Optimization design of pump motor based on genetic algorithm and neural network

This paper presents a pump motor modeling and optimization method based on electromagnetic calculation and BP neural network. This method, according to the structure parameters of three-phase asynchronous motor in drainage pump, through the motor electromagnetic calculation, calculates the motor performance. It establishes a BP neural network model to fitting flow and water head for hydraulic performance of the pump. Combine motor electro-magnetic calculation model with pump hydraulic fitting model, an integrated model of pump and motor is built. Use simulated annealing genetic algorithm as optimization method, optimizes the structure parameters of the motor. The main function of this method is, maintain the input shaft power put into the pump, and intelligently search the optimal motor parameters. At the same time, accurate matches the motor performance with the pump load characteristic, reduce the supporting motor power.