Power Converters and Control of Renewable Energy Systems Paper

7/29/2019 Power Converters and Control of Renewable Energy Systems Paper

1/20

1

Abstract The global electrical energy consumption is steadily

rising and consequently there is a demand to increase thepower generation capacity. A significant percentage of therequired capacity increase can be based on renewable energy

sources. Wind turbine technology, as the most cost effectiverenewable energy conversion system, will play an importantpart in our future energy supply. But other sources like

microturbines, photovoltaics and fuel cell systems may also beserious contributors to the power supply. Characteristically,power electronics will be an efficient and important interface

to the grid and this paper will first briefly discuss threedifferent alternative/renewable energy sources. Next, variousconfigurations of the wind turbine technology are presented, as

this technology seems to be most developed and cost-effective.

Finally, the developments and requirements from the grid arediscussed.

I. INTRODUCTION

The energy comsumption is steadily increasing and thederegulation of electricity has caused that the amount ofinstalled production capacity of classical high power stationscannot follow the demand. A method to fill out the gap is tomake incentives to invest in alternative energy sources likewind turbines, photovoltaic systems, microturbines and alsofuel cell systems. The wind turbine technology is one of themost promising alternative energy technology [1]-[3]. Themodern development started in the 1980s with sites of a

few tens of kW to Multi-MW range wind turbines today.E.g. Denmark has a high penetration (> 20%) of windenergy in major areas of the country and in 2003 15% of thewhole electrical energy consumption was covered by windenergy. A higher penetration level will even be seen in thenear future. The technology used in the early developedwind turbines was based on a squirrel-cage inductiongenerator connected directly to the grid. That almost directlytransfers the wind power pulsations to the electrical grid.Furthermore, there was no fast control of the active andreactive power, which typically are the key parameters tocontrol the frequency and the voltage of the grid. As thepower range of the wind turbines increases those parametersbecome more and more important. The power electronics is

the key-technology to change the basic characteristic of thewind turbine from being an energy source to be an activepower source. Such possibilities are also used to interfaceother renewable energy sources [4]-[8].

This paper will first explain the basic principles of windpower conversion, fuel cells and photovoltaics. Next, thetrend in power electronics is outlined. Diffe-rent windturbine configurations are reviewed, as they are the mostpromising alternative energy technologies today.

Finally, a general discussion about interface issues ofrenewable energy sources is done.

II. RENEWABLE ENERGY SOURCES

Three different renewable energy sources are brieflydescribed. They are wind power, fuel cell and photovoltaic.

A. Wind power conversion

The function of a wind turbine is to convert the linearmotion of the wind into rotational energy that can be used todrive a generator, as illustrated in Fig. 1. Wind turbinescapture the power from the wind by means ofaerodynamically designed blades and convert it into rotatingmechanical power. At present, the most popular windturbine is the Horizontal Axis Wind Turbine (HAWTs)where the number of blades is typically three.

Wind turbine blades use airfoils to develop mechanicalpower. The cross-sections of wind turbine blades have theshape of airfoils as the one shown in Fig. 2.

Airflow over an airfoil produces a distribution of forcesalong the airfoil surface. The resultant of all these pressureand friction forces is usually resolved into two forces and amoment, lift force, drag force and pitching moment, asshown in Fig. 2.

The aerodynamic power,P, of a wind turbine is given by:

pCvRP32

2

1= (1)

whereis the air density, R is the turbine radius, v is thewind speed and CP is the turbine power coefficient whichrepresents the power conversion efficiency of a wind turbine.CP is a function of the tip-speed ratio (), as well as theblade pitch angle () in a pitch controlled wind turbine. isdefined as the ratio of the tip speed of the turbine blades towind speed, and given by:

v

R = (2)

where is the rotational speed of the wind turbine.The Betz limit, CP,max (theoretical) =16/27, is the maximumtheoretically possible rotor power coefficient. In practicethree effects lead to a decrease in the maximum achievablepower coefficient [1]:

Rotation of the wake behind the rotor Finite number of blades and associated tip losses Non-zero aerodynamic drag

Frede Blaabjerg, Remus Teodorescu, Zhe ChenAalborg University, Institute of Energy Technology,

[email protected], [email protected], [email protected]

Power Converters and Control of Renewable Energy Systems

Marco LiserrePolitecnico di Bari, CEMD research group

[email protected]


2/20

2

A typical CP-curve for a fixed pitch angleis shown inFig. 3. It can be seen that there is a practical maximumpower coefficient, CP,max. Normally, a variable speed windturbine follows the CP,max to capture the maximum power upto the rated speed by varying the rotor speed to keep thesystem at the optimum tip-speed ratio, opt.

As the blade tip-speed typically should be lower than half

the speed of sound the rotational speed will decrease as theradius of the blade increases. For MW wind turbines therotational speed will be 10-15 rpm. A common way toconvert the low-speed, high-torque power to electricalpower is to use a gear-box and a normal speed generator asillustrated in Fig. 1. The gear-box is optional as multipolegenerator systems are alternative solutions.

Fig. 2. A simple airfoil used in wind turbines.

Fig. 3. Typical Cp- curve for a wind turbine for a fixed angle .

The development in the wind turbine systems has beensteady for the last 25 years and four to five generations ofwind turbines exist. It is now a proven technology.

It is important to be able to control and limit the power athigher wind speeds, as the power in the wind is a cube of thewind speed.

Wind turbines have to be cut out at a high wind speed toavoid damage. A turbine could be designed in such a waythat it converts as much power as possible in all windspeeds, but then it would have to be too heavy. The highcosts of such a design would not be compensated by theextra production at high winds, since such winds are rare.Therefore, turbines usually reach maximum power at amuch lower wind speed, the rated wind speed (9-12 m/s).

The power limitation may be done by one of theaerodynamic mechanisms: stall control (the blade position isfixed but stall of the wind appears along the blade at higherwind speed), active stall (the blade angle is adjusted in orderto create stall along the blades) or pitch control (the bladesare turned out of the wind at higher wind speed).

B. Fuel Cell power conversion

The fuel cell is a chemical device, which produceselectricity directly without any intermediate stage and hasrecently received much attention [7]. The most significantadvantages are low emission of green house gases and highpower density. For example, a zero emission can beachieved with hydrogen fuel. The emission consists of onlyharmless gases and water. The noise emission is also low.The energy density of a typical fuel cell is 200 Wh/l, whichis nearly ten times of a battery. Various fuel cells areavailable for industrial use or currently being investigatedfor use in industry, including

Proton Exchange Membrane Solid Oxide Molten Carbonate Phosphoric Acid Aqueous Alkaline

The efficiency of the fuel cell is quite high (40%-60%). Alsothe waste heat generated by the fuel cell can usually be usedfor cogeneration such as steam, air-conditioning, hot air andheating, then the overall efficiency of such a system can beas high as 80%.

Angle of attack:

Trailing edge

wind

Leading edge

Lift force

Drag force

Pitching moment

Pitch angle:

Fig. 1. Conversion from wind power to electrical power in a wind turbine [11].

Power conversion &power con trol

Wind power

Power converter(optional)

Power conversion &power con trol

Power conversionPower transmission Power transmission

Supply grid

Consumer

Rotor Gearbox (optional) Generator

Electrical Power

Fig. 1. Conversion from wind power to electrical power in a wind turbine [11].


3/20

3

Fig. 4. V-I characteristics of a fuel cell [12].

A typical curve of the cell electrical voltage against currentdensity is shown in Fig. 4. It can be seen that there exists aregion where the voltage drop is linearly related with thecurrent density due to the Ohmic contact.

Beyond this region the change in output voltage variesrapidly. At very high current density, the voltage dropssignificantly because of the gas exchange efficiency. At lowcurrent level, the Ohmic loss becomes less significant, theincrease in output voltage is mainly due to the activity of thechemicals. Although the voltage of a fuel cell is usuallysmall, with a theoretical maximum being around 1.2 V, fuelcells may be connected in parallel and/or in series to obtainthe required power and voltage.

The power conditioning systems, including inverters andDC/DC converters, are often required in order to supplynormal customer load demand or send electricity into thegrid.

C. The photovoltaic cellPhotovoltaic (PV) power supplied to the utility grid isgaining more and more visibility due to many nationalincentives [7]. With a continuous reduction in system cost(PV modules, DC/AC inverters, cables, fittings and man-power), the PV technology has the potential to become oneof the main renewable energy sources for the futureelectricity supply.

