Abstract: Based on BP neural network technique, taking crude oils sulphur content, acid value, temperature, pressure and flow rate as input parameters and internal corrosion rate as output parameter, prediction model of internal corrosion rate for oil pipeline is built.Influence of different factors on internal corrosion rules of pipeline is predicted.Prediction results fit simulated test data perfectly, which indicates that BP neural network model can accurately predict internal corrosion rate of oil pipelines.