Overview of TIMES (MARKAL-EFOM)

Overview of TIMES: Parameters, Primal
Variables & Equations
Uwe Remme
Institute of Energy Economics and the Rationale Use of Energy, Universität Stuttgart
ETSAP Workshop
November 21, 2007, Brasilia
1
Overview
● Introduction to TIMES
● The Reference Energy System (RES):
i. Building blocks of the RES
ii. Characteristics
● Time dimension
i. Time horizon of model analysis
ii. Time segments within a year
● Mathematical formulation:
i. Decision variables.
ii. Basic equations and related input data
iii. Objective function
Uwe Remme Overview of TIMES November 21, 2007 2

Cost and emissions balance ws
f lo ons
Pr i c
es rgy is s i
Ene Em
Energy prices, Resource availability
GDP
Coal processing
Process energy
Domestic Industry
sources Heating area
Refineries
Population
Commercial and
Power plants tertiary sector Light
and
Transportation Communication
Households Power
Demands
CHP plants Person
and district kilometers
heat networks
Imports Transportation Freight
kilometers
Gas network
Cos
Primary energy Cap Final energy ts
a Demand services
citie
s

Development
• By ETSAP
• Implementation in GAMS
• Model generator Methodology
• Bottom-up Model
• Perfect competition
• Perfect foresight (or myopic variant)
TIMES • Optimisation (LP/MIP/NLP)

Min/Max Objective function
(The Integrated MARKAL s.t.
EFOM System) Equations, Constraints
Decision Variables <=> Solution
Input parameters
Features
● Multi-region
● Elastic demands
● Vintaging
● Load curve
● Endogeneous learning
● Discrete capacity expansion
● Macroeconomic linkage
● Stochastic programming
● Trade-off analysis
● Damage functions for external costs of
pollutants
● Climate extension

Model formulation of TIMES
Objective Minimizing discounted system costs
function: =
Sum of
• Import-/Extraction costs,
• variable and fix OM costs,
• Investment costs,
•…
Input data Decision variables

Model equations: Transformation relationship
(→ e.g. Efficiency relationship for power plant)
Cost data
Energy and emission balances
Efficiencies Process activities
Share constraints on input/output side of technology
Full load hours Energy/Emission flows
(→ e.g. max. ratio of electricity to heat for CHP)
Emission factors
Capacity-activity constraint New capacities
Demand (→ e. g. available capacity limits elec gen of power plant)
Fundamental prices
…
Cumulated constraints over time
(→ e. g. available fossil resources)
Peaking constraint
(→ Ensuring reserve capacity at peak load)
Load curve equations
Storage equations (e.g. pump storage)
Scenario specific constraints

(→ e. g. bound on CO2 emissions, quota for renewables)

Linear programming (LP) model
Decision variables (positvie, continuous):
• Activity variables (Production level of technologies)
Optimization problem • Energy flows
• Investment decisions
Objective function
Min c1 x1 + c2 x2 + K + cn x n
Model constraints a11 x1 + a12 x2 + L + a1n xn ≥ b1

(linear constraints):
a21 x1 + a22 x2 + L + a 2 n xn ≥ b2
• Energy/ emission balances M M O M M
• Efficiency relationships
• Utilization constraints ai1 x1 + ai 2 x2 + L + ain xn ≥ bi
• Peaking eqn (reserve capacity)
• GHG mitigation targets, quota for
M M O M
renewables,… am1 x1 + am 2 x2 + L + amn xn ≥ bm
xj ≥ 0
● Standard TIMES model: Linear programming

● Implemented in modeling environment GAMS (General Algebraic Modeling System) for optimization/equilibrium problems
● Solution by interior point solvers (CPLEX, XPRESS)
● Variants of TIMES:
i. Macro economic module -> Non-linear eqns -> Non-linear programming
ii. Block-wise capacity expansion -> Binary variables -> Mixed-integer programming
Results of TIMES model run
● All decision variables accessible in a browsable database (data cube):
i. Energy flows, resource extraction, emissions
ii. Activity levels
iii. Total capacities and new investments
iv. Import, export flows
● Energy balance tables constructed from detailed results
● Total system costs, annual costs (discounted + undiscounted) for each technology
split-up by cost category (variable, fix O&M, investment, …)
● Price information from so-called dual solution, e.g.
i. Long-run marginal costs for energy carriers (prices including capital costs)
ii. Necessary subsidy or cost reduction, so that non-competitive technology will be used
iii. Marginal CO2 abatement costs = CO2 certificate price

Reference energy system
t
ea
End-use
tH
ity
ter
ine
at
tric
tric
wa
m
sectors
sel
he
sol
ke
a
c
2
s
Di s
Ste
Co
Di e
rm
Ele
Ga
om
Ga
LH
as
Conversion
Wa
Lig oal
al g
Ro
a ss
Residential
nite
c
Primary sector
tur
nd
m
rd
Bio
Gas heating
Na
Ha
Wi
Oil
energy
supply Coal processing Oil heating
Elc. heating
Lignite resources Refinery
Local heat grid
Gas processing Gas water boiler

Oil resources
Elc. water boiler
Oil import
Elec sector Elc. heat pump
Coal cond. PP
Coal resources
at
ter
he
at
Lignite cond. PP
wa
he
Coal import
ess
rm
om
Coal IGCC PP
c
Pro
Wa
Commercial
Ro
Gas CC PP Room heat boilers
Gas resources
Wind converter Process heat boilers
Gas import
Warm water boilers
CHP sector
V
NP
Coal CHP
Residual wood
Industry
Coal CHP Industrial boilers
Gas CC CHP
M
Area PV,
M
Transport
PK
TK
Solarthermal, Biomass CC CHP
Energy crops Gasoil car
Gasoline car
Area Wind
2
CH
LH2 car
Electrolysis
Busses
Liquefication
Gasification
Trucks

Representation of a simple technology
Commodities
Coal PC Supercrit. Unit 2005 2010 2020 2030
Process Size MWel 600 600 600 600
Coal Construction time Years 3 3 3 3

Electricity CO2
Lifetime Years 35 35 35 35
Efficiency (LHV) % 46 47 48 50
Flows Flows
Max. availability h/a 7500 7500 7500 7500
Supercrititcal 0.736 kg Spec. Investment

