Erin Boyd
Department of Energy Office of Energy Policy and Systems Analysis
erin.boy[email protected].gov
DOE’s Technical Assistance Website
www.energy.gov/ta
Power Sector Modeling 101
2
Presentation Description DOE Power Sector Modeling 101
With increased energy planning needs and new regulations, environmental agencies, state energy
offices and others have expressed more of an interest in electric power sector models, both for
(a) interpreting the results and potential applications of modeling from other groups, and (b)
informing future modeling efforts a state air agency may want to initiate. This presentation covers
the basics of power sector capacity expansion modeling, and briefly touches on other types of
modeling and analytical tools available to provide data on the electric power system. Capacity
expansion models simulate generation and transmission capacity investment, given assumptions
about future electricity demand, fuel prices, technology cost and performance, and policy and
regulation.
Capacity expansion modeling topics covered in this presentation include:
typical model outputs,
needed model inputs,
types of questions these models are well suited to answer and those they are not,
key considerations when selecting a model, and
key considerations when comparing model results or designing modeling scenarios.
For more information on technical assistance resources available to state, local and tribal officials,
visit DOE’s Technical Assistance website at www.energy.gov.ta, or submit a request at
TechnicalAssistance@hq.doe.gov.
3
Overview
1. Power System Questions
2. Model Types
Data and Analysis Tools
Capacity Expansion Models
Production Cost (Grid Operations/Unit Commitment and Dispatch) Models
Network Reliability Models
3. Summary
Key Consideration: Identify the question(s) you want to answer, and
then pick the tool that will most effectively provide this information.
(As opposed to picking a tool, and then finding out its not the
appropriate resource to provide the information you need.)
4
Examples of Power System Questions
Data and Resource Assessment
What are the local wind and solar resources? How much available natural gas
capacity is in the region? What is the cost of avoided/saved energy consumption?
Generation and Transmission Capacity Expansion
How to plan a resource portfolio for the future (i.e., generation, retirements)?
What type of generation should be built to meet demand? How much of it? Will
it necessitate development of new transmission capacity? How does the optimal
system change with constraints on emissions, or with local economic
development goals? How can the system be optimized to deliver reliable power
at least-cost under specified environmental constraints? What are the costs, rate
impacts, and welfare implications of alternative power sector policies or
regulations? What are the key drivers of the system?
Generation and Transmission System Operation
Given a generation and transmission system what is the lowest cost way to
operate the system while maintaining reliability under uncertainty and meeting
other types of constraints (e.g., emissions)?
5
Power System Questions
Network Reliability
Will the transmission system work under periods of high load? Will the system
be able to remain stable after a loss of a large power plant? Will a loss of
transmission line or power system cause instability and cause individual
generators or sections of the network to disconnect from the rest of the grid?
6
Model Types
1. Data and Analysis Tools generally accessible
2. Capacity Expansion Models
3. Production Cost (Grid Operations/Unit Commitment and Dispatch)
Models
4. Network Reliability Models
7
Model Types
1. Data and Analysis Tools generally accessible
2. Capacity Expansion Models
3. Production Cost (Grid Operations/Unit Commitment and
Dispatch) Models
4. Network Reliability Models
Requires expert modeler to manage inputs,
run the model, and interpret the outputs
8
Data
Power System Tracking - capacity, generation, fuel use, fuel prices, electricity
price, electricity consumption, energy efficiency savings, policies (e.g., state
renewable portfolio standards, state energy efficiency policies)
Resource Assessment - spatially and temporally explicit assessment of renewable
energy resources
Key Resources
U.S. Energy Information Agency (EIA)
Database of State Incentives for Renewables and Efficiency (DSIRE)
DOE's State and Local Energy Data (SLED)
National Electric Energy Data System (NEEDS)
EPAs Emissions & Generation Resource Integrated Database (eGRID)
EPAs Air Markets Program Data (AMPD)
National Renewable Energy Lab’s Report - Estimating Renewable Energy Economic
Potential in the United States
ABB (ASEA Brown Boveri) Velocity Suite
SNL Energy
Data and Analysis Tools (1 of 2)
Zero Cost
9
Data and Analysis Tools (2 of 2)
Analysis Tools: spreadsheet tools or simple calculators that allow users to conduct
high-level gross analyses of the power sector
EPA's AVoided Emissions and geneRation Tool (AVERT) - estimates the emissions implications
of energy efficiency measures and new renewable generation capacity
ACEEE’s State and Utility Pollution Reduction Calculator Version 2 (SUPR 2) 19 different
policies and technologies to choose from to build a compliance scenario to EPAs Clean Power
Plan, including energy efficiency, renewable energy, nuclear power, emissions control, and
natural gas.
