.00001

8 49.5 433.33

36 1 0.60496

7 1 1514

18 1.987562 1150.00001

8 49.5 433.33

36 1 0.63104 1 1515

19 1.987562 1150.00001

8 49.5 433.33

36 1 0.60053

5 1 1515

20 1.987562 1150.00001

8 49.5 433.33

36 1 0.56090

3 1 1513

Avg 1.987562 1150.00001

8 49.5 433.33

36 1 0.61709

5 1 1513

4.3.3.10. glcSolve

4.3.3.10.1. Stopping Criteria:

Max.Iterations is exceeded > No. of variables*1000.

Max.function evaluations > No. of variables*2000.

If the difference of objective function is < 10-6

212

Table.4.12 Results of glcSolve in 20 trails for office IEQ

Trail

s

No.

Thermal Co2 Soun

d Illum IEQ Time Iter Func.Coun

t 1 -1.92593 1727.77777

8 70.5 1522.22

2 1 0.93305

3 222 1503

2 -1.92593 1727.77777

8 70.5 1522.22

2 1 1.29115

4 222 1503

3 -1.92593 1727.77777

8 70.5 1522.22

2 1 0.87620

1 222 1503

4 -1.92593 1727.77777

8 70.5 1522.22

2 1 1.23000

4 222 1503

5 -1.92593 1727.77777

8 70.5 1522.22

2 1 0.99029

7 222 1503

6 -1.92593 1727.77777

8 70.5 1522.22

2 1 0.90531

8 222 1503

7 -1.92593 1727.77777

8 70.5 1522.22

2 1 0.94392

1 222 1503

8 -1.92593 1727.77777

8 70.5 1522.22

2 1 0.95278

3 222 1503

9 -1.92593 1727.77777

8 70.5 1522.22

2 1 1.09855 222 1503

10 -1.92593 1727.77777

8 70.5 1522.22

2 1 0.95906

5 222 1503

11 -1.92593 1727.77777

8 70.5 1522.22

2 1 1.10017

7 222 1503

12 -1.92593 1727.77777

8 70.5 1522.22

2 1 1.01511

2 222 1503

13 -1.92593 1727.77777

8 70.5 1522.22

2 1 0.90944

4 222 1503

14 -1.92593 1727.77777

8 70.5 1522.22

2 1 0.96646

6 222 1503

15 -1.92593 1727.77777

8 70.5 1522.22

2 1 1.01381

2 222 1503

16 -1.92593 1727.77777

8 70.5 1522.22

2 1 1.01043

5 222 1503

17 -1.92593 1727.77777

8 70.5 1522.22

2 1 1.09151

1 222 1503

18 -1.92593 1727.77777

8 70.5 1522.22

2 1 1.19118

3 222 1503

19 -1.92593 1727.77777

8 70.5 1522.22

2 1 0.86412 222 1503

20 -1.92593 1727.77777

8 70.5 1522.22

2 1 0.92683 222 1503

Avg -1.92593 1727.77777

8 70.5 1522.22

2 1 1.01347

2 222 1503

Table.4.13. Comparative results of optimization methods for office IEQ

Method Thermal

sensation

Carbon

dioxide

Sound

level

Horizontal

illumination

IEQ TIME iterations

GA 1.785 1613.71 63.695 1381.06 1 1.02834 51

SA 0.2 1272.77 60.715 962.05 1 2.49694 2010.05

PS 2 1800 70.5 1600 1 0.1482 60

PSO -0.2 1581.18 58.0872 1188.64 1 0.094 56.8

GL -0.2 1792.31 62.17 1509.08 1 1.19686 4

Fmincon 2 1800 72 1600 1 10.73477 5624

DE 0.38385 1189.74 56.9655 1023.34 1 0.00337 40

LGO 2 1759.81 68.7652 1561.97 1 0.82818 2699

glcCluster 1.98756 1150 49.5 433.334 1 0.6171 1513

glcSolve -1.9259 1727.78 70.5 1522.22 1 1.01347 1503

213

Fig.4.4. Comparative graph for office Indoor Environmental Quality

4.3.4. Discussion of the comparative results

All the ten methods yield the IEQ value as 1 which is the acceptable optimum value.

The time taken by Fmincon is maximum. Number of iterations was maximum for DE and

minimum for GL. Carbon dioxide, sound and illumination level are more or less the same for

all 10 methods. The thermal sensation alone was different for different methods

0.001

0.01

0.1

1

10

100

1000

10000

GA

SA

PS

PSO

GL

Fmincon

DE

LGO

glcCluster

glcSolve

214

4.3.5. PARAMETERS.

4.3.5.1. THERMAL SENSATION

Thermal comfort is that condition of mind which expresses satisfaction with the

thermal environment. Thermal environment encompasses characteristics which affect a

person's heat loss. In terms of bodily sensations, thermal comfort is a sensation of hot, warm,

slightly warmer, neutral, slightly cooler, cool and cold. From the physiological point of view,

thermal comfort occurs when there is a thermal equilibrium in the absence of regulatory

sweating between the heat exchange between the human body and the environment.

TABLE.4.14. Thermal sensation results in all 10 methods

Trial no. GA SA PS PSO GL DE NL LGO glcCluster glcSolve

1 2 2 2 2 -2 1.8902 2 2 -1.92592593 -1.92592593

2 2 -2 2 2 2 2 2 2 -1.92592593 -1.92592593

3 2 2 2 -2 -2 -2 2 2 -1.92592593 -1.92592593

4 2 2 2 -2 2 -1.9906 2 2 -1.92592593 -1.92592593

5 1.9 -2 2 2 -2 -2 2 2 -1.92592593 -1.92592593

6 2 2 2 2 -2 1.9 2 2 -1.92592593 -1.92592593

7 2 -2 2 -1.9997 2 -2 2 2 -1.92592593 -1.92592593

8 2 -2 2 2 2 1.9833 2 2 -1.92592593 -1.92592593

9 2 2 2 -2 -2 2 2 2 -1.92592593 -1.92592593

10 2 -2 2 -2 -2 1.9283 2 2 -1.92592593 -1.92592593

11 2 -2 2 -2 2 -1.9913 2 2 -1.92592593 -1.92592593

12 2 2 2 2 -2 2 2 2 -1.92592593 -1.92592593

13 2 2 2 -2 -2 1.9 2 2 -1.92592593 -1.92592593

14 2 -2 2 -2 2 1.9933 2 2 -1.92592593 -1.92592593

15 1.9 -2 2 2 -2 -2 2 2 -1.92592593 -1.92592593

16 2 2 2 -2 -2 -1.9362 2 2 -1.92592593 -1.92592593

17 2 2 2 -2 -2 -2 2 2 -1.92592593 -1.92592593

18 1.9 -2 2 -2 2 2 2 2 -1.92592593 -1.92592593

19 -2 2 2 2 2 2 2 2 -1.92592593 -1.92592593

20 2 2 2 2 2 2 2 2 -1.92592593 -1.92592593

AVG 0.2 2 -0.19999 -0.2 0.38385 2 2 -1.92592593 -1.92592593

215

FIG.4.5. Graph of Thermal sensation results in all 10 methods

0 2 4 6 8 10 12 14 16 18 20 22

-202

TRAILS

GA

-202

SA

123

PS

-202

PS

O

-202

GL

-202

DE

123

NL

123

LG

O

-2

glc

Clu

-2

glc

So

l

From the graph we can see that Pattern search, Fmincon, LGO, glcCluster and

glcSolve are consistent in all 20 trials.

216

4.3.5.2. CARBON DI-OXIDE

Carbon dioxide (CO2) is the chief greenhouse gas that results from human activities

and causes global warming and climate change. Though carbon dioxide is not toxic in itself,

the amount found in the indoor environment is used as an indicator for human comfort.

Elevated levels of carbon dioxide indicate that an insufficient amount of fresh, outdoor air is

being delivered to the occupied areas of the building. This also indicates that other pollutants

in the building may exist at elevated levels since there is not enough fresh air to dilute them.

Since carbon dioxide is an unavoidable, predictable, and easily measured product of human

occupancy, it is used as a marker for other pollutants emanating from humans or other

sources in the building. However, carbon dioxide is mostly a threat to health, when the

concentration is high enough to displace the oxygen, which can lead to suffocation in a

confined space.

Table.4.15. Carbon Dioxide results in all 10 methods

Tria

no. GA SA PS PSO GL DE NL LGO GLCc GLCs

1 1799.9 1379.6 1800 1528.

9 1791.1 858.20

17 1800 1759.806

548

1150.000

018

1727.77

778 2 1740.6 1213.4 1800 1436.

2 1799.9 1060.1 1800 1759.806

548 1150.000

018 1727.77

778 3 1799.7 1513.7 1800 1769.

3 1799.9 1729.5 1800 1759.806

548 1150.000

018 1727.77

778 4 1070.5 1160.6 1800 1687.

9 1799.6 897.30

13 1800 1759.806

548

1150.000

018

1727.77

778 5 1662.3 1341.8 1800 1335.

3 1756.2 1583 1800 1759.806

548

1150.000

018

1727.77

778 6 1713.5 1467.3 1800 1758.

5 1799.8 1234.7 1800 1759.806

548

1150.000

018

1727.77

778 7 1217.2 1318.7 1800 774.4

371 1799.5 1305.5 1800 1759.806

548

1150.000

018

1727.77

778 8 1775.7 1219.8 1800 1073.

6 1798.7 690.94

03 1800 1759.806

548 1150.000

018 1727.77

778 9 1724.6 1127 1800 1785.

2 1799.5 1652.5 1800 1759.806

548 1150.000

018 1727.77

778 10 1771 1486.7 1800 1718.

7 1795.6 725.13

15 1800 1759.806

548

1150.000

018

1727.77

778 11 1295.4 982.3 1800 1770.

7 1799.5 649.27

97 1800 1759.806

548

1150.000

018

1727.77

778 12 1668.8 1330.5 1800 1753.

2 1799.1 1251.8 1800 1759.806

548

1150.000

018

1727.77

778 13 1549.2 1041.9 1800 1777.

2 1799.1 1020.9 1800 1759.806

548

1150.000

018

1727.77

778 14 1510.7 1284.2 1800 1585.

4 1799.9 970.85

8 1800 1759.806

548 1150.000

018 1727.77

778 15 1799.9 1303.7 1800 1712.

2 1776.1 1697.5 1800 1759.806

548 1150.000

018 1727.77

778 16 1722.7 1174.6 1800 1675.

8 1794.9 708.00

32 1800 1759.806

548

1150.000

018

1727.77

778 17 1611.4 1410.5 1800 1641.

3 1800 1610.8 1800 1759.806

548

1150.000

018

1727.77

778 18 1756.2 1313 1800 1455.

5 1769.3 1770.6 1800 1759.806

548

1150.000

018

1727.77

778 19 1789.5 1228.1 1800 1706.

3 1787.9 1225.4 1800 1759.806

548

1150.000

018

1727.77

778 20 1295.4 1157.9 1800 1678 1780.6 1152.7 1800 1759.806

548 1150.000

018 1727.77

778 AVG

1613.7

1272.7 1800 1581.182

1792.31

1189.736

1800 1759.806548

1150.000018

1727.77778

217

FIG.4.6. Graph for Carbon Dioxide concentration results in all 10 methods

0 2 4 6 8 10 12 14 16 18 20 22

1000

1500

TRAILS

GA

1000

1500

SA

1800PS

10001500

PSO

176017801800

GL

50010001500

DE

1800NL

1600

1800

LGO

11001200

glcC

lu

1600

1800

glcS

ol

From the graph we can see that Pattern search, Fmincon, LGO, glcCluster and

glcSolve are consistent in all 20 trials.

