Fault Diagnosis Method of Power System Based on the Adaptive Fuzzy Petri Net

Power system is one of the complex systems. Diagnosis for it is a difficult task. It is very important to design a quick, simply programming inference method to diagnose this system. Aiming at this object, a method based on the adaptive fuzzy Petri net (AFPN) is proposed to model and diagnose power system. AFPN not only takes the descriptive advantages of fuzzy Petri net, but also has learning ability like neural network. By this mean, firstly set up a fuzzy Petri net using the fuzzy production rule. Then the weights of the fuzzy Petri net are trained by neural network. At last, when the weights of the fuzzy Petri net are fixed, the fault origin can be found through the fault inference. The method has advantages in scientifically selecting model weights and parallel inference.