Fault diagnosis for substation based on optimizing the combination of fault-masses

In view of the uncertainties of the causalities from fault sources to fault symptoms in fault diagnosis for substation, a fault diagnosis method based on optimizing the combination of fault-masses is proposed. The Petri-net models of protection scheme are established according to the protection scheme and the operation mode of substation, and all couples of fault-masses and symptom-masses are obtained with the help of the transferring-state equation of the Petri-net models. A fitness function is presented to evaluate the feasibility that the combination of fault-masses acts as the optimal answer to the known fault symptoms, and the genetic algorithm is employed to search out the optimal answer. The trials prove that the developed methodology is effective and can supply detailed diagnosis results: fault sources, protective device non-operations, circuit breaker non-operations and missing signals. It is beneficial for the operators to confirm broken-down components quickly.