Artificial Neural Network Model to Predict Yield Stress of Waxy Crude Oil

In allusion to the problem of shutting down and restarting of waxy crude oil pipeline and based on the data obtained from simulating experimental loop, a calculation on startup yield stress for waxy crude oil under similar running condition at different shutdown temperatures is made with back propagation algorithm (BP algorithm), and startup yield value is predicted for waxy crude oil at different startup temperatures based on calculation results. The computational results are compared with the experimental results. The results show that as for the startup yield stress value predicted by the artificial neural network and the experimental value, the maximum value is 1.9%, and the minimum error is 0.12%.