PRIMES model Design and Features European Bridges of

1
PRIMES modelDesign and Features

• European Bridges of Knowledge Program
• Energy Policy of the EU and implications for
  Turkey
• Ankara 20/06/2003
• Dr. L. Mantzos
• E3M-Lab / ICCS-NTUA
• contact Kapros_at_central.ntua.gr

2
History

• PRIMES Outcome of JOULE research projects
• Focus of model design
• market mechanisms
• modularity for demand and supply
• detailed technology representation
• Structural formulation consistent with
• engineering evidence and
• economic optimisation behaviour of each economic
  agent acting in the energy market
• Older models
• EFOM, MARKAL (global optimisation) lacked market
  mechanisms and individual behaviour
• MIDAS (econometric) lacked engineering evidence
  in the demand side
• PRIMES in the stream of models developed in the
  US
• IFFS, NEMS (US DOE)
• but also the simpler models GEMS, GEMINI, ENPEP
• Characterised as
• Partial Equilibrium, or generalised equilibrium
  model for the energy system

3
History

• Development started in 1994
• several versions of the sub-models
• integration proved complex (algorithm, economic
  equilibrium paradigm, consistency between
  sub-models)
• PRIMES ver. 1, operational early 1997
• extensively used (March-October 1997)
• evaluation of policies and measures of the EU for
  the Kyoto conference for climate change
• based on that experience
• Development of PRIMES ver. 2
• completely new design of the sub-models and
  interfaces
• further integration between centralised and
  independent power and steam production
• major change in the model mathematical
  formulation
• Non-linear mixed complementarity formulation
• Solution in GAMS/PATH
• Advantages Completeness, Consistency of
  interactions, introduction of non-linearities

4
PRIMES model overview (1)

• An Energy-System Model covering market-driven
  behaviour of energy/economic agents
• Solving for the whole energy system
• Modular structure
• Economic decision of agents / Price-driven
  clearing of energy markets
• Explicit technologies in both demand and supply
• Environment integrated when emission constraints
  apply to the whole energy system the model
  suggests least cost allocation of effort to
  agents
• Dynamic model includes vintages of equipment
• Long term 2030
• Covers all EU15 member states, 13 EU candidate
  countries, Norway and Switzerland, individually

5
PRIMES model overview (2)

• Produces long term (up to 2030) projections of
• production, imports, conversion, consumption and
  prices of energy
• investments, technology choice and cost of
  policies
• given exogenous assumptions for
• macroeconomic and financial factors
• world energy markets
• resources, technologies and costs
• behavioral and technology choice characteristics
  of the different energy agents
• Linked to
• POLES model (IEPE - world)
• GEM-E3 model (NTUA - economic growth)
• PRIMES-Refinery model (IFP - refineries)

6
PRIMES model overview (3)

• PRIMES integrates two levels
• sub-models each represents the demand and/or
  supply economic behaviour of an agent acting in
  the energy market
• market integration level exchange of price and
  quantity signals. Determines prices/quantities of
  equilibrium that balance all energy markets
  simultaneously
• Economic behaviour considers the influence of
  policies and regulation including the environment
• Some of the markets clear at national level,
  others may clear at the EU-wide level
• Dynamic simulation, time forward, myopic
  anticipation assumption

7
PRIMES model overview (4)

• Demandf(price)
• SupplyDemand
• PriceInverse function (Supply)

• Detailed engineering-oriented demand sub-models
  mimic the economic behaviour of the consumer
• Complex engineering model optimising energy
  supply sub-system
• Expresses financial and pricing attitudes of
  suppliers reflecting market competition regimes

The above can take any mathematical form
8
PRIMES model overview (5)

• Technology dynamics
• vintages
• penetration of new technologies
• competition between generations
• inertia from past structures and
• rhythm of capital turnover
• Explicit technologies in demand and supply
• Chronological load to synchronise electricity,
  steam, renewables, pipeline fuels (gas)
• in both demand and supply
• Non-linearities
• Economies of scale
• Learning by doing
• Consumer acceptance

9
Short description of PRIMES model (3)
10
PRIMES Integration

• Non-linear mixed complementarity approach
• Original formulation of sub-models NLP
• Transformation into MCP
• Objective function replaced by KKT first order
  conditions
• Constraints as before
• MCP problem is a system of non-linear
  inequalities
• Integrated model Single system of inequalities
• Sub-models in MCP
• Demand/supply equality constraints
• Prices, linked to market regime and marginal
  costs
• Global environmental constraints

11
PRIMES Integration

• Consistency demonstrated in the theoretical
  literature
• Model solution leading to the maximisation of
  consumer and producer surplus
• Powerful algorithm (GAMS/PATH) facilitates
  solution even in the presence on non-linearities,
  and very large models
• No Gauss-Seidel or Jacobi as in IFFS and NEMS
• No flip-flops
• Abandon of linear programming and related
  limitations
• Possibility to introduce non-linear cost curves,
  technology dynamics, pressures on use of
  capacities

12
Reporting of PRIMES model

• Model report files include in full detail
• Demographic assumptions
• Macroeconomic and sectoral assumptions
• International fuel prices assumptions
• Transport activity results by mode (both for
  passenger and freight transport)
• Energy production and net imports
• Energy conversion in power plants, CHP plants,
  district heating plants, refineries, etc
• Energy consumption by sector and fuel
• Fuel prices by sector and fuel
• CO2 emissions by sector and fuel

13
Policy issues covered

• PRIMES supports policy analysis in the following
  fields
• standard energy policy issues security of
  supply, strategy, costs etc
• environmental issues
• pricing policy, taxation, standards on
  technologies
• new technologies and renewable sources
• energy efficiency in the demand-side
• fuel efficiency and modal split in transport
• alternative fuels
• conversion decentralisation, electricity market
  liberalisation