218

4.3.5.3. SOUND PRESSURE LEVEL

Acoustics is the interdisciplinary science that deals with the study of all mechanical

waves in gases, liquids, and solids including vibration, sound, ultrasound and infrasound. The

perception of sound in any organism is limited to a certain range of frequencies. Hearing loss

due to prolonged exposure to noise is well documented. Excessive noise also has an adverse

effect on personal health and wellbeing, ability to perform quiet tasks, and productivity in

general. Because land is becoming scarcer, buildings are being constructed closer together

and closer to noise sources such as highways, railways, and airports. As a result, sound or

acoustic control is becoming increasingly important. The reduction of airborne sound through

a wall is called sound transmission loss (STL).

TABLE.4.16. Sound Pressure level results in all 10 methods

Trial

no.

GA SA PS PSO GL DE NL LGO GLCc GLCs

1 59.6 68.4 70.5 61.2 69.7 45.1112 72 68.76515

98

49.500000

9 70.5

2 62.8 70.6 70.5 53.6 58.7 60.6 72 68.76515

98

49.500000

9 70.5

3 71.7 50.5 70.5 45.2 70.6 48.6 72 68.76515

98

49.500000

9 70.5

4 64.3 64.2 70.5 54.7 54 53.8414 72 68.76515

98

49.500000

9 70.5

5 57 72 70.5 60.5 56.8 58.2 72 68.7651598

49.5000009

70.5

6 55.7 53.6 70.5 50.7 71.7 54.9 72 68.7651598

49.5000009

70.5

7 60.2 62.4 70.5 49.843 58.7 66.3 72 68.76515

98

49.500000

9 70.5

8 69.7 58.4 70.5 49.9 66.2 48.5607 72 68.76515

98

49.500000

9 70.5

9 57.4 71.3 70.5 60.3 61.4 59.7 72 68.76515

98

49.500000

9 70.5

10 63.8 56.2 70.5 60.1 63 46.8846 72 68.76515

98

49.500000

9 70.5

11 71.3 64.3 70.5 55.8 65.8 63.5242 72 68.7651598

49.5000009

70.5

12 60.2 45.3 70.5 65.1 53.2 64.7 72 68.7651598

49.5000009

70.5

13 71.4 61.3 70.5 61 55.6 71.7 72 68.76515

98

49.500000

9 70.5

14 52.7 71.5 70.5 66.4 66.7 47.7356 72 68.76515

98

49.500000

9 70.5

15 69 45.5 70.5 49.8 58.6 52.6 72 68.76515

98

49.500000

9 70.5

16 65.6 46.5 70.5 50 59.7 65.3522 72 68.76515

98

49.500000

9 70.5

17 63.8 52.1 70.5 67.9 71.1 60.2 72 68.7651598

49.5000009

70.5

18 54.5 68.8 70.5 70.2 48.3 59.1 72 68.7651598

49.5000009

70.5

19 71.9 71.2 70.5 58 62.7 45.9 72 68.76515

98

49.500000

9 70.5

20 71.3 60.2 70.5 71.5 70.9 65.8 72 68.76515

98

49.500000

9 70.5

AV

G 63.695 60.7

15 70.5 58.087

16 62.17 56.9655 72 68.76515

98

49.500000

9 70.5

219

FIG.4.7. Graph for Sound Pressure level results in all 10 methods

0 2 4 6 8 10 12 14 16 18 20 22

6070

TRAILS

GA

506070

SA

657075

PS

506070

PS

O

506070

506070

DE

GL

65707580

NL

657075

LG

O

455055

glc

Clu

657075

glc

So

l

From the graph we can see that Pattern search, Fmincon, LGO, glcCluster and

glcSolve are consistent in all 20 trials.

220

4.3.5.4. HORIZONTAL ILLUMINATION

Lighting or illumination is the deliberate application of light to achieve some aesthetic

or practical effect. In some design instances, materials used on walls and furniture play a key

role in the lighting effect. Surfaces or floors that are too reflective create unwanted glare.

Specification of illumination requirements is the basic concept of deciding how

much illumination is required for a given task. Clearly, much less light is required to

illuminate a hallway or bathroom compared to that needed for a word processing work

station. Generally speaking, the energy expended is proportional to the design illumination

level. Beyond the energy factors being considered, it is important not to over-design

illumination, lest adverse health effects such as headache frequency, stress, and

increased blood pressure be induced by the higher lighting levels. In addition, glare or excess

light can decrease worker efficiency.

TABLE.4.17.Horizontal illumination results in all 10 methods

Trial

no. GA SA PS PSO GL DE NL LGO GLCc GLCs

1 1553.

7

1014

.9 1600 1302.

9

1565.

3

555.2

472 1600 1561.970

895

433.3335

754

1522.22

222 2 1370.

9 1125

.6 1600 991.6 1548.

5 865.3 1600 1561.970

895 433.3335

754 1522.22

222 3 1569.

3 970.

4 1600 945.7 1565.

7 1302.

6 1600 1561.970

895 433.3335

754 1522.22

222 4 1196.

7

846.

1 1600 1191.

7

1560.

7

790.6

261 1600 1561.970

895

433.3335

754

1522.22

222 5 1513.

8

746.

3 1600 1222.

3

1380.

8 877.1 1600 1561.970

895

433.3335

754

1522.22

222 6 1520.

9

1100

.5 1600 1192.

7

1580.

9 550.2 1600 1561.970

895

433.3335

754

1522.22

222 7 1524.

2

1029

.3 1600 966.6

852

1557.

4

1110.

2 1600 1561.970

895

433.3335

754

1522.22

222 8 1395.

2 878.

4 1600 1359.

2 1592.

6 478.5444

1600 1561.970895

433.3335754

1522.22222

9 1503.8

959.8

1600 1181.9

1491.2

1587.9

1600 1561.970895

433.3335754

1522.22222

10 1018.

5

1042

.4 1600 1148.

9 1541 491.9

793 1600 1561.970

895

433.3335

754

1522.22

222 11 1440.

9

1116

.7 1600 1139.

2 1589 584.1

255 1600 1561.970

895

433.3335

754

1522.22

222 12 1434 789.

4 1600 1049.

7

1435.

2

1380.

1 1600 1561.970

895

433.3335

754

1522.22

222 13 1232.

4

854.

1 1600 1218.

2 1599 1450.

5 1600 1561.970

895

433.3335

754

1522.22

222 14 792.1 964.

4 1600 1365.

9 1599.

2 803.1949

1600 1561.970895

433.3335754

1522.22222

15 1538.8

791.1

1600 1128.5

1473.2

1197.2

1600 1561.970895

433.3335754

1522.22222

16 1495 961.

3 1600 1152.

2

1164.

1

799.9

138 1600 1561.970

895

433.3335

754

1522.22

222 17 1265.

3

1145

.8 1600 1459.

9

1556.

6

1566.

2 1600 1561.970

895

433.3335

754

1522.22

222 18 1234.

4

1081

.6 1600 1466.

1

1451.

6 957.6 1600 1561.970

895

433.3335

754

1522.22

222 19 1580.

4

828.

1 1600 1308 1446.

4 1537 1600 1561.970

895

433.3335

754

1522.22

222 20 1440.

9 994.

8 1600 981.6 1483.

2 1581.

2 1600 1561.970

895 433.3335

754 1522.22

222 AVG 1381.

06 962.05

1600 1188.644

1509.08

1023.337

1600 1561.970895

433.3335754

1522.22222

221

FIG.4.8.Graph for horizontal illumination results in all 10 methods

0 2 4 6 8 10 12 14 16 18 20 22

1000

1500

TRAILS

GA

80010001200

SA

1600

PS

100012001400

PS

O

120014001600

GL

50010001500

DE

1600NL

1600

LG

O

400

450

glc

Clu

1400

1600g

lcS

ol

From the graph we can see that Pattern search, Fmincon, LGO, glcCluster and

glcSolve are consistent in all 20 trials.

222

4.3.5.5. IEQ.

TABLE.4.18. IEQ results in all 10 methods

Trial no. GA SA PS PSO GL DE NL LGO GLCc GLCs

1 1 1 1 1 1 1 1 1 1 1

2 1 1 1 1 1 1 1 1 1 1

3 1 1 1 1 1 1 1 1 1 1

4 1 1 1 1 1 1 1 1 1 1

5 1 1 1 1 1 1 1 1 1 1

6 1 1 1 1 1 1 1 1 1 1

7 1 1 1 1 1 1 1 1 1 1

8 1 1 1 1 1 1 1 1 1 1

9 1 1 1 1 1 1 1 1 1 1

10 1 1 1 1 1 1 1 1 1 1

11 1 1 1 1 1 1 1 1 1 1

12 1 1 1 1 1 1 1 1 1 1

13 1 1 1 1 1 1 1 1 1 1

14 1 1 1 1 1 1 1 1 1 1

15 1 1 1 1 1 1 1 1 1 1

16 1 1 1 1 1 1 1 1 1 1

17 1 1 1 1 1 1 1 1 1 1

18 1 1 1 1 1 1 1 1 1 1

19 1 1 1 1 1 1 1 1 1 1

20 1 1 1 1 1 1 1 1 1 1

AVG 1 1 1 1 1 1 1 1 1 1

223

FIG.4.9. Graph for IEQ results in all 10 methods

0 2 4 6 8 10 12 14 16 18 20 22

0

2

GA

TRAILS

0

2

SA

0

2

PS

0

2

PS

O

0

2

GL

0

2

DE

0

2

NL

0

2

LG

O

0

2

glc

Clu

0

2g

lcS

ol

From the graph we can see that all ten methods are consistent in all 20 trials with

value 1 which is the optimum acceptable value.

224

4.3.5.6. ELAPSED TIME

CPU time is the time for which the CPU was busy executing the task. It does not take

into account the time spent in waiting for I/O (disk IO or network IO). Since I/O operations,

such as reading files from disk, are performed by the OS, these operations may involve

noticeable amount of time in waiting for I/O subsystems to complete their operations. This

waiting time will be included in the elapsed time, but not CPU time. Hence CPU time is

usually less than the elapsed time.

