Upload
phamngoc
View
217
Download
0
Embed Size (px)
Citation preview
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
OSMOSE-‐Lua – A Unified Approach to Energy Systems Integration with Life Cycle
AssessmentMin-‐Jung Yooa, Lindsay Lessardb, c, Maziar Kermanib, François Maréchalb
a EPFL, Laboratory for Computer-‐aided Design and Production (LICP), Lausanne, Switzerland b EPFL, Industrial Process & Energy Systems Engineering Group (IPESE), Lausanne, Switzerland
c Quantis, EPFL Innovation Park, Apt. D, 1015 Lausanne, Switzerland
PSE2015-‐ESCAPE25, Copenhagen, Denmark June 3, 2015
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
The system vision of the energy usage
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
Socio - ecologic - economic environment
The system vision of the energy usage
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
Socio - ecologic - economic environment
The system vision of the energy usage
Products &Energy services products
products
Demand
Price
t
t
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
Socio - ecologic - economic environment
The system vision of the energy usage
Raw materials
Resources
Price
Availability
t
t
Products &Energy services products
products
Demand
Price
t
t
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
Socio - ecologic - economic environment
The system vision of the energy usage
Conversion systems
Raw materials
Resources
Price
Availability
t
t
Products &Energy services products
products
Demand
Price
t
t
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
Socio - ecologic - economic environment
The system vision of the energy usage
Conversion systems
Raw materials
Resources
Price
Availability
t
t
Products &Energy services products
products
Demand
Price
t
t
Waste Emissions
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
Socio - ecologic - economic environment
Energy system design
Conversion systems
Raw materials
Resources
Price
Availability
t
t
Products &Energy services products
products
Demand
Price
t
t
Waste Emissions
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
Problem definition
• Characterise and localise the technologies to be used to supply the energy services and the products needed over the lifetime of the system’s element at the time they are needed with the resources that are available • Decisions to be taken • Technology : choice, size and location • The operating strategy • The services that are provided • The resources that are used
• Criteria • Minimum cost = OPEX + APEX • Maximum profit = NPV (i,lifetime,cost) • Minimum environmental impact = LCIA • Max renewable energy integration
• Constraints • Mass & Energy Balances • Context specs • Bounds
Yu,l, Su,l
fu,l,t(p) 8p, u, lfu!s,l,t(p) 8p, u, l, sfr!u,l,t(p) 8p, u, l, r
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
The energy systems engineering methodology
Solutions
Energy services Resources Context & Constraints
Superstructure
System Boundaries
System performances indicators•Economic•Thermodynamic• Life cycle environmental impact
Results analysis•Exergy analysis•Composite curves•Sensitivity analysis•Multi-criteria
Technology options
Models
250
300
350
400
450
500
550
600
0 5000 10000 15000 20000 25000 30000
T(K
)
Q(kW)
Cold composite curve
Hot composite curveHeat & Mass integration
Decision variables
Solving method
Thermo-economic Pareto
Multi-objectiveOptimization
Out=F(In,P)
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
Energy systems integration : multi-‐scale problem
6
EnA / EnMS
Common element is the energy planning
• Energy consumption profile determination
• Energy performance evaluation
• Energy baseline generation
• Identification and evaluation of improvement options
• Implementation and monitoring
Step 1
Step 2
• Systematic approach
• Methodologies
• Tools
Context Literature review Problematic 1st year research Research plan Conclusion
Step 3
Details
Scale
Technologies
Processes
Industrial Clusters
Districts
Regions
* Model the interactions * Identify the synergies * Implement the interactions => Network or Grid
* Multi scale problem * Systematic framework
Materials
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
• Model of the interconnectivity (mass, heat, energy) • Model of the emissions (Equipment, Emissions) • Model of the cost (size => cost, maintenance)
The energy technology building block
-‐ Equipment sizing model -‐ Cost estimation
Heat transfer requirement
Heat transfer
Thermo-‐chemical conversion Model
Material streamsProduct streams
ElectricityHeat transfer requirement
Water streamsWater streams
Waste streamsWaste streams
ElectricityUnit parameters Decision Variables
Life cycle emissions
Life cycle of equipment -‐ Production -‐ Dismantling
MaintenanceInvestment cost
LCA models
LCA models
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
Energy technology approach
• Advantages • Expert developed model • Flowsheeting model and tool • Level of detail
• Independent of the energy systems integration • Model data base can be developed
• Difficulties • Remote access of models • Use of surrogate models
• Consistency of the models in the energy systems • from black box to white box
• Model data base • Confidence for reusing the models • Documentation • Certification of the model quality
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
Energy Technologies Encapsulation
Unit documentation DescriptionPurposeLimitationAuthors
Unit model report Summary of resultsImportant dataErrorsValidity
Unit execution handler Input dataCalculation engines
Existing flowsheeting softwareMimetic models
Output dataIntegration/interconnectivityCostingLCA => Ecoinvent LCI
Model ontologyModel encapsulation
Common interface of a module for system integration
Flowsheeting tools •BELSIM-VALI•gPROMS•ASPEN plus•HYSYS•Matlab•Simulink•(CITYSIM)•MODELICA•Others possible
•CAPE-OPEN ?•PROSIM
Unit connectivity Layers connections
Layer ontologyData base
ModelData base
KnowledgeData base
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
Ontology of a value in OSMOSE LUA
A variable is the smallest data structure handled by OSMOSE. A variable is defined by1 Name or tagname that litteraly identifies the variable 2 ID that is tagging the variable 3 Parent indicates the module to which the variable belongs 4 Value that defines one instance of the variable => Matrix
A priori value of the variable will exist for any time in the time sequence, for any period and for any scenario in which the system is calculated.
5 MinimumValue and a MaximumValue that defines the bounds for the value of the variables 6 DistributionType and DistributionParameters that defines the uncertainty of the value
7 DefaultValue that defines the value of the variable when it is created. this is also the recommended order of magnitude of the value recommended for the calculations
8 PhysicalUnit that defines the physical unit of the value of the variable 9 DefaultPhysicalUnits that defines the physical unit with which the calculation have to be made, It means that a
preprocessing phase will have to convert the value from the PhysicalUnits to the DefaultPhysicalUnits as well as a post-processing phase that is going to convert from the DefaultPhysicalUnits to the PhysicalUnits .
