A self-learning propotional–integral–derivative control of grouting pressure using the back-propagation model

Abstract

For the security of grouting process of dam foundation, grouting pressure control is one of the most important problems. In order to avoid dangerous grouting pressure fluctuation and improve the control precision, a feedback propotional–integral–derivative control method was presented for the whole grouting system. Because the grouting pressure is affected by many factors such as grouting flow, grouts density, and geological conditions, the parameters of propotional–integral–derivative must be tuned. In this article, the adaptive tuning method is presented. The back-propagation artificial neural networks model was proposed to simulate the grouting control process, and sensitivity analysis algorithm based on orthogonal test method was adopted for the selection of input variables. To obtain the optimal propotional–integral–derivative parameters, an iteration algorithm was used in each sampling interval time and the discrete Lyapunov function of the tracking error. The simulation results showed that self-learning propotional–integral–derivative tuning was robust and effective for the realization of the automatic control device in the grouting process.