The PV cell is an all-electrical device, which produceselectrical power when exposed to sunlight and connected toa suitable load. Without any moving parts inside the PVmodule, the tear-and-wear is very low. Thus, lifetimes ofmore than 25 years for modules are easily reached.However, the power generation capability may be reduced to75% ~ 80% of nominal value due to ageing. A typical PVmodule is made up around 36 or 72 cells connected in series,encapsulated in a structure made of e.g. aluminum andtedlar. An electrical model of the PV cell is depicted in Fig.5.

iSC

iPV

id uPV

(a)

IPV

PPV

pMPP

UPVuOC

iSC

(uMPP

, iMPP

)

(b)

Fig. 5. Model and characteristics of a PhotoVoltaic (PV) cell.(a) Electrical model with current and voltages defined.(b) Electrical characteristic of the PV cell, exposed to a given amount

of sunlight at a given temperature.

Several types of proven PV technologies exist, where thecrystalline (PV module light-to-electricity efficiency: =10% - 15%) and multi-crystalline ( = 9% - 12%) silicon

cells are based on standard microelectronic manufacturingprocesses. Other types are: thin-film amorphous silicon ( =10%), thin-film copper indium diselenide ( = 12%), andthin-film cadmium telluride ( = 9%). Novel technologiessuch as the thin-layer silicon ( = 8%) and the dye-sensitisednano-structured materials ( = 9%) are in their earlydevelopment. The reason to maintain a high level ofresearch and development within these technologies is todecrease the cost of the PV-cells, perhaps on the expense ofa somewhat lower efficiency. This is mainly due to the factthat cells based on todays microelectronic processes arerather costly, when compared to other renewable energysources.

The series connection of the cells benefits from a high

voltage (around 25 V ~ 45 V) across the terminals, but theweakest cell determines the current seen at the terminals.


4/20

4

0 0.1 0.2 0.3 0.4 0.5 0.6 0.70

2

4

6

Cell voltage [V]

Cellcurrent[A]

15oC

40oC

75oC

0 0.1 0.2 0.3 0.4 0.5 0.6 0.70

0.5

1

1.5

2

2.5

Cell voltage [V]

Cellpower[W]

15 oC

40 oC

75 oC

1000 W/m2

600 W/m2

200 W/m2

(a)

(b) Fig. 6. Characteristics of a PV cell. Model based on the British PetroleumBP5170 crystalline silicon PV module. Power at standard test condition(1000 W/m2 irradiation, and a cell temperature of 25 C): 170 W @ 36.0 V.Legend: solid at 15 oC, dotted at 40 oC, and dashdot at 75 oC.

The series connection of the cells benefits from a highvoltage (around 25 V ~ 45 V) across the terminals, but theweakest cell determines the current seen at the terminals.This causes reduction in the available power, which to someextent can be mitigated by the use of bypass diodes, inparallel with the cells. The parallel connection of the cellssolves the weakest-link problem, but the voltage seen atthe terminals is rather low. Typical curves of a PV cellcurrent-voltage and power-voltage characteristics are plottedin Fig. 6a and Fig. 6b respectively, with insolation and celltempera-ture as parameters. The graph reveals that the

captured power is determined by the loading conditions(terminal voltage and current). This leads to a few basicrequirements for the power electronics used to interface thePV module(s) to the utility grid.

The job for the power electronics in renewable energysystems is to convert the energy from one stage into anotherstage to the grid (alternative voltage) with the highestpossible efficiency, the lowest cost and to keep a superiorperformance. The basic interfacing is shown in Fig. 10.

Fig. 10. Power electronic system with the grid, load/source, powerconverter and control.

This power electronic system can be used with many loadsand generators.

III. SINGLE-PHASE PV-INVERTERS

The general block diagram for single-phase grid connectedphotovoltaic systems is presented in Fig. 1a. It consists ofPV array, PV inverter, controller and grid.

a)

b) c) d)

Fig 1. General schema for single-phase grid connected photovoltaicsystems. a) Block diagram; b) Central inverter; c) String inverter; d)

Module integrated inverter

The PV array can be a single panel, a string of PV panelsor a multitude of parallel strings of PV panels. Centralizedor decentralized PV systems can be used as depicted in theFig. 1,b,c,d .

Central invertersIn this topology the PV plant (> 10 kW) is arranged in

many parallel strings that are connected to a single centralinverter on the DC-side (Fig. 1b). These inverters arecharacterized by high efficiency and the lowest specific cost.However, the energy yield of the PV plant decreases due tomodule mismatching and partial shading conditions. Also,the reliability of the plant is limited due to the dependence

of power generation on a single component: the failure ofthe central inverter results in the whole PV plant being outof operation.

String inverter

Similar to the central inverter, the PV plant in this concept isdivided into several parallel strings. Each of the PV stringsis assigned to a designated inverter, the so-called "stringinverter"(Fig. 1c). String inverters have the capability ofseparate MPP tracking of each PV string. This increases the

Power converter

Reference (local/centralized)

Control

Power flow

Load /

generator

Loads

ApplianceIndustryCommunication

Generators

WindPhoto-voltaicFuel cellOther sources

2-3 2-3

PVArray

PV Inverter+ LP Filter Grid

Control


5/20

5

energy yield via the reduction of mismatching and partialshading losses. These superior technical characteristics leadto a reduction in the system cost, increase the energy yieldand enhance the supply reliability. String inverters haveevolved as a standard in PV system technology for gridconnected PV plants.

An evolution of the string technology applicable for higherpower levels is the multi-string inverter [1]. It allows theconnection of several strings with separate MPP trackingsystems (via DC/DC converter) to a common DC/ACinverter. Accordingly, a compact and cost-effective solutionwhich combines the advantages of central and stringtechnologies is achieved. This multi-string topology allowsthe integration of PV strings of different technologies and ofvarious orientations (south, west and east). Thesecharacteristics allow time shifted solar power whichoptimizes the operation efficiencies of each stringseparately. The application area of the multi-string invertercovers PV plants of 3-10 kW.

Module integrated inverterThis system uses one inverter is used for each module (Fig1d). This topology optimizes the adaptability of the inverterto the PV characteristics, since each module has its ownMPP tracker. Although the module integrated inverteroptimizes the energy yield, it has a lower efficiency than thestring inverter. Module integrated inverters are characterizedby more extended AC-side cabling, since each module ofthe PV plant has to be connected to the available AC grid(e.g. 230 V/ 50 Hz). Also, the maintenance processes arequite complicated, especially for facade-integrated PVsystems. This concept can be implemented for PV plants ofabout 50- 400 W peak.

PV inverterThe PV inverter technology has evolved quite a lot duringthe last years towards maturity [2]. Still there are differentpower configurations possible as shown in the Fig. 2.

PV

Inverters

with DC-DC

converter

without DC-DC

converter

with isolation

without isolation

on the LF side

on the HF side

with isolation

without isolation

Fig. 2: Power configuration for PV inverters

The question of having or not a dc-dc converter is first ofall related to the PV string configuration. Having morepanels in series and lower grid voltage, like in US andJapan, it is possible to avoid the boost function. Thus asingle stage PV inverter can be used leading to higherefficiency.

The issue of isolation is mainly related to safety standardsand is for the moment only required in US. The drawback ofhaving so many panels in series is that MPPT is harder to

achieve especially during partial shading, as demonstrated in[3]. In the following, the different PV inverters powerconfigurations are described in more detail.

PV inverters with DC-DC converter with isolationThe isolation is typically acquired using a transformer that

can be placed on either the grid frequency side (LF) asshown in the Fig. 3a or on the high-frequency (HF) side inthe dc-dc converter as shown in the Fig. 3b. The HFtransformer leads to more compact solutions but high careshould be taken in the transformer design in order to keepthe losses low.

DC

ACGrid

PVArray

DC

DC

(a)

DC

AC

GridPVArray

DC

AC

AC

DC

(b)

Fig. 3. PV inverter system with DC-DC converter and isolation transformera) on the LF side b) n the HF side

In the Fig.4 is presented a PV inverter with HFtransformer using an isolated push-pull boost converter [4]

Fig. 4. PV inverter with HF transformer in the dc-dc converter

Also, the dc-ac inverter in this solution is a low costinverter switched at the line frequency. The new solutionson the market are using PWM dc-ac inverters with IGBTswitched typically at 10-20 kHz leading to better powerquality performances.

Other solutions for high frequency dc-dc converters withisolations includes: full-bridge isolated converter, single-inductor push-pull converter (SIC) and double-inductorconverter (DIC) as depicted in Fig. 5 [5]


6/20

6

L

Fig. 5 dc-dc converter topologies with isolation. a) full-bridge; b) single-

inductor pushh-pull; c) double-inductor push-pull

In order to keep the magnetic components compact highswitching frequencies in the range of 20 100 kHz areemployed.The full-bridge converter is usually utilized at power levelsabove 750W. The advantages of this topology are: goodtransformer utilization bipolar magnetization of the core,good performance with current programmed control reduced DC magnetization of transformer. The maindisadvantages in comparison with push-pull topology are thehigher active part count and the higher transformer rationeeded for boosting the dc voltage to the grid level.

