Application of Similarity Modeling in Failure Prognostic System

Abstract

Based on similarity modeling, the failure prognostic system is rooted in mass of historical data that can provide easy and rapidly implementary and reliable failure prognostic system for power plant .The system can find equipment fault symptom before fault formation, so measures can be taken in advance to avoid the occurrence of failure. In this paper, the similarity theory, modeling and applications of failure prognostic systems are described detially. The prognostic example of a power plant in Shanxi is used to demonstrate that the system can detect abnormal changes of equipment in a timely manner, and the prognostic effect is achieved .