2.2 kWh coal plant (SCP) costs (overnight)
€/kWel 1175 1175 1140 1140
0.27 kgSKE 1 kWh Fix O&M €/(kW a) 40.5 40.5 40.5 40.5
Var. O&M €/MWhel 2.6 2.6 2.6 2.6
Efficiency eqn η SPC ⋅ FLOSCP ,COL = FLOSCP , ELC
Emission eqn ε SCP ,COL ⋅ FLOSCP ,COL = FLOSCP ,CO 2
Activity definition ACTSCP = FLOSCP , ELC
Utilization eqn ACTSCP , ELC ≤ α SCP ⋅ CAPSCP , ELC

Horizontal linkages: Technology chains
LV
ty
ci
tri
ec
d
El
V
V
he
M
H
er
as
C icity
ow
ty
w
ci
p
tri
Appliances
tr
l,
l,
oa
ec
ec
oa
2
O
C
El
El
C
Coal Lighting
imports Coal
Supercritical Grid MV &
transport Transf.
coal plant Urban trains
Domestic Grid HV &
mining Transf.
Industry
Backward Loops
η SPC ⋅ FLOSCP ,COL = FLOSCP , ELC

η SPC ⋅ FLOSCP ,COL = FLOSCP , ELC η SPC ⋅ FLOSCP ,COL = FLOSCP , ELC
η SPC ⋅ FLOSCP ,COL = FLOSCP , ELC η SPC ⋅ FLOSCP ,COL = FLOSCP , ELC η SPC ⋅ FLOSCP ,COL = FLOSCP , ELC η SPC ⋅ FLOSCP ,COL = FLOSCP , ELC ε SCP ,COL η⋅ FLO⋅ FLO
η SPC ⋅ FLOSCP ,COL = FLOSCP , ELC SCP ,COL = FLO
SCP ,COL = FLO SCP , ELC
SCP ,CO 2
SPC
ε SCP ,COL ⋅ FLOSCP ,COL = FLOSCP ,CO 2 ε SCP ,COL η⋅ FLO⋅ FLO
SCP ,COL = FLO
ε SCP ,COL ⋅ FLOSCP,COL = FLOSCP ,CO 2 ε SCP ,COL ⋅ FLOSCP,COL = FLOSCP ,CO 2 SCP ,COL = FLO SCP , ELC
SCP ,CO 2
ε SCP ,COL ⋅ FLOSCP,COL = FLOSCP ,CO 2 ε SCP ,COL ⋅ FLOSCP ,COL = FLOSCP,CO 2 SPC
ε SCP ,COL ⋅ FLOSCP,COL = FLOSCP ,CO 2 ε SCP ,COLACT
⋅ FLO SCP = FLO
SCP ,COL = FLOSCP ,CO 2
SCP , ELC
ACTSCP = FLOSCP , ELC ACTSCP = FLO

ACT ε SCP ,COL≤ ⋅αFLO⋅SCP ,COL = FLOSCP ,CO 2
SCP , ELC
ACTSCP = FLOSCP , ELC ACTSCP = FLOSCP , ELC ACTSCP = FLOSCP , ELC ACTSCP = FLOSCP , ELC SCP , ELC SCP CAP
SCP , ELC
ACTSCP = FLOSCP , ELC ACTSCP = FLOSCP , ELC
ACTSCP , ELC ≤ α SCP ⋅ CAPSCP , ELC ACTSCP , ELC ≤ α SCP ⋅ CAPSCP , ELC
ACTSCP , ELC ≤ α SCP ⋅ CAPSCP , ELC ACTSCP , ELC ≤ α SCP ⋅ CAPSCP , ELC ACTSCP , ELC ≤ α SCP ⋅ CAPSCP , ELC ACTSCP , ELC ≤ α SCP ⋅ CAPSCP , ELC ACTSCP = FLOSCP , ELC
ACTSCP , ELC ≤ α SCP ⋅ CAPSCP , ELC ACTSCP , ELC ≤ α SCP ⋅ CAPSCP , ELC
FLOSCP , ELC = FLODEM , ELC FLOSCP , ELC = FLODEM , ELC FLOSCP , ELC = FLODEM , ELC FLOSCP , ELC = FLODEM , ELC FLOSCP , ELC = FLODEMACT
, ELCSCP , ELC ≤ α SCP ⋅ CAPSCP , ELC

Vertical linkages: Competition
LV
ty
ci
tri
ec
d
El
V
V
he
M
H
er
as
C icity
ow
ty
w
ci
p
tri
Appliances
tr
l,
l,
oa
ec
ec
oa
2
O
C
El
El
C
Supercritical Lighting
Domestic coal plant (exst)
mining Coal Grid MV &
transport Ultrasupercritical Transf.
coal plant (new) Urban trains
Coal Grid HV &
imports Transf.
s
ga
al
Industry
ur
at
Natural gas
N
GT (exist)
Natural gas Competing options to produce

CC (new) electricity:
• between technologies
ite
gn
• between old and new plants

Li
Supercritical Influenced by:

lignite plant (exst)
• Technology costs
Supercritical • Efficiencies
lignite plant (new)
• Emission factors
• Fuel prices…

Vertical linkages: Substitution
t
ea
End-use
tH
Dis ity
ter
ine
at
tric
tric
wa
m
sectors
he
sel
sol
ke
a
c
2
s
Ste
Co
Di e
rm
Ele
om
Ga
Ga
LH
as
Conversion
Wa
Lig oal
al g
Ro
a ss
Residential
nite
c
Primary sector
tur
nd
m
rd
Bio
Gas heating
Na
Ha
Wi
Oil
energy
supply Coal processing Oil heating
Elc. heating
Lignite resources Refinery Subsitution
Local heat grid

options
Oil resources
Elc. water boiler
Oil import
Coal cond. PP
Coal resources
at
ter
he
at
Lignite cond. PP
wa
he
Coal import
ess
rm
om
Coal IGCC PP
c
Pro
Wa
Commercial
Ro
Gas resources
Gas import
Warm water boilers
CHP sector
P
GD
Coal CHP
Residual wood
Industry
Gas CC CHP
M
Area PV,
M
Transport
PK
TK
Gasoline car
Area Wind
2
CH
LH2 car
Electrolysis
Busses
Liquefication
Gasification
Trucks

Interdependencies in the energy system
At optimal solution:
t
ea
End-use
ity
ter
ine
tH
at
• Equilibrium between electricity supply and demand
tric
wa
m
sectors
he
sel
tric
sol
ke
a
c
2
s
Ste
Co
Di e
rm
Ele
om
Ga
Ga
LH
Di s
as
Conversion
Wa
Lig oal
al g
Ro
a ss
Changes inPrimary
the system (e.g. phase-out of nuclear) sector Residential
nite
c
tur
nd
m
rd
Bio
Wi
Na
Gas heating
Ha
Oil
yield new equilibrium
energy e.g.:
1) Missing
supply Coal processing
nuclear substituted by coal (or natural gas, Oil heating
renewables in CO2 reduction scenario)