The single inductor push-pull converter can provideboosting function on both the boosting inductor and

transformer, reducing thus the transformer ratio. Thus higherefficiency can be achieved together with smoother inputcurrent. On the negative side it can be mentioned that highervoltage blocking switches are required and the transformerwith tap point puts some construction and reliabilityproblems.

Those shortcomings can be alleviated using the doubleinductor push-pull converter (DIC) where the boost inductorhas been split in two. Actually this topology is equivalentwith two interleaved boost converters leading to lowerripple in the input current. The transformer construction ismore simple not requiring tap point. The single disadvantageof this topology remains the need for an extra inductor.

PV inverters with DC-DC converter without isolationIn some countries as the grid-isolation is not mandatory,

more simplified PV inverter design can be used, as shown inFig. 6

DC

DC

DC

ACGrid

PVArray

(a)

(b)

Fig 6. PV inverter system with DC-DC converter without isolationtransformer

a) General diagramb) Practical example with boost converter and full-bridge inverter [4]

In Fig. 6b a practical example [4] using a simple boost

converter is shown.Another novel transformerless topology [4] featuring ahigh efficiency time-sharing dual mode single-phasepartially controlled sinewave PWM inverter composed ofquasi time-sharing sinewave boost chopper with a newfunctional bypass diode Db in the boost chopper side andcomplementary sinewave PWM full-bridge inverter (Fig. 8).

a)

b)

Fig.8. Time-sharing dual-mode sinewave modulatedsingle-phase inverter with boost chopper [6]

a) Circuit system configuration. b) Operating principle


7/20

7

PV inverters without DC-DC converterThe block diagram of this topology is shown in the Fig.

9a.

DC

AC

GridPV

Array

(a)

(b)

Fig. 9. PV inverter system without DC-DC converter

and with isolation transformera) general diagram b) practical example with full-bridge inverter and grid-side transformer [4]

In Fig.9b are presented two topologies of PV inverterswhere the line frequency transformer is used. For higherpower levels, self-commutated inverters using thyristors arestill being used on the market [4]

PV inverters without DC-DC converter without isolationThe block diagram of this topology is shown in the

Fig.10a

DC

AC

GridPV

Array

(a)

(b)

(c )

Fig. 10. Transformerless PV inverter system without DC-DC convertera) general diagram b) typical example with full-bridge inverter [4]

c) multilevel [7]

In Fig. 10b, a typical transformerless topology is shownusing PWM IGBT inverters. This topology can be usedwhen there are available a large nr. of solar panelsconnected in series producing in exces of the grid voltagepeak at all times.

Another interesting PV inverter topololgy without boost

and isolation can be achieved using multilevel concept. Gridconnected photovoltaic systems with a five level cascadedinverter is presented in Fig 10c [7]. The redundant inverterstates of the five level cascaded inverter allow for a cyclicswitching scheme which minimizes the switching frequency,equalizes stress evenly on all switches and minimizes thevoltage ripple on the DC capacitors.

IV. CONTROL OF SINGLE-PHASE PV-INVERTERS

Control DC-DC boost converterIn order to control the output dc-voltage to the desired value,a control system is needed which can automatically adjustthe duty cycle, regardless of the load current or input

changes. There two types of control for the dc-dcconverters: the direct duty-cycle control and the currentcontrol[8]. As shown in the Fig. 11.

CompensatorPulse-width

modulator

Converter

Sensor gain

vref

vFC

(t)

iload

(t)

d(t)+

-

vDC(t)Errorsignal

Controlsignal

Referenceinput

(a)

CompensatorComparator and

controller

Converter

Sensor gain

vref

vFC

(t)

iload

(t)

d(t)+

-

vDC

(t)

Errorsignal

Controlsignal

Referenceinput

iswitch

(t)

iswitch

(t)iswitch_ref

(t)

(b)

Fig. 11. Control strategies for switched dc-dc convertersa) direct duty-cycle control b) current control

Duty-Cycle controlThe output voltage is measured and then compared to the

reference. The error signal is used as input in thecompensator, which will compute from it the duty-cyclereference for the pulse-width modulator.

Current ControlThe converter output is controlled by choice of the

transistor peak current. The control signal is a current and asimple control network switches on and off the transistorsuch its peak current follows the control input. The currentcontrol, in the case of an isolated boost push-pull converterhas some advantages against the duty-cycle control likesimpler dynamics (removes one pole from the control-tooutput transfer function) . Also as it uses a current sensor it


8/20

8

can provide better protection of the switch by limiting thecurrent to acceptable levels.

Another issue is the transformer saturation. In thetransformer it can be induced a dc bias current generated bysmall voltage imbalances due to the small differences inboost inductors and/or switches. This dc current bias

increases or decreases the transistor currents. The currentcontrol will alter the switch duty cycles in a way that theseimbalances tend to disappear and the transformer volt-second balance to be maintained. Finally, the currentcontrol is better suited to modularity where current sharingneeds to be solved when running in parallel.

Among the drawbacks of the current control it can bementioned that it required an extra current sensor and it hasa susceptibility to noise and thus light filtering of feedbcaksignals is required. The current control become unstablewhenever the duty-cycle becomes larger than 0.5 but thisdrawback can be overcome by ramping the reference currentsignal.

Considering the above-enumerated arguments, the current

control seems to be more attractive for PV inverterapplications and it is widely used.

Control of DC-AC grid converterFor the grid-connected PV inverters in the range of 1-5

kW, the most common control structure for the dc-ac gridconverter is using a current-controlled H-bridge PWMinverter having a low-pass output filter. Typically L filtersare used bu the new trend is to use LCL filters that being ahigher order filter (3rd) leads to more compact design. Thedrawback is that due to its own resonance frequency it canproduce stability problems and special control design isrequired [9]. A typical dc-ac grid converter with LCL filteris depicted in the Fig. 12

Fig.12. The H-bridge PV inverter connected to the gridthrough an LCL filter

The harmonics level in the grid current is still acontroversial issue for PV inverters. The IEEE 929 standard

from 2000 allows a limit of 5% for the current totalharmonic distortion (THD) factor with individual limits of4% for each odd harmonic from 3rd to 9th and 2% for 11thto 15th while a recent draft of European IEC61727 suggestssomething similar. These levels are far more stringent thanother domestic appliances such as IEC61000-3-2 as PVsystems are viewed as generation sources and so are subjectto higher standards than load systems.

Classical PI control with grid voltage feed-forward[10],[11] as depicted in Fig. 13a is commonly used forcurrent-controlled PV inverters, but this solution exhibits

two well known drawbacks: inability of the PI controller totrack a sinusoidal reference without steady-state error andpoor disturbance rejection capability. This is due to the poorperformance of the integral action.

ii*

iiGPI(s)

Gd(s) Gf(s)

iiui*

ug

(a)

ii*

ii

Gc(s)

Gh(s)

Gd(s) Gf(s)iiui

*

(b)

Fig. 13.The current loop of PV inverter.a) with PI controller; b) with P+Resonant (PR) controller

The PI current controllerGPI(s) is defined as:

( ) IPI PK

G s Ks

= + (1)

In order to get a good dynamic response, a grid voltagefeed-forward is used, as depicted in Fig. 15a. This leads inturn to stability problems related to the delay introduced inthe system by the voltage feedback filter.

In order to alleviate these problems, a second ordergeneralized integrator (GI) as reported in [12] can be used.The GI is a double integrator that achieves an infinite gain at

a certain frequency, also called resonance frequency, andalmost no attenuation exists outside this frequency. Thus, itcan be used as a notch filter in order to compensate theharmonics in a very selective way. This technique has beenprimarily used in three-phase active filter applications asreported in [12] and also in [13] where closed-loopharmonic control is introduced. Another approach reportedin [14] where a new type of stationary-frame regulatorscalled P+Resonant (PR) is introduced and applied to three-phase PWM inverter control. In this approach the PI dc-compensator is transformed into an equivalent ac-compensator, so that it has the same frequency responsecharacteristics in the bandwidth of concern. The current loopof the PV inverter with PR controller is depicted in the Fig.

13b.The P+Resonant (PR) current controller Gc(s) is definedas [12], [15]:

2 2( )c P I

o

sG s K K

s = +

+

The harmonic compensator (HC) Gh(s) as defined in [10]:


9/20

9

( )22

3,5,7

( )h Ih

h o

sG s K

s h==

+

is designed to compensate the selected harmonics 3rd, 5th and7th as they are the most prominent harmonics in the currentspectrum . A processing delay typical equal to T

sfor the

PWM inverters [8] is introduced in ( )dG s . The filter transfer

function Gf(s) is expressed in (4) [16].