TABLE.4.19. Elapsed time results in all 10 methods

Trail GA SA PS PSO GL DE NL LGO GLCc GLCs

1 1.208

193

2.5824

72

0.156

075

0.144

904

1.546

532

0.003

03

8.86601

9

0.912

226

0.597

778

0.933

053 2 1.133

192

2.2277

51

0.148

124

0.118

412

1.885

316

0.003

735

13.4064

07

0.732

615

0.715

455

1.291

154 3 0.775

846

2.3222

16

0.148

314

0.081

629

1.012

803

0.002

558 14.7796 0.731

219

0.612

695

0.876

201 4 1.089

181

2.4827

02

0.146

55

0.090

009

1.505

692

0.010

54

12.6790

64

0.724

432

0.552

044

1.230

004 5 1.059

298

1.9079

2

0.146

723

0.086

346

1.191

908

0.003

255

7.11649

2

0.715

838

0.728

062

0.990

297 6 1.140

209

3.0296

57

0.150

499

0.087

901

1.099

644

0.004

085

11.6128

6

0.753

451

0.635

165

0.905

318 7 0.768

988

2.8660

03

0.148

128

0.095

911

1.044

684

0.003

35 7.61832 0.736

378

0.626

25

0.943

921 8 0.919

29

2.2797

64

0.147

931

0.085

106

1.153

72

0.001

541

6.78551

4

0.743

917

0.562

734

0.952

783 9 1.172

938

2.0897

66

0.148

575

0.088

058

1.076

416

0.001

507

8.38266

8

0.740

182

0.598

343

1.098

55 10 0.917

832

2.1340

44

0.147

136

0.079

668

1.219

463

0.003

296

13.3524

94

0.961

585

0.737

171

0.959

065 11 0.913

941

2.9115

1

0.149

625

0.102

006

1.070

33

0.002

578

9.97782

8

0.713

963

0.584

185

1.100

177 12 1.169

327

3.0143

31

0.146

856

0.094

814

0.935

505

0.002

499

11.5853

06

0.943

391

0.611

783

1.015

112 13 1.059

507

2.2839

27

0.146

429

0.097

502

1.032

483

0.003

78

9.27154

5

0.971

514

0.563

611

0.909

444 14 0.887

249

2.4240

2

0.146

925

0.093

003

1.285

674

0.003

476

8.60228

6

0.969

471

0.586

987

0.966

466 15 1.115

581

3.0542

13

0.147

402

0.098

596

0.933

089

0.003

66

12.3196

61

1.006

35

0.570

589

1.013

812 16 1.195

485

2.0850

67

0.147

624

0.089

77

0.953

743

0.003

21

14.4853

07

0.962

908

0.661

609

1.010

435 17 1.032

374

2.5689

22

0.146

341

0.083

28

1.379

827

0.002

003

10.1844

98

0.745

92

0.604

967

1.091

511 18 0.932

092

2.9044

54

0.146

18

0.089

962

1.021

947

0.003

846

7.90361

3 0.983 0.631

04

1.191

183 19 1.162

234

2.8653

76

0.149

296

0.093

909

1.075

74

0.003

841

13.6878

91 0.811 0.600

535

0.864

12 20 0.913

941

1.9046

84

0.149

342

0.079

203

1.512

711

0.001

516

12.0780

18

0.704

146

0.560

903

0.926

83 AVG

VG

1.028

335

2.4969

3995

0.148

204

0.093

999

1.196

861

0.003

365

10.7347

6955

0.828

1753

0.617

0953

1.013

4718

225

FIG.4.10.Graph for Elapsed time results in all 10 methods

0 2 4 6 8 10 12 14 16 18 20 22

0.81.01.2

TRAILS

GA

2.02.53.0

SA

0.1450.1500.155

PS

0.10

0.15

PS

O

1.01.52.0

GL

0.0000.0050.010

DE

10

15

NL

0.8

1.0

LG

O

0.60.7

glc

Clu

0.81.01.2

glc

Sol

From the graph, we can see that none of the methods was consistent in all 20 trials but

Pattern search method alone got a more or less same value – average of 0.148 seconds.

Which is second best in terms of time. Therefore it is the best method.

226

4.3.5.7. ITERATIONS

In a computational procedure, a cycle of operations is repeated, often to approximate

the desired result more closely. Iteration means the act of repeating a process usually with the

aim of approaching a desired goal or target or result. Iteration in computing is the repetition

of a process within a computer program. It may also refer to the process of iterating a

function i.e. applying a function repeatedly, using the output from one iteration as the input to

the next. Another use of iteration in mathematics is in iterative methods which are used to

produce approximate numerical solutions to certain mathematical problems.

TABLE.4.20. Iterations results in all 10 methods

Trial no. GA SA PS PSO GL DE NL LGO GLCc GLCs

1 51 2004 60 57 4 40 5002 2699 1515 1503

2 51 2003 60 73 4 40 7277 2699 1511 1503

3 51 2001 60 53 4 40 7042 2699 1514 1503

4 51 2006 60 56 4 40 6897 2699 1512 1503

5 51 2005 60 53 4 40 4067 2699 1513 1503

6 51 2017 60 55 4 40 5507 2699 1511 1503

7 51 2031 60 59 4 40 5312 2699 1513 1503

8 51 2005 60 51 4 40 4672 2699 1511 1503

9 51 2007 60 54 4 40 4847 2699 1513 1503

10 51 2006 60 52 4 40 6477 2699 1511 1503

11 51 2011 60 66 4 40 4662 2699 1516 1503

12 51 2002 60 62 4 40 6032 2699 1513 1503

13 51 2013 60 63 4 40 5712 2699 1513 1503

14 51 2007 60 53 4 40 4612 2699 1513 1503

15 51 2007 60 59 4 40 6592 2699 1511 1503

16 51 2015 60 57 4 40 5987 2699 1513 1503

17 51 2016 60 53 4 40 4567 2699 1514 1503

18 51 2004 60 51 4 40 4912 2699 1515 1503

19 51 2016 60 58 4 40 6077 2699 1515 1503

20 51 2025 60 51 4 40 6227 2699 1513 1503

AVG 51 2010.05 60 56.8 4 40 5624 2699 1513 1503

227

FIG.4.11. Graph for Iterations results in all 10 methods

0 2 4 6 8 10 12 14 16 18 20 22

455055

TRAILS

GA

2000

2020

SA

556065

PS

506070

PS

O

345

GL

35

40

45

DE

4000

6000

NL

26002800

LGO

1510

1515

glcC

lu1400

1600gl

cSol

From the graph we can see that GA, Pattern search, GL, DE, LGO and glcSolve are

consistent in all 20 trials.

228

Table 4.21- Comparative table for parameters in all 10 methods

Variables GA SA PS PSO GL fmincon DE LGO Glc

Cluster

Glc

Solve

PMV X X 2

X X 2

X 2

-1.9

-1.9

CO2 X X 1800

X X 1800

X 1759

1150

1727.7

Sound X X 70.5

X X 72

X 68.76

49.5

70

Illumina tion

X X 1600

X X 1600

X 1561.9

433.3

1522.2

IEQ 1

1

1

1

1

1

1

1

1

1

TIME 0.148 0.09

3

0.003

ITERS X 60

X 4

40

X X

- Represents the parameters which are consistent for all the 20 trials and the

corresponding parameter values are given in the respective cell.

X - Represents the parameters which are not consistent for all the 20 trials

In case of iterations and elapsed time only the two or three minimum values alone are given.

4.3.6. Results and Discussion

With the two extreme values of parameters from survey, the optimization is carried

out with different solvers. As they are of stochastic type, their results may vary from trial to

trial and so the problem is made to run for 20 trials (Elbeltagi, Tarek Hegazy, & & Grierson,

2005) and an average of all trials is taken as the final value of the parameter, by the solver.

The solvers are compared with three different criteria

1. Consistency

The consistency table gives the parameters that remain constant for all the

trails. All the solvers give the same value of IEQ for all the runs, which in turn

indicates that the quality requirements are acceptable.

Thermal - P.S (2), NL(2), LGO (2),glcSolve (-1.9), glcCluster (-1.9)

CO2 - P.S (1800), NL(1800), LGO (1759),glcSolve(1727.7), glcCluster

229

(1150)

Sound - P.S(70.5), NL(72), LGO (68.76),glcSolve (70), glcCluster (49.5)

Illumination -P.S (1600), NL(1600), LGO (1561.91),glcSolve (1522.2),

glcCluster (433.3)

So we see that the solvers Pattern Search,Fmincon, glcSolve, glcCluster&

LGO remain constant throughout their runs.

2. Minimum Run Time

For a minimum run time of the problem we got PS (0.093 seconds), Pattern

Search (0.148 seconds), DE (0.003 seconds).

3. Minimum Evaluation

This criterion will determine the effectiveness of the algorithm. From the

result table, we see that the Pattern Search, GODLIKE and DE algorithms have

minimum evaluation of 60 , 4 and 40 respectively

4. Simplicity of Algorithm

Of all the algorithms we have taken the Pattern Search algorithm is the

simplest followed by GA, PSO, DE, Simulated Annealing, GODLIKE, Non-

Linear, and Direct algorithm.

5. Results according to Standards

This is the most important criterion that determines whether the solver is

practical or not. We got the standard values from ASHRAE, IES, Guidance for

employers on the Control of Noise at Work Regulations 2005 as:

Thermal comfort: -3 to 3

Carbon dioxide: less than 1000ppm

Sound level: 40 dBA to 70dBA

Illumination level: 1000 lux to 2000 lux

With the above standards the solvers which adhere to the standard are:

Thermal comfort: GA, SA, PS, PSO, FMINCON, DE, GL,LGO,

glcCluster, glcSolve

230

Carbon dioxide: GA, SA, PS, PSO, FMINCON,GL, LGO, glcCluster,

glcSolve

Sound level: GA, SA, PS, PSO, FMINCON, DE, GL, LGO,glcCluster,

glcSolve

Illumination level: DE, glcCluster, glcSolve

The following table gives a summary of all the criteria for the solvers:

Table.4.22. Summary of all the criteria for the solvers

Criteria GA SA PS PSO Fmincon DE GL LGO glcClus glcSolve

Result

according

to

ASHRAE

¾

=75%

¾

=75%

¾

=75%

¾

=75%

¾

=75%

¾

=75%

¾

=75%

¾

=75%

4/4

=100%

4/4

=100%

Consistency - - - - - -

Min-Run

Time - - - - - - - -

Min-

Evaluation - - - - - - - -

Simple

Algorithm - - - - - - - - -

Thus it is seen that the Pattern Search solver satisfies all the criteria and scores 75%

for its practicality in giving result according to ASHRAE, IES and Guidance for

employers on the Control of Noise at Work Regulations 2005, So the appropriate

algorithm, for optimization of thermal comfort is suggested as Direct search

algorithm & the solver is PATTERN SEARCH

4.3.7. Conclusion

Office environment is generally designed to the design guides and practises for the

occupant‘s comfort. In this work, the overall IEQ of offices of Karunya University in

Coimbatore was evaluated by 220 occupants in four aspects, namely thermal comfort, indoor

air quality, equivalent noise level and illumination level. All the offices considered are

naturally ventilated buildings. The results showed that the operative temperature, carbon

231

dioxide concentration, equivalent noise level and illumination level had important effects on

the overall IEQ acceptance. Empirical expressions have been proposed to approximate the

occupant‘s acceptance of the IEQ in office. The non traditional algorithms are used to find

the optimum value of the IEQ.

Here, ten non-traditional optimization algorithms were presented. These

include: GA, SA, PS, PSO, GL, FMINCON, EA, LGO, glcCluster, glcSolve. A brief

description of each method is presented along with a Pseudo code to facilitate their

implementation. MATLAB programs were written to implement each algorithm. The IEQ

problem for the offices of the Karunya University was solved using all algorithms, and the

comparative results were presented.

4.4. IEQ Residence

4.4.1. Introduction

Indoor environmental quality (IEQ) and occupant comfort are closely related. IEQ

parameters are interdependent and must be considered interactively. Other than the office

people spend most of the times in their residences. If the residences are also thermally

comfortable then their health and other related things like production and so on will not be

affected. In the case of residences also, we have considered only residences where natural

ventilation is preferred and closing and opening of the windows and using of fans is

maximum.

4.4.2. Field measurements

Subjective as well as objective evaluations of indoor environmental conditions

made by 102 occupants from 11 typical residential apartments in Karunya University were

collected through individual interviews. The interviewees were mainly those occupants

staying at the quarters of Karunya University the longest time (as compared with other

activities), and the housing samples covered 11 residential flats. The inclusion of various

apartment types could cover a wide range of probable indoor environmental conditions. The

apartments varied in size from 330 m2 to 1336.17 m

2 and were equipped with window.

232

The indoor physical parameters describing the indoor environmental quality of a

space were CO2 concentration (ppm), horizontal illuminance level (lux) and sound pressure

level (dBA). Therefore, the measured IEQ data could sufficiently reflect the real-time

exposed indoor environment to the respondent. The number of measurements was determined

based on the arrangement and partitioning of the specific apartment and the distribution of

occupants.

Unlike the measurement approach in office, each assessment sample in residential

housing did not necessarily require a separate 15 min physical measurement due to the small

living area. In this case, two physical measurements carried out in both dining/living room

and bedrooms were considered representative for the assessment of the example apartment

case. The IEQ data logged in the dining/living room were considered to be applicable to the

individual occupant within the group. Such condition was judged, depending on the space

between each occupant.