Elc. heating
2) Increase in electricity price (and CO2 certificate price) Local heat grid
3) Substitution of electricity in the end-use sectors
Oil resources
Elc. water boiler
Oil import
Coal cond. PP
Coal resources
at
ter
he
at
Lignite cond. PP
wa
he
Coal import
ess
rm
om
c
Coal IGCC PP
Pro
Wa
Commercial
Ro
Gas resources
Gas import
Warm water boilers
CHP sector
P
GD
Coal CHP
Residual wood
Industry
Gas CC CHP
M
Area PV,
M
Transport
PK
TK
Gasoline car
2
Area Wind
CH
LH2 car
Electrolysis
Busses
Liquefication
Gasification
Trucks

Interdependencies in the energy system
t
ea
End-use
ity
ter
ine
tH
ss
at
tric
wa
m
sectors
ma
he
tric
sol
ke
a
c
2
s
Ste
Co
Bio
rm
Ele
om
Ga
Ga
LH
Di s
as
Conversion
Wa
Lig oal
al g
Ro
a ss
Residential
nite
c
Primary sector
tur
nd
m
rd
Bio
Wi
Na
Gas heating
Ha
Oil
energy
supply Coal processing Biomass heating
Elc. heating
Local heat grid

Oil resources
Elc. water boiler
Oil import
Coal cond. PP
Coal resources
at
ter
he
at
Lignite cond. PP
wa
he
Coal import
ess
rm
om
c
Biomass IGCC PP
Pro
Wa
Commercial
Ro
Gas resources
Gas import
Warm water boilers
CHP sector
P
GD
Coal CHP
Residual wood
Industry
Gas CC CHP
M
Area PV,
M
Transport
PK
TK
Gasoline car
2
Area Wind
CH
LH2 car
Electrolysis
Busses
Liquefication
Gasification
Trucks

Dynamic model
Model horizon
Period
03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32
Milestoneyear
eat
eat
eat
eat
eat
eat
End-use End-use End-use End-use End-use End-use
tH
tH
tH
tH
tH
tH
ity
ity
ity
ity
ity
ity
ter
ter
ter
ter
ter
ter
e
Lig oal oline
Lig oal oline
e
tricat
tricat
tricat
tricat
tricat
at
coa asolin
coa asolin
coa asolin
solin
ctric
tric
tric
tric
tric
tric
tric
l ga am
l ga am
l ga am
l ga am
l ga am
am
sectors sectors sectors sectors sectors sectors
kwea
kwea
kwea
kwea
kwea
wa
el
el
el
sel
el
sel
leche
leche
leche
leche
leche
he
ke
ies
ies
ies
ies
2
2
as
as
as
as
as
s
s
s
Dis
Dis
Dis
Dis
Dis
Dis
tura Ste
tura Ste
tura Ste
tura Ste
tura Ste
Ste
Co
Co
Co
Co
d Die
Co
Co
Die
Ele
Ga
rm
rm
rm
rm
rm
rm
Ga
Ga
Ga
oEm
oEm
om
oEm
om
om
LH
LH
LH
LH
LH
LH
s
s
nite G
nite G
nite G
nite G
nite G
G
G
Conversion Conversion Conversion Conversion Conversion Conversion
d D
d D
d D
d D
E
E
l ga
Lig oal
l
Wa
Wa
Wa
Wa
Wa
Wa
Ro
Ro
Ro
Ro
Ro
Ro
ss
ss
ss
ss
ss
ss
Residential Residential Residential Residential Residential Residential
tura
c
nite
ma
ma
ma
ma
ma
ma
Primary sector Primary sector Primary sector Primary sector Primary sector Primary sector
d
rd
rd
rd
rd
rd
rd
Win
Win
Win
Win
Win
Win
Bio
Bio
Bio
Bio
Bio
Bio
Gas heating Gas heating Gas heating Gas heating Gas heating Gas heating
Lig
Lig
Lig
Ha
Na
Ha
Na
Ha
Na
Ha
Na
Ha
Na
Ha
Na
Oil
Oil
Oil
Oil
Oil
Oil
energy energy energy energy energy energy
supply Coal processing supply Coal
Oil heating
processing supply Coal
Oil heating
Oil heating
Oil heating
Oil heating
processing Oil heating
Elc. heating Elc. heating Elc. heating Elc. heating Elc. heating Elc. heating
Lignite resources Refinery Lignite resources Refinery Lignite resources Refinery Lignite resources Refinery Lignite resources Refinery Lignite resources Refinery
Local heat grid Local heat grid Local heat grid Local heat grid Local heat grid Local heat grid
Gas processing Gas