( )

( )

2 2

2 2

( ) 1( )

( )

LCif

i i res

s zi sG s

u s L s s

+= =

+

where 12LC g fz L C

=

and ( )2

2 i g LC

res

i

L L z

L

+ =

The current error - disturbance ratio rejection capability atnull reference is defined as:

( )* 0

( )( )

( ) 1 ( ) ( ) ( ) ( )i

f

g c c d fi

G ss

u s G s G s G s G s

=

=+ +

where: is current error and the grid voltage ug isconsidered as the disturbance for the system.

The Bode plots of disturbance rejection for the PI and PRcontrollers are shown in Fig 14. As it can be observed, ThePR provides much higher attenuation for both fundamentaland lower harmonics then PI. The PI rejection capability at5th and 7th harmonic is comparable with that one of a simple

proportional (P) controller, the integral action being

irrelevant.

-150

-100

-50

0

101

102

103

-540

-450

-360

-270

PR+HC

PI

P

Fig. 14. Bode plot of disturbance rejection (current error ratio disturbance)of the PR+HC, P and PR current controllers.

Thus it is demonstrated the superiority of the PR controllerrespect to the PI in terms of harmonic current rejection. In[15] the discrete implementation on a low-cost fixed-pointDSP is demonstrated. In Fig. 15 some experimental resultswith a 3kW PV inverter are shown demonstrating theharmonic compensation.

0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04-25

-20

-15

-10

-5

0

5

10

15

20

25

time[sec]

Ig (exp) [5A/div]

Ug (exp) [100/div]

(a)

0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04-25

-20

-15

-10

-5

0

5

10

15

20

25

time[sec]

Ig (exp) [5A/div]

Ug (exp) [100/div]

(b)

0 0. 005 0.01 0. 015 0. 02 0. 025 0. 03 0. 035 0. 04-25

-20

-15

-10

-5

0

5

10

15

20

25

time[sec]

Ig (exp) [5A/div]

Ug (exp) [100/div]

(c)

Fig.15. Experimental results at 3kW. Grid voltage and current. a) with PIcontroller. B) with PR; c) with PR+HC

The issue of stability when several PV inverters arerunning in parallel on the same grid is becoming more andmore important especially when LCL filters are used. In[17] it is shown that in the case of a concentration of severalhundreds of solar roofs in Holland, resonance frequencies in

the range of 1-2 kHz are occurring as a resulat of the gridinteraction with the LCL filters. Thus, special attention isrequired when designing the current control. A method fordesigning both the controller and LCL filter ensuringstability is shown in [9].

MPPT

In order to capture the maximum power, a maximumpower point tracker (MPPT) is required. The maximumpower point of solar panels is a function of solar irradiance


10/20

10

and temperature as depicted in Fig..16. This function canbe implemented either in the dc-dc converter or in the dc-acconverter. Several algorithms can be used in order toimplement the MPPT as followings.

(a)

(b)Fig. 16: PV characteristics.

a) Irradiance dependence b) temperature dependence

Perturb and Observe

The most commonly used MPPT algorithm is Perturb andObserve (P&O), due to its ease of implementation in itsbasic form [17]. Figure 16 shows the P vs. V and I curves ofa PV array, which has a global maximum at the MPP. Thus,if the operating voltage of the PV array is perturbed in agiven direction and dP/dV > 0, it is known that the

perturbation moved the array's operating point toward theMPP. The P&O algorithm would then continue to perturbthe PV array voltage in the same direction. If dP/dV < 0,then the change in operating point moved the PV array awayfrom the MPP, and the P&O algorithm reverses the directionof the perturbation. A problem with P&O is that it oscillatesaround the MPP in steady state operation. It also can track inthe wrong direction, away from the MPP, under rapidly

increasing or decreasing irradiance levels. There are severalvariations of the basic P&O that have been designed tominimize these drawbacks. These include using an averageof several samples of the array power and dynamicallyadjusting the magnitude of the perturbation of the PVoperating point.

Incremental Conductance

The incremental conductance algorithm seeks to overcomethe limitations of the P&O algorithm by using the PV array'sincremental conductance to compute the sign of dP/dV

without a perturbation [17]. It does this using an expressionderived from the condition that, at the MPP, dP/dV = 0.Beginning with this condition, it is possible to show that, atthe MPP dI/dV = -I/V. Thus, incremental conductance candetermine that the MPPT has reached the MPP and stop

perturbing the operating point. If this condition is not met,

the direction in which the MPPT operating point must beperturbed can be calculated using the relationship betweendI/dV and -I/V. This relationship is derived from the factthat dP/dV is negative when the MPPT is to the right of theMPP and positive when it is to the left of the MPP. Thisalgorithm has advantages over perturb and observe in that itcan determine when the MPPT has reached the MPP, where

perturb and observe oscillates around the MPP. Also,incremental conductance can track rapidly increasing anddecreasing irradiance conditions with higher accuracy than

perturb and observe. One disadvantage of this algorithm isthe increased complexity when compared to perturb andobserve. This increases computational time, and slows downthe sampling frequency of the array voltage and current.

Parasitic Capacitance

The parasitic capacitance method is a refinement oftheincremental conductance method that takes into accountthe parasitic capacitances of the solar cells in the PV array[17]. Parasitic capacitance uses the switching ripple of theMPPT to perturb the array. To account for the parasiticcapacitance, the average ripple in the array power andvoltage, generated by the switching frequency, are measuredusing a series of filters and multipliers and then used tocalculate the array conductance. The incrementalconductance algorithm is then used to determine thedirection to move the operating point of the MPPT. Onedisadvantage of this algorithm is that the parasitic

capacitance in each module is very small, and will onlycome into play in large PV arrays where several modulestrings are connected in parallel. Also, the DC-DC converterhas a sizable input capacitor used filter out small ripple inthe array power. This capacitor may mask the overall effectsof the parasitic capacitance of the PV array.

Constant Voltage

This algorithm makes use of the fact that the MPP voltagechanges only slightly with varying irradiances, as depictedin Fig. 16a. The ratio of VMP/VOC depends on the solarcell parameters, but a commonly used value is 76% [17]. Inthis algorithm, the MPPT momentarily sets the PV arraycurrent to zero to allow a measurement of the array's opencircuit voltage. The array's operating voltage is then set to76% of this measured value. This operating point ismaintained for a set amount of time, and then the cycle isrepeated. A problem with this algorithm is available energyis wasted when the load is disconnected from the PV array,also the MPP is not always located at 76% of the arraysopen circuit voltage.

Anti-islanding

In addition to the typical power quality regulationsconcerning the harmonic distortion and EMI limits, the grid-


11/20

11

HarmonicInjection

Grid

Anti-aliasing Sample&Hol A / D Windowing Pre-processing DFT

Post-processing

Anti-aliasing Sample&Hol A / D Windowing Pre-processing DFT

Voltage Signa

Current Signal

U

Z=

I

Impedancer pp ng

&Display

Internal Logic

Harmonic Current

Harmonic Voltage

connected PV inverters must also meet specific powergeneration requirements like the islanding detection, or evencertain country-specific technical recommendations forinstance the grid impedance change detection (in Germany).Such extra-requirements contribute to a safer grid-operationespecially when the equipment is connected in dispersed

power generating networks but impose additional effort toreadapt the existing equipments.The European standard EN50330-1 (draft) [19] describes

the ENS (the German abbreviation of Mains monitoringunits with allocated Switching Devices) requirement, settingthe utility fail-safe protective interface for the PVconverters. The goal is to isolate the supply within 5 secondsafter an impedance change of Z = 0.5 W, which is associatedwith a grid failure. The main impedance is typically detected

by means of tracking and step change evaluation at thefundamental frequency. Therefore, a method of measuringthe grid impedance value and its changes should beimplemented into existing PV-inverters.

One solution is to attach a separate device developed only

for the measuring purpose as depicted in Fig. 17a.

(a)

(b)Fig. 1. Grid-impedance measurement for PV inverters. a) using external

device; b) embedded on the inberter control using harmonic injection

This add-on option is being commonly used in thecommercial PV inverters, but the new trend is to implementthis function embedded in the inverter control without extrahardware. Numerous publications exist in this field, whichoffer measuring solutions for the grid impedance for a widefrequency range from dc up to typically 1 kHz [20].Unfortunately, not always can these methods easily beembedded into a non-dedicated platform, i.e. PV-inverters

featuring typically a low-cost DSP. Specific limitations like

real-time computation, A/D conversion accuracy and fixed-point numerical limitation, are typically occurring.