10 Status that identifies if the variable has a fixed value or a value that will be calculated. The status should identify as well if the value is calculated for each time or if it is constant or if it requires recalculation for each time. Status is equivalent to ISVIT in OSMOSE.i x : value is unique for the scenario (i.e. 1 value/scenario) ii x0 : value is period dependent (i.e. 1 value/period and / period)The value may calculated during
preprocessing phase using the model (i.e. constant that is period dependent).iii x00 : the value is time dependent (i.e. 1 value/time and period and scenario).
where x = 1 : for constant to be fixed by the user or the optimizer
2 : for constant fixed by the user or the optimizer otherwise use the default value -‐1 : for values calculated by the model
11 Dependence Matrix a dependence matrix that identifies the dependence of the value with respect to the other values of the problem is stored. Each of the dependence includes the mention if the dependence is linear, in which case the value of the dependence is stored in the field as a couple (ID,numerical value), or it can be stored as a reference to a non linear dependence that could be a surrogate model or the access to the model that calculates the dependence.
mu,l(t), Yi,u,l(t), Xi,u,l(t),⇧i,u,l(t),KPIk, m⇤u,l, Y
⇤i,u,l, X
⇤i,u,l,⇧
⇤i,u,l
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
Superstructure definitions
Luc Girardin page 25/77
Travail de diplôme : Méthodologie de conception optimale d'un réseau hydrogène d'un site industriel.
7.2 DÉFINITION DE LA SUPERSTRUCTURE
Le réseau d'hydrogène est conçu comme une superstructure qui contient toutes les sources et puits desunités productrices/consommatrices d'hydrogène, et qui incorpore toutes les connections possibles entre cessources et ces puits (Figure 7.2). Toutes les connections de la superstructure ne sont pas admissibles, cequi revient à introduire dans le modèle des contraintes sur l'établissement des nouvelles connexions.
Le point de fonctionnement de la structure est défini par l'ensemble des données requises pour chaqueconsommateur, chaque producteur, et chaque unité.
Chaque source de la structure est caractérisée par :
! la pureté de l'hydrogène produit Xp
! la pureté en impuretés produites x p , j
! un débit d'hydrogène disponible délivré "mp
! le niveau de pression total du flux délivré P p .
En contrepartie, chaque puits est défini par :
! la pureté requise du flux d'entrée Xc
! les limites de composition en impuretés du mélange admis xc , jmax
! un besoin de débit d'hydrogène "mc
! la pression totale requise du mélange à l'entrée Pc .
Le nombre des puits et sources, regroupé au sein d'une même unité, est égal au nombre de niveaux distinctsde pureté des flux d'hydrogène admis, respectivement rejetés. Dans un premier temps, on peut admettre quechaque unité possède au maximum 1 consommateur et 2 producteurs. Un nombre plus élevé de ceséléments par unité correspond à un affinement du modèle, et entraîne une augmentation de la densité deconnexions au sein de la superstructure
On a vu que les unités regroupent, dans une même structure, des puits et des sources qu'on appelleconsommateurs et producteurs.
Elles sont définies par :
! l'identité des puits et sources qui la compose,
! leurs coordonnées (X,Y) géo-référencées,
! un identifiant, qui permet de les regrouper en sous-systèmes pour l'analyse dans les diagrammes dePincement (cf. Chapitre "Représentation graphiques").
25/77 18/02/05
Figure 7.1 : Schéma de la superstructure complexe
250
300
350
400
450
500
550
600
0 5000 10000 15000 20000 25000 30000
T(K
)
Q(kW)
Cold composite curve
Hot composite curve
• Explicit • In flowsheeting software
• Automatic • Based on the connectivity description • Restricted matches • Routing
• Implicit • Heat cascades
• Combined • Mass and Energy
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
Geograhical informationAt a given location : combination of energy technologies
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
Geograhical information
Location i, (xi, yi, zi)
Location 1, (x1, y1, z1)
Location 2, (x2, y2, z2)
Geneva
At a given location : combination of energy technologies
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
Geograhical information
• Coordinates attributed to locations • GIS data base => POSTGRE SQL • GIS Level of details
• Roads, district heating network, electricity grid • GIS Tools : Routing, Layers, Representations • GIS Data
• Buildings • Resources • Topology
Location i, (xi, yi, zi)
Location 1, (x1, y1, z1)
Location 2, (x2, y2, z2)
Geneva
At a given location : combination of energy technologies
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
Geograhical information
• Coordinates attributed to locations • GIS data base => POSTGRE SQL • GIS Level of details
• Roads, district heating network, electricity grid • GIS Tools : Routing, Layers, Representations • GIS Data
• Buildings • Resources • Topology
•Link between GIS data and Models •Link with unit models parameters •Access to GIS tools in the workflow
Location i, (xi, yi, zi)
Location 1, (x1, y1, z1)
Location 2, (x2, y2, z2)
Geneva
At a given location : combination of energy technologies
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
Geograhical information
• Coordinates attributed to locations • GIS data base => POSTGRE SQL • GIS Level of details
• Roads, district heating network, electricity grid • GIS Tools : Routing, Layers, Representations • GIS Data
• Buildings • Resources • Topology
•Link between GIS data and Models •Link with unit models parameters •Access to GIS tools in the workflow
Location i, (xi, yi, zi)
Location 1, (x1, y1, z1)
Location 2, (x2, y2, z2)
Geneva
T TTT -20
0
20
40
60
80
0 50 100 150 200 250 300 350 400
T(C
)Q(kW)
AirWaste Water
Heating Hot water
recoveryRecoverable
Heat requirement
At a given location : combination of energy technologies
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
Energy system design : problem definition
Given a set of energy conversion technologies : Where to locate the energy conversion technologies ? How to connect the buildings ? How to operate the energy conversion technologies ?
