Research on Predicting Mud Flow Rate of Trailing Suction Dredger Based on PSO-LSTM

Aiming at the difficulty of applying the mathemati-cal model of the dredger with rake suction in actual engineering, a data-driven method was used to predict the mud flow rate in the pipeline. In order to improve the prediction accuracy, the offline construction data are preprocessed on the basis of the measurement data from the relevant sensors on the rake suction dredger. Then, the mud prediction model was established using Particle Swarm Optimization (PSO) algorithm for optimizing Long Short Term Memory (LSTM) network and compared with LSTM network and BP neural network. The results show that the prediction performance of this method is better than that of LSTM network and BP neural network, which can accurately predict the mud flow rate and provide theoretical guidance for dredging operations.