FUZZY EXPERT SYSTEM FOR DISTRIBUTION SYSTEM RESTORATION AND CONTINGENCY OPERATION

Abstract

This thesis comprises seven chapters. The first three chapters provide the background information in order to appreciate the later chapters. People with distribution operation experience and with basic working knowledge on expert system and fuzzy logic may skip the first three chapters. Chapters 4 and 5 are the key chapters. They deal with the strategies employed by the fuzzy expert system as well as the implementation issues. Chapter 6 is devoted to the testing of the system while chapter 7 provides the conclusion and future directions of this project. The chapter summaries are as follows. Chapter 1 introduces the desire of electric utilities to operate their distribution networks in the most effective, efficient and reliable manner as well as the different computer technologies available to alleviate these problems. Towards the latter portion, it also brings out the purpose of the project. Chapter 2 provides some, information about the common operating problems encountered by todays electric utilities from the perspective of the Manila Electric Company (MERALCO). The chapter: also cites typical operating procedures applied by operators when such operating problems arise. Chapter 3 gives a brief description of expert systems and fuzzy logic. At the beginning of the chapter, a description of the benefits, application areas as well as the architecture of typical expert systems is provided. Towards the end, fuzzy logic concepts, basic operations and their usefulness when integrated with expert systems are discussed. Chapter 4 is devoted to the strategies implemented in the fuzzy expert system to address the common problems encountered in the day-to-day operation of the distribution network. The chapter highlights the salient features of the strategies, that is, object-oriented, multi-level and hierarchical, as well as the procedures and criteria applied to identify and to assess solutions for the problems within the domain. Chapter 5 gives an overview on the implementation of the fuzzy expert system. Specifically, the chapter deals on how fuzzy logic, traditional numerical methods and expert system technologies were combined in the system, the software and hardware requirements, the platform for the prototype and ultimate implementation and the ·functions implemented in the software. Furthermore, desirable design features of the implementation such as, the automatic and interactive modes of operation, the ability to plan on four planning periods, etc., are also highlighted. Chapter 6 describes the different tests that were conducted in the fuzzy expert system. For clarity, starting each test is a brief description of the scenario as well as its relevance in distribution operation. Single line diagrams of the test scenarios are also provided. Finally, Chapter 7 describes the conclusions arrived in this project as well as the future directions that it will have to take in order to make it more tuned to solving the common problems in the operation of the distribution system.