Least-cost planning for Pacific Gas and Electric: Stage 2, Case studies.

摘要

This project is an attempt to evaluate the usefulness of mainframe electric utility planning models for conducting least-cost planning (LCP). In the project's first stage, the Load Management Strategy Testing Model (LMSTM), written by Decision Focus, Inc. (DFI) for the Electric Power Research Institute (EPRI) (DFI, 1982, 1984, 1985), was calibrated to the Pacific Gas and Electric Company (PG&E) system. The calibration attempted to replicate PG&E as portrayed in the company's 1985 Long Term Plan (LTP) (Kahn et al., 1987), in the company's previous LMSTM input files, and in other data sources provide by PG&E. The attraction of LMSTM rests on its being an integrated model that attempts to solve the whole planning problem simultaneously. Only if the whole system is modeled as one organic entity can perturbations on opposite sides of the planning problem be compared. The stage 1 report describes the LMSTM calibration process in detail and discusses many of the issues and problems involved in LCP. In this, the second stage of the project, the stage 1 calibration is updated to incorporate more recent company information. This modified calibration is used as the ''base case'' against which some modest test cases are evaluated using LMSTM and other computer models, which solve parts of the problem exogenously to LMSTM. In general, the approach of the case study is to specify a demand-side program using data from PG&E and exogenous sources, to observe how this program would change the use of supply-side resources in the planning horizon (1989--2003), and then, to make a supply-side adjustment, based entirely on the judgment of the modeler, that seems to be the best cost-saving alternative available. 37 refs., 26 figs., 20 tabs.