New approaches to integrated resource planning for electric utilities: Dealing with uncertainty and combined utility and cogeneration planning

Integrated Resource Planning (IRP) is a new approach to electric utility planning that (among other factors) emphasizes the explicit consideration of cogeneration plants as alternatives to conventional utility power plants, and the analysis of the uncertainties and risks posed by different resource portfolios. The purpose of this thesis is to provide insight into these issues and to develop methods for the more efficient solution of large-scale IRP models.;Using an analytic planning model, one chapter considers the impact of resource lead time when planning under uncertainty. Among other theoretical results, the analysis demonstrates that some "short lead time" technologies screened out by standard deterministic methods may enter the optimal solution when uncertainty is considered. Conversely, some "long lead time" technologies, that do enter the optimal solution when uncertainty is ignored, may leave. A numerical example demonstrates that short lead time technologies can have a major impact on the optimal capacity mix and the expected cost of meeting demand.;Another chapter derives optimality conditions for peak load pricing and capacity planning of combined electric utility and cogeneration systems under conditions of demand uncertainty, extending earlier analyses that examined electricity-only systems in isolation. It is shown that under these more general conditions, peak load pricing (i.e., charging higher prices in periods of higher expected demand) is not necessarily optimal under all circumstances. Furthermore, the optimal levels of electricity and thermal energy reliability may be greater. A detailed numerical example illustrates these and other theoretical results, and demonstrates the importance of the relationship between electricity and thermal energy demand distributions.;A linear programming model for the cooperative planning of electric utility and cogeneration systems is the subject of a third chapter. It is shown that under mild conditions, the problem may be formulated more efficiently than has previously been done, by exploiting the special structure of the optimal solution. Numerical tests indicate that this reformulation typically allows a reduction in the time required for problem solution by a factor of two or more.