1 Symbols
Roman lettersAn Annuities of a given investment [-]Ai,j,p Surface of the pipe of network p between nodes i and j [m2]B Arbitrarily big valueCaw Investment costs for air/water heat pump(s) [CHF]Cboiler Annual investment costs of the boiler(s) [CHF/year]Cfix Fix part of investment costs for a given device [CHF]Cgas Total gas costs [CHF]cgas Gas costs [CHF/kWh]Cgrid Total grid costs [CHF]cgrid Grid costs [CHF/kWh]C inv Investment costs of a given device [CHF]C inv
an Annual investment costs of a given device [CHF]Cpipes Total costs for the piping [CHF]Cprop Fix part of investment costs for a given device [CHF/kW]Cref Investment costs of a device chosen as reference [CHF]Cww Investment costs for water/water heat pump(s) [CHF]CO2
gas Total CO2 emissions for one year due to the combustion of gas [kg]co2
gas CO2 emissions due to the combustion of gas per kWh [kg/kWh]CO2
grid Total CO2 emissions for one year due to the consumption of electricity from the grid [kg]co2
grid CO2 emissions due to the consumption of electricity from the grid [kg/kWh]COP aw
t,k Coe⇤cient of performance for the air/water heat pump during period t at node k [-]COPww
t,k Coe⇤cient of performance for a water/water heat pump during period t at node k [-]cp Isobaric specific heat [kJ/(K· kg)]Disti,j Distance between nodes i and j [m]Ht Number of hours in period t [hour]Econs
t,k Electricity consumption during period t at node k [kW]Eexp
t,k Eletricity exported to the grid during period t by device e [kW]Egrid
t,k Electricity bought from the grid during period t [kW]Eaw
t,k Electricity consumed by the air/water heat pump during period t at node k [kW]Eww
t,k Electricity consumed by the water/water heat pump during period t at node k [kW]Eloss
t,k Electricity losses during period t [kW]Epump
t Pumping power for the network during period t [kW]Etech
t,k,e Electricity produced or consumed during period t at node k by device e [kW]Fm Maintenance factor [-]Gast Overall gas consumption during period t [kg]Lmin
e Minimum allowable part-load of device e [-]
Mbuildt,k
Mass flow of water from the network, circulating during period t through thebuilding at node k to heat up the building [kg/s]
Mpipet,i,j,p
Mass flow of water flowing during period t, in network p, from node i to nodej [kg/s]
Mmaxt,p,i,j
Maximum mass flow of water flowing in network p, from node i to node j, overall periods of time [kg/s]
M techt,k
Mass flow from the network being heated up by the device(s) at node k duringperiod t [kW]
MT buildt,k
Mass flow from the network flowing through a device at node k during periodt, to be re-heated, times its temperature [(kg/s)K]
2
MT pipet,i,j,p
Mass flow flowing during period t from node i to node j in the network p, timesits temperature [(kg/s)K]
N Expected lifetime for a given investment [year]Ploss Pressure losses in the pipes [Pa/m]Qaw
t,k Heat delivered by the air/water heat pump to the consumer at node k during period t [kW]Qboiler
t,k Heat delivered by the boiler at node k during period t [kW]Qcons
t,k Heat consumption at node k during period t [kW]Qnet
t,k Heat delivered by the network to the consumer at node k during period t [kW]Qnet ww
t,k Heat delivered by the network to the water/water heat pump at node k during period t [kW]Qtech
t,k,e Heat produced by device e located at node k during period k [kW]Qww
t,k Heat delivered by an water/water heat pump to the consumer at node k during period t [kW]r Interest rate [-]S Size of a given device [kW]Saw
k Nominal size of the air/water heat pump located at node k [kW]Sboiler
k Nominal size of the boiler at node k [kW]Snom
e Nominal power of a device [kW]Sref Size of a device chosen as reference [kW]Sww
k Nominal power of the water/water heat pump located at node k [kW]T atm Atmospheric temperature [K]T cold Temperature of the heat source for heat pumps [K]T cons
t,k Temperature at which the heat is required by the consumer at node k during period t [K]T hot Temperature of the heat sink for heat pumps [K]T net Design temperature of the network [K]v Velocity of the water through the pipes [m/s]X = 1 if a device exists, 0 otherwiseXaw
k = 1 if there is an air/water heat pump at node k, 0 otherwiseXboiler
k = 1 if there is a boiler at node k, 0 otherwiseXgas
e = 1 if device e needs gas to operate, 0 otherwiseXnode
k = 1 if a device e is implemented at node k, 0 otherwiseXtech
k,e = 1 if a device can be implemented at node k, 0 otherwiseXww
k = 1 if there is an water/water heat pump at node k, 0 otherwiseY i,j,p = 1 if a connection exists between i and j for network p , 0 otherwise
Greek letters�Theat Pinch at the heat-exchangers [K]�Tnet ww Temperature di⇥erence of the water in the network when it serves as heat source for water/water heat pump(s) [K]�boiler Thermal e⌅ciency of the boiler(s) [-]�ele Electric e⌅ciency of device e [-]�grid E⌅ciency of the grid [-]�the Thermal e⌅ciency of device e [-]⇥ Exergetic e⌅ciency [-]⇤ Density [kg/m3]
Indicesk nodesi, j connection from node i to node jt timee technologies
3
Networks superstructure
1 Symbols
Roman lettersAn Annuities for a given investment [-]Ai,j,p Area of the pipe between nodes i and j of network p [m2]B Arbitrarily big valueCaw Investment costs for air/water heat pump(s) [CHF]Cboiler Annual investment costs for the boiler(s) [CHF/year]Cfix Fix part of the investment costs for a given device [CHF]Cgas Total annual natural gas costs [CHF/year]cgas Natural gas costs [CHF/kWh]Cgrid Total annual grid costs [CHF/year]cgrid Grid costs [CHF/kWh]C inv Investment costs of a given device [CHF]C inv
an Annual investment costs of a given device [CHF/year]Cpipes Total annual costs for the piping [CHF/year]Cprop Fix part of the investment costs for a given device [CHF/kW]Cref Investment costs of a device chosen as reference [CHF]Cww Investment costs for water/water heat pump(s) [CHF]CO2
gas Total annual CO2 emissions due to the combustion of natural gas [kg/year]co2
gas CO2 emissions due to the combustion of natural gas [kg/kWh]CO2
grid Total annual CO2 emissions due to the consumption of electricity from the grid [kg/year]co2
grid CO2 emissions due to the consumption of electricity from the grid [kg/kWh]COPhp Coe⇤cient of performance of the central heat pumpCOP aw
