Research on the prediction and control method of offshore pipeline plug based on UKF-WNN hybrid model
In order to solve the problem that it is difficult for oil and gas production system to accurately control slug flow in real time under deep sea conditions,a prediction and control method for slug flow in offshore pipelines based on UKF-WNN Mixture model was established.The results show that the UKF-WNN Mixture model can better fit the relationship between top pressure,flow,valve opening and bottom pressure.Compared with field tests,its Mean squared error is 44%and 47%lower than that of UKF and WNN,respectively.It can still provide more accurate estimates for the controller when there is interference in the system input,making the system stable.The addition of WNN effectively improves the stability of the control system,demonstrating the possibility of application of artificial intelligence methods in the field of slug flow control.The method proposed in this study has certain theoretical and practical significance in improving the prediction and control accuracy of slug flow in offshore pipelines,and ensuring the safety of offshore pipelines.
offshore pipeline slugstate estimationUKFWNNhybrid model
万方数据
维普