A novel approach presented in [21], [22] estimates the gridimpedance on-line with the purpose of detection the stepchange of 0.5 as required in [19] as shown in Fig. 17b..The solution is found by injecting a test signal trough the

inverter modulation process. This signal, an interharmoniccurrent with a frequency close to the fundamental,determines a voltage drop due to the grid impedance, whichis measured by the existing PV-inverter sensors. Then, thesame CPU unit that makes the control algorithm carries outthe calculations and gives the grid impedance value. The

principle of this method is shown in Fig. 18.This approach provides a fast and low cost solution to meet

the required standards and was succesfully implemented ona TMS320F24x 16-bit fixed point DSP platform as an add-on to the existing control.

V. CONTROL OF THREE-PHASE INVERTERS

The control of a three-phase inverter connected to the gridhas more in common with the control of an activerectifier/filter rather than with the control of an ac drive. Infact with the first the distributed inverter shares thecharacteristic to be connected to the grid on the ac side,while with second it shares the common characteristic tohave less responsibilities in the management of the dc-linkvoltage that is usually controlled by another converter stage.Hence from the control perspective the three-phasedistributed inverter as an advantage over the rectifier and adisadvantage over the motor inverter.

Its control issues will be discussed starting from itsmathematical model both with L-filter and LCL-filter on thegrid side.

Then simple controls as well as few advanced ones will beintroduced and briefly discussed. Finally some advancedtopics and some experimental results will close this Section.

Mathematical Model of the L-filter inverter

The state of the three-phase inverter is modelled by meansof a switching space-vector defined with the switching

functions )(tpj (j = a, b, c)

( )2a b c2p(t) p (t) p (t) p (t)3= + +

then if the inverter is connected to the grid through an L-filter (Fig. 1)

Fig. 18. Implementation structure of grid impedance estimation


12/20

12

d ( )( ) ( ) ( )

d

i tv t e t Ri t L

t= + + (2)

o1v(t) p(t)v (t)2

=

Fig. 1 L-filter inverter connected to the grid.

assuming to neglect the dc voltage dynamic such as the dcvoltage vo(t) is ain input for the system. Moreover )(tv is

the space-vector of the inverter input voltages; )(ti is the

space-vector of the inverter input currents; )(te is the

space-vector of the input line voltages.The mathematical model written in the state space form is

d ( ) 1 1( ) ( ) ( ) ( )

d 2o

i tRi t e t p t v t

t L

= +

(4)

A commonly used approach in analysing three-phasesystems is to adopt a dq-frame that rotates at the angularspeed (where = 2fandfis the fundamental frequencyof the power grids voltage waveform). The space-vectorswhich express the inverter electrical quantities are projectedon the d-axis and q-axis. As a consequence if a space-vectorwith constant magnitude rotates at the same speed of theframe, it has constant d- and q- components while if rotatesat a different speed or it has a time-variable magnitude it has

pulsating components. Thus in a dq-frame rotating at the

angular speed (2) becomes

( )( ) ( ) ( ) ( ) ( )

( )( ) ( ) ( ) ( ) ( )

1 1

2

1 1

2

dq d d d o

q

d q q q o

di ti t Ri t e t p t v t

dt L

di ti t Ri t e t p t v t

dt L

= +

+ = +

(5)

(5) shows how in the dq-frame the d- and q- differentialequations for the current are dependent due to the cross-coupling terms iq(t) and id(t).

Mathematical Model of the LCL-filter inverter

In the following the LCL-filter based inverter model isreported in order to highlight the increased complexity ofthe system. The system is shown in Fig. 2.

Fig. 2 LCL-filter inverter connected to the grid.

AC Current controlThe ac current control (CC) is usually adopted because the

current controlled converter exhibits, in general, bettersafety, better stability and faster response [1].

This solution ensures several additional advantages. The

feedback loop also results in some limitations, such as thatfast-response voltage modulation techniques must beemployed, like PWM. Optimal techniques, which use

precalculated switching patterns within the ac period, cannotbe used, as they are not oriented to ensure current waveformcontrol [1].

Generally the current control is the most inner loop of acascade control that employ a dc-link voltage levelmanagement system and active and reactive powercontroller as reported in Fig. 3

The use of an ac LCL-filter claims for a deep dynamic andstability analysis of the current control loop [2]. In order tohighlight the stability problems that arise from the use of anLCL-filter it is sufficient to show the d or q system plant in

Laplace domain. If the converter side current is sensed, thesystem plant is

( )

( )

2 2

2 22

( ) 1( )

( )

LC

res

s zi sG s

v s L s s

+= =

+(7)

If the grid side current is sensed, the plant for control is

( )

2

2 22

( ) 1( )

( )LC

res

zi sG s

v s L s s = =

+(8)

where12

1LC fz L C

= and ( )2 2

1 2 2res LC L L z L = + .

1

1 1

11 11 1

1 1

2 2

22 2

2 2

2

2 2

10 0 0

10 0 0

1 10 0 0

1 10 0 0

10 0 0

10 0 0

= +

f f

f f

d d

q q

C d C d f f

C q C q

f fd d

q q

R

L L

Ri iL Li i

v vC Cd

v vdt

C Ci i

Ri i

L L

R

L L

1

1

2

2

1 0 000 0

1 0 00

0 0

0 0 10

0 0

0 0 10

0 0

+

d d

q q

L

e vL

e v

L

L

(6)


13/20

13

In both the cases the two poles related to the resonance ofthe LCL-filter challenge the current control instability,

particularly the second one (sensing of the grid current)generally lead to a more stable behaviour.

Two axis-based current control

The most used control technique is the two axis-based [1].Then if the two-axis system is a stationary -frame, theproportional plus resonant controller can be adopted and it is

2 20

2 20

0

( )

0

ip

PRi

p

K sK

sD s

K sK

s

+

+ =

+ +

(9)

If the frame is a rotating dq-frame, classical PI can be used

0( )

0

ip

PI dqi

p

KK

sD s

K

K s

+

=

+

(10)

If this controller is transformed into an -frame then

02 2 2 2

0 0

02 2 2 2

0 0

( )

i ip

PIi i

p

K s KK

s sD s

K K sK

s s

+

+ + =

+ + +

(11)

(11) it is equal to (9) except for non-diagonal terms. Hencethe PI controller in dq-frame and PR controller in -framecan achieve similar performances.

Fig. 1. The block diagram of a typical three-phase distributed inverter.

In the case ofdq-frame, if it is oriented such as the d-axisis aligned on the grid voltage vector the control is calledVoltage Oriented Control (VOC) (Fig 4). The referencecurrent d-component i*d is controlled to manage the active

power flow while the reference current q-component i*q iscontrolled to mange the reactive power flow. To have thegrid current vector in phase with the grid voltage vector, i*qshould be zero.Grid voltage harmonic compensators

The grid voltage is usually affected by a backgrounddistortion that can result in a high harmonic distortion of the

grid current. This problem can be solved both in a stationary-frame both in a rotating dq-frame. In the first case it issufficient to plug in other resonant controller also calledharmonic compensators

( )

22

3,5,7 0

( )R ihh

sG s k

s h

=

=

+

(12)

where h is the order of the harmonic to be compensated.If the controller adopts a rotating dq-frame approach it is

possible to introduce other dq-frame rotating at multiplespeed respect to the fundamental one and adopting standardPI in each of them. In both the cases it is necessary that theharmonics to be compensated stay into the bandwidth of thecurrent controller otherwise stability problems arise.

Current control active damping

Fig. 4. Voltage Oriented Control based on the use of a rotating dq-frame.

This solution seems very attractive especially in applicationsabove several kW, where the use of a damping resistorincreases the encumbrances, the losses could claim forforced cooling and the efficiency decrement becomes a key

point. In [3] a lead-lag network has been used on the filtercapacitor voltage and it is possible to avoid the use of newsensors because this voltage is near to the grid which isnormally sensed. Moreover, in [4] an interesting approach to

perform active damping has been proposed: a virtual resistoris added. The virtual resistor is an additional controlalgorithm that makes the LCL-filter behaving as if there wasa real resistor connected to it. However, an additionalcurrent sensor is needed if the virtual resistor is connected inseries to the filter inductor or capacitor and an additionalvoltage sensor is needed, if it is connected in parallel.Basically all these approaches are multiloop-based [5] whilean alternative solution consists in adopting a more complexcontroller acting as a digital filter around the resonancefrequency of the LCL-filter [2].

Direct power control

In the last years the most interesting emerging technique hasbeen the direct power control developed in analogy to thewell known direct torque control used for drives. In DPCthere are no internal current loops and no PWM modulator

block because the converter switching states are


14/20

14

Some tests results, obtained on the set-up shown in Fig. 6,are reported in order to evaluate the impact of the non-idealconditions on the behaviour of a PR-based controller in a-frame (Fig. 7), the use of harmonic compensator in astationary -frame to mitigate these effects (Fig. 8) andfinally the effect of active damping (Fig. 9).

appropriately selected by a switching table based on theinstantaneous errors between the commanded and estimatedvalues of active and reactive power [1], [6] Fig 5. The mainadvantage of the DPC is in its simple algorithm instead themain disadvantage is in the need for high samplingfrequency required to obtain satisfactory performances.

