Design of expert system for predicting and evaluating centrifugal pump operation based on GRU-BP neural network

At present, in actual factory production, the running status of the equipment will affect the quality and output of the product. The timely and effective prediction of the running status to the equipment in Short-term can improve the stability and reliability of production. Firstly, a Gated Recurrent Unit (GRU) model is proposed to deal with the prediction of the future short-term operation status. Secondly, through the idea of combining neural network and expert system, an expert system model based on Gated Recurrent Unit-Back Propagation (GRU-BP) Neural Network is designed to evaluate the short-term running state of the centrifugal pump in the future. After the simulation results of field equipment operation parameters, it is verified that the GRU-BP model can not only judge the running state of the machine pump, but also predict the state of the machine pump in a short period of time with high accuracy, and has a great potential in the direction of pump state prediction.