Gas
water
processing
boiler Gas
Gas
water
processing
boiler Gas
Gas
water
processing
boiler Gas
Gas
water
processing
boiler Gas
Gas
water
processing
boiler Gas water boiler
Oil resources Oil resources Oil resources Oil resources Oil resources Oil resources
Elc. water boiler Elc. water boiler Elc. water boiler Elc. water boiler Elc. water boiler Elc. water boiler
Oil import Oil import Oil import Oil import Oil import Oil import
Elec sector Elec sector
Elc. heat pump
Elec sector
Elc. heat pump
Elec sector
Elc. heat pump
Elec sector
Elc. heat pump
Elec sector
Elc. heat pump Elc. heat pump
Coal cond. PP Coal cond. PP Coal cond. PP Coal cond. PP Coal cond. PP Coal cond. PP
Coal resources Coal resources Coal resources Coal resources Coal resources Coal resources
t
t
ea
ea
ea
ea
ea
ea
ter
ter
ter
ter
ter
ter
at
at
at
at
at
at
Lignite cond. PP Lignite cond. PP Lignite cond. PP Lignite cond. PP Lignite cond. PP Lignite cond. PP
sh
sh
sh
sh
sh
sh
wa
wa
wa
wa
wa
wa
he
he
he
he
he
he
Coal import Coal import Coal import Coal import Coal import Coal import
ces
ces
ces
ces
ces
ces
rm
rm
rm
rm
rm
rm
om
om
om
om
om
om
Coal IGCC PP Coal IGCC PP Coal IGCC PP Coal IGCC PP Coal IGCC PP Coal IGCC PP
Pro
Pro
Pro
Pro
Pro
Pro
Wa
Wa
Wa
Wa
Wa
Wa
Commercial Commercial Commercial Commercial Commercial Commercial
Ro
Ro
Ro
Ro
Ro
Ro
Gas CC PP RoomGas
heatCC
boilers
PP RoomGas
heatCC
boilers
PP RoomGas
heatCC
boilers
PP RoomGas
heatCC
boilers
PP RoomGas
heatCC
boilers
PP Room heat boilers
Gas resources Gas resources Gas resources Gas resources Gas resources Gas resources
Wind converter Process
Windheat
converter
boilers Process
Windheat
converter
boilers Process
Windheat
converter
boilers Process
Windheat
converter
boilers Process
Windheat
converter
boilers Process heat boilers
Gas import Gas import Gas import Gas import Gas import Gas import
Warm water boilers Warm water boilers Warm water boilers Warm water boilers Warm water boilers Warm water boilers
CHP sector CHP sector CHP sector CHP sector CHP sector CHP sector
P
P
GD
GD
GD
GD
GD
GD
Coal CHP Coal CHP Coal CHP Coal CHP Coal CHP Coal CHP
Residual wood Residual wood
Industry Residual wood
Industry Industry
Coal CHP Industrial
Coalboilers
CHP Industrial
Coalboilers
CHP Industrial
Coalboilers
CHP Industrial
Coalboilers
CHP Industrial
Coalboilers
CHP Industrial boilers
Gas CC CHP Gas CC CHP Gas CC CHP Gas CC CHP Gas CC CHP Gas CC CHP
Area PV, Area PV, Area PV,
M
Area PV,
M
Area PV,
M
Area PV,
M
M
M
Transport Transport Transport Transport Transport Transport
PK
PK
PK
PK
PK
PK
TK
TK
TK
TK
TK
TK
Solarthermal, Biomass CC CHP Solarthermal, Biomass CC CHP Solarthermal, Biomass CC CHP Solarthermal, Biomass CC CHP Solarthermal, Biomass CC CHP Solarthermal, Biomass CC CHP
Energy crops Energy crops Gasoil car Energy crops Gasoil car Energy crops Gasoil car Energy crops Gasoil car Energy crops Gasoil car Gasoil car
Gasoline car Gasoline car Gasoline car Gasoline car Gasoline car Gasoline car
Area Wind Area Wind Area Wind Area Wind Area Wind Area Wind
2
2
CH
CH
CH
CH
CH
CH
LH2 car LH2 car LH2 car LH2 car LH2 car LH2 car
Electrolysis Electrolysis Electrolysis Electrolysis Electrolysis Electrolysis
Busses Busses Busses Busses Busses Busses
Liquefication Liquefication Liquefication Liquefication Liquefication Liquefication
Gasification Gasification Gasification Gasification Gasification Gasification
Trucks Trucks Trucks Trucks Trucks Trucks
2005 2010 2015 2020 2025 2030

Vintaging
Added capacity
γ (v = 2000) = 0.38 = konst.
γ (v = 1990) = 0.35 = konst. Modelyears
2000
2010
1990
Period t 1 Period t 2 Period t 3
● Process can be specified as vintaged one by entry in set PRC_VINT.

● The characteristics of a vintaged process can be distinguished by its vintage
year, e.g. process flow variables have as additional index of the current period
t the construction period v: VAR_FLO(r,v,t,p,c,s)

Time horizon Model horizon
1st period 2nd period 3rd period
94 95 96 97 98 99 00 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15
Pastyear
Milestoneyears
● Different period durations are possible

● Different type of years:
i. MILESTONYR
ii. PASTYEAR
iii. MODLYEAR = MILESTONYR + PASTYEAR
iv. DATAYEAR: years with input data, input data are inter-/extrapolated to
milestoneyears
Past investments
Existing capacity of technology ECOASTM000
100 MW
50 MW
88 89 90 91 92 93 94 95 96 97 98 99 00 01 02 03 04 05 06 07 08 09 10 11 12
Pastyears Milestoneyears
● Specification of existing capacity by past investments in their vintage/past years

(NCAP_PASTI(r,t,prc)):
PARAMETER NCAP_PASTI(REG,ALLYEAR,PRC)
/ WEU.1991.ECOASTM000 100
WEU.1996.ECOASTM000 50 /
● Alternative specification of residual curve for Milestoneyears PRC_RESID(r,t,prc)
PARAMETER PRC_RESID(REG,ALLYEAR,PRC)
/ WEU.2000.ECOASTM000 150
WEU.2005.ECOASTM000 50 /
Foresight
Myopic foresight
Perfect foresight
(Dynamic-recursive)
Model horizon
Period Run 6
Run 5 28 29 30 31 32
03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32
Run 4 23 24 25 26 27
Milestoneyear Run 3 18 19 20 21 22
One optimization run over entire horizon Run 2 13 14 15 16 17
Run1 08 09 10 11 12
03 04 05 06 07
Sequence of model runs
● Perfect foresight:
i. Decisions take into account entire future model horizon
ii. Model gives optimal strategy under assumed conditions
● Myopic foresight:
i. Decisions are based on only limited knowledge of the future
ii. Implicitly assumed that current conditions will last forever; to some extent
ignorant about future
Timeslices
Model horizon
Period 1 Period 2 Period 3 Period 4
Winter
SP SU FA WI Seasons (Saisonal)
Representative year
(milestoneyear)
Winter Week-end
SU_WD
SU_WE
SP_WD
FA_WD
SP_WE
FA_WE
WI_WD
WI_WE
Week day (Weekly)
Winter Week-end Night

SU_WDD
SU_WDN
SP_WDD
SP_WDN
SU_WED
SU_WEN
FA_WDD
FA_WDN
SP_WED
SP_WEN
FA_WED
FA_WEN
WI_WDD
WI_WDN
WI_WED
WI_WEN
Part of the day (Daily)

Reserve capacity in the power sector
Electric load/
Capacity
Secure capactiy
Reserve capacity (com_pkrsv)
Peak load
(reality)
Peak load factor (com_pkflx)
Peak load
due to approximated
load curve Real load curve
Average day load
Average night
load
0 4 8 12 16 20 24
Daytime

Timeslices (2) SPG(Shadow Primary Commodity Group) is
defined as the group of commodities on the
DAYNITE opposite side of the PCG having the same
ANNUAL Process commodity type as the PCG.
DAYNITE Primary commodity
COAHAR group (NRG) SET COM_TSL
ELC
= defining Activity /
WEU.ELC.DAYNITE
...
Daynite Daynite
WEU.COAHAR.ANNUAL
Supercrititcal
coal plant (SCP) /
SET PRC_TSL
Flow has DAYNITE resolution /
because flow(s) of primary ...
Flow has DAYNITE commodity group are modelled at
resolution because WEU.SCP.DAYNITE
PRC_TSL level
PRC_TSL (DAYNITE) ...
is finer than COM_TSL /
of COAHAR
● Modeller specifies process and commodity timeslice levels, timeslice levels of commodtiy balances
and process flows are deteremined according to the following rules:
● Commodity timeslice level COM_TSL ⇒ timeslice level of commodity balance equation (default
ANNUAL)
● Process timeslice level PRC_TSL ⇒ timeslice level of activity variable and corresponding flow
variables (default ANNUAL)
● All other flows are modelled at the finest level of COM_TSLevel of the SPG or PRC_TSL.