Reduction of the number of sensors

The basic number of needed sensors is 4 (two ac currentsand two ac voltages). However this number can be reducedavoiding the use of grid voltage vector implementing avirtual sensor or using a zero crossing detector in order tohave the phase reference for the current in order to haveunity power factor. Moreover if a feedforward currentcontrol technique is adopted the grid current sensors can beavoided but it is essential to provide a method forovercurrent protection in industrial applications.

Fig. 5. Direct Power Control based on the active and reactive powercalculation.

In [8] an algorithm to estimate position of line voltage ispresented. The proportional-plus-integral current regulator ismodified to obtain the angle error signal driving anobserver, similar in structure to a phase-locked loop, which

provides the angle of line voltages.

Non-ideal conditions

The non-ideal conditions are many and they can affectvery much the overall system performance. They are toolong computation time, presence of acquisition filters, ac

phase unbalance, location of the grid voltage sensors after adominant reactance and passive damping if an LCL-filter isused. A proper design to take into consideration them should

be provided [9].It is well known that the grid unbalance causes even

harmonics at the dc output and odd harmonics in the inputcurrent [10]. Some solutions have been studied such as theuse of negative sequence in the reference current thatunfortunately leads to uncontrollability of the power factoror the use of two current controllers for positive andnegative sequences, which also can create stability

problems.

EMC-issues

The main EMC-issues are related to the low frequencyrange and thus to the correct control to the current. Thus theuse of a LCL-filter on the ac side is an interesting solution:reduced values of the inductance can be used and the gridcurrent is almost ripple free. The design of the LCL-filter

has been investigated [11].

Future research topics

Some intriguing topics of research are:

Results

Fig. 6. Laboratory set-up The block diagram of a typical three-phase

distributed inverter.

the immunity of the inverter to the presence ofpolluting loads connected to the same PCC

compliance with international standards/need for newstandards respect to the harmonics due to theswitching;

to reduce the number of components; whether use or not the Phase Locked Loop; whether to use grid voltage feedforward or not; grid current sensor position


15/20

15

Fig. 7. Compensation of grid background distortion: grid currents [2 A/div]and grid voltage [100 V/div] (sampling/switching 10 kHz, active power 2

kW, PR-controllers in a -frame).

Fig. 8. Compensation of grid background distortion: grid currents [2 A/div]and grid voltage [100 V/div] (sampling/switching 10 kHz, active power 2

kW, PR-controllers in a -frame with 5th and 7th harmonic compensators).

Fig. 9. Control change from active damping to no damping (t=40 ms): gridcurrents [2 A/div] (sampling/switching 10 kHz, active power 2 kW, PR-

controllers in a -frame).

VI. CONVERTERTOPOLOGIES FORWIND TURBINES

In a fixed speed wind power conversion system, thepower may be limited aerodynamically either by stall, activestall or by pitch control [6], [7]. Normally induction

generators are used in fixed speed systems, which are almostindependent of torque variation and operate at a fixed speed(slip variation of 1-2%). Fig. 11 shows different topologiesfor the first category of wind turbines.

All three systems are using a soft-starter (not shown inFig. 11) in order to reduce the inrush current and thereby

limit flicker problems on the grid. They also need a reactivepower compensator to reduce (almost eliminate) the reactivepower demand from the turbine generators to the grid.

It is usually done by continuously switching capacitorbanks following the production variation (5-25 steps). Thosesolutions are attractive due to cost and reliability but theyare not able (within a few ms) to control the active powervery fast. The generators have typically a pole-shift

possibility in order to maximize the energy capture.The next category is variable speed systems [6]-[35] where

pitch control is typically used. Variable speed wind turbinesmay be further divided into two parts, one with partiallyrated power electronic converters and one with fully rated

power electronic converters.

Gear

Induction

generator

Pitch

Grid

Reactive

compensator

I

(a)

Gear

Induction

generator

Stall

Grid

Reactive

compensator

II

(b)

(c)

Fig. 11. Wind turbine systems without power converter but withaerodynamic power control.

Pitch controlled (System I) b) Stall controlled (System II) c) Active stallcontrolled (System III)

Gear

Induction

generator

Active

Stall

Grid

Reactive

compensator

III


16/20

16

(a)

(b)

Fig. 12. Wind turbine topologies with partially rated power electronics andlimited speed range, (a) Rotor-resistance converter (System IV) (b) Doubly-

fed induction generator (System V).

Fig. 12 shows wind turbines with partially rated powerelectronic converters that are used to obtain an improvedcontrol performance. Fig. 12a shows a wind turbine systemwhere the generator is an induction generator with awounded rotor. An extra resistance is added in the rotor,which can be controlled by power electronics. This is adynamic slip controller and it gives typically a speed rangeof 2-10 %. The power converter for the rotor resistancecontrol is for low voltage but high currents. At the sametime an extra control freedom is obtained at higher windspeeds in order to keep the output power fixed. This solutionstill needs a soft-starter and a reactive power compensator.

A second solution of using a medium scale powerconverter with a wounded rotor induction generator isshown in Fig. 12b [18]-[26]. Slip-rings are making theelectrical connection to the rotor. A power convertercontrols the rotor currents. If the generator is running super-synchronously electrical power is delivered through both therotor and the stator. If the generator is running sub-synchronously electrical power is only delivered into therotor from the grid. A speed variation of 30 % aroundsynchronous speed can be obtained by the use of a powerconverter of 30 % of nominal power.

Furthermore, it is possible to control both active (Pref) andreactive power (Qref), which gives a better grid performance,and the power electronics enable the wind turbine to act

more as a dynamic power source to the grid. The solutionshown in Fig. 12b needs neither a soft-starter nor a reactive

power compensator. The solution is naturally a little bitmore expensive compared to the classical solutions shownin Fig. 11 and Fig. 12a. However, it is possible to savemoney on the safety margin of gear, reactive powercompensation units and it is possible to capture more energyfrom the wind.

The wind turbines with a full-scale power converterbetween the generator and grid give extra losses in thepower conversion but it may be gained by the addedtechnical performance [9]. Fig. 13 shows four possiblesolutions with full-scale power converters.

The solutions shown in Fig. 13a and Fig. 13b are

characterized by having a gear. A synchronous generatorsolution shown in Fig. 13b needs a small power converterfor field excitation. Multi-pole systems with thesynchronous generator without a gear are shown in Fig. 13cand Fig. 13d.

The last solution uses permanent magnets, which are stillbecoming cheaper and thereby more attractive. All foursolutions have the same controllable characteristics since thegenerator is decoupled from the grid by a dc-link. The

power converter to the grid enables the system very fast tocontrol active and reactive power. However, the negativeside is a more complex system with a more sensitiveelectronic part.

By introducing power electronics many of the wind

turbine systems get a performance like a power plant. Inrespect to control performance they are faster but of coursethe produced real power depends on the available wind. Thereactive power can in some solutions be delivered withouthaving any wind.

(a)

(b)

(c)

Gear

Wounded Rotor

Induction

generator

Pitch

Grid

Reactive

compensator

IV

Resistance

control

with PE

Gear

Doubly-fed

induction generator

Pitch

Grid

V

DC

AC

AC

DC

Pref

Qref

Gear

Inductiongenerator

Pitch

GridDC

AC

AC

DC

Pref

Qref

VI

Pi tch

V II

G e a r

S y n c h r o n o u sGenerator

GridD C

A C

A C

D C

Pref

Qre f

D C

A C

Grid

Pref Qref

Synchronous

Generator

Multi-polePitch

DC

AC

AC

DC

VIII

DC

AC


17/20

17

PM-synchronous

Generator

Multi-pole

Pitch

GridDC

AC

AC

DC

Pref

Qref

IX

(d)Fig. 13. Wind turbine systems with full-scale power converters.a) Induction generator with gear (System VI)b) Synchronous generator with gear (System VII)c) Multi-pole synchronous generator (System VIII)d) Multi-pole permanent magnet synchronous generator (System IX)

Fig. 13 also indicates other important issues for windturbines in order to act as a real power source for the grid.They are able to be active when a fault appears at the gridand so as to build the grid voltage up again quickly; thesystems have the possibility to lower the power productioneven though more power is available in the wind andthereby acting as a rolling capacity. Finally, some are able tooperate in island operation in the case of a grid collapse.

