Distribution system service restoration using the artificial neural network approach and pattern recognition method

Service restoration of a distribution system is investigated by using artificial intelligence. The purpose is to reach a proper restoration plan for the unfaulted zone after a fault has been identified and isolated. To reduce the outage period and improve service reliability, the restoration plan must be devised in a very short period. In the paper, two approaches using artificial intelligence, i.e. the artificial neural network (ANN) approach and the pattern recognition method, are developed to determine the restoration plan in a very efficient manner. The effectiveness of the proposed approaches is demonstrated by the restoration of electricity service following a fault in a distribution system in Taipei, Taiwan. It is concluded from the example that a proper restoration plan can be reached very efficiently using the proposed approaches. Therefore, it can be used by distribution system operators to reach a restoration plan