t,k Coe⇤cient of performance for the air/water heat pump during period t at node k [-]COPww
t,k Coe⇤cient of performance for a water/water heat pump during period t at node k [-]cp Isobaric specific heat [kJ/(K· kg)]Disti,j Distance between nodes i and j [m]Ht Number of hours in period t [hour]Econs
t,k Electricity consumption during period t at node k [kW]Eexp
t,e Eletricity exported to the grid during period t by device e [kW]Egrid
t,k Electricity bought from the grid during period t by node k [kW]Eaw
t,k Electricity consumed by the air/water heat pump during period t at node k [kW]Eww
t,k Electricity consumed by the water/water heat pump during period t at node k [kW]Eloss
t Electricity losses during period t [kW]Epump
t Pumping power for the network during period t [kW]Etech
t,k,e Electricity produced or consumed during period t at node k by device e [kW]Fs Scaling factor [-]Fm Maintenance factor [-]Gast Natural gas consumption during period t [kg]Lmin
e Minimum allowable part-load of device e [-]Mbuild
t,k Water circulating during period t through the building at node k, to heat it up [kg/s]Mpipe
t,i,j,p Water flowing during period t, from node i to node j, in network p [kg/s]Mmax
i,j,p Maximum flow of water between nodes i and j, in network p, over all periods [kg/s]M tech
t,k Water being heated up by the device(s) during period t, at node k [kg/s]MT tech
t,k Water flowing through a device during period t at node k to be re-heated, times its temperature [(kg/s)K]
2
Q1..T
E1..T
Q1..T
E1..T
Q1..T
E1..T
Q1..T
E1..T
Q1..T
E1..TQ1..T
E1..T
Q1..T
E1..T
Q1..T
E1..T
Cinv =ne�
e=1
nk�
k=1
ae + ae � Se,k
Investment
Cgas =ne�
e=1
nk�
k=1
np�
t=1
HtQe,k,t
�the
1 Symbols
Roman lettersAn Annuities for a given investment [-]Ai,j,p Area of the pipe between nodes i and j of network p [m2]B Arbitrarily big valueCaw Investment costs for air/water heat pump(s) [CHF]Cboiler Annual investment costs for the boiler(s) [CHF/year]Cfix Fix part of the investment costs for a given device [CHF]Cgas Total annual natural gas costs [CHF/year]cgas Natural gas costs [CHF/kWh]Cgrid Total annual grid costs [CHF/year]cgrid Grid costs [CHF/kWh]C inv Investment costs of a given device [CHF]C inv
an Annual investment costs of a given device [CHF/year]Cpipes Total annual costs for the piping [CHF/year]Cprop Fix part of the investment costs for a given device [CHF/kW]Cref Investment costs of a device chosen as reference [CHF]Cww Investment costs for water/water heat pump(s) [CHF]CO2
gas Total annual CO2 emissions due to the combustion of natural gas [kg/year]co2
gas CO2 emissions due to the combustion of natural gas [kg/kWh]CO2
grid Total annual CO2 emissions due to the consumption of electricity from the grid [kg/year]co2
grid CO2 emissions due to the consumption of electricity from the grid [kg/kWh]COPhp Coe⇤cient of performance of the central heat pumpCOP aw
t,k Coe⇤cient of performance for the air/water heat pump during period t at node k [-]COPww
t,k Coe⇤cient of performance for a water/water heat pump during period t at node k [-]cp Isobaric specific heat [kJ/(K· kg)]Disti,j Distance between nodes i and j [m]Ht Number of hours in period t [hour]Econs
t,k Electricity consumption during period t at node k [kW]Eexp
t,e Eletricity exported to the grid during period t by device e [kW]Egrid
t,k Electricity bought from the grid during period t by node k [kW]Eaw
t,k Electricity consumed by the air/water heat pump during period t at node k [kW]Eww
t,k Electricity consumed by the water/water heat pump during period t at node k [kW]Eloss
t Electricity losses during period t [kW]Epump
t Pumping power for the network during period t [kW]Etech
t,k,e Electricity produced or consumed during period t at node k by device e [kW]Fs Scaling factor [-]Fm Maintenance factor [-]Gast Natural gas consumption during period t [kg]Lmin
e Minimum allowable part-load of device e [-]Mbuild
t,k Water circulating during period t through the building at node k, to heat it up [kg/s]Mpipe
t,i,j,p Water flowing during period t, from node i to node j, in network p [kg/s]Mmax
i,j,p Maximum flow of water between nodes i and j, in network p, over all periods [kg/s]M tech
t,k Water being heated up by the device(s) during period t, at node k [kg/s]MT tech
t,k Water flowing through a device during period t at node k to be re-heated, times its temperature [(kg/s)K]
2
1 Symbols
Roman lettersAn Annuities for a given investment [-]Ai,j,p Area of the pipe between nodes i and j of network p [m2]B Arbitrarily big valueCaw Investment costs for air/water heat pump(s) [CHF]Cboiler Annual investment costs for the boiler(s) [CHF/year]Cfix Fix part of the investment costs for a given device [CHF]Cgas Total annual natural gas costs [CHF/year]cgas Natural gas costs [CHF/kWh]Cgrid Total annual grid costs [CHF/year]cgrid Grid costs [CHF/kWh]C inv Investment costs of a given device [CHF]C inv
an Annual investment costs of a given device [CHF/year]Cpipes Total annual costs for the piping [CHF/year]Cprop Fix part of the investment costs for a given device [CHF/kW]Cref Investment costs of a device chosen as reference [CHF]Cww Investment costs for water/water heat pump(s) [CHF]CO2
gas Total annual CO2 emissions due to the combustion of natural gas [kg/year]co2
gas CO2 emissions due to the combustion of natural gas [kg/kWh]CO2
grid Total annual CO2 emissions due to the consumption of electricity from the grid [kg/year]co2
grid CO2 emissions due to the consumption of electricity from the grid [kg/kWh]COPhp Coe⇤cient of performance of the central heat pumpCOP aw
t,k Coe⇤cient of performance for the air/water heat pump during period t at node k [-]COPww
t,k Coe⇤cient of performance for a water/water heat pump during period t at node k [-]cp Isobaric specific heat [kJ/(K· kg)]Disti,j Distance between nodes i and j [m]Ht Number of hours in period t [hour]Econs
t,k Electricity consumption during period t at node k [kW]Eexp
t,e Eletricity exported to the grid during period t by device e [kW]Egrid
t,k Electricity bought from the grid during period t by node k [kW]Eaw
t,k Electricity consumed by the air/water heat pump during period t at node k [kW]Eww
t,k Electricity consumed by the water/water heat pump during period t at node k [kW]Eloss
