Application of PSO-LSTM for Forecasting of Ship Traffic Flow

Ship traffic flow prediction is facing many problems at present, such as high randomness, many influencing factors and low accuracy. In this paper, a long short-term memory (LSTM) prediction model based on particle swarm optimization (PSO) is proposed to improve the accuracy of ship traffic flow prediction. Aiming at the problem that the parameters of LSTM model are difficult to select, PSO is used to optimize the parameters of LSTM model. The PSOLSTM model was constructed based on the data of the intersections of inland waterways in Huaian section of the Beijing-Hangzhou Grand Canal. The prediction results of this model are compared with the LSTM model and Support Vector Regression (SVR) model. The experimental results show that the root mean square error (RMSE) of the model is 2.211, the mean absolute error (MAE) is 2.068. Compared with other prediction models, the prediction accuracy is improved obviously.