A Multi-Agent Intelligent Interpretation System for Power System Disturbance Diagnosis

摘要

A large volume of data is generated following a fault in the electrical power system. As a result of extensive research, hybrid intelligent systems have been identified as a method of supporting the online analysis and interpretation of the data. Hybrid intelligent systems employ multiple artificial intelligence techniques to perform various analytical and interpretation functions in order to generate a comprehensive diagnosis and meaningful information for the engineers to act upon. The authors have developed a multi-agent architecture as a platform for hybrid intelligent systems within the electrical power system. This paper will discuss the design of the architecture, and how it is being used to provide online power system protection analysis for ScottishPower's engineers. The paper will also discuss the integration of a novel model based diagnosis agent within the overall multi-agent architecture. In addition, it will consider the engineering issues surrounding the application.