Multi-regional TIMES model
Export Import
region region ● Inter-regional exchange
process between two internal
Aluminium regions similar to
Coal Gas
import/export process; thus:
i. easy linkage of different
Steel
regions
Region 1 Region 2
ii. modelling of trade
Money
● Trade processes can be
Oil described similar to regular
technologies, e.g. capacities,
Electricity
Inter-regional investment costs, losses, etc.
exchange process
Region 3
Area of study

SET PRC
/ Defining processes
...
SSCCDRFLX00 Process name
...
OILDST
/;
SET TOP (REG,PRC,COM,IO)
OILLPG
/
OILCRD SSCDRFLX00
...
OILGSL
WEU.SSCCDRFLX00.OILCRD.IN
WEU.SSCCDRFLX00.OILDST.OUT Topology OILKER
WEU.SSCCDRFLX00.OILLPG.OUT
...
/;
SET COM_GRP
SET PRC_CAPUNT(REG,PRC,CG,UNITS_CAP)
/ Capacity unit
/
...
User defined ...
SSCDRFLX00_NRGO
commodity group WEU.SSCCDRFLX00.SSCCDRFLX00_NRGO.PJA
...
...
/;
/; PCG is
SET COM_GMAP(REG,CG,COM) SSCCDRFLX00
SET PRC_ACTUNT(REG,PRC,CG,UNITS_ACT)
/ and activity unit
/
... PJ
Members of ...
WEU. SSCDRFLX00_NRGO.OILDST
WEU.SSCCDRFLX00.SSCCDRFLX00_NRGO.PJ
WEU. SSCDRFLX00_NRGO.OILLPG commodity group ...
... No Conversion
/;
/; factor needed
PARAMETER PRC_ACTFLO(REG,ALLYEAR,PRC,CG)
SET PRC_CG(REG,PRC,CG)
/
/
/;
...
WEU. SSCCDRFLX00.SSCDRFLX00_NRGO
PARAMETER PRC_CAPACT(REG,PRC)
...
/ Conversion
/; Commodity groups WEU.SSCCDRFLX00 1.0 factor is 1
related to /;
SSCCDRFLX00
Process definition in VEDA-FE
OILDST
OILLPG
OILCRD SSCDRFLX00
OILGSL
OILKER
Capacity unit: PJ/a

Topology Activity unit: PJ
Activity = sum of all Efficiency from
eNRGy output flows output flows NRGO
to crude oil input
~FI_ST: TCH, PRC, PRE
~FI_UT: TCAP=PJ/a; TACT=PJ
Existing Flexible Refinery ~FI_T

Share~U
Sets TechName TechDesc Comm-IN Comm-OUT EFF~NRGO Life BNDACT~UP BNDACT~UP~2050 ENVACT
P
\D_TC: PCG=NRGO; SSCDRFLX00 Flexible Refinery OILCRD 1.05 50
OILLPG 0.5000
OILGSL 0.5000
OILKER 0.5000
OILDST 0.5000

Basic equations and decision variables
Electricity
Coal (COL) (ELC) CO2 Input parameter
ηt , SPC ,ts Plant efficiency
FLOt , SCP ,COL,ts ACTt , SCP ,ts εt,SCP,COL,CO2,ts CO2 Emission factor
FLOt,SCP,CO2,ts α t , SCP ,ts Availability in time segment ts
Supercrititcal α t , SCP , ANNUAL Annual availability
coal plant (SCP)
cap_past t,SCP Existing capacity
FLOt , SCP , ELC ,ts Δts
NCAPt , SCP Duration of time segment ts
t Index model period
ts Index time segment
Basic process equations
Decision variables
Efficiency eqn ηt , SPC ,ts ⋅ FLOt , SCP ,COL,ts = FLOt , SCP , ELC ,ts
Emission eqn εt,SCP,COL,CO2,ts ⋅ FLOt,SCP,COL,ts = FLOt,SCP,CO2,ts Process flows

FLOt,SCP,COL,ts , FLOt,SCP,ELC,ts , FLOt,SCP,CO2,ts
Activity definition ACTt , SCP ,ts = FLOt , SCP , ELC ,ts Activity ACTt,SCP,ts
Capacity-Activity ∑ ACT
ts
t , SCP ,ts ≤ α t , SCP , ANNUAL ⋅ CAPt , SCP
New capacity
constraints NCAPt,SCP
ACTt , SCP ,ts ≤ α t , SCP ,ts ⋅ CAPt , SCP ⋅ Δts
Total capacity
CAPt , SCP = cap_pastt,SCP + ∑ cptyrv ,t , SCP ⋅ NCAPv , SCP CAPt,SCP
Capacity definition v <t

Variables vintage
actual
Associated bound
parameters
period
region
● Process oriented: period
process
timeslice lo,up or fx
i. VAR_ACT(r,v,t,p,s): activity of a process COAHAR ELC ACT_BND(r,t,p,s,l)

VAR_FLO
ii. VAR_CAP(r,v,p): installed capacity = VAR_ACT
previous investments Coal plan CAP_BND(r,v,p,l)

SCP
+ new investments VAR_NCAP
+ past investments still existing
VAR_FLO
iii. VAR_NCAP(r,v,p): new investment in period v NCAP_BND(r,v,p,l)
● Flow oriented:
i. VAR_FLO(r,v,t,p,c,s): flow level of commodity c linked to process p

imp or exp IRE_BND(r,t,c,s,all_reg,ie,l)
ii. VAR_IRE(all_reg,v,t,p,c,s,ie): inter-regional exchange variable IRE_XBND(all_reg,t,c,s,ie,l)
iii. VAR_SIN / VAR_SOUT(r,v,t,p,c,s): flows entering/leaving a process p storing a

commodity c
STG_IN/OUTBND(r,t,c,s,all_reg,ie,l)
Variables contd.
Associated bound
parameters
● Commodity oriented (only created if bound provided):
i. VAR_COMPRD(r,t,c,s): total production of a commodity COM_BNDPRD(r,t,c,s,l)
ii. VAR_COMCON(r,t,c,s): total consumption of a commodity COM_BNDCON(r,t,c,s,l)
iii. VAR_COMNET(r,t,c,s): net level of a commodity COM_BNDNET(r,t,c,s,l)