t Electricity losses during period t [kW]Epump
t Pumping power for the network during period t [kW]Etech
t,k,e Electricity produced or consumed during period t at node k by device e [kW]Fs Scaling factor [-]Fm Maintenance factor [-]Gast Natural gas consumption during period t [kg]Lmin
e Minimum allowable part-load of device e [-]Mbuild
t,k Water circulating during period t through the building at node k, to heat it up [kg/s]Mpipe
t,i,j,p Water flowing during period t, from node i to node j, in network p [kg/s]Mmax
i,j,p Maximum flow of water between nodes i and j, in network p, over all periods [kg/s]M tech
t,k Water being heated up by the device(s) during period t, at node k [kg/s]MT tech
t,k Water flowing through a device during period t at node k to be re-heated, times its temperature [(kg/s)K]
2
Maintenance
Gas Grid
Electricity
Electricity
Emissions
1 Symbols
Roman lettersAn Annuities for a given investment [-]Ai,j,p Area of the pipe between nodes i and j of network p [m2]B Arbitrarily big valueCaw Investment costs for air/water heat pump(s) [CHF]Cboiler Annual investment costs for the boiler(s) [CHF/year]Cfix Fix part of the investment costs for a given device [CHF]Cgas Total annual natural gas costs [CHF/year]cgas Natural gas costs [CHF/kWh]Cgrid Total annual grid costs [CHF/year]cgrid Grid costs [CHF/kWh]C inv Investment costs of a given device [CHF]C inv
an Annual investment costs of a given device [CHF/year]Cpipes Total annual costs for the piping [CHF/year]Cprop Fix part of the investment costs for a given device [CHF/kW]Cref Investment costs of a device chosen as reference [CHF]Cww Investment costs for water/water heat pump(s) [CHF]CO2
gas Total annual CO2 emissions due to the combustion of natural gas [kg/year]co2
gas CO2 emissions due to the combustion of natural gas [kg/kWh]CO2
grid Total annual CO2 emissions due to the consumption of electricity from the grid [kg/year]co2
grid CO2 emissions due to the consumption of electricity from the grid [kg/kWh]COPhp Coe⇤cient of performance of the central heat pumpCOP aw
t,k Coe⇤cient of performance for the air/water heat pump during period t at node k [-]COPww
t,k Coe⇤cient of performance for a water/water heat pump during period t at node k [-]cp Isobaric specific heat [kJ/(K· kg)]Disti,j Distance between nodes i and j [m]Ht Number of hours in period t [hour]Econs
t,k Electricity consumption during period t at node k [kW]Eexp
t,e Eletricity exported to the grid during period t by device e [kW]Egrid
t,k Electricity bought from the grid during period t by node k [kW]Eaw
t,k Electricity consumed by the air/water heat pump during period t at node k [kW]Eww
t,k Electricity consumed by the water/water heat pump during period t at node k [kW]Eloss
t Electricity losses during period t [kW]Epump
t Pumping power for the network during period t [kW]Etech
t,k,e Electricity produced or consumed during period t at node k by device e [kW]Fs Scaling factor [-]Fm Maintenance factor [-]Gast Natural gas consumption during period t [kg]Lmin
e Minimum allowable part-load of device e [-]Mbuild
t,k Water circulating during period t through the building at node k, to heat it up [kg/s]Mpipe
t,i,j,p Water flowing during period t, from node i to node j, in network p [kg/s]Mmax
i,j,p Maximum flow of water between nodes i and j, in network p, over all periods [kg/s]M tech
t,k Water being heated up by the device(s) during period t, at node k [kg/s]MT tech
t,k Water flowing through a device during period t at node k to be re-heated, times its temperature [(kg/s)K]
2
1 Symbols
Roman lettersAn Annuities for a given investment [-]Ai,j,p Area of the pipe between nodes i and j of network p [m2]B Arbitrarily big valueCaw Investment costs for air/water heat pump(s) [CHF]Cboiler Annual investment costs for the boiler(s) [CHF/year]Cfix Fix part of the investment costs for a given device [CHF]Cgas Total annual natural gas costs [CHF/year]cgas Natural gas costs [CHF/kWh]Cgrid Total annual grid costs [CHF/year]cgrid Grid costs [CHF/kWh]C inv Investment costs of a given device [CHF]C inv
an Annual investment costs of a given device [CHF/year]Cpipes Total annual costs for the piping [CHF/year]Cprop Fix part of the investment costs for a given device [CHF/kW]Cref Investment costs of a device chosen as reference [CHF]Cww Investment costs for water/water heat pump(s) [CHF]CO2
gas Total annual CO2 emissions due to the combustion of natural gas [kg/year]co2
gas CO2 emissions due to the combustion of natural gas [kg/kWh]CO2
grid Total annual CO2 emissions due to the consumption of electricity from the grid [kg/year]co2
grid CO2 emissions due to the consumption of electricity from the grid [kg/kWh]COPhp Coe⇤cient of performance of the central heat pumpCOP aw
t,k Coe⇤cient of performance for the air/water heat pump during period t at node k [-]COPww
t,k Coe⇤cient of performance for a water/water heat pump during period t at node k [-]cp Isobaric specific heat [kJ/(K· kg)]Disti,j Distance between nodes i and j [m]Ht Number of hours in period t [hour]Econs
t,k Electricity consumption during period t at node k [kW]Eexp
t,e Eletricity exported to the grid during period t by device e [kW]Egrid
t,k Electricity bought from the grid during period t by node k [kW]Eaw
t,k Electricity consumed by the air/water heat pump during period t at node k [kW]Eww
t,k Electricity consumed by the water/water heat pump during period t at node k [kW]Eloss
t Electricity losses during period t [kW]Epump
t Pumping power for the network during period t [kW]Etech
t,k,e Electricity produced or consumed during period t at node k by device e [kW]Fs Scaling factor [-]Fm Maintenance factor [-]Gast Natural gas consumption during period t [kg]Lmin
e Minimum allowable part-load of device e [-]Mbuild
t,k Water circulating during period t through the building at node k, to heat it up [kg/s]Mpipe
t,i,j,p Water flowing during period t, from node i to node j, in network p [kg/s]Mmax
i,j,p Maximum flow of water between nodes i and j, in network p, over all periods [kg/s]M tech
t,k Water being heated up by the device(s) during period t, at node k [kg/s]MT tech
t,k Water flowing through a device during period t at node k to be re-heated, times its temperature [(kg/s)K]
2
Industry
Powerplant
Wastetreat.
