Fuzzy neural network approach for fault diagnosis of power plant thermal system under different operating points

Developing a practical fault diagnosis system for a power plant thermal system is important in ensuring safe and reliable operation of the whole power generation unit. It is a difficult task due to the complex structure of the thermal system and the variable operating points of the system. Although artificial neural networks are very efficient in thermal system fault diagnosis, their applications for fault diagnosis usually consider only a single operating point (usually at the full rated load). Thus a fuzzy neural network approach for thermal system fault diagnosis under different operating points is investigated in this paper. An extended fuzzy fault symptom calculation method suitable for different load is discussed. The feedwater heater system of a 300 MW coal-fired power generation unit is taken as a fault diagnosis example. Several diagnosis tests for different faults under different operating points are carried out on a full-scope simulator of the power unit. The results under different operating points demonstrate the fault diagnosis ability of the fuzzy neural network method.