Brattle has experience with several models to inform clients on integrated resource planning (IRP) and related decisions, as well as asset retirement and investment strategies.
- Long-Term Planning and Capacity Expansion/Retirement Models: These models determine optimal electric system capacity investment and retirement decisions based on expected capital costs of new units, load growth, fuel costs, environmental constraints/costs, and other drivers. For example, Brattle’s proprietary Xpand model is capable of modeling multi-sector carbon policies while simulating the operation of the electric sector in details. By incorporating emissions and abatement cost curves for sectors beyond electricity, the Xpand model considers the economic trade-offs between reducing emissions in the electric sector and non-electric sectors to comply with an economy-wide greenhouse gas (GHG) cap, thus allowing GHG prices to be determined endogenously at economy-wide equilibrium. Similarly, the capacity expansion module of PSO allows us to run long-term planning models stochastically in addition to the traditional deterministic method. When it is run stochastically, the model will recommend a portfolio that gives the optimal results for the portfolio of futures, even if the results are not ideal for any specific future. The PSO capacity expansion model also enables us to represent the electrical system at the nodal level when making capacity expansion and retirement decisions. This feature allows us to make very detailed recommendations on the locational impacts of new units and potential retirement decisions.
- Nodal Energy Market Simulation Models: Our production cost models account for unit commitment and dispatch, transmission constraints, and the variability of renewable generation in more detail than the long-term planning model. This enables us to evaluate production costs, pricing dynamics, and emissions with greater precision. Often, we populate our production cost model with the optimal expansion plan from the long-term planning model. Separately, we analyze alternative future scenarios with different load and generation profiles.