(production – consumption)
● Blending variables
i. VAR_BLND(r,t,ble,opr): amount of blending stock opr needed
for the production of blending product ble

Basic equations
● EQ(l)_COMBAL r,t,c,s Commodity balance
● EQ_ACTFLO r,v,t,p,s Definition of activity variable
● EQ_CAPACT r,v,t,p,s Utilization constraint
● EQ_PTRANS r,v,t,p,cg1,cg2,s Transformation equation
● EQ(l)_INSHR/OUTSHR r,t,p,c,cg,s Share constraints on in/output side of

process
● EQ_OBJ Objective function

Commodity balance equation
COM Commodity type MATerial, FINancial
=
≥ Commodity type NRG, ENV, DEM
Imports Exports
VAR_IRE VAR_IRE
Production Consumption Consumption
VAR_FLO VAR_FLO
Production
Blending
Blending
VAR_BLND, emissions from blending
VAR_BLND
ENV_BL(r,com,ble,opr,t) or energy input
for blending BL_INP(r,ble,c)
Output storage Input storage
VAR_SOUT VAR_SIN
Projected demand
COM_PROJ(r,t,c)
Capacity related commodity flows Capacity related commodity flows
NCAP_COM(r,v,p,com,‘out‘) NCAP_COM(r,v,p,com,‘in‘)
Auxiliary emission of exchange process Auxiliary production of exchange process

IRE_FLOSUM IRE_FLOSUM
Commodity released at decommissioning Commodity required at investment

NCAP_OCOM(r,t,p,c) NCAP_ICOM(r,v,p,c)

Commodity balance equation contd.
● Commodity balance is created for timeslices on timeslice level specified by COM_TSL or
individual timeslices given by COM_TS (note: then commodity is only available in COM_TS
timeslices)
● Commodity efficiency COM_IE: Commodity production × COM _ IE ≥ Commodity consumption
● Annual demand given by COM_PROJ(r,t,c) Electric load/
Capacity
● Load curve of demand described by COM_FR(r,t,c,s):
Secure capactiy
PARAMETER COM_FR Reserve capacity (com_pkrsv)
/ Peak load
(reality)
Peak load factor (com_pkflx)
WEU.2000.RH.ID 0.12000000 Peak load
due to approximated
WEU.2000.RH.IN 0.06000000 load curve Real load curve
Average day load
WEU.2000.RH.SD 0
WEU.2000.RH.SN 0
WEU.2000.RH.WD 0.54670000 Average night
load
WEU.2000.RH.WN 0.27330000
/ 0 4 8 12 16 20 24
Daytime
● Cumulative commodity bounds between two periods t1 and t2:
● COM_CUMNET(r,t1,t2,c,l) limit on net amount of commodity
● COM_CUMPRD (r,t1,t2,c,l) limit on production of commodity
Transformation equation
● Transformation equations establishes relationship between the flows of two commodity groups.
● Example 1: Efficiency of coal plant ECOASTM000
COAHAR ELC
Coal plant
ECOASTM000
VAR_FLOCOAHAR VAR_FLOELC
Efficiency
FLO_FUNCr,t,ECOASTM000,COAHAR,ELC, s = 0.33
EQ_PTRANS r,v,t,ECOASTM000,COAHAR,ELC,s
Transformation
equation
FLO_FUNCr,t,ECOASTM000,COAHAR,ELC , s × VAR_FLOr,v,t,ECOASTM000,COAHAR,s = VAR_FLOr,v,t,ECOASTM000,ELC,s
Process 2nd commodity

1st commodity group
group
Transformation equation contd.
● Example 2: Simple refinery RFLX00
OILCRD Primary Commodity Group OILDSL OILGSL

NRGO
Refinery
RFLX00
1st commodity is now on the output side:

FLO_FUNC appears always on the side of
Efficiency FLO_FUNCr,t,RFLX00,NRGO,OIL,s = 1 the process indicated by the first
commodity group!!
EQ_PTRANS r,v ,t,RFLX00,NRGO,OILCRD,s
FLO_FUNC r,t,RFLX00,NRGO,OILCRD,s × (VAR_FLOr,v ,t,RFLX00,OILDSL,s + VAR_FLOr,v ,t,RFLX00,OILGSL,s ) = VAR_FLOr,v,t,RFLX00,OILCRD,s
Process 1st commodity 2nd commodity

group group

● Example 3: Oil/Gas power plant
Overall Efficiency
FLO_FUNCr,t,EGASOIL000,NRGI,ELC,s = 1
OILDST GASNAT Commodity group ELC
NRGI Fuel-dependent Efficiency
FLO_SUM r,t,EGASOIL000,NRGI,OILDST,ELC,s = 0.35
EGASOIL000
FLO_SUM r,t,EGASOIL000,NRGI,GASNAT,ELC,s = 0.32
Process 2nd commodity

1st commodity Member of group
group 1st commodity
group
EQ_PTRANS r,v,t,EGASOIL000,NRGI,ELC,s
⎛VAR_FLOr,v,t,EGASOIL000,ELC,s × FLO_SUM r,t,EGASOIL000,NRGI,OILDST,ELC,s ⎞

⎜ ⎟
FLO_FUNCr,t,EGASOIL000,NRGI,ELC,s × ⎜ ⎟ = VAR_FLOr,v,t,EGASOIL000,ELC,s
⎜ + VAR_FLO ⎟
⎝ r,v,t,EGASOIL000,ELC,s × FLO_SUM r,t,EGASOIL000,NRGI,GASNAT,ELC,s ⎠