Being the base for evaluating the energy benchmarking models, effective

measurements were essential and thus accurate and reliable data could be obtained through a

dichotomous assessment scale, the occupant acceptance of the perceived indoor environment

was recorded in the form of direct feedback using the question ‗Is the thermal environment/

indoor air quality/noise level/illumination level being perceived in the residential

environment acceptable to you?‘ [42] The ranks ‗(1) Yes, acceptable‘ and ‗(0) No, not

acceptable‘ were self-explanatory. In order to confirm the validity of a response, each

respondent had to use a semantic differential evaluation scale (ref) for the subjective

assessment of thermal environment and IAQ, and a visual analogue assessment scale for the

evaluation of aural and visual comforts. At the end of the survey, an overall IEQ acceptance

was determined.

233

Table 4.23.Occupant’s votes on acceptance of a perceiving indoor environmental quality

U – Unacceptable; A - Acceptable

The overall IEQ acceptance θ for a resident environment perceived by an occupant

expressed by a multivariate logistic regression model is proposed

1

θ = 1 - -----------------------------------------

1 + exp (C0,0 + )

Where the regression constants determined from the 102 occupant evaluations; Values of the

constants confirm the relative importance of the four contributors to θ, the larger the value,

the greater the importance. Occupants were very sensitive to the operative temperature when

compared to the other three parameters. Regression coefficients can be evaluated with

surveyed occupant responses from residential environment for the overall IEQ acceptance.

C0,0 and Ci,o are the regression constants which can be determined from filed measurements,

φ1 is the occupant acceptance correlated with the thermal sensation vote 1, CO2

concentration 2 (ppm). The equivalent sound pressure level 3 (dBA) and the horizontal

illumination level 4 (lux)

The thermal environment acceptance φ1, with the maximum acceptance = 0.95, is

given below, where C0,1 and C1.1 are the regression coefficients,

The acceptances φ2 , φ3 and φ4 are expressed by logistic regression models with regression

coefficients C0,j and C1,j

Overall

acceptance

Θ

Votes Thermal

environment

φ1

Air quality

φ2

Noise level

φ3

Illumination

level

φ4

U

A

8

95

U A U A U A U A

6

5

2

89

3

3

4

92

2

6

5

89

5

5

2

90

TOTAL 102 11 91 6 96 8 94 10 92

234

Table 4.24. Regression coefficients of logistic regression model.

No variable C0,i C1,i C2,i C3,i C4,i

0 Φ0 -33.24 21.95 1.614 11.779 21.90

1 Φ1 0.03353 0.2179 --- --- ---

2 Φ2 45.21 -0.0257 --- --- ---

3 Φ3 23.82 -0.2981 --- --- ---

5 Φ4 -14.08 0.9043 --- --- ---

Various combinations of contributors i=1, 2, 3, 4 and the corresponding overall IEQ

acceptance were considered. A total of 24 possibilities were found. Taking the binary notation

for the acceptance i.e., 0 for ‗unacceptable‘ and 1 for ‗acceptable‘ the predicted acceptance of

IEQ (θ) is calculated.

Table 4.25.Overall IEQ acceptance

Case No. Survey Sample Contributors Predicted

acceptance of IEQ θ Φ1 Φ2 Φ3 Φ4

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

1

0

1

2

0

1

2

5

1

0

0

2

0

6

6

75

0

0

0

0

0

0

0

0

1

1

1

1

1

1

1

1

0

0

0

0

1

1

1

1

0

0

0

0

1

1

1

1

0

0

1

1

0

0

1

1

0

0

1

1

0

0

1

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1.188 10-5

4.8 10-10

0.608

1.18876 10-5

1.2497 10-5

2.47 10-9

0.8889

1.2497 10-5

0.9999

0.61987

0.9999

6.2768 10-5

0.9999

0.891194

1.00000

235

4.4.3. Algorithms

4.4.3.1. Genetic algorithm

4.4.3.1.1. Stopping Criteria reached:

The options and the stopping criteria which are set are same as that for the IEQ Office

Buildings problem. This case also reaches the final solution by the stopping condition,

‖ the change in the final value of the system is less than 10-6‖. Hence we say that the global

optimum solution is obtained naturally.

TABLE.4.26. Results of GA in 20 trails for resident IEQ

Trails Thermal

sensation

vote

CO2 Sound

pressure

level

Illumination IEQ TIME ITERATIONS

1 0 744.2 67 1264.3 1 0.381663 51

2 -0.0005 803.7882 67.0003 591.4519 1 0.390658 51

3 0 1045.3 67 672.3 1 0.41928 51

4 0.0303 648.083 67.0001 508.4575 1 0.383461 51

5 0.0007 382.3028 67 988.4221 1 0.403158 51

6 -0.0004 951.0608 67.0219 261.205 1 0.377546 51

7 0.0066 354.1662 67 893.4416 1 0.391053 51

8 0.0016 623.0983 67.2189 768.8616 1 0.385192 51

9 0 1153 67 187.1 1 0.405951 51

10 0.0335 863.3774 67.0058 707.8476 1 2.318421 51

11 0 1108.4 68.5 1221.7 1 4.457991 51

12 0 609.3 67.1 1217.3 1 0.400942 51

13 0 369.5 67.3 1059.3 1 0.395878 51

14 0.0026 330 67 187.0001 1 0.41974 51

15 0 330 67 1475.6 1 0.396296 51

16 0.0007 684.0009 67.1748 187.001 1 0.415877 51

17 0 347.5 67.2 1381.9 1 0.405463 51

18 0.0033 485.9581 67.0087 89.5593 1 0.3879 51

19 0 684.3 67 1458.9 1 0.382807 51

20 0 1062.3 67 349 1 0.380528 51

Avg 0.00392 678.9818 67.12653 773.532385 1 0.69499 51

236

Fig.4.12.Convergence of GA

4.4.3.2. Simulated annealing

4.4.3.2.1. Stopping Criteria Reached:

The options and the stopping criteria which are set are same as that for SA in the IEQ

Office Buildings problem. Though the iterations are of large number, this case also reaches

the final solution by the stopping condition,‖ the change in the final value of the system is

less than 10-6‖. The large number of iterations is because that the SA algorithm is

Metaheuristics type . Hence we say that the global optimum solution is obtained naturally.

0 10 20 30 40 50 60 70 80 90 100-1

-1

-1

-1

-1

-1

-1

Generation

Fitness v

alu

eBest: -1 Mean: -1

Best f itness

Mean fitness

237

TABLE4.27. Results of SA in 20 trails for resident IEQ

Trails Thermal

sensation

vote

CO2 Sound

pressure

level

Illumination IEQ Time Iterations

1 0 915 67 854.5 1 2.741429 2000

2 0 915 67 854.5 1 1.024044 2000

3 0 915 67 854.5 1 1.05276 2000

4 0 915 67 854.5 1 1.031611 2000

5 0 915 67 854.5 1 1.035856 2000

6 0 915 67 854.5 1 1.034596 2000

7 0 915 67 854.5 1 1.029433 2000

8 0 915 67 854.5 1 1.067096 2000

9 0 915 67 854.5 1 1.044101 2000

10 0 915 67 854.5 1 1.042642 2000

11 0 915 67 854.5 1 1.026046 2000

12 0 915 67 854.5 1 1.021253 2000

13 0 915 67 854.5 1 1.018455 2000

14 0 915 67 854.5 1 1.02123 2000

15 0 915 67 854.5 1 1.018589 2000

16 0 915 67 854.5 1 1.007763 2000

17 0 915 67 854.5 1 1.027883 2000

18 0 915 67 854.5 1 1.018682 2000

19 0 915 67 854.5 1 1.010296 2000

20 0 915 67 854.5 1 1.021411 2000

Avg 0 915 67 854.5 1 1.114759 2000

Fig.4.13.Convergence of SA

0 200 400 600 800 1000 1200 1400 1600 1800 2000-2

-1.5

-1

-0.5

0

0.5

Iteration

Function v

alu

e

Best Function Value: -1

1 2 3 40

200

400

600

800

1000Best point

Number of variables (4)

Best

poin

t

0 10 20 30 40 50 60 70 80 90 100

Time

Iteration

f-count

% of criteria met

Stopping Criteria

0 200 400 600 800 1000 1200 1400 1600 1800 2000-1

-0.8

-0.6

-0.4

-0.2

0

Iteration

Function v

alu

e

Current Function Value: -0.99134

238

4.4.3.3. Pattern search

4.4.3.3.1. Stopping Criteria Reached:

The solution is reached by the stopping condition, ―difference in function value less

than 10-6‖ and also comparatively the iterations are of less in number, this indicates quick

convergence. The final value of the solution is naturally obtained.

TABLE.4.28. Results of PS in 20 trails for resident IEQ

Trails Thermal CO2 Sound Illumination IEQ Time Iterations

1 0 915 67 854.5 1 0.979432 20

2 0 915 67 854.5 1 0.053695 20

3 0 915 67 854.5 1 0.062384 20

4 0 915 67 854.5 1 0.054444 20

5 0 915 67 854.5 1 0.060776 20

6 0 915 67 854.5 1 0.06348 20

7 0 915 67 854.5 1 0.054188 20

8 0 915 67 854.5 1 0.050838 20

9 0 915 67 854.5 1 0.062327 20

10 0 915 67 854.5 1 0.043569 20

11 0 915 67 854.5 1 0.07328 20

12 0 915 67 854.5 1 0.053983 20

13 0 915 67 854.5 1 0.62883 20

14 0 915 67 854.5 1 0.054073 20

15 0 915 67 854.5 1 0.062371 20

16 0 915 67 854.5 1 0.052222 20

17 0 915 67 854.5 1 0.04451 20

18 0 915 67 854.5 1 0.063024 20

19 0 915 67 854.5 1 0.070971 20

20 0 915 67 854.5 1 0.075022 20

Avg 0 915 67 854.5 1 0.133171 20

239

Fig.4.14.Convergence of PS

0 2 4 6 8 10 12 14 16 18 20-2

-1.5

-1

-0.5

0

0.5

Iteration

Function v

alu

e

Best Function Value: -1

0 2 4 6 8 10 12 14 16 18 200

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

Iteration

Mesh s

ize

Current Mesh Size: 9.5367e-007

240

4.4.3.4. Particle swarm optimization

4.4.3.4.1. Stopping Criteria Reached:

The options and the stopping criteria which are set are same as that for PSO in the

IEQ Office Buildings problem. This case also the final solution reaches by the stopping

condition,‖ the change in the final value of the system is less than 10-6‖ but the specialty is

the elapsed time which is less than other solvers. The global optimum solution is obtained

without any other stopping conditions.

TABLE.4.29. Results of PSO in 20 trails for resident IEQ

Trails Thermal CO2 Sound level Illumination IEQ Time Iterations

1 0 1076.8 67 437.1 1 0.120159 70

2 0 1221.6 67 425.9 1 0.121556 56

3 0 849 67 1288.8 1 0.130095 52

4 0 849 67 1288.8 1 0.130094 52

5 -0.0282 920.0475 67.0023 246.3722 1 0.121258 51

6 0 1204.9 67 998.7 1 0.10458 51

7 0 1111.5 67 489.9 1 0.140347 54

8 0.0069 534.5203 67.0005 344.6397 1 0.147119 64

9 0 1279.8 67 1276.6 1 0.145017 68

10 0.0043 827.5092 67 544.1189 1 0.120778 52

11 0 1021.7 67 1253.8 1 0.104316 51

12 0 1085.6 67 1367.2 1 0.123623 57

13 0 1305 67 871.2 1 0.126779 60

14 0 843.1 67 1220.5 1 0.138437 52

15 0.002 672.6782 67.0011 372.4912 1 0.114034 51

16 0 1384 67 780 1 0.114465 52

17 0.0064 650.2896 67.0001 459.1902 1 0.157973 58

18 -0.0115 645.3098 67.0017 763.8825 1 0.094197 54

19 0 1414.4 67 511 1 0.112067 51

20 0.152 453.1459 67.0029 458.3543 1 0.152493 51

Avg 0.006595 967.495 67.00043 769.92745 1 0.125969 55.35

241

4.4.3.5. GODLIKE

4.4.3.5.1. Stopping Criteria Reached:

The options and the stopping criteria which are set are same as that for GODLIKE in

the IEQ Office Buildings problem. This case also the final solution reaches by the stopping

condition,‖ the change in the final value of the system is less than 10-6‖.The solver exchanges

the population among the solvers hence the iteration indicates number of times the population

is exchanged. The global optimum solution is obtained without any other stopping conditions.