Synapse's Clean Power Plan Planning Tool (CP3T) and MJ Bradley's & Associates CPP
Compliance Tool - both Excel-based spreadsheet tools for performing first-pass planning of
statewide compliance with EPA's Clean Power Plan
Advanced Energy Economy (AEE) State Tool for Electricity Emissions Reduction (STEER) - an
open access integrated resource planning model that constructs a merit order for dispatch
from generator-level cost data and simulates generation based on least-cost strategies
(currently available for PA, MI, AR, VA, and IL (more will be forthcoming).
These analysis tools are free and publicly available.
10
Capacity Expansion Models
Capacity expansion models simulate generation and transmission capacity
investment, given assumptions about future electricity demand, fuel prices,
technology cost and performance, and policy and regulation
What mix of generators should we build to meet load?
Does a policy affect cost of service regions and competitive regions in different
ways?
0 2000 4000 6000 8000
0
1000
2000
3000
4000
5000
6000
7000
Cumulative Hours at Load
Load (MW)
Baseload
Intermediate
Load
Peak Load
11
Capacity Expansion Models
Typical Outputs
Annual generation, generation and transmission capacity builds/retirements,
emissions, fuel consumption, electricity prices, credit/allowance prices
12
Capacity Expansion Models (1 of 2)
Examples of Capacity Expansion Models:
National-Scale: National Energy Modeling System (NEMS), Regional Energy
Deployment System (ReEDS), Integrated Planning Model (IPM), Haiku, MARKAL
(MARKet Allocation)
Utility-Scale: Resource Planning Model (RPM), Aurora, System Optimizer,
Strategist, PLEXOS
Typically have higher spatial and temporal resolution
Often used for Integrated Resource Plans (IRPs)
In addition to having staff or paying an expert to run these models, the
commercial models also require a licensing fee
13
Capacity Expansion Models (2 of 2)
What do these models do particularly well?
Examine the impacts of power sector policies (or alternative technology/fuel
trajectories) on the generation and capacity mix in the mid- to long-term
What don't they do?
Many do not have chronological unit commitment (i.e., every hour of the year
chronologically); some use aggregate (model) plants for dispatch; transmission
and power flow are a stylized representation (pipe flow or DC)
What kinds of questions/analyses can the model answer/address?
Quantifying the impacts of environmental policies on generation and capacity?
What are the cost implications of alternative pathways to a low greenhouse gas
emissions future? How will alternative future prices of natural gas impact
capacity investment? What is the change in consumption and expenditures?
What are the efficiency and distributional effects of various policy designs?
14
Capacity Expansion Model Capabilities - Key Considerations in
Model Selection (1 of 2)
Regionality
Geographic scope (state, regional, national)
Cost-of-service vs. competitive regions
Temporal Resolution
Time of day, Seasons
Time Steps
Building new capacity, dispatch
Time Horizon
Near-term: 2015-2020, Long-term: 2015-2050
Representation of Generating Units
Individual Plants or Model Plants
Representation of capital costs and other production costs
Representation of Transmission and Associated Constraints
Pipeflow or DC Powerflow; Individual transmission lines or aggregated
15
Capacity Expansion Model Capabilities - Key Considerations in
Model Selection (2 of 2)
Representation of Renewable Energy (RE)
RE technologies represented in the model
Underlying resource dataset spatial and temporal resolution
Accessibility cost (connecting RE resources to load)
Accounting for variability and uncertainty in generation (e.g., representation
and treatment of curtailments and capacity value of RE technologies)
Consideration of other parameters (e.g., electric power sector model vs.