VII. CONVERTERTOPOLOGIES FORWIND TURBINES

Controlling a wind turbine involves both fast and slowcontrol. Overall the power has to be controlled by meansof the aerodynamic system and has to react based on a set-

point given by dispatched center or locally with the goal tomaximize the production based on the available wind power.

The power control system should also be able to limit thepower. An example of an overall control scheme of a windturbine with a doubly-fed generator system is shown in Fig.10.

Below maximum power production the wind turbine willtypically vary the speed proportional with the wind speed

and keep the pitch angle fixed. At very low wind the speedof the turbine will be fixed at the maximum allowable slip inorder not to have overvoltage.

A pitch angle controller will limit the power when theturbine reaches nominal power. The generated electrical

power is done by controlling the doubly-fed generatorthrough the rotor-side converter. The control of the grid-sideconverter is simply just keeping the dc-link voltage fixed.Internal current loops in both converters are used whichtypically are linear PI-controllers, as it is illustrated in Fig.11a. The power converters to the grid-side and the rotor-sideare voltage source inverters.

Another solution for the electrical power control is to usethe multi-pole synchronous generator. A passive rectifier

and a boost converter are used in order to boost the voltageat low speed. The system is industrially used today. It ispossible to control the active power from the generator. Thetopology is shown in Fig. 11b. A grid inverter is interfacingthe dc-link to the grid. Here it is also possible to control thereactive power to the grid. Common for both systems arethey are able to control reactive and active power very fastand thereby the turbine can take part in the power systemcontrol.

FIG control

ower controller Speed controller

Wind turbine control

otor side

converter controller

Grid side


Measurementgrid point M

C

C C

C

meas

gen

PWM PWM

N

T

refconv

gridP

,

refconv

gridQ,

meas

dcU

meas

gridP

meas

gridP

meas

gridQ

meas

ac

ref

dcU

refrated

gridP,

cross-couplingGrid

operatorscontrolsystem

meas

rotorI

FIG control

ower controllerower controller Speed controllerSpeed controller

Wind turbine control

otor side


otor side


Grid side


Measurementgrid point M

C

C C

C

meas

gen

PWM

N

T

refconv

gridP

,

refconv

gridQ,

meas

dcU

meas

gridP

meas

gridP

meas

gridQ

ac

ref

dcU

refrated

gridP,

cross-couplingGrid

operatorscontrolsystem

meas

rotorI

Fig. 10. Control of wind turbine with doubly-fed induction generator system [35 ].


18/20


19/20

19

Generators and Power Electronics for Wind Turbines, 2001, Ris-R-1205 (EN).

[11] M.P. Kazmierkowski, R. Krishnan, F. Blaabjerg. Control in PowerElectronics Selected problems. Academic Press, 2002.

[12] Z. Chen, E. Spooner, Wind turbine power converters: a comparativestudy,Proc. of Power Electronics and Variable Speed Drives, 1998,

pp. 471 476.[13] L.H. Hansen, P.H. Madsen, F. Blaabjerg, H.C. Christensen, U.

Lindhard, K. Eskildsen, Generators and power electronicstechnology for wind turbines, Proc. of IECON '01, Vol. 3, 2001, pp.2000 2005.

[14] J. Rodriguez, L. Moran, A. Gonzalez, C. Silva, High voltagemultilevel converter with regeneration capability, Proc. of PESC99, 1999, Vol.2, pp.1077-1082.

[15] E.N. Hinrichsen, Controls for variable pitch wind turbinegenerators, IEEE Trans. On Power Apparatus and Systems, Vol.103, No. 4, 1984, pp. 886-892.

[16] T. A. Lipo, Variable Speed Generator Technology Options for WindTurbine Generators, NASA.Workshop on HAWTT Technology,May 1984, pp. 214-220.

[17] Z. Chen, E. Spooner, Grid Interface Options for Variable-Speed,Permanent-Magnet Generators, IEE Proc. -Electr. PowerApplications, Vol. 145, No. 4, July 1998, pp. 273-283.

[18] J.D. van Wyk, J.H.R. Enslin, A Study of Wind Power Converterwith Microcomputer Based Maximal Power Control Utilising anOversynchronous Electronic Schertives Cascade, Proc. of IPEC '83,Vol. I, 1983, pp. 766-777.

[19] D. Arsudis, W. Vollstedt, Sensorless Power control of a Double-FedAC-Machine with nearly Sinusoidal Line Currents, Proc. of EPE '89,Aachen 1989, pp. 899-904

[20] D. Arsudis, Doppeltgespeister Drehstromgenerator mitSpannungszwischenkreis Umrichter in Rotorkreis fr WindKraftanlagen, Ph.D. Thesis, 1998, T.U. Braunschweig, Germany.

[21] D. Arsudis, Sensorlose Regelung einer doppelt-gespeistenAsynchronmaschine mit geringen Netzrckwirkungen, Archiv frElektrotechnik, Vol. 74, 1990, pp. 89-97.

[22] S. Bhowmik, R. Spee, J.H.R. Enslin, Performance optimization fordoubly fed wind power generation systems, IEEE Transactions onIndustry Applications, Vol. 35, No. 4 , July-Aug. 1999, pp. 949 958.

[23] J.B. Ekanayake, L. Holdsworth, W. XueGuang, N. Jenkins, Dynamicmodelling of doubly fed induction generator wind turbines, Trans. onPower Systems, Vol. 18, No. 2, May 2003, pp. 803- 809.

[24] Yamamoto, O. Motoyoshi, Active and Reactive Power control for

Doubly-Fed Wound Rotor Induction Generator, Proc. of PESC '90,Vol. 1, pp. 455-460.[25] E. Bogalecka, Power control of a doubly fed induction generator

without speed or position sensor, Proc. of EPE93, Vol. 8, pp. 224-228, 1993.

[26] R. Pena, J.C. Clare, G.M. Asher, Doubly fed induction generatorusing back-to-back PWM converters and its application to variablespeed wind-energy generation. IEE Proc. on electronic powerapplication, 1996, pp. 231-241.

[27] Z. Chen, E. Spooner, Voltage Source Inverters for High-Power,Variable-Voltage DC Power Sources, IEE Proc. Generation,Transmission and Distributions, Vol. 148, No. 5, September 2001, pp.439-447.

[28] M. Dahlgren, H. Frank, M. Leijon, F. Owman, L. Walfridsson, Windpower goes large scale, ABB Review, 2000, Vol.3, pp. 31-37.

[29] A.K. Wallace, J.A. Oliver, Variable-Speed Generation Controlled byPassive Elements, Proc. of ICEM 98, Vol. 3, 1998, pp. 1554-1559.

[30] Z. Chen, E. Spooner, Grid Power Quality with Variable-Speed Wind

Turbines, IEEE Transactions on Energy Conversion, Vol. 16, No.2,June 2001, pp. 148-154.

[31] S. Bolik, Grid Requirements Challenges for Wind Turbines, Proc.of Fourth International Workshop on Large-Scale Integration of WindPower and Transmission Networks for Offshore Windfarms, 2003.

[32] . Larsson, The Power quality of Wind Turbines, Ph.D. thesis,Chalmers University of Technology, Gteborg, Sweden, 2000.

[33] Z. Saad-Saoud, N. Jenkins, The application of advanced static VArcompensators to wind farms, Power Electronics for RenewableEnergy (Digest No: 1997/170), 1997, pp. 6/1 - 6/5.

[34] P. Kazmierkowski, R. Krishnan, F. Blaabjerg, Control in PowerElectronics, Academic Press, 2002, ISBN 0-12-40277205.

[35] M. Liserre, A. DellAquila and F. Blaabjerg, Genetic algorithmbased design of the active damping for a LCL-filter three-phase activerectifier, IEEE Trans. on Power Electr. Vol. 19, pp. 76 -86, Jan 2004.

[36] V. Blasko and V. Kaura, A novel control to actively damp resonancein input LC filter of a three-phase voltage source converter, IEEETrans. on Ind. App., vol. 33, 1997, pp. 542-550.

[37] P. A. Dahono, A control method to damp oscillation in the input LCfilter of AC-DC PWM converters, in Proc. of PESC 2002, June

2002, pp. 1630-1635,.[38] P. C. Loh, M. J. Newman, D.N .Zmood, D.G. Holmes, A

Comparative Anlysis of multiloop Voltage Regulation Strategies forSingle and Three-Phase UPS Systems IEEE Trans. on Power Elect,vol. 18, no. 5. Sept. 2003.

[39] T. Ohnishi, Three phase PWM converter/inverter by means ofinstantaneous active and reactive power control in Proc of IECON91, pp. 819-824.

[40] T. Noguchi, H. Tomiki, S. Kondo, I. Takahashi: Direct powercontrol of PWM converter without power-source voltage sensors,IEEE Trans. on Ind. App., vol. 34, May/June 1998, pp. 473-479.