TABLE.4.30. Results of GL in 20 trails for resident IEQ

Trails Thermal vote CO2 Sound Illumination IEQ Time Iterations

1 -0.0005 880.7416 67 510.4552 1 1.421388 4

2 0 707.1 67 1040.7 1 1.483061 4

3 0 885 67 1093.7 1 1.011846 4

4 -0.0002 945.5216 67 356.0738 1 1.652818 4

5 0 922.4489 67 561.348 1 1.396136 4

6 0 1194.2 67 927.2 1 1.606247 4

7 0 1273.4 67 1101.4 1 1.888101 4

8 0.0001 833.6192 67 640.5575 1 1.345902 4

9 0 1088.7 67 1021.5 1 3.324052 4

10 -0.0006 728.5694 67 800.8187 1 1.21074 4

11 0 768.5 67 1095.1 1 2.176719 4

12 0.0013 868.9358 67 575.0751 1 1.179232 4

13 0 451.4 67 1049 1 1.23686 4

14 -0.001 859.5101 67.00006 948.2773 1 1.025074 4

15 0 946.1 67 1248 1 2.032979 4

16 0 800.9 67 1250.9 1 1.237546 4

17 0 1127.9 67 1159.2 1 1.085968 4

18 0.0006 753.0979 67 557.074 1 1.954029 4

19 0 1053.7 67 616.5 1 1.552567 4

20 0 525.6 67 1382 1 1.325332 4

Avg -0.000015 880.7472 67 896.74398 1 1.55733 4

242

4.4.3.6. Fmincon.

4.4.3.6.1. Stopping Criteria Reached:

The options and the stopping criteria which are set are same as that for

Fmincon in the IEQ Office Buildings problem. This case also the final solution reaches by the

stopping condition,‖ the change in the final value of the system is less than 10-6‖. The global

optimum solution is obtained without any other stopping conditions. The exception is that the

elapsed time is high comparatively; this is due to the traditional technique modified version

of using lagranges multipliers.

TABLE.4.31. Results of Fmincon in 20 trails for resident IEQ

Trails Thermal vote CO2 Sound Illumination IEQ Time Iterations

1 0 915 67.0001 854.5 1 4.758093 2972

2 0 915 67.0001 854.5 1 3.753865 2694

3 0 915 67.0001 854.5 1 6.801059 4607

4 0 915 67.0001 854.5 1 4.662614 3188

5 0 626.5 67 1199.6 1 4.542072 2728

6 0 915 67.0001 854.5 1 5.275708 3777

7 0 915 67.0001 854.5 1 5.449035 3034

8 0 915 67.0001 854.5 1 4.073841 2881

9 0 915 67.0001 854.5 1 4.747727 3214

10 0 915 67.0001 854.5 1 5.406024 4080

11 0 915 67.0001 854.5 1 5.538329 3780

12 0 915 67.0001 854.5 1 3.909716 2722

13 0 915 67.0001 854.5 1 3.911251 2953

14 0 915 67.0001 854.5 1 4.54207 2728

15 0 915 67.0001 854.5 1 8.01847 3505

16 0 915 67.0001 854.5 1 4.79385 2929

17 0 928.2 67 1111.2 1 4.99689 3108

18 0 915 67.0001 854.5 1 3.32011 2562

19 0 915 67.0001 854.5 1 9.42467 3923

20 0 915 67.0001 854.5 1 4.21304 2783

AVG 0 901.235 67.0000 884.59 1 5.10692 3208

243

4.4.3.7. Direct evolution

4.4.3.7.1. Stopping Criteria Reached:

The options and the stopping criteria which are set are same as that for DE in the IEQ

Office Buildings problem. This case also the final solution reaches by the stopping

condition,‖ the change in the final value of the system is less than 10-6‖. It is seen from the

results that the final vectors (parameter values) is not consistent, this is because DE uses

different type of cross over method. The global optimum solution is obtained without any

other stopping conditions.

TABLE.4.32. Results of DE in 20 trails for resident IEQ

Trails Thermal CO2 Sound Illumination IEQ Time Iterations

1 -0.4 1295.6 67.9 364.8 1 0.015565 40

2 0.2 710.4 68.2 1002.6 1 0.001693 40

3 0.3 1173.6 72.7 1513.6 1 0.002347 40

4 -0.2 1088.4 67.3 1311.3 1 0.004192 40

5 0.3 696.7 68.8 1018 1 0.001568 40

6 0.3 1012.8 72.6 297.3 1 0.00224 40

7 0.3 1457.3 67.5 1485.9 1 0.002525 40

8 0.1019 981.3327 68.7214 937.3448 1 0.004635 40

9 0.1427 639.9009 67.7519 769.4966 1 0.0045 40

10 -0.4 1206.5 67.4 1450.3 1 0.004322 40

11 -0.2 802.6 69 1480.5 1 0.004393 40

12 0 940.3 68 1394.7 1 0.004464 40

13 0.1 1446 67.8 463.4 1 0.002196 40

14 -0.6 1427.8 69.8 898.8 1 0.004563 40

15 -0.2 1441.2 67.9 560.6 1 0.004486 40

16 -0.1 1396.9 67.2 396.2 1 0.004362 40

17 0.209 668.6973 67.7521 300.4359 1 0.004467 40

18 0.1 1255.1 69.2 1092.3 1 0.004208 40

19 0 718.4 68.9 1022.2 1 0.002391 40

20 0.1 1235.9 68.3 1022 1 0.002312 40

Avg 0.00268 1079.772 68.63627 939.088865 1 0.004071 40

244

4.4.3.8. LGO

4.4.3.8.1. Stopping Criteria Reached:

The options and the stopping criteria which are set are same as that for LGO in the

IEQ Office Buildings problem. The global solution reaches by the stopping condition,‖ the

change in the final value of the system did not improve‖. The elapsed time is close to that of

other Direct algorithm solvers but it does not use Lipchitz constant

TABLE.4.33 .Results of LGO in 20 trails for resident IEQ

Trails Thermal CO2 Sound Illumination IEQ Time Iterations

1 0.0027 497.0895 67 931.8641 1 1.166753 3223

2 0.0027 497.0895 67 931.8641 1 0.42993 3223

3 0.0027 497.0895 67 931.8641 1 0.42055 3223

4 0.0027 497.0895 67 931.8641 1 0.43747 3223

5 0.0027 497.0895 67 931.8641 1 0.442481 3223

6 0.0027 497.0895 67 931.8641 1 0.414941 3223

7 0.0027 497.0895 67 931.8641 1 0.421402 3223

8 0.0027 497.0895 67 931.8641 1 0.438488 3223

9 0.0027 497.0895 67 931.8641 1 0.422659 3223

10 0.0027 497.0895 67 931.8641 1 0.431362 3223

11 0.0027 497.0895 67 931.8641 1 0.427368 3223

12 0.0027 497.0895 67 931.8641 1 0.435591 3223

13 0.0027 497.0895 67 931.8641 1 0.440821 3223

14 0.0027 497.0895 67 931.8641 1 0.431039 3223

15 0.0027 497.0895 67 931.8641 1 0.426922 3223

16 0.0027 497.0895 67 931.8641 1 0.435226 3223

17 0.0027 497.0895 67 931.8641 1 0.431876 3223

18 0.0027 497.0895 67 931.8641 1 0.428309 3223

19 0.0027 497.0895 67 931.8641 1 0.437069 3223

20 0.0027 497.0895 67 931.8641 1 0.429926 3223

Avg 0.0027 497.0895 67 931.8641 1 0.467509 3223

245

4.4.3.9. glcCluster

4.4.3.9.1. Stopping Criteria Reached:

The default options are taken from the solver from the previous run of the IEQ Office

Buildings problem. The global solution reaches by the stopping condition,‖ the change in the

final value of the system is less than 10-7‖.Though glcCluster uses Clustering algorithm in

addition it has very less elapsed time.

TABLE4.34. Results of glcCluster in 20 trails for resident IEQ

Trails Thermal CO2 Sound Illumination IEQ Time Iterations

1 0 915 67.6111 1299.5 1 0.027167 1516

2 0 525 67.6111 1299.5 1 0.027167 1516

3 0 655 67.6111 1299.5 1 0.027167 1516

4 0 1045 67.6111 1299.5 1 0.027167 1516

5 0 1045 67.6111 1299.5 1 0.027167 1516

6 0 655 67.6111 1299.5 1 0.027167 1516

7 0 1175 67.6111 1299.5 1 0.027167 1516

8 0 525 67.6111 1299.5 1 0.027167 1516

9 0 655 67.6111 1299.5 1 0.027167 1516

10 0 655 67.6111 1299.5 1 0.027167 1516

11 0 915 67.6111 1299.5 1 0.027194 1516

12 0 525 67.6111 409.5 1 0.027194 1516

13 0 655 67.6111 854.5 1 0.027194 1516

14 0 1045 67.6111 1299.5 1 0.027194 1516

15 0 1045 67.6111 1299.5 1 0.027194 1516

16 0 655 67.6111 1299.5 1 0.027194 1516

17 0 1175 67.6111 1299.5 1 0.027194 1516

18 0 525 67.6111 1299.5 1 0.027194 1516

19 0 655 67.6111 409.5 1 0.027194 1516

20 0 655 67.6111 1299.5 1 0.027194 1516

Avg 0 785 67.6111 1188.25 1 0.02718 1516

246

4.4.3.10. glcSolve

4.4.3.10.1. Stopping Criteria Reached:

The options and the stopping criteria are taken from the previous run of IEQ Office

Building problem. The final solution reaches by the stopping condition,‖ the change in the

final value of the system is less than 10-6‖. glcSolve uses one of the complex algorithm and

even after giving long range values for parameters (which is not recommended) it takes little

time to complete optimization.

TABLE.4.35. Results of glcSolve in 20 trails for resident IEQ

Trails Thermal CO2

concen

tration

Sound Illumination IEQ Time Fun eval Iterations

1 0 915 67.611 1299.5 1 0.31502 1529 110

2 0 915 67.611 1299.5 1 0.360233 1529 110

3 0 915 67.611 1299.5 1 0.332232 1529 110

4 0 915 67.611 1299.5 1 0.334579 1529 110

5 0 915 67.611 1299.5 1 0.355238 1529 110

6 0 915 67.611 1299.5 1 0.388838 1529 110

7 0 915 67.611 1299.5 1 0.342736 1529 110

8 0 915 67.611 1299.5 1 0.332214 1529 110

9 0 915 67.611 1299.5 1 0.333896 1529 110

10 0 915 67.611 1299.5 1 0.336873 1529 110

11 0 915 67.611 1299.5 1 0.336153 1529 110

12 0 915 67.611 1299.5 1 0.323308 1529 110

13 0 915 67.611 1299.5 1 0.333478 1529 110

14 0 915 67.611 1299.5 1 0.320211 1529 110

15 0 915 67.611 1299.5 1 0.340051 1529 110

16 0 915 67.611 1299.5 1 0.337327 1529 110

17 0 915 67.611 1299.5 1 0.333373 1529 110

18 0 915 67.611 1299.5 1 0.343034 1529 110

19 0 915 67.611 1299.5 1 0.325855 1529 110

20 0 915 67.611 1299.5 1 0.334273 1529 110

Avg 0 915 67.611 1299.5 1 0.337946 1529 110

247

TABLE.4.36 Comparative results of optimization methods for Resident IEQ.