Building plant.
. [5] Francois Marechal, Celine Weber, and Daniel Favrat. Multi-Objective Design and Optimisation of Urban Energy Systems, pages 39–81. Number ISBN: 978-3-527-31694-6. Wiley, 2008.
Water
Fuel
Water
FuelIndustrial plant.
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
Project data base
Predefined data
Project data base
OSMOSE LUA
Saved user dataEnergy technologies
Problems
ModulesDefault values
ThermodynamicSizing
Costing Life Cycle Inventory
EcoInvent
Work flow
ModelsSolving methods
in locations
GIS data base
GUI
Results data base
Layer Ontology
Superstructure
Yi,u,l(p(t)) = Fu(Xi,u,l(p(t)),⇧i,u,l(p(t)))
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
MILP process integration model @location k
Water: mii, Tin, hin
return : moo, Tout, hout
Water: mio, Tin, hin
return : moi, Tin, hin
Technologies w, @location k period s and time t(s)Demand @location k, period s and time t(s)
Storage tank
2
Tst,2
HEX h1(t)
heat demand
Storage tank
1
Tst,1
...
Tst,...
Storage tank
l
Tst,l
...
Tst,...
Storage tank
nl
Tst,nl
HEX h2(t) HEX h
..(t) HEX h
l(t) HEX h
nl-1(t)
heat demand heat demand heat demand heat demand
heat excess
HEX c1(t)
heat excess
HEX c2(t)
heat excess
HEX c..(t)
heat excess
HEX cl(t)
heat excess
HEX cnl-1
(t)
Units for heat integration
HL HL HL HL HL HL
HEX: Heat exchangers
HL: Heat losses
Storage system
S. Fazlollahi, G. Becker and F. Maréchal. Multi-Objectives, Mul.-Period Optimization of district energy systems: II- Daily thermal storage, in Computers & Chemical Engineering, 2013a
Buildings inertia
Heat exchange by heat cascade model
Electricity balance
Existence : w in location k
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
• Mass balance
• Mass/heat distribution
Layers models / automatic superstructure
nuX
u=1
k+u,l,t · fu,l,t(p) =nlX
j=1
eu,l,j,t(p) 8t(p) 8p8l
nuX
u=1
k�u,l,t · fu,l,t(p) =nlX
j=1
eu,j,l,t(p) 8t(p) 8p8l
nuX
u=1
nlX
l=1
k+u,l,t · fu,l,t(p) = 0 8t(p) 8p
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
Process integration models• MILP (Mixed Integer Linear Programming) models
• Integer => Selection/Usage • Continuous => Level of usage in t(period) • Parametrised (T)
• Discretisation • Decomposition
• Applications • Utility integration (Papoulias et al. , 1983, Marechal, et al., 1991)
• Combined Heat and Power + Steam network (Marechal et al. , 1997) • Refrigeration Cycle (Marechal et al., 2000) • ORC systems (Bendig, 2015)
• Multi-period (Marechal et al., 2003) • Multi period + Multi time (Weber et al., 2008) • Multi period + Multi time + Storage (Fazlollahi et al., 2012) • Multi period + Multi time + Multi-location+Storage (Fazlollahi et al., 2014) • Multi period + Multi time + Multi-location+Storage+Seasonal Storage (Rager et al., 2015) • Bio-refineries (Ensinas, 2015) • Industrial ecology (Gerber, 2013) • Restricted Matches (Becker et al., 2012) • Heat and Mass (Kermani et al., 2014) • Heat load distribution (Marechal et al., 1989) • Heat load Distribution multi-period (Mian et al. , 2014) • Heat load Distribution site scale (Pouransari et al., 2014) • Hydrogen Network (Girardin et al., 2006)
• MINLP (Mixed Integer Non Linear Programming) models • Heat Exchanger Network Design (Grossmann et al., Mian et al, 2014) • Heat Exchanger Network Design Multi-period (Mian et al., 2014)
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
Access to solvers
• MILP/MINLP solvers • Mathematical Programming Language • GLPK (MILP-Open source) • AMPL (MILP and MINLP - Proprietary)
• Solvers • LPSOLVE (Open Source) • Gurobi • CPLEX • BONMIN • Baron • NLPSOL
• Evolutionary algorithms • MOO (EPFL) • Dakota
• Parallelisation • MPI => grid computing
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
Computational platform -‐ OSMOSEevolutionary, multi-‐objective optimisation algorithm
(MINLP master problem)
Decision variables
performances
economic model
LCA models
LCI database (ecoinvent)
componentsunit processes elementary flows
energy-‐ and material-‐ flow models superstructure
detailed models: flowsheeting software
average models: LCI database
energy & mass integration (MILP slave sub-‐problem)
process integration software
supply chain synthesis
state variables
multi-‐time approach
Decision variables
background processes
process units
impact assessmentstate variables
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
OSMOSE : Computer Aided Platform
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
OSMOSE : Computer Aided Platform
GIS data base Industrial ecologyUrban systems
GUI : Spreadsheets, Matlab
Data Structuring
Technology models data base Energy conversionSharing knowledge
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
OSMOSE : Computer Aided Platform
Flowsheeting tools •BELSIM-VALI•gPROMS•ASPEN plus•HYSYS•Matlab•Simulink•(CITYSIM)•MODELICA•Others possible
•CAPE-OPEN ?•PROSIM•UNISIM ?
Energy technology data base •Data/models interfaces•Simulation•Process integration interface•Costing/LCIA performances•Reporting/documentation•Certified dev procedure
Modeling tools integration
GIS data base Industrial ecologyUrban systems
GUI : Spreadsheets, Matlab
Data Structuring
Technology models data base Energy conversionSharing knowledge
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
OSMOSE : Computer Aided Platform
Flowsheeting tools •BELSIM-VALI•gPROMS•ASPEN plus•HYSYS•Matlab•Simulink•(CITYSIM)•MODELICA•Others possible
•CAPE-OPEN ?•PROSIM•UNISIM ?