[41] I. Agirman and V. Blasko, A novel control method of a VSC withoutAC line voltage sensors, IEEE Trans. on Ind. App., vol. 39,March/April 2003, pp.519-524.

[42] M. Liserre A. DellAquila, F. Blaabjerg. Design and control of athree-phase active rectifier under non-ideal operating conditions inProc. of IAS 2002, pp. 1181-1188.

[43] A. Moran, P. D. Ziogas, G. Joos: Design aspects of synchronousPWM rectifier-inverter system under unbalanced input voltageconditions, IEEE Trans. on Ind. App., vol. 28, Nov./Dec. 1992, pp.1286-1293.

[44] M. Liserre, F. Blaabjerg, S. Hansen: Design and Control of an LCL-filter Based Active Rectifier, Conf. Rec. 36th IAS Ann. Meeting,Chicago (USA), Sept./Oct. 30-4, 2001

[45] Teodorescu,R., Blaabjerg,F., Liserre.M Stability of Grid-Connected PV Inverters with Large Grid Impedance Variation Proceedings of PESC04, Aachen

[46] T.Shimizu, M.Hirakata, T.Kamezawa, H.Watanabe - GenerationControl Circuit for Photovoltaic Modules IEEE Trans. On PowerElectronics, Vol. 16, No. 3, May, 2001, pp 293 300

[47] Calais, M.; Myrzik, J.; Spooner, T.; Agelidis, V.G.: Inverters forsingle-phase grid connected photovoltaic systems-an

overview, Power Electronics Specialists Conference, 2002. pesc 02.2002 IEEE 33rd Annual, Volume: 4, 23-27 June 2002 Pages: 1995 2000.

[48] K. Ogura; T. Nishida; E. Hiraki; M. Nakaoka and Shinichiro Nagai:Time-Sharing Boost Chopper Cascaded Dual Mode Single-PhaseSinewave Inverter for Solar Photovoltaic Power Generation System,35th Annuak IEEE Power Electronics Specialists Conference, Aachen,Germany, 2004, Pages: 4763-4766.

[49] Calais, M.; Agelidis, V.G.; Borle, L.J.; Dymond, M.S.: Atransformerless five level cascaded inverter based single

phase photovoltaic system, Power Electronics SpecialistsConference, 2000. PESC 00. 2000 IEEE 31st Annual, Volume: 3, 18-23 June 2000 Pages: 1173 - 1178 vol.3.

[50] Frede Blaabjerg; Zhe Chen; Soeren Baekhoej Kjaer: PowerElectronics as Efficient Interface in Dispersed Power Generation

Systems, IEEE Transactions on Power Electronics, Vol. 19, No.5,September 2004

[51] H.Haeberlin Evolution of Inverters for Grid connected PVsystems from 1989 to 2000- Proceedings of 17-th EuropeanPhotovoltaic Solar Energy Conference, Munich, Oct. 2001

[52] [1]Frede Blaabjerg; Zhe Chen; Soeren Baekhoej Kjaer: PowerElectronics as Efficient Interface in Dispersed Power GenerationSystems, IEEE Transactions on Power Electronics, Vol. 19, No.5,September 2004

[53] [2]H.Haeberlin Evolution of Inverters for Grid connected PVsystems from 1989 to 2000- Proceedings of 17-th EuropeanPhotovoltaic Solar Energy Conference, Munich, Oct. 2001

[54] [3]T.Shimizu, M.Hirakata, T.Kamezawa, H.Watanabe - GenerationControl Circuit for Photovoltaic Modules IEEE Trans. On PowerElectronics, Vol. 16, No. 3, May, 2001, pp 293 300

[55] [4]Calais, M.; Myrzik, J.; Spooner, T.; Agelidis, V.G.: Inverters forsingle-phase grid connected photovoltaic systems-an overview,


20/20

20

Power Electronics Specialists Conference, 2002. pesc 02. 2002 IEEE33rd Annual, Volume: 4, 23-27 June 2002 Pages: 1995 2000.

[56] [5]Ned Mohan, Tore Undeland, William P. Robbins: PowerElectronics. Converters, Applications and Design. [Mohan] JohnWiley & Sons 2003, ISBN: 0-471-22693-9

[57] [6]K. Ogura; T. Nishida; E. Hiraki; M. Nakaoka and ShinichiroNagai: Time-Sharing Boost Chopper Cascaded Dual Mode Single-Phase Sinewave Inverter for Solar Photovoltaic Power Generation

System, 35th Annuak IEEE Power Electronics SpecialistsConference, Aachen, Germany, 2004, Pages: 4763-4766.

[58] [7]Calais, M.; Agelidis, V.G.; Borle, L.J.; Dymond, M.S.: Atransformerless five level cascaded inverter based single phase

photovoltaic system, Power Electronics Specialists Conference,2000. PESC 00. 2000 IEEE 31st Annual, Volume: 3, 18-23 June 2000Pages: 1173 - 1178 vol.3.

[59] [8]Robert W. Erickson, Dragan Maksimovic: Fundamentals of PowerElectronics [Erickson] - Kluwer Academic Pub; March 1, 1997,ISBN: 0-412-08541-0, 773 pages.

[60] [9]Teodorescu,R., Blaabjerg,F., Liserre.M Stability of Grid-Connected PV Inverters with Large Grid Impedance Variation Proceedings of PESC04, Aachen

[61] [10]M. Kazmierkowski, R.Krishnan, F.Blaabjerg, Control in PowerElectronics. Selected Problems, Academic Press 2002, ISBN 0-12-402772-5.

[62] [11]C. Cecati, A. Dell'Aquila, M. Liserre and V. G. Monopoli,"Design of H-bridge multilevel active rectifier for traction systems",IEEE Trans. on Ind. Applicat., vol. 39, Sept./Oct. 2003, pp. 1541-1550.

[63] [12]X. Yuan, W. Merk, H. Stemmler, J. Allmeling Stationary-Frame Generalized Integrators for Current Control of Active PowerFilters with Zero Steady-State Error for Current Harmonics ofConcern Under Unbalanced and Distorted Operating ConditionsIEEE Trans. on Ind. App., vol. 38, no. 2, Mar./Apr. 2002, pp.523 532.

[64] [13]P. Mattavelli A Closed-Loop Selective Harmonic Compensationfor Active Filters IEEE Trans. on Ind. App., vol. 37, no. 1,

january/february, 2001, pp. 81 89.[65] [14]D. N. Zmood, D. G. Holmes, Stationary Frame Current

Regulation of PWM Inverters with Zero Steady-State Error IEEETrans. on Power Electr., vol. 18, no. 3, May 2003, pp. 814 822.

[66] [15]Teodorescu,R., Blaabjerg,F., Liserre.M., Borup,U.,- A NewControl Structure for Grid-Connected PV Inverters with Zero Steady-State Error and Selective Harmonic Compensation Proceedings of

APEC04, Anaheim, CA[67] [16]M. Liserre, F. Blaabjerg, and S. Hansen, Design and control ofan LCL-filter based active rectifier, IEEE Trans. on Ind. App., vol.38, no. 2 Sept./Oct. 2001, pp. 299-307.

[68] [17]J. H. R. Enslin and P. J. M. Heskes, Harmonic interactionbetween a large number of distributed power inverters and thedistributed network, PESC 2003, pp. 1742-1747.

[69] [18]Hohm, D.P.; Ropp, M.E. Comparative study of maximumpower point tracking algorithms using an experimental,programmable, maximum power point tracking test bed,Photovoltaic Specialists Conference, 2000. Conference Record of theTwenty-Eighth IEEE, 15-22 Sept. 2000 Pages:1699 1702.

[70] [19]European Standard EN 50330-1, Photovoltaic semiconductorconverters Part 1: Utility interactive fail safe protective interface forPV-line commutated converters - Design qualification and typeapproval, 1999.

[71] [20]M. Sumner, B. Palethorpe, D.W.P. Thomas, P. Zanchetta, M.C.DiPiazza, "A technique for power supply harmonic impedance

estimation using a controlled voltage disturbance", Trans. on PowerElectronics, Vol. 17, Issue 2, 2002, pp. 207 215.

[72] [21]Teodorescu,R., Asiminoei,L., Blaabjerg, Borup,U.,- A newmethod of on-line grid impedance estimation for PV inverters Proceedings of APEC04, Anaheim, CA

[73] [22]A. Timbus, R. Teodorescu, F. Blaabjerg, U. Borup. Online Gridmeasurement and ENS detection for PV inverters running on highlyinductive grid IEEE Trans. On Power Electronics. Letters (in

process

Documents

Power Converters and Control of Renewable Energy Systems Paper