IEQ-RESIDENCE

Methods

Thermal

sensation

vote

Co2

concentration

Sound

pressure

level

Horizontal

Illumination

IEQ Time Iterations

Genetic

Algorithm 0.00392 678.9818 67.1265 773.53238 1 0.6949 51

Simulated

annealing

ANNEALING

0 915 67 854.5 1 1.1147 2000

PATTERN

SEARCH 0 915 67 854.5 1 0.1331 20

PSO 0.00659 967.495 67.0004 769.92745 1 0.1259 55.35

GODLIKE 0.00001 880.7472 67 896.74398 1 1.5573 4

fmincon 0 901.235 67.0000 884.59 1 5.1069 3208.4

DE

SOLUTION 0.00268 1079.772 68.6362 1605.31387 1 0.004 40

LGO 0.0027 497.0895 67 931.8641 1 0.4675 3223

glcCluster 0.00105 882.688 67.4026 1111.1647 1 0.770 1286.98

glcSolve 0 915 67.611 1299.5 1 0.3379 1529

Fig .4.15.Comparative graph for residence thermal comfort

4.4.4. Comparison of results

The IEQ values for all the ten optimization is 1 and that is the optimum value. The

elapsed time is maximum for DE and minimum for PS. The carbon dioxide concentration,

0.001

0.01

0.1

1

10

100

1000

10000 GA

SA

PS

PSO

GL

Fmincon

DE

LGO

glcCluster

glcSolve

248

sound pressure level and illumination level are more or less the same for all methods. The

thermal sensation is neutral for few methods.

4.4.5. PARAMETERS.

4.4.5.1. THERMAL SENSATION

Thermal comfort is that condition of mind which expresses satisfaction with the

thermal environment. Thermal environment encompasses characteristics of the environment

which affects a person's heat loss. In terms of bodily sensations, thermal comfort is a

sensation of hot, warm, slightly warmer, neutral, slightly cooler, cool and cold

TABLE.4.37. Thermal sensation results in all 10 methods

Trials GA SA PS PSO GODLIKE fmincon DE LGO glcClu glcSol

1 0 0 0 0 -0.0005 0 -0.4 0.0027 0 0

2 -0.0005 0 0 0 0 0 0.2 0.0027 0 0

3 0 0 0 0 0 0 0.3 0.0027 0 0

4 0.0303 0 0 0 -0.0002 0 -0.2 0.0027 0 0

5 0.0007 0 0 -0.0282 0 0 0.3 0.0027 0 0

6 -0.0004 0 0 0 0 0 0.3 0.0027 0 0

7 0.0066 0 0 0 0 0 0.3 0.0027 0 0

8 0.0016 0 0 0.0069 0.0001 0 0.1019 0.0027 0 0

9 0 0 0 0 0 0 0.1427 0.0027 0 0

10 0.0335 0 0 0.0043 -0.0006 0 -0.4 0.0027 0 0

11 0 0 0 0 0 0 -0.2 0.0027 0 0

12 0 0 0 0 0.0013 0 0 0.0027 0 0

13 0 0 0 0 0 0 0.1 0.0027 0 0

14 0.0026 0 0 0 -0.001 0 -0.6 0.0027 0 0

15 0 0 0 0.002 0 0 -0.2 0.0027 0 0

16 0.0007 0 0 0 0 0 -0.1 0.0027 0 0

17 0 0 0 0.0064 0 0 0.209 0.0027 0 0

18 0.0033 0 0 -0.0115 0.0006 0 0.1 0.0027 0 0

19 0 0 0 0 0 0 0 0.0027 0 0

20 0 0 0 0.152 0 0 0.1 0.0027 0 0

avg 0.00392 0 0 0.00659

5

-0.000015 0 0.0026

8

0.0027 0 0

249

FIG.4.16. Graph for Thermal sensation results in all 10 methods

250

4.4.5.2. CARBON DI-OXIDE

Carbon dioxide (CO2) is the chief greenhouse gas that results from human activities

and causes global warming and climate change. Though carbon dioxide is not toxic in itself,

the amount found in the indoor environment is used as an indicator for human comfort.

Elevated levels of carbon dioxide indicate that an insufficient amount of fresh, outdoor air is

being delivered to the occupied areas of the building. This also indicates that other pollutants

in the building may exist at elevated levels since there is not enough fresh air to dilute them.

Since carbon dioxide is an unavoidable, predictable, and easily measured product of human

occupancy, it is used as a marker for other pollutants emanating from humans or other

sources in the building. Carbon dioxide is mostly a threat to health, when the concentration is

high enough to displace the oxygen, which can lead to suffocation in a confined space.

TABLE.4.38. Carbon Dioxide results in all 10 methods

Trials GA SA PS PSO GODLIKE fmincon DE LGO glcClu glcSol

1 744.2 915 915 1076.8 880.7416 915 1295.6 497.0895 915 915

2 803.7882 915 915 1221.6 707.1 915 710.4 497.0895 525 915

3 1045.3 915 915 849 885 915 1173.6 497.0895 655 915

4 648.083 915 915 849 945.5216 915 1088.4 497.0895 1045 915

5 382.3028 915 915 920.0475 922.4489 626.5 696.7 497.0895 1045 915

6 951.0608 915 915 1204.9 1194.2 915 1012.8 497.0895 655 915

7 354.1662 915 915 1111.5 1273.4 915 1457.3 497.0895 1175 915

8 623.0983 915 915 534.5203 833.6192 915 981.332

7

497.0895 525 915

9 1153 915 915 1279.8 1088.7 915 639.900

9

497.0895 655 915

10 863.3774 915 915 827.5092 728.5694 915 1206.5 497.0895 655 915

11 1108.4 915 915 1021.7 768.5 915 802.6 497.0895 915 915

12 609.3 915 915 1085.6 868.9358 915 940.3 497.0895 525 915

13 369.5 915 915 1305 451.4 915 1446 497.0895 655 915

14 330 915 915 843.1 859.5101 915 1427.8 497.0895 1045 915

15 330 915 915 672.6782 946.1 915 1441.2 497.0895 1045 915

16 684.0009 915 915 1384 800.9 915 1396.9 497.0895 655 915

17 347.5 915 915 650.2896 1127.9 928.2 668.697

3

497.0895 1175 915

18 485.9581 915 915 645.3098 753.0979 915 1255.1 497.0895 525 915

19 684.3 915 915 1414.4 1053.7 915 718.4 497.0895 655 915

20 1062.3 915 915 453.1459 525.6 915 1235.9 497.0895 655 915

avg 678.9818 915 915 967.4950

3

880.747225 901.235 1079.77

1545

497.0895 785 915

251

FIG.4.17. Graph for Carbon Dioxide results in all 10 methods

252

4.4.5.3. SOUND PRESSURE LEVEL

Acoustics is the interdisciplinary science that deals with the study of all mechanical

waves in gases, liquids, and solids including vibration, sound, ultrasound and infrasound. The

perception of sound in any organism is limited to a certain range of frequencies. Hearing loss

due to prolonged exposure to noise is well documented. Excessive noise also has an adverse

effect on personal health and wellbeing, ability to perform quiet tasks, and productivity in

general. Because land is becoming scarcer, buildings are being constructed closer together

and closer to noise sources such as highways, railways, and airports. As a result, sound or

acoustic control is becoming increasingly important. The reduction of airborne sound through

a wall is called sound transmission loss (STL).

TABLE.4.39. Sound Pressure level results in all 10 methods

Trials GA SA PS PSO GODLIKE fmincon DE LGO glcClu glcSol

1 67 67 67 67 67 67.0001 67.9 67 67.6111 67.611

2 67.0003 67 67 67 67 67.0001 68.2 67 67.6111 67.611

3 67 67 67 67 67 67.0001 72.7 67 67.6111 67.611

4 67.0001 67 67 67 67 67.0001 67.3 67 67.6111 67.611

5 67 67 67 67.0023 67 67 68.8 67 67.6111 67.611

6 67.0219 67 67 67 67 67.0001 72.6 67 67.6111 67.611

7 67 67 67 67 67 67.0001 67.5 67 67.6111 67.611

8 67.2189 67 67 67.0005 67 67.0001 68.721

4

67 67.6111 67.611

9 67 67 67 67 67 67.0001 67.751

9

67 67.6111 67.611

10 67.0058 67 67 67 67 67.0001 67.4 67 67.6111 67.611

11 68.5 67 67 67 67 67.0001 69 67 67.6111 67.611

12 67.1 67 67 67 67 67.0001 68 67 67.6111 67.611

13 67.3 67 67 67 67 67.0001 67.8 67 67.6111 67.611

14 67 67 67 67 67.00006 67.0001 69.8 67 67.6111 67.611

15 67 67 67 67.0011 67 67.0001 67.9 67 67.6111 67.611

16 67.1748 67 67 67 67 67.0001 67.2 67 67.6111 67.611

17 67.2 67 67 67.0001 67 67 67.752

1

67 67.6111 67.611

18 67.0087 67 67 67.0017 67 67.0001 69.2 67 67.6111 67.611

19 67 67 67 67 67 67.0001 68.9 67 67.6111 67.611

20 67 67 67 67.0029 67 67.0001 68.3 67 67.6111 67.611

avg 67.12653 67 67 67.0004

3

67.000003 67.00009 68.636

27

67 67.6111 67.611

253

FIG.4.18. Graph for Sound Pressure level results in all 10 methods

254

4.4.5.4. HORIZONTAL ILLUMINATION.

Lighting or illumination is the deliberate application of light to achieve some aesthetic

or practical effect. In some design instances, materials used on walls and furniture play a key

role in the lighting effect. Surfaces or floors that are too reflective create unwanted glare.

Specification of illumination requirements is the basic concept of deciding how

much illumination is required for a given task. Clearly, much less light is required to

illuminate a hallway or a bathroom compared to that needed for a word processing work

station. Generally speaking, the energy expended is proportional to the design illumination

level. Beyond the energy factors being considered, it is important not to over-design

illumination, lest adverse health effects such as headache frequency, stress, and

increased blood pressure be induced by the higher lighting levels. In addition, glare or excess

light can decrease worker efficiency.

TABLE.4.40.Horizontal illumination results in all 10 methods.

