Fitting of a neural network to control the intelligent operation of a high voltage circuit breaker

´Intelligent operation (IO)´ can improve the reliability of a high voltage circuit breaker and prolong its life. In this paper an artificial neural network (ANN) is used in the control of circuit breaker intelligent operation. Thus, during real-time control, it can save a lot of calculating time spent in the very complicated opening process of the circuit breaker. In the design of the controller of a circuit breaker IO, the structure of feedforward multilayer network is used, and two kinds of back-propagation learning algorithms, the self-adapting adjusting learning and the momentum method, are applied to the supervised training of the neural network. Both algorithms greatly enhance the training speed, shorten the training time and speed up the convergence. After training, an artificial neural network controller (ANNC) of the system is formed. It is proved that the ANNC has a higher accuracy and can meet the controlling requirement of the circuit breaker IO. This method can be used for reference by other control systems for solving complicated nonlinear control equations.