14
The demand side in PRIMES (1)

• Industry 9 sectors according to EUROSTAT Energy
  Balances definitions further decomposed to
  sub-sectors, for each one different energy uses
  defined
• Households decomposition along typical patterns
  of household energy/technology behaviour
• Tertiary decomposition along types of services
  (market services, non-market services, trade),
  agriculture
• Transport decomposition along passenger and
  freight transport
• Passenger transport private cars, motorcycles,
  public road transport, rail, aviation, inland
  navigation
• Freight transport trucks, rail, inland
  navigation
• Fuels detail at the level of EUROSTAT Energy
  Balances
• Alternative technologies defined at the level of
  energy uses

15
The demand side in PRIMES (2)

• Structure of the demand side model

16
The demand side in PRIMES (3)

• Minimise total energy and environmental costs
  subject to
• Total useful energy needs
• Energy use capacities
• Technology availability
• Emissions constraints
• Through
• Changes in the fuel mix
• Capacity replacement
• Technological choice
• Pollution permits and other
• Three types of mechanisms are considered
  simultaneously
• Economic optimality
• Dynamics I.e. constraints from existing capacity
• Gradual market penetration and acceptance
• Different formulation by sector so as to reflect
  inherent characteristics

17
The demand side in PRIMES (4)
18
Power and steam generation in PRIMES (1)

• Electricity and steam generation in PRIMES
• Three different types of generators considered
  utilities, industrial autoproducers, other
  generators
• Different characteristics and decisions
• Economies of scale, market privileges
• Installed capacity categorised in 45 different
  plant types
• Capacity expansion 88 different plant types for
  new plants (technical and economic
  characteristics evolve over time) possibility
  for re-powering of existing plants
• Chronological load curves synchronisation of
  four loads demand of electricity/steam,
  intermittent, fuel pricing
• Simultaneous decision on electricity/steam
  production
• Strategic capacity expansion problem
• Operational plant selection and utilisation
  problem
• Cost evaluation and pricing policy

19
Power and steam generation in PRIMES (2)Design
Principles

• ENTITIES
• existing plants, candidate plants (not discrete)
• network nodes and links
• companies
• exchange contracts
• fuel contracts and prices
• intermittent sources
• abatement
• LOAD (chronological)
• synchronization of four loads electricity,
  steam, intermittent, fuel pricing

• Generic Code
• The model code accommodates for different
  structural features
• Dynamics
• possibility for myopic anticipation or perfect
  foresight
• Regions
• possibility for single country or multiple
  countries runs
• Non linearity
• technical-economic features, economies of scale,
  learning by doing

20
Power and steam generation in PRIMES (3)The
plants

• Technology types
• 8 conventional thermal according to thermodynamic
  cycle
• 6 GTCC technologies
• 6 clean coal
• 3 peak devices, 3 fuel cells, 3 nuclear, 2
  boilers
• 10 renewables
• CHP 9 types
• Fuel type 13 of which 3 multiple fuel
• Companies
• Utilities
• Industrial generators
• Tertiary generators

• A plant is an element of the Cartesian product of
  the following elements
• Technology
• Multiple fuel or single fuel capability
• CHP technique, electricity only or steam only
• Size of Plant
• Company
• Technical-economic characteristics, potential
  etc. differ according to five dimensions
• Restrictions on possible choices of companies
• e.g. industrial generator only small size plants

21
Power and steam generation in PRIMES (4)The
network

• Nodes Production, Transmission, Distribution
• groupings by company and country as some
  constraints apply only at a group level (e.g.
  reserve margin, environmental regulation, fuel or
  exchange contracts)
• Flows electricity and steam over different
  topology of the network
• Plants are linked to Production nodes, owned by
  companies according to the grouping
• Customers (from PRIMES demand side) are connected
  to Distribution nodes
• multiple connections are possible, as well as
  privileges
• Exchanges are conveyed via transmission nodes
• constrained by capacity, losses, contracts etc.

22
Power and steam generation in PRIMES (5)

• Structure of power/steam generation model

Links to other Utilities (e.g. other countries)
23
Power and steam generation in PRIMES (6) Flows
over the Network

• Electricity
• Consumption
• Exchanges
• Thermal Power
• Intermittent Power
• Steam
• Industrial
• District Heating
• Fuel Purchased
• stackable (coal, oil)
• load related (natural gas)
• Contracts vs. Spot for the above items

• Determined over Chronological Load
• time segments of typical days
• linked to demand
• two seasons
• Items with load pattern
• electricity
• consumption, production, exchanges
• steam
• consumption, production, exchanges
• intermittent supply
• inflow to reservoir hydro
• contracts, some fuels

24
Conversion Decentralisation

• Comparative advantage
• availability of fuel (e.g. waste) or site for RES
• avoiding costs of self-supply of energy uses
  (e.g. steam)
• Obstacles
• market relationships
• supply contracts
• support (fixed costs, risk)
• Benefits
• Efficiency, Economics

Type of Technological Progress facilitating
decentralisation
25
Mathematical Form

• Minimize
• Total system cost for expansion, operation, trade
• both for electricity and steam

Non-linear or Mixed-Complementarity Formulation

• under constraints expressing
• demand of electricity per time segment
• demand for steam per time segment
• network flow equilibrium over topology
• generation capacities
• reservoir hydro, intermittent supply
• CHP possibilities
• transmission capacities
• fuel and electricity/steam exchange contracts
• fuel availability restrictions
• environmental regulation
• reliability, reserve margins

Linear Constraints