Energy technology data base •Data/models interfaces•Simulation•Process integration interface•Costing/LCIA performances•Reporting/documentation•Certified dev procedure
Modeling tools integration
MILP/MINLP models Heat/mass integrationSub systems analysisSuperstructureHEN synthesis models
Optimal control models MILP/ AMPL or GLPKMulti-period problems
Sizing/costing data base LCIA database (ECOINVENT)
Process integration
GIS data base Industrial ecologyUrban systems
GUI : Spreadsheets, Matlab
Data Structuring
Technology models data base Energy conversionSharing knowledge
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
OSMOSE : Computer Aided Platform
Flowsheeting tools •BELSIM-VALI•gPROMS•ASPEN plus•HYSYS•Matlab•Simulink•(CITYSIM)•MODELICA•Others possible
•CAPE-OPEN ?•PROSIM•UNISIM ?
Energy technology data base •Data/models interfaces•Simulation•Process integration interface•Costing/LCIA performances•Reporting/documentation•Certified dev procedure
Modeling tools integration
MILP/MINLP models Heat/mass integrationSub systems analysisSuperstructureHEN synthesis models
Optimal control models MILP/ AMPL or GLPKMulti-period problems
Sizing/costing data base LCIA database (ECOINVENT)
Process integration
Grid computing
Multi-objective optimisation Evolutionary - HybridProblem decompositionUncertainty
Decision support
GIS data base Industrial ecologyUrban systems
GUI : Spreadsheets, Matlab
Data Structuring
Technology models data base Energy conversionSharing knowledge
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
Results analysis
• Graphical representations • Composite curves • Sankey diagrams
• PDF reports • Sensitivity analysis • Pareto curves => data base of solutions • Decision variables values
• Uncertainty analysis => most probable optimal solutions
250
300
350
400
450
500
550
600
0 5000 10000 15000 20000 25000 30000
T(K
)
Q(kW)
Cold composite curve
Hot composite curve
64 66 68 70 72 74 76 78 80 82 84 86 88600
700
800
900
1000
1100
1200
1300
SNG efficiency equivalent εchem [%]
Spec
ific
inve
stm
ent
cost
cG
R [U
SD/k
W]
pressurisation
pressurisation, gas turbine integration,steam drying
& hot gas cleaning
hot gas cleaning & steam dryingwith heat cogeneration
453 K
513 K
5 %
35 %
1073 K
1173 K
573 K
873 K
1 bar
50 bar
573 K
673 K
573 K
673 K
0
1
40 bar
100 bar
623 K
823 K
323 K
523 K
SNG Outputminimal
maximal
min
max
min
max
min
max
Td,in Φw,out Tg Tg,p
pm Tm,in Tm,out yRankine
ps,p Ts,s Ts,u2
0 20 40 60 80 100 1200
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
0.1
SNG proccess configuration number
Pro
babi
lity
to b
e in
top
10 [−
]
Prod. costRes. profitabilityBM Break even
4.3. COMPARISON WITH CONVENTIONAL LCA 45
-500 %
-400 %
-300 %
-200 %
-100 %
0 %
100 %
200 %
scale-independent,conventional LCIA
(average lab/pilot tech.)
without cogenerationIntegrated industrial base case scenario (8 MWth)
with cogeneration
(a) Overall impacts
0
5
10
15
20
25
Remaining processes
Rape methyl ester
NOx emissions
Solid waste
Charcoal
Olivine
Infrastructure
Wood chips production
Electricity cons./prod.
Avoided NG extraction
Avoided CO2 emissions
UBP/
MJ w
ood
scale-independent,conventional LCIA
(average lab/pilot tech.)
harm
ful
bene
ficia
l
harm
ful
bene
ficia
l
harm
ful
bene
ficia
l
without cogenerationIntegrated industrial base case scenario (8 MWth)
with cogeneration
(b) Detailed contribution
Figure 4.3: Comparison of LCIA results obtained by conventional LCA and the proposedmethodology for Ecoscarcity06, single score, with detailed process contributions (Gerberet al. (2011a))
The differences between the conventional LCA and the base case scenario without and witha Rankine cycle are mainly due to the improved conversion efficiency obtained by applyingprocess design method, which has a direct influence on the quantity of SNG produced perunit of biomass. In the base case scenarios, the efficiency is considerably higher becausethe energy integration is optimized in the process design. As a direct consequence, the
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
Case study
PF(E)3 project: Calculation PlatForm for Energy Efficiency and Environmental optimisation
• Started in January 2013, duration of 30 months • Tool for design and decision making based on the notion of cost/benefit • Use of OSMOSE platform from EPFL’s IPESE group
EPFL, IFP Energies Nouvelles, IDEEL, EDF R&D, Armines CEP, …
Context of the project
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
Case study
A city of 100,000 habitants
Resource Price Units Uncertainty
Biomass (wood) 0.05 €/kWh ±30%
Biomass import 0.10 €/kWh ±30%
Natural gas 0.078 €/kWh ±30%
Electricity (grid) 0.20 €/kWh ±30%
Oil 1.05 €/kg ±40%
Diesel 1.80 €/kg ±40%
Petrol 1.88 €/kg ±40%
DME 0.65 €/lit ±50%
FT diesel 0.125 €/kWh ±50%
MeOH 0.323 €/lit ±50%
Services Value Units
Mobility 36.1 pkm/s
Electricity 22,000 kW
Hot water 24,210 kW
Space heating 117,623 kW
Wastewater 0.95 m3/s
Municipal waste 2.23 kg/s
Municipal organic waste 0.51 kg/s
Industry Size Units
Sawmill 100,000 m3 woodin/y
Greenhouse 12,000 tonnes tomatoes produced/y
Laundry service 20,000 tonnes dirty clothes washed/y
DME plant 50 MWthBM
FT plant 100 MWthBM
MeOH plant 50 MWthBM
Input data
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
Case study Models
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
Case study
Urban system
Models
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
Case study
Urban system
Biogas Diesel Electricity Heat
Mobility Natural gas (biogenic) Natural gas (fossil) Oil
Organic waste Petrol Solid waste Industry main product(s)
Wastewater Water Wood biomassUnit U
Required 1 Alternative 2 Alternative 3
Product 4 Product 5
Energy services
Waste treatment
Space heating
Domestic hot water
Electricity Transport
Municipal solid waste
Wastewater
Municipal organic waste
Models
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
Case study
Urban system
Biogas Diesel Electricity Heat
Mobility Natural gas (biogenic) Natural gas (fossil) Oil
Organic waste Petrol Solid waste Industry main product(s)
Wastewater Water Wood biomassUnit U
Required 1 Alternative 2 Alternative 3
Product 4 Product 5
Energy services
Waste treatment
Space heating
Domestic hot water
Electricity Transport
Municipal solid waste
Wastewater
Municipal organic waste
Industries
Sawmill
Greenhouse
Industrial laundry
Biorefineries
Models
