Development and application of improved techniques for the evaluation of electric energy alternatives for New York State

Abstract

The product of electric energy has an unlimited number of implications. Some implications are well understood while others are subject to orders of magnitude levels of uncertainty. The technique that has been developed provides the user with a reference set of implication coefficients that he can either accept or modify. Thus, an unlimited number of implications can be identified and evaluated. The mathematical model calculates implications in terms of the contribution from each generating unit type and the total contribution from all generating units on both a yearly and cumulative basis. Specific implications that are evaluated are health effects, cost of external damage, barrels of oil, and tons of coal. These implications are reported for a representative set of cases.

Observations, conclusions, and recommendations that have been derived form the development of these methods and their application to the evaluation of electric energy alternatives for New York State are summarized. These methods are shown to be most valuable for the purpose of evaluation the wide range of questions associated with the evaluation of the electric generation alternatives on a well defined and consistent basis. The results from applying this method confirm the validity of some traditional electric generation planning assumptions and general guidelines, while some conventional wisdom is contradicted as a result of applying these methods to the New York State electric electric situation. The underlying condition that forces a shift in traditional planning guidelines in New York State is the extreme increase in oil price and the high fraction of New York State generating capacity that is oil fueled.

The mathematical model is described. The policy calculating portion of the model includes preprocessing, optimization, and post processing steps. The preprocessing step includes identification of variables, development of simulation equations, solution of simulation equations for variables that are not subject to optimization, and transforming the remaining simulation equations to a standard input format for linear programming. The optimization step solves the linear programming problem. The post processing step combines the results of the preprocessing and linear programming calculation into a standard form schedule of installed generating capacity and operating policy. The implications of the policy are then calculated in yearly and cumulative schedules from the policy and user selected implication coefficients.

Traditional assumptions and guidelines that are contradicted as a result of the high price and large fraction of oil fueled capacity include generation reserve criteria, value of load management, policies for load leveling, and the value of energy storage. The most favorable case that was evaluated placed no upper limit on electric generation reserve relative to a reference case that limited reserve to 35%. The unlimited generation reserve case adds an additional 5853 Mw of nuclear generation in 1987 which results in 53% system reserve. This results in a cumulative discounted savings of .5% over the 1980 to 1999 planning period, along with cumulative 15% decrease in expected excess deaths, a 16% decrease in external property damage, a 14% decrease in oil consumption, and a 17% decrease in coal consumption. It is noted that current generation planning decisions will have no significant effect on the 1980 consumer, but will have a major effect on the price paid by the 1999 consumer and a corresponding effect on the future New York State social and economic conditions.

The interactive graphics computing system is described. This includes: reasons for developing an interactive graphics analysis system, a description of the computing system hardware including a foreground computer, a mid-ground computer, and a background computer along with the peripheral equipment on the foreground computer, the user procedure for operating the interactive graphics system, a functional description of the computer programs, and a summary of the computational system performance in the terms of computational time, computational cost, and turnaround time for each case.

The cases that are evaluated and described. Each case can be considered a response to a question. These questions are classified in categories that include reference, sensitivity to nuclear parameters, reserve incentive, lead forecast uncertainty, nuclear moratorium, specific alternatives, new generation type, new fuel type, load management, storage, sensitivity to present worth, ideal generating system, ideal generating system with load management, no load growth, and sensitivity to length of planning period. Results are summarized in terms of cost and described in terms of installed generation strategy, operating strategy, and special observations.

The subject of this thesis is the development and application of improved techniques for the evaluation of electric generation alternatives for New York State. The development phase combines the best available computer software, computer hardware, computer methods, and analytical techniques. The application phase transforms questions into case studies that are evaluated by calculating policy and the implications of policy.