economy wide model, representation of environmental constraints)
16
Capacity Expansion Model Analyses- Key Considerations when
Comparing Model Results or Designing Modeling Scenarios (1 of 2)
Input Assumptions
Fuel prices (exogenous or endogenous)
Technology cost and performance assumptions: e.g., capital cost, fixed and variable O&M
costs, capacity factors (exogenous or endogenous)
Constraints on deployment or use of specific technologies
Representation of Electricity Demand
Baseline
Exogenous or endogenous; demand elasticity
Energy efficiency representation
Cost/Benefit Metrics
Welfare, total cost, allowance/credit prices
Distributional impacts - consumer/producer surplus, regional cost metrics
Electricity Bills and Prices
Competitive vs. cost-of-service
Wholesale vs. retail
17
Capacity Expansion Model Analyses- Key Considerations when
Comparing Model Results or Designing Modeling Scenarios (2 of 2)
Retirements
Exogenous or endogenous
Detailed Representation of Policies
Ex: Carbon Policies
Rate vs. mass vs. price
Covered sources
Total emissions vs. covered emissions
Treatment of nuclear plants, existing renewables, biomass, etc.
National uniform vs. patchwork policy
Trading parameters
Over what time period are constraints applied (annually, compliance periods, etc.)
If applicable, how is auction revenue used?
If applicable, are allowances freely allocated and to who? On what basis (historic or
updating)?
18
Energy Efficiency (EE) in Capacity Expansion Models
EE is an energy planning resources that can reduce energy
bills and lower regulatory compliance costs
EE representation in capacity expansion models
Endogenous - rebates
Exogenous
Resources: EE Potential studies and EERS Goals
Source: EIA Energy Today 10/23/14; 2013 Annual CO
2
Analysis
19
Production Cost (Unit Commitment and Dispatch) Models
Simulate operation of a specified power system over a relatively short period
compared to Capacity Expansion Model (1-week to 1-year), but at higher temporal
resolution (hours to 5-minutes)
What is the least cost dispatch of a complex system of interconnected generators
to reliably meet load in every hour of the day at every location?
Capacity Expansion Output -> Production Cost Input
20
Production Cost (Unit Commitment and Dispatch) Models
Typical Outputs: Sub-hourly unit level generation, powerflow, locational
marginal prices, emissions, fuel consumption, loss of load, ancillary service
prices (prices of ancillary services to balance transmission system),
curtailments
21
Production Cost (Unit Commitment and Dispatch) Models
Examples of Production Cost Models:
PROMOD, GE-Maps, PLEXOS, GridView
What do these models do particularly well?
Simulate detailed (hourly to sub-hourly) operation of a given system; Assess
resource adequacy and other aspects of reliability of a system; Analyze the
impact of changes in the system (e.g., retirement/addition of capacity) on system
operation; Assess transmission congestion and locational marginal prices;
Describe the daily pattern of emissions
What don't these models do?
Build/invest in new generation or transmission capacity; Typically cannot model
the entire US simultaneously - regional focus is required; Do not address all
aspects of reliability
What kinds of questions/analyses can the model answer/address?
What are the operations, emissions, and resource adequacy impacts of
retirement of coal or nuclear units in a given region? What is the maximum
potential for redispatch from coal steam units to NGCCs? What is the value of
storage, demand response, and solar power to the power system?
22
Network Reliability Models: AC Powerflow and Dynamics
Perform “deep-dive” simulations of the transmission network to address
specific situations. Simulates very short time periods (~ 30 sec to 1 min).
