Hydro-generator units operating condition forecasting and fault diagnosis based on BP neural network

In this paper, from the Angle to predict , take hydro generating operation condition parameters (head, power) as input sample, take vibration, shaft waggling and pulse pressure, bearings temperature and so on parameter as output sample, create neural network prediction model. Train the established models, through comparing a different designs scheme, chose one smaller error model. Predict through the trained neural network modes ,and compare with the measurement values.