● Two possible cases:
CG1 CG2 CG1 CG2
C1 C2 flosum r,v,t,p,CG1,C1,CG2,s C3 C4 C1 C2 flosum r,v,t,p,CG2,C4,CG1,s C3 C4
flofunc r,v,t,p,CG1,CG2,s flofunc r,v,t,p,CG2,CG1,s
Prozess p Prozess p
Commodity Commodity
specific flosum r,v,t,p,CG2,C3,CG1,s specific
flosum r,v,t,p,CG1,C2,CG2,s efficiencies
efficiencies
on the input side on the output side
⎛ flosumr,v,t,p,CG1,C1,CG2,s FLOr ,v,t,p,C1,s ⎞
flofuncr,v,t,p,CG1,CG2,s ⎜⎜ ⎟
⎟ FLOr ,v,t,p,C1,s + FLOr ,v ,t,p,C2,s
⎝ + flosum r,v,t,p,CG1,C2,CG2,s FLO r , v , t,p,C2 , s ⎠
= =
⎛ flosum r,v,t,p,CG1,C3,CG2,s FLOr,v,t,p,C3,s ⎞
FLOr,v,t,p,C3,s + FLOr,v,t,p,C4,s flofuncr,v,t,p,CG2,CG1,s ⎜⎜ ⎟
⎟
⎝ + flosum r,v,t,p,CG1,C4,CG2,s FLO r,v,t,p,C4,s ⎠
● It is not possible to use commodity specific efficiencies on both sides of the process at the same time!
● EQ_PTRANS is created on the finer timeslice level of SPG or PRC_TSL. SPG (Shadow Primary
Commodity Group) is defined as the group of commodities on the opposite side of the PCG having the
same commodity type as the PCG.

Definition of activity variable
● Activity of a process equals the sum of the flows specified in the Primary
Commodity Group (PRC_ACTUNT).
● Activity variable is created on the timeslice level specified by PRC_TSL.
OILCRD Primary Commodity Group OILDSL OILGSL

NRGO
Refinery
RFLX00
EQ_ACTFLOr,v,t,RFLX00,s
VAR_ACTr,v,t,RFLX00,s = VAR_FLOr,v,t,RFLX00,OILDSL,s + VAR_FLOr,v,t,RFLX00,OILGSL,s

Capacity utilization constraint
EQ(l)_CAPACT r,v,t,p,s
≤
VAR_ACT(r,v,t,p,s) Available capacity of process p in period t and timeslice s
=
Capacity:
timeslice s member of
• Past investments before model
PRC_TSL
horizon: NCAP_PASTI
• New investments from previous
periods
• New investments in actual period t
Three availabilities:
• NCAP_AFA(r,v,p,l) : Annual availability

• NCAP_AFS(r,v,p,s,l) : Seasonal availability
• NCAP_AF(r,v,p,s,l) : Availability in timeslices
which can be combined.

Capacity utilization constraint contd.
● Effective capacity of a new investment with construction time in construction period:
TIMES assumes that 80 MW
NCAP_ILED = 2 are existing during the entire
period!
VAR_NCAP = 100 MW 80 MW
Period duration=10a Period duration=10a
● Similar assumption for capacity being decommissioned:
VAR_NCAP = 100 MW 70 MW
Period duration=10a Period duration=10a

Share constraints on the input/output of a process
● Possibility to limit the share of a commodity flow within a commodity group on the input or output
side of a process.
● Example
OILCRD Commodity Group OILDSL OILGSL
NRGO
Refinery
RFLX00
FLO_SHARr,t,FRLX 00 ,NRGO,OILDSL,s ,UP = 0.3 ● Fixed, upper or lower bounds may be

specified.
Process Commodity
Commodity ● Commodity group must not necessarily
group defining the total flow compromise all output/input flows, one can
EQ(l)_IN/OUTSHR identify a subgroup as commodity group.
∑VAR_FLO r,v,t,RFLX00,OILDSL,s
v
≤ FLO_SHARr,t,RFLX00,NRGO,OILDSL,s,UP
∑ (VAR_FLOr,v,t,RFLX00,OILDSL,s + VAR_FLOr,v,t,RFLX00,OILGSL,s )
v

Objective function: Minimizing total system costs
● Objective function = total discounted energy system costs over the entire model horizon
● Typical cost components in the objective function:

Electricity
i. Variable O&M costs in period t Coal (COL) (ELC) CO2
K + varomt,SCP ⋅ ACTt , SCP ,ts ⋅ Δt + K FLOt , SCP ,COL,ts ACTt , SCP ,ts
FLOt,SCP,CO2,ts
Supercrititcal
ii. Fixed O&M costs in period t coal plant (SCP)
FLOt , SCP , ELC ,ts
NCAPt , SCP
K + fixomt,SCP ⋅ CAPt , SCP ,ts ⋅ Δt + K
iii. Levelized investment costs in period t (capital recovery factor crf depends on economic lifetime
and/or technology specific discount rate)
K + crf t,SCP ⋅ invcostt,SCP ⋅ NCAPt,SCP,ts + K
iv. Costs/Revenues from imports/exports in period t

K + pricet,COL, IMP ⋅ IMPt ,COL ⋅ Δt + K
● Costs in real terms

● General discount rate used to discount costs from different periods to base year
● Investment costs spread over construction time to mimick interest cost during construction
Objective function
● Discounted sum of the annual costs minus revenues:
+ Investment costs
Construction
+ Costs for sunk material during construction time
+ Variable costs
+ Fix operating and maintenance costs
Operation
+ Imports
+ Taxes
+ Surveillance costs
Decommissioning
+ Decommissioning costs
Operation - Subsidies
- Exports
Decommissioning - Recuperation of sunk material

- Salvage value
● Distinction between technical and economic lifetime
● General discount rate (discounting to base year) and technology specific
discount rate (calculating annuities)
● Investment and decommissioning lead-times
Objective function
Annualizing of the investment costs
Annualized
costs
Model years
Construction Technical lifetime= Decommissioning
time Economic lifetime time
t1= v vintage Period t2 Period t3 Period t4

Technical lifetime
Investment costs Decommissioning costs

Summary of basic model equations
⎡∑∑ ⋅ varom y,p,ts ⋅ ACT y , p ,ts + ∑ crf y,p ⋅ invcost y,p ⋅ NCAPy,p + ∑ fixom y,p ⋅ CAPy,p ⎤
● Objective function ⎢ p ts p p ⎥
Min ∑ disc y ⎢ ⎥
⎢ ⎥
⎢ + ∑∑ impprice y ,c ,ts ⋅ IMPy ,c ,ts − ∑∑ expprice y,c,ts ⋅ EXPy ,c ,ts + ∑∑∑ flocost y , p ,c ,ts ⋅ FLO y , p ,c ,ts ⎥
y
⎣ c ts c ts c p ts ⎦
● Commodity balance ∑ ∑ FLO

p∈Production ts
t , p , c ,ts + ∑ IMPt ,c ,ts =
ts
∑ ∑ FLO
p∈Consumption ts
t , p , c ,ts + ∑ EXPt ,c ,ts
ts
● Transformation equation ηt , p ,cin ,cout ,ts ⋅ FLOt , p ,cin ,ts = FLOt , p ,cout ,ts
FLOt , p ,com,ts
● Input/Output shares on process flows ≤ (=, ≥ ) floshart , p ,com,cg ,ts ,bd
∑ FLO
c∈cg
t , p , c ,ts
● Activity definition ACTt , p ,ts = FLOt , p ,c ,ts
● Utilization constraints ACTt , p ,ts ≤ α t , p ,ts ⋅ CAPt , p
∑ FLO t , p , c , ts
● Market share constraints ts
≤ (=, ≥ ) mrkshrt , p ,c ,bd
∑ ∑ FLO
p∈Production / Consumption ts
t , p , c ,ts