Analysis includes:
AC Powerflow: Simulates the AC transmission network to check operational
feasibility (steady state)
System Dynamics: Simulates dynamic events in the power system to examine
reliability under fault conditions
Simulation of contingency events to examine frequency response Example
output: frequency nadir (lowest frequency), settling frequency
Simulation of transient stability
Will generators remain
synchronized with voltage spike?
(e.g., lightening)
Production Cost Output/Input <-->
Network Reliability Input/Output
23
Network Reliability Models: AC Powerflow and Dynamics
Network Reliability Models are typically run by ISOs/RTOs, reliability
organizations, large utilities, and consultants
Examples: Positive Sequence Load Flow (PSLF), Power System Simulator for
Engineering (PSSE)
What do these models do particularly well?
Detailed simulations of the transmission network including dynamic events that
can occur in seconds (and cause big problems); these models aren't run on a day
to day basis they are only run to examine significant changes to an existing
system
What don't these models do?
Anything related to system operation in economic” time frames (typically more
than about 30 seconds)
What kinds of questions/analyses can the model answer/address?
Will the system remain able to maintain frequency after retirement of large
synchronous machines with set generation and transmission mix? Will the system
maintain reliability to respond to a voltage swing or other transients?
24
What Aspects of Reliability do Different Types of Models
Address?
Model Type Generator
(Resource)
Adequacy
Flexibility
Requirement
Transmission
Adequacy
Generator
Contingencies
Transmission
Contingencies
Frequency
Stability
Voltage
Stability,
Voltage Control
Capacity
Expansion
Often Somewhat/
depends
Typically No No No No No
Production
Cost (Unit
Commitment
and Dispatch)
Yes Yes Partially Somewhat Somewhat Somewhat No
Network
Reliability
(AC Power
Flow,
Dynamics)
No No Yes Yes Yes Yes Yes
25
Summary
Wide range of tools/models to address a wide range of questions
Within a particular type or class of model or tool, different models have
different strengths and weaknesses
Selection of a specific model or tool needs to be closely tied to the analytical
application
Many assumptions to consider when comparing results from different models
or modeling analyses
Results of a modeling analysis need to be interpreted in the context of the
limitations of the model
Questions?
Erin Boyd
DOE Technical Assistance Website
www.energy.gov/ta
26
Appendix
27
Key Data Resources (1 of 2)
Key Data Resources
U.S. Energy Information Agency (EIA) - independent agency that collects, analyzes, and
disseminates independent and impartial energy information
Database of State Incentives for Renewables and Efficiency (DSIRE) national and state
incentives and policies that support renewable energy and energy efficiency
DOE's State and Local Energy Data (SLED) - Get basic energy market information that can
help state and local governments plan and implement clean energy projects, including:
Electricity generation; Fuel sources and costs; Applicable policies, regulations, and
financial incentives
National Electric Energy Data System (NEEDS) - database contains the generation unit
records used to construct the "model" plants that represent existing and
planned/committed units in EPA modeling applications of IPM
EPAs Emissions & Generation Resource Integrated Database (eGRID) - data on the
environmental characteristics of almost all electric power generated in the United States.
These environmental characteristics include: air emissions for nitrogen oxides, sulfur
dioxide, carbon dioxide, methane, and nitrous oxide; emissions rates; net generation;
resource mix; and many other attributes
28
Key Data Resources (2 of 2)
Key Data Resources
EPAs Air Markets Program Data (AMPD) tool which allows users to search EPA data to
answer scientific, general, policy, and regulatory questions about industry emissions
National Renewable Energy Lab’s Report - Estimating Renewable Energy Economic
Potential in the United States geographic economic potential for renewable energy
technologies
ABB (ASEA Brown Boveri) Velocity Suite commercial data set that includes fuel
production, costs, electricity generation and capacity, market data, and more
SNL Energy commercial energy information that integrates news, data and analytics in
real time on a Web-based platform. Access to comprehensive financials, breaking news,
proprietary regulatory research, market pricing and fundamentals of supply and demand
29
What Aspects of Reliability do Different Types of Models
Address?