Trials GA SA PS PSO GODLIKE fmincon DE LGO glcClu glcSol

1 1264.3 854.5 854.5 437.1 510.4552 854.5 364.8 931.86 1299.5 1299.5

2 591.4519 854.5 854.5 425.9 1040.7 854.5 1002.6 931.86 1299.5 1299.5

3 672.3 854.5 854.5 1288.8 1093.7 854.5 1513.6 931.86 1299.5 1299.5

4 508.4575 854.5 854.5 1288.8 356.0738 854.5 1311.3 931.86 1299.5 1299.5

5 988.4221 854.5 854.5 246.3722 561.348 1199.6 1018 931.86 1299.5 1299.5

6 261.205 854.5 854.5 998.7 927.2 854.5 297.3 931.86 1299.5 1299.5

7 893.4416 854.5 854.5 489.9 1101.4 854.5 1485.9 931.86 1299.5 1299.5

8 768.8616 854.5 854.5 344.6397 640.5575 854.5 937.344

8

931.86 1299.5 1299.5

9 187.1 854.5 854.5 1276.6 1021.5 854.5 769.496

6

931.86 1299.5 1299.5

10 707.8476 854.5 854.5 544.1189 800.8187 854.5 1450.3 931.86 1299.5 1299.5

11 1221.7 854.5 854.5 1253.8 1095.1 854.5 1480.5 931.86 1299.5 1299.5

12 1217.3 854.5 854.5 1367.2 575.0751 854.5 1394.7 931.86 409.5 1299.5

13 1059.3 854.5 854.5 871.2 1049 854.5 463.4 931.86 854.5 1299.5

14 187.0001 854.5 854.5 1220.5 948.2773 854.5 898.8 931.86 1299.5 1299.5

15 1475.6 854.5 854.5 372.4912 1248 854.5 560.6 931.86 1299.5 1299.5

16 187.001 854.5 854.5 780 1250.9 854.5 396.2 931.86 1299.5 1299.5

17 1381.9 854.5 854.5 459.1902 1159.2 1111.2 300.435

9

931.86 1299.5 1299.5

18 89.5593 854.5 854.5 763.8825 557.074 854.5 1092.3 931.86 1299.5 1299.5

19 1458.9 854.5 854.5 511 616.5 854.5 1022.2 931.86 409.5 1299.5

20 349 854.5 854.5 458.3543 1382 854.5 1022 931.86 1299.5 1299.5

avg 773.5324 854.5 854.5 769.92745 896.74398 884.59 939.088

865

931.86 1188.2

5

1299.5

255

FIG.4.19. Graph for Horizontal Illumination results in all 10 methods

256

4.4.5.5. IEQ

TABLE.4.41. IEQ results in all 10 methods

Trails GA SA PS PSO GOD

LIKE

fmincon DE LGO glcCluster glcSolve

1 1 1 1 1 1 1 1 1 1 1

2 1 1 1 1 1 1 1 1 1 1

3 1 1 1 1 1 1 1 1 1 1

4 1 1 1 1 1 1 1 1 1 1

5 1 1 1 1 1 1 1 1 1 1

6 1 1 1 1 1 1 1 1 1 1

7 1 1 1 1 1 1 1 1 1 1

8 1 1 1 1 1 1 1 1 1 1

9 1 1 1 1 1 1 1 1 1 1

10 1 1 1 1 1 1 1 1 1 1

11 1 1 1 1 1 1 1 1 1 1

12 1 1 1 1 1 1 1 1 1 1

13 1 1 1 1 1 1 1 1 1 1

14 1 1 1 1 1 1 1 1 1 1

15 1 1 1 1 1 1 1 1 1 1

16 1 1 1 1 1 1 1 1 1 1

17 1 1 1 1 1 1 1 1 1 1

18 1 1 1 1 1 1 1 1 1 1

19 1 1 1 1 1 1 1 1 1 1

20 1 1 1 1 1 1 1 1 1 1

Avg 1 1 1 1 1 1 1 1 1 1

257

FIG.4.20 Graph for IEQ results in all 10 methods

258

4.4.5.6. ELAPSED TIME

CPU time is the time for which the CPU was busy executing the task. It does not take

into account the time spent in waiting for I/O (disk IO or network IO). Since I/O operations,

such as reading files from disk, are performed by the OS, these operations may involve

noticeable amount of time in waiting for I/O subsystems to complete their operations. This

waiting time will be included in the elapsed time, but not CPU time. Hence CPU time is

usually less than the elapsed time.

TABLE.4.42. Elapsed time results in all 10 methods

Trials GA SA PS PSO G-L fminco

n

DE LGO glcClu

ster

glcSol

ve 1 0.3816

63

2.74 0.97 0.12 1.42 4.75 0.0155

65

1.16 0.0271

67

0.31

2 0.3906

58

1.02 0.05 0.12 1.48 3.75 0.0016

93

0.42 0.0271

67

0.36

3 0.4192

8

1.05 0.06 0.13 1.01 6.80 0.0023

47

0.42 0.0271

67

0.33

4 0.3834

61

1.03 0.05 0.13 1.65 4.66 0.0041

92

0.43 0.0271

67

0.33

5 0.4031

58

1.03 0.06 0.12 1.39 4.54 0.0015

68

0.44 0.0271

67

0.35

6 0.3775

46

1.03 0.06 0.10 1.60 5.27 0.0022

4

0.41 0.0271

67

0.38

7 0.3910

53

1.02 0.05 0.14 1.88 5.44 0.0025

25

0.42 0.0271

67

0.34

8 0.3851

92

1.06 0.05 0.14 1.34 4.07 0.0046

35

0.43 0.0271

67

0.33

9 0.4059

51

1.04 0.06 0.14 3.32 4.74 0.0045 0.42 0.0271

67

0.33

10 2.3184

21

1.04 0.04 0.12 1.21 5.40 0.0043

22

0.43 0.0271

67

0.33

11 4.4579

91

1.02 0.07 0.10 2.17 5.53 0.0043

93

0.42 0.0271

94

0.33

12 0.4009

42

1.02 0.05 0.12 1.17 3.90 0.0044

64

0.43 0.0271

94

0.32

13 0.3958

78

1.01 0.09 0.12 1.23 3.91 0.0021

96

0.44 0.0271

94

0.33

14 0.4197

4

1.02 0.05 0.13 1.02 4.54 0.0045

63

0.43 0.0271

94

0.32

15 0.3962

96

1.01 0.06 0.11 2.03 8.01 0.0044

86

0.42 0.0271

94

0.34

16 0.4158

77

1.00 0.05 0.11 1.23 4.79 0.0043

62

0.43 0.0271

94

0.33

17 0.4054

63

1.02 0.04 0.15 1.08 4.99 0.0044

67

0.43 0.0271

94

0.33

18 0.3879 1.01 0.06 0.09 1.95 3.32 0.0042

08

0.42 0.0271

94

0.34

19 0.3828

07

1.01 0.07 0.11 1.55 9.42 0.0023

91

0.43 0.0271

94

0.32

20 0.3805

28

1.02 0.07 0.15 1.32 4.21 0.0023

12

0.42 0.0271

94

0.33

Avg 0.6949

9

1.11 0.13 0.12 1.55 5.10 0.0040

7145

0.46 0.0271

8

0.33

259

FIG.4.21. graph for Elapsed time results in all 10 methods

260

4.4.5.7. ITERATIONS

Iteration is a computational procedure in which a cycle of operations is repeated, often

to approximate the desired result more closely. Iteration means the act of repeating a process

usually with the aim of approaching a desired goal or target or result. Iteration in computing

is the repetition of a process within a computer program. It may also refer to the process of

iterating a function i.e. applying a function repeatedly, using the output from one iteration as

the input to the next. Another use of iteration in mathematics is in iterative methods which

are used to produce approximate numerical solutions to certain mathematical problems.

TABLE.4.43 . Iterations results in all 10 methods

Trials GA SA PS PSO GODLIKE fmincon DE LGO glcCluster glcSolve

1 51 2000 20 70 4 2972 40 3223 1516 1529

2 51 2000 20 56 4 2694 40 3223 1516 1529

3 51 2000 20 52 4 4607 40 3223 1516 1529

4 51 2000 20 52 4 3188 40 3223 1516 1529

5 51 2000 20 51 4 2728 40 3223 1516 1529

6 51 2000 20 51 4 3777 40 3223 1516 1529

7 51 2000 20 54 4 3034 40 3223 1516 1529

8 51 2000 20 64 4 2881 40 3223 1516 1529

9 51 2000 20 68 4 3214 40 3223 1516 1529

10 51 2000 20 52 4 4080 40 3223 1516 1529

11 51 2000 20 51 4 3780 40 3223 1516 1529

12 51 2000 20 57 4 2722 40 3223 1516 1529

13 51 2000 20 60 4 2953 40 3223 1516 1529

14 51 2000 20 52 4 2728 40 3223 1516 1529

15 51 2000 20 51 4 3505 40 3223 1516 1529

16 51 2000 20 52 4 2929 40 3223 1516 1529

17 51 2000 20 58 4 3108 40 3223 1516 1529

18 51 2000 20 54 4 2562 40 3223 1516 1529

19 51 2000 20 51 4 3923 40 3223 1516 1529

20 51 2000 20 51 4 2783 40 3223 1516 1529

avg 51 2000 20 55.35 4 3208 40 3223 1516 1529

261

FIG.4.22. Graph for Iterations results in all 10 methods

262

TABLE.4.44 Comparative table for parameters in all 10 methods

Variables GA SA PS PSO GL fminc

on

DE LGO Glc

Cluster

Glc

Solve

PMV X

0

0 X X

0 X X

0

0

CO2 X

915

915 X X X X

497

785

915

Sound X

67

67 X

67 X X

67

67.6

67.6

Illumina

tion

X

854.5

854.5 X X X X

931.8 X

1299.5

IEQ

1

1

1

1

1

1

1

1

1

1

TIME 0.13 0.12 0.027

ITERS X 20

X 4

X

- Represents the parameters are consistent for all the 20 trials and the corresponding

parameter values are given in the respective cell.

X - Represents the parameters are not consistent for all the 20 trials

In case of iterations and elapsed time only the two or three minimum values alone are given.

4.4.6. Result and Discussion

With the two extreme values of parameters from survey, the optimization is carried

out with different solvers. As they are of stochastic type their results may vary from trial to

trial so and the problem is made to run for 20 trials (Elbeltagi, Tarek Hegazy, & & Grierson,

2005) and an average of all trials is taken as the final value of the parameter, by the solver.

The solvers are compared with different criteria

263

1. Consistency

The consistency table gives the parameters that remain constant for all the

trails. All the solvers give the same value of IEQ for all the runs. Which in

turn indicate that the quality requirements are in the acceptable range.

Thermal – PS (0),NL (0),SA(0), glcSolve (0), glcCluster (0), LGO (0.0027)

CO 2 - PS(915), SA(915), glcSolve (915), LGO (497)

Sound - PS(67),SA(67), glcSolve (67.611), glcCluster (67.611), LGO (67)

Illumination - PS(854.5),SA(854.5), glcSolve (1299.5), LGO (931.86)

So we see that the solvers SA, Pattern Search, glcSolve, glcCluster& LGO

remain constant throughout their runs.

2. Minimum Run Time

For minimum run time of the problem we got PSO (0.12 seconds), Pattern

Search (0.13 seconds).

3. Minimum Evaluation

This criterion will determine the effectiveness of the algorithm. From the

result table we see that the Pattern Search and GODLIKE algorithms have

minimum evaluation of 20 and 4 respectively

4. Simplicity of Algorithm

Of all the algorithms we have taken the Pattern Search algorithm is the most

simplest followed by GA, PSO, DE, Simulated Annealing, GODLIKE, Non-

Linear, Direct algorithm.