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
Case study
Urban system
Biogas Diesel Electricity Heat
Mobility Natural gas (biogenic) Natural gas (fossil) Oil
Organic waste Petrol Solid waste Industry main product(s)
Wastewater Water Wood biomassUnit U
Required 1 Alternative 2 Alternative 3
Product 4 Product 5
Energy services
Waste treatment
Space heating
Domestic hot water
Electricity Transport
Municipal solid waste
Wastewater
Municipal organic waste
Industries
Sawmill
Greenhouse
Industrial laundry
Biorefineries
Available resourcesIndigenous resources
Imported resources
Water
Biomass (wood)
Natural gas
Diesel
Electricity import
Oil
Petrol
Imported biomass (wood)
Models
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
Case study
Urban system
Biogas Diesel Electricity Heat
Mobility Natural gas (biogenic) Natural gas (fossil) Oil
Organic waste Petrol Solid waste Industry main product(s)
Wastewater Water Wood biomassUnit U
Required 1 Alternative 2 Alternative 3
Product 4 Product 5
Energy services
Waste treatment
Space heating
Domestic hot water
Electricity Transport
Municipal solid waste
Wastewater
Municipal organic waste
Conversion technologies
Boilers
Cars
Engines
Turbines
Dryers
Gasifiers
Biogas purification
Biometha-‐nation
Incineration
Wastewater treatment
Synthetic natural gas production
Industries
Sawmill
Greenhouse
Industrial laundry
Biorefineries
Available resourcesIndigenous resources
Imported resources
Water
Biomass (wood)
Natural gas
Diesel
Electricity import
Oil
Petrol
Imported biomass (wood)
Models
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
Case study User-‐interface
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
Business as usual (BAU)
• Space heating1 – Heat recovery from MSWI – 57% fuel oil, 18% natural gas, 14%
biomass (wood), 11% electricity
• Hot water – Assumed same as space heating
• Transport2 – 24% diesel, 76% petrol
• Industries – Biorefineries self-‐sufficient – Others ! heat from natural gas
• Electricity – From MSWI – From grid (UCTE)
Case study Assumptions
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
Business as usual (BAU)
• Space heating1 – Heat recovery from MSWI – 57% fuel oil, 18% natural gas, 14%
biomass (wood), 11% electricity
• Hot water – Assumed same as space heating
• Transport2 – 24% diesel, 76% petrol
• Industries – Biorefineries self-‐sufficient – Others ! heat from natural gas
• Electricity – From MSWI – From grid (UCTE)
Optimisation constraints
• Space heating and hot water • Possible heat recovery from MSWI,
WWTP, industries (refineries, laundry service), organic waste (OW) biogas
• Transport • 56% diesel / petrol • 44% non fossil fuel driven
• Industries • Heat recovery when possible
• Electricity • From MSWI, WWTP, OW biogas • Remainder from grid (UCTE)
Case study Assumptions
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
Case study Result of optimization
Monte Carlo simulation • Prices of resources are subjected to uncertainty • Finding the most resilient solutions
Probability of being the best solution in 800 Monte Carlo simulations
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
Case study Result of optimization
Monte Carlo simulation • Prices of resources are subjected to uncertainty • Finding the probability of being among the top 10 best solutions
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
Case study Result of optimization
Monte Carlo simulation • Prices of resources are subjected to uncertainty • Finding the probability of being among the top 10 best solutions
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
Case study
Total environmental impacts = 492,000 tons of CO2-eq/yr Total operating cost = 226.8 million CHF/yr
Result of optimization (one solution in the set)
The thicknesses represent the size of the connection in “kW” (except for connections to mobility which are scaled up to be visible) Each colour represents one type of connection, e.g. electricity, heat, natural gas, wood, mobility, …
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
Case study Climate change (tons CO2-‐eq/yr)
Electricity (grid)
Petrol
Diesel
Fuel oil
Natural gas
Biomass (import) for biorefineries
Biomass (import) for SNG producpon
Biomass (import) for sawmill
Biomass (indigenous) for sawmill
Climate change (tons CO2-‐eq/y)0 62'500 125'000 187'500 250'000
Oppmised systemBusiness as usual
Impo
rted
resources
Total
0 200'000 400'000 600'000 800'000
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
Case study Climate change (tons CO2-‐eq/yr)
Electricity (grid)
Petrol
Diesel
Fuel oil
Natural gas
Biomass (import) for biorefineries
Biomass (import) for SNG producpon
Biomass (import) for sawmill
Biomass (indigenous) for sawmill
Climate change (tons CO2-‐eq/y)0 62'500 125'000 187'500 250'000
Oppmised systemBusiness as usual
Impo
rted
resources
Total
0 200'000 400'000 600'000 800'000
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
Conclusions and future work
• Data management platform • Data bases • Technology Models => Ontology • Connectivity => Ontology • Geographical information system => SQL • Life Cycle Inventory => SQL • Costing
• Tools integration and handlers • Flowsheeting tools • Routing algorithms • Data analysis and processing • Clustering • Statistical analysis
• MILP problem formulation • Problem decomposition & methods • MILP/MINLP solvers
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
Conclusions
• Workflow organisation => list of tasks • Aggregation/disaggregation • Super-structure definition & models • Optimisation problems • Multi-objective • MI(N)LP
• Parallel computing • Decision support & uncertainty
• Applications • Process design • Process integration & HEN design • Site scale integration • Industrial Symbiosis development • Urban Energy Systems development • Industrial ecology • Region-City scale development • Autonomous systems (Ships, cars)
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
Thank you for your attention!
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
Appendices
EPFL-‐SCI-‐STI-‐FM (IPESE) June 2015 ‹#›
A1. Life Cycle Impact Assessment (LCIA) in Osmose-‐Lua
Source: Gerber, L., Gassner, M., and Maréchal, F. (2011). Systematic integration of LCA in process systems design: Application to combined fuel and electricity production from lignocellulosic biomass. Computers & Chemical Engineering 35, 1265–1280.