Failure prognostic system of power generating equipment based on similarity modeling

CHANG Jian1， GAO Ming2
（1. Shanxi Datong International YunGang Thermal Power Co.， Ltd.， Datong 037039， China；
2. Automation Department， North China Electric Power University， Baoding 071003， China）

Abstract： Aiming at unplanned downtime of the power plant which caused by generating equipments failure caused unnecessary economic loss，the power plant launched failure prognostic warning system based on fault diagnosis was studied. Failure prognostic system was consisted of two model，state estimation and monitoring. By analyzing the modeling method of failure prognostic system，the method  was proposed that using the similarity modeling to realize equipment state estimation. By comparing the similarity relationship between real-time data and historical normal data of equipment，the possible state of equipment was estimated. After monitoring module receiving equipment real-time data and estimate state data，the equipment status was judged，and the failure information was shown for the staff to provide reference to analyzing related reason of the failure. Finally the failure prognostic case of a power plant in Shanxi was introduced.The results show that，the failure prognostic system can analyze the equipment information and identify equipment performance anomalies，so that generating companies can detect the signs of equipment failure before failure happened and take measures to avoid equipment failure early.
Key words： similarity；failure prognostic；modeling；power plant