Multi-stage investment planning

Added in 0.3

GenX can be used to study the long-term evolution of the power system across multiple investment stages, in the following two ways:

• The user can formulate and solve a deterministic multi-stage planning problem with perfect foresight i.e. demand, cost, and policy assumptions about all stages are known and exploited to determine the least-cost investment trajectory for the entire period. The solution of this multi-stage problem relies on exploiting the decomposable nature of the multi-stage problem via the implementation of the dual dynamic programming algorithm, described in Lara et al. 2018 here. This algorithm splits up a multi-stage investment planning problem into multiple, single-period sub-problems. Each period is solved iteratively as a separate linear program sub-problem (“forward pass”), and information from future periods is shared with past periods (“backwards pass”) so that investment decisions made in subsequent iterations reflect the contributions of present-day investments to future costs. The decomposition algorithm adapts previous nested Benders methods by handling integer and continuous state variables, although at the expense of losing its finite convergence property due to potential duality gap.
• The user can formulate a sequential, myopic multi-stage planning problem, where the model solves a sequence of single-stage investment planning problems wherein investment decisions in each stage are individually optimized to meet demand given assumptions for the current planning stage and with investment decisions from previous stages treated as inputs for the current stage. We refer to this as "myopic" (or shortsighted) mode since the solution does not account for information about future stages in determining investments for a given stage. This version is generally more computationally efficient than the deterministic multi-stage expansion with perfect foresight mode.

More information on this feature can be found in the section Multi-stage setup.