Model Type Generator
(Resource)
Adequacy
a
Flexibility
Requirement
b
Transmission
Adequacy
c
Generator
Contingencies
d
Transmission
Contingencies
e
Frequency
Stability
f
Voltage Stability,
Voltage Control
g
Capacity
Expansion
Often
1
Somewhat/
Depends
3
Typically No
4
No No No No
Production
Cost (Unit
Commitment
and
Dispatch)
Yes
2
Yes Partially
5
Somewhat
6
Somewhat Somewhat
8
No
Network
Reliability
(AC Power
Flow,
Dynamics)
No No Yes Yes
7
Yes Yes
9
Yes
30
Footnotes for Reliability Table (slide 1 of 2)
a) Defined as having adequate capacity to meet demand measured by loss of load probability in each
time interval or cumulative loss of load expectation. Considers resource availability in each time
period but typically does not measure flexibility, or the ability to ramp from one state to another
b) Defined as having adequate generator flexibility to ramp from one demand period to another
c) Defined as not violating transmission thermal, voltage and stability limits
d) Defined as maintaining system reliability upon the failure of large generators, including the single
largest generator
e) Defined as maintaining system reliability upon the failure of transmission lines, including the single
largest line
f) Defined as maintaining system frequency by provision of inertia, primary frequency (governor)
response and regulating reserves.
g) Defined as the ability to maintain system voltage by provision of reactive power. NOTE we are not
including the third leg of the stability trifecta transient/rotor angle stability. Network reliability
models are the only class of models that do this.
31
1) Large-area capacity expansion models will often inforce a peak capacity requirement that is a proxy
for resource adequacy. Commercial models may often include resource adequacy calculations
2) Most production cost models can provide full Loss of Load Expectation/Loss of Load Probability
(LOLE/LOLP) calculations. There are also several dedicated models that perform resource adequacy
calculations, including stochastics
3) Most capacity expansion models do not include full chronology. However many have the ability to
capture some aspect of the need for flexibility, by either running chronological dispatch during some
sub set of time periods, or imposing a flexibility constraint in the objective function
4) Most capacity expansion models have a very rudimentary treatment of transmission. However some
will enforce constraints that require additional transmission when new resources are added
5) Most production cost models now include DC optimized power flow which measures some aspects of
transmission adequacy (such as thermal limits on certain lines) However due to computational
complexity, full transmission power flow analysis is not possible
6) PCMs simulate the holding of reserves in each time period. These reserves should be adequate to
meet contingency events, but the actual consequences of contingencies cannot be tested with a PCM.
7) Actually simulates a contingency event to ensure reliable operation
8) As with contingency reserves, PCMs can enforce holding of regulating reserves (and potentially even
primary frequency response) but cannot simulate the operation of reserves to check frequency
stability
9) Most network reliability models don’t simulate regulation reserves, which kick in after inertia and
primary frequency response. There is a limited set of models (I only know of one) that can simulate
the full response to frequency deviations, because they combine physics-based dynamic models with
semi-economics/rule-based automatic generation control models.
32
Models writers or owners (1 of 2)
Capacity Expansion Models
National Energy Modeling System (NEMS) U.S. Energy Information Agency
Regional Energy Deployment System (ReEDS) National Energy Renewable
Laboratory
Integrated Planning Model (IPM) - ICF
Haiku Resources for the Future
MARKAL (MARKet Allocation) International Energy Agency
Resource Planning Model (RPM) - National Energy Renewable Laboratory
Aurora - EPIS
System Optimizer - ABB
Strategist - ABB
PLEXOS Energy Exemplar
33
Models writers or owners (2 of 2)
Production Cost Models
PROMOD ABB
GE-Maps General Electric
PLEXOS Energy Exemplar
GridView ABB
Network Reliability Models
PSLF General Electric
PSSE Siemens