Further model equations
● Description of exchange processes in multi-regional models
● Elastic demands
● Product/market share constraints, e.g. share of hydrogen cars in total person kilometer
demand
● Peaking equation: Ensures enough available secure capacity during peak demand
● Storage equation: Modeling of storage between timeslices (e.g. pump storage) or
between periods (e.g. stockpiling)
● Commodity-specific availabilities, e.g. full load hours of CHP plant in backpressure and
condensing mode
● User constraints: Flexible framework to formulate constraints being not part of the
standard portfolio of TIMES equations, e.g. growth constraints, renewable quota

Model variants and extensions
● Block-wise formulation of new investments (lumpy Damage functions
investments)
Cost of damage
● Endogenous technological learning for that pollutant
b
c>
=
pe
● Damage functions for external costs of pollutants
Slo
● Sensitivity analysis algorithm = b>
a
pe
Sl o
● Stochastic programming =a
Slope
● MACRO economic extension Total Emission

Threshold of one pollutant
● Climate module
Climate Module – Main features Stochastic TIMES – Example
High Growth
Non-CO2 gases:
Conversion to CO2-eq exogenous forcing 0.4
High Mitigation
CO2 EMISSIONS CONCENTRATION RADIATIVE GLOBAL MEAN 0.5
(sum of flows) (stock GtC) FORCING (W/m2) TEMP INCREASE
(°C) 0.6 Low Growth
From TIMES Linear 3 reservoirs One log equation
processes Atmosphere Linear 2 reservoirs
Atmosphere & 0.5 High Growth
Biosphere &
ocean surface ocean surface 0.5
Deep ocean Deep ocean
Low Mitigation
Environmental constraints Reporting parameters / 0.5
endogenous variables
Radiative forcing
sensitivity to CO2 Temperature
Stage 1 Stage 2 Low Growth Stage 3
concentration: sensitivity to CO2
concentration:
γ = 4.1 W/m2
High uncertainty 2000 2005 2010 2015 2020 2030 2040 2050
γ = 5.35•ln(2) W/m2

Overview of TIMES (MARKAL-EFOM)

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Overview of TIMES (MARKAL-EFOM)

Uploaded by

Copyright:

Available Formats

Overview of TIMES: Parameters, Primal

Variables & Equations

Uwe Remme Overview of TIMES November 21, 2007 2

Uwe Remme Overview of TIMES November 21, 2007 3

TIMES • Optimisation (LP/MIP/NLP)

Uwe Remme Overview of TIMES November 21, 2007 4

Input data Decision variables

Scenario specific constraints

Uwe Remme Overview of TIMES November 21, 2007 5

Model constraints a11 x1 + a12 x2 + L + a1n xn ≥ b1

● Standard TIMES model: Linear programming

Uwe Remme Overview of TIMES November 21, 2007 7

Gas processing Gas water boiler

Uwe Remme Overview of TIMES November 21, 2007 8

Coal Construction time Years 3 3 3 3

Supercrititcal 0.736 kg Spec. Investment

Var. O&M €/MWhel 2.6 2.6 2.6 2.6

Efficiency eqn η SPC ⋅ FLOSCP ,COL = FLOSCP , ELC

Emission eqn ε SCP ,COL ⋅ FLOSCP ,COL = FLOSCP ,CO 2

Activity definition ACTSCP = FLOSCP , ELC

Utilization eqn ACTSCP , ELC ≤ α SCP ⋅ CAPSCP , ELC

Uwe Remme Overview of TIMES November 21, 2007 9

η SPC ⋅ FLOSCP ,COL = FLOSCP , ELC

ACTSCP = FLOSCP , ELC ACTSCP = FLO

Uwe Remme Overview of TIMES November 21, 2007 10

Natural gas Competing options to produce

• between old and new plants

Supercritical Influenced by:

Uwe Remme Overview of TIMES November 21, 2007 11

Gas processing Gas water boiler

Uwe Remme Overview of TIMES November 21, 2007 12

renewables in CO2 reduction scenario)

Uwe Remme Overview of TIMES November 21, 2007 13

Gas processing Gas water boiler

Uwe Remme Overview of TIMES November 21, 2007 14

Lig oal oline

Lig oal oline

Gas processing Gas

2005 2010 2015 2020 2025 2030

Uwe Remme Overview of TIMES November 21, 2007 15

γ (v = 2000) = 0.38 = konst.

γ (v = 1990) = 0.35 = konst. Modelyears

Period t 1 Period t 2 Period t 3

● Process can be specified as vintaged one by entry in set PRC_VINT.

Uwe Remme Overview of TIMES November 21, 2007 16

1st period 2nd period 3rd period

● Different period durations are possible

● Specification of existing capacity by past investments in their vintage/past years

One optimization run over entire horizon Run 2 13 14 15 16 17

Winter Week-end Night

Uwe Remme Overview of TIMES November 21, 2007 20

Reserve capacity (com_pkrsv)

Uwe Remme Overview of TIMES November 21, 2007 21

Uwe Remme Overview of TIMES November 21, 2007 22

Uwe Remme Overview of TIMES November 21, 2007 23

Capacity unit: PJ/a

Existing Flexible Refinery ~FI_T

Uwe Remme Overview of TIMES November 21, 2007 25

Emission eqn εt,SCP,COL,CO2,ts ⋅ FLOt,SCP,COL,ts = FLOt,SCP,CO2,ts Process flows

Uwe Remme Overview of TIMES November 21, 2007 26

i. VAR_ACT(r,v,t,p,s): activity of a process COAHAR ELC ACT_BND(r,t,p,s,l)

previous investments Coal plan CAP_BND(r,v,p,l)

i. VAR_FLO(r,v,t,p,c,s): flow level of commodity c linked to process p