5. Results according to Standards

This is the most important criterion that determines whether the solver is

practical or not. We got the standard values from ASHRAE, IES, Guidance for

employers on the Control of Noise at Work Regulations 2005 as:

Thermal comfort: -3 to 3

Carbon dioxide: less than 1000ppm

264

Sound level: 40 dBA to 70dBA

Illumination level: 800 lux to 1200 lux

With the above standards the solvers which adhere to the standard are:

Thermal comfort: GA, SA, PS, PSO, FMINCON, DE, GL,LGO,

glcCluster, glcSolve

Carbon dioxide: GA, SA, PS, PSO, FMINCON,GL, LGO, glcCluster,

glcSolve

Sound level: GA, SA, PS, PSO, FMINCON, DE, GL, LGO,

glcCluster, glcSolve

Illumination level: SA, PS, FMINCON, GL, LGO, glcCluster,

The following table gives a summary of all the criteria for the solvers:

Table.4.45. Summary of all the criteria for the solvers

Criteria GA SA PS PSO Fmincon DE GL LGO glcClus glcSolve

Result

according

to

ASHRAE

¾

=75%

4/4

=100%

4/4

=100%

¾

=75%

4/4

=100%

2/4

=50%

4/4

=100%

4/4

=100%

4/4

=100%

3/4

=75%

Consistency - - - - - -

Min-Run

Time - - - - - - - -

Min-

Evaluation - - - - - - - -

Simple

Algorithm - - - - - - - - -

Thus it is seen that the Pattern Search solver satisfies all the criteria and scores 100%

for its practicality in giving result according to ASHRAE, IES, Guidance for

employers on the Control of Noise at Work Regulations 2005, So the appropriate

algorithm, for optimization of thermal comfort is suggested as Direct search

algorithm & the solver is PATTERN SEARCH

265

4.4.7. CONCLUSION

Overall indoor environmental quality (IEQ) in terms of an occupant‘s acceptance has

not been considered in many residential buildings neither it is optimized. In this study, the

overall IEQ of residential apartments which are naturally ventilated at the Karunya University

quarters in Coimbatore was evaluated by 102 occupants in four aspects, namely thermal

comfort, indoor air quality , equivalent noise level and illumination level. All the offices

considered are naturally ventilated buildings. The results showed that the operative

temperature, carbon dioxide concentration, equivalent noise level and illumination level had

important effects on the overall IEQ acceptance. Empirical expressions were proposed to

approximate the occupant acceptance. The values are optimized using ten different non-

traditional optimization techniques.

Here, ten non-traditional optimization algorithms were presented. These include: GA,

SA, PS, PSO, GL, FMINCON, EA, LGO, glcCluster, glcSolve. A brief description of each

method is presented along with a pseudo code to facilitate their implementation. MATLab

programs were written to implement each algorithm. The IEQ problem for the Residential

buildings of the Karunya University was solved using all algorithms, and the comparative

results were presented.

Bibliography ANSI/ASHRAE55-2004. (2004). Thermal Environmental conditions for Human occupancy. Atlanda,

USA: American Society of Heating, Refrigerating and Air-Conditioning Engineers.

ANSI/ASHRAEstandard62-2007. (2007). Design for acceptable indoor air quality,. Atlanta: American

Society of Heating, Refrigerating and AIr-conditioning Engineers, Inc.,.

ASHRAE. (1989). Handbook-fundamentals, chapter 8,, Physiological Principles, Comfort and Health.

ASTM. (2003). litStandard guide for using indoor carbon dioxide concentrations to evaluate indoor air

quality and ventilation. D6245-98.

Auliciems.A. (1983). Psychophysical criteria for global thermal zones of building design. International

Journal of Biometerology , 69 -83.

Auliciems.A. (1984). Thermobile controls for human comfort. Heating and ventlating Engineeers ,

April/May 31-33.

Ayr, u., Cirillo, e., Fato, I., & Martellotta, F. (2003). A new approach to assessing the performance of

noise indices in buildings. Applied Acoustics , 64(2), 129-145.

266

Bean, A., & Bell, R. (1992). the CSP index : A practical measure of office jighting quality as perceived

by the office workers. Lighting Research and Technolgy , 24(2), 214-225.

Brager,G.S, & Dedear,R.J. (1998). Thermal adaptation in the building environment: a literature

review. Energy and Buildings , 27(1) 83 -96.

Bulysssen, P., & Cox, C. (2002). Indoor environmental quality and upgrading of European office

buildings. Energy and Buildings , 34(2),155-162.

Clausen, G., & Wyon, D. (2008). The combined effects of many different indoor environmental

factors on acceptability and office work performance. HVAC&R Research , 14(1), 103-113.

Clausen, G., Carrick, L., Fanger, P., Kim, S., Poulson, T., & Rindel, J. (1993). A comparative study of

discomft caused by indoor air pollution, thermal load and noise. Indoor Air , 3, 255-262.

Dan, J. (1993). Total Environmental Quality. In ASHRAE Transaction (pp. 960-967).

Elbeltagi, E., Tarek Hegazy, 1., & & Grierson, D. (2005). . Comparison among five evolutionary-based

optimization algorithms. . Advanced Engineering Informatics , 19, 43-53.

Fanger, P. (1970). Thermal comfort: Analysis and applications in Environmental engineering,. New

York: McGraw-Hill.

Fanger, P. (1995). Comments on 'a comparison of the predicted and reported thermal sensation vote

in homes during winter and summer'. Energy and Buildings , 22(1) 89.

Fanger, P., Olf, & Decipol. (1988). OLf and decipol: new units for perceived air qualtiy,. Building

Services engineering Research and Technolgy , 9, 155-157.

Fanger.P.O., T. (2002). Extension of the PMV model to non-air-conditioned buildings in warm

climates. Energy and Buildings , 34(6), 533-536.

Federspiel, C. (1882). Used-adaptable and minimum-power thermal comfort Control. Ph.D

Thesiis,MIT,Department of Mechanicla Engineering.

Fountain, m., Brager, G., Arens, E., Bauman, F., & Benton, C. (1994). Comfort Control for short-term

occupancy. Energy and Buildings , 211-213.

Gan, G., & Croome, D. (1994). Thermal comfort models based on Field measurements. ASHRAE

Transactions , 782-704.

Goldman, R. (1999). Extrapolating ASHRAE's comfort model. HVAC&R Research , 5(3),189-194.

Haghighat, F., & Donnini, G. (1999). Impace of psycho-social factors on perception of the indoor air

environment studies in 12 office buildings. Buildinh and Environment , 34, 479-503.

Han, J., Yang, W., Zhou, J., Zhang, G., Zhang, Q., & Moschandreas, D. (2009). A comparative analysis

of Urban and Rural residential thermla comfort under natural ventilation environment. Energy and

Buildings , 41(2), 139-145.

267

Hikmat, H., Hind, M., & Muna, H. (2009). Evaluating indoor environmental quality of public school

buildings in Jordan. Indoor and Built Environment , 18(1),66-76.

Houser, K., & Tiller, D. (2003). Measuring and Subjective response to interior lighting; paired

comparisons and semantic differential scaling. Lighting Research Technology , 35(3), 183-198.

Hui, P., Wong, L., & Mui, K. (2008). Using Carbon dioxide concentration to assess indoor air quality in

offices. Indoor Built Environment , 17(3), 213-219.

Hui, P., Wong, L., & Mui, L. (2006). Feasibility study of an express assesment protocol for the indoor

air quality of air-conditioned offices. Indoor and Built Environment , 15(4), 373 -378.

InternationalStandardISO7730. (1994). Moderate thermal environments- Determinatiob of the PMV

and PPD indices and specification of the conditions for thermal comfort.

ISO7730, B. (1995). Moderate thermla environments, Determination of PMV and PPF indices and

specification of the conditions for thermal comfort.

living-smartly.com. (n.d.). Retrieved September 09, 2009, from http://living-

smartly.com/2010/01/ac-and-ventilation-for-india-%E2%80%93-white-paper/

MacArthur, J. (1986). Humidity and predicted-mean-vote-based PMV-Based Comfort control.

ASHRAE Transactions , 15-17.

Mendell, M. (2003). Indices for IEQ and building related symposiums. Indoor Air , 13(4), 364-368.

Mui, K., & Wong, L. (2006). A method of assessing the acceptability of noise levels in air conditioned

offices. Building Serviecs Engineering Research and Technology , 27(3), 249-243.

Mui, K., & Wong, L. (2006). Acceptable illumination levels for office occupants. Architectural Science

review , 49(2), 116-119.

Mui, K., & Wong, L. (2006). Minimum Acceptable noise level for office occupants. Building Servies

Engineering Research and Technology , 27(3);249-254.

Mui, K., Wong, L., & Wong, Y. (2009). Acceptable noise levels for construction site offices. Building

Servies Engineering Research and Technology , 30, 87-94.

Mui, K., Wong, L., HUI, P., & Law, K. (2008). Epistemic evalution of policy influence on workplace

indoor air qualityof Hong Kong in 1996-2005. Building services Engineering Research and Technology

, 29(2),157-164.

Mui,K.M, & Wong,L.T. (2007). Neutral Temperature in subtropical climates- a filed survey in air-

conditioned offices. Building and environment , 42(2) 699 -706.

Mui.K.W, & Wong,L.T. (2007). Evaluation of neutral criterion of indoor air quality for iar conditioned

offices in subtropical climates. Building Services Engineering Research and Technology , 28(1), 23-33.

Naganoa, K., & Horikoshib, T. (2005). New comfort index during combined conditions of moderate

low ambient temperature and traffic noise . Energy and Buildings , 37, 287- 294.

268

Nicol, J., & Humphreys.A'. (2001). Adaptive thermal comfort and sustainable thermal standards for

buildings. conference Moving thermal comfort Standards into the 21st century, Windor,UK , 5 -8.

Olesen, & Bjarne. (2009). Productivity and Indoor Air qualtiy. Lyngby, Denmark: International centre

for Indoor Environment and Energy.

Oseland.N.A. (1995). Predicted and reported thermal sensation in climate chambers, offices and

homes. Energy and Buildings , 23(2) 105-115.

Pellerin, N., & Candas, V. (2004). Effects of steady-state noise and temperature conditions on

environmental perception and acceptability. Indoor Air , 14(2), 129-136.

Persily, A. (1997). Evaluating Building IAQ and Ventilation wiht indoor carbon dioxide. ASHRAE

Transactions , 103(2), 193=203.

Portney, L., & Watkins, M. (2000). Foundations of Clinical Research - Applications to Practise, second

ed.,. NJ: Prentice Hall Health.

Sherman, M. (1985). A Simplified Model of Thermal Comfort. Energy and Buildings , 8, 37-50.

Sofuoglu, S., & Moschandreas, D. (2003). The link between symptoms of office building occupants

and in-office air pollution:the indoor air pollution index. Indoor Air , 13(4), 332-343.

Toftem, J. (2002). Human response to combined indoor environment exposures. Energy and

Buildings , 34(6),601-606.

Viollin, S. (2003). Two examples of audio-visual interactions in an urban context. Proceedings of the

5th European Conference on noise control, (p. 73). Euronoise.

Wallace, L., Nelson, C., & Dunteman, G. (1991). Workplace characteristics associated wiht health and

comfort concerns in three office buildings in Washington,DC,Healthy Buildings 1991. American

Society of Heating, Refrigerating and Air-conditioning Engineers , 36-39.

Wang, Z. (2005). A field study of the thermal comfort in residential buildings in Harbin. Building and

Environment , 41(8(, 1034 - 1039.

WilliamFisk, & OlliSeppanen. (2007). Providing better indoor Environmental qualtiy brings economic

benefits. Proceedings of Clima 2007 well being indoors. Helsinki University of Technology, Espoo,

Finland.

Wong, L., & Leung, L. (2005). Minimum fire alarm sound pressure level for elder centers. Building

and Environment , 40(1), 125-133.

Wong, L., Mui, K., & Hui, P. (2008). A multivariate-logistic model of acceptance for indoor

environmental quality(IEQ) in offices,. Building and Environment , 43(1),1-6.

Wong, L., Mui, K., & Hui, P. (2006). A statistical model for characterizing common air Pollutants in

air-conditioned offices. Atmospheric Environment , 40(23),4246-4257.

269

Wong,L.T, Mui,K.W, Fong,N.K, & Hui,P.S. (2007). Bayesian Adaptive Comfort Temperature of air-

condioning system in subtropical climate. Building and Environment , 42(5) 1983-1988.

Yang, W., & Kang, J. (2005). Acoustic comfort evaluation in urban open public spaces. Applied

Acoustics , 66(2), 211-229.

Yoshino, H., Yoshino,Y, Zhang,Q, Mochida,A, Li,N, LI,Z, et al. (2006). Indoor thermal environmental

and energy saving for urban residential buildings in China. Energy and Buildings , 38 (11) 1308 -1319.