基于SSA-CNN-BILSTM组合优化的时空序列预测模型OA

In order to extract the latent features and hidden information of spatiotemporal series data more effectively and improve the prediction accuracy of the model,a spatiotemporal series prediction method based on SSA-CNN-BILSTM combination optimization is proposed.Using convolution and pooling of convolutional neural network(CNN)to extract features from input data,and using BILSTM to predict,the sparrow search algorithm(SSA)was used to optimize the parameters of the model.The simulation experiment was carried out with the passenger flow data of Chengdu Rail Transit,and two evaluation indexes were selected and compared with LSTM,BILSTM,CNN-BILSTM and PSO-CNN-BILSTM respectively.The results show that the combined model has the best prediction precision,the SSA-CNN-BILSTM combined optimization model presented in this paper is proved to be an effective time-space series prediction method.