Electricity supply planning when demand is uncertain
A brief taxonomy of the different kinds of uncertainties that beset electricity supply planners is presented and how these various classes of uncertainty affect the decision about what is the least-cost expansion decision is described. In particular, it is argued that the effects of short-term uncertainties caused by weather or forced outages or planning decisions can be separated from the longer-range uncertainties caused by unavoidable errors in demand forecasting. The problems of incorporating long-run uncertainties into supply planning decisions are discussed; it is argued that the method used in this study, dynamic programming, is the only one that practically allows true-least-cost supply planning and even then places fairly strong constraints on the detail with which the generating system can be modeled. It is shown that, when demand is uncertain, the least-cost supply plan will often include a substantial fraction of short-lead-time base-load plants even though the deterministic bus-bar cost of the base-load electricity that they produce exceeds that of longer-lead-time plants. Finally, appended to the report is a discussion of the basic dynamic programming algorithm, a description of the computer program used to obtain these results, and a short user's guide to the version that is currently implemented at Teale Data Center for use by the commission staff.
- OSTI ID:
- 5764683
- Report Number(s):
- NP-24132
- Country of Publication:
- United States
- Language:
- English
Similar Records
DIAMOND: A model of incremental decision making for resource acquisition by electric utilities
Effect of demand uncertainty on the relative economics of electrical-generation technologies with differing lead times