An SSA-CNN-AM-BILSTM method for valve opening prediction

Benzene production is an important chemical industry process for producing organic compounds such as styrene and phthalic acid. In this paper, a neural network method based on SSA-CNN-AM-BILSTM is proposed to predict the valve opening in the benzene making process. The method consists of Sparrow Search Algorithm (SSA), Convolutional Neural Network (CNN), Attention Mechanism (AM) and Bidirectional Long and Short-Term Memory Network (BiLSTM). The CNN performs the readout of the data features, and the AM is utilized to capture the effect of the feature states on the single loop at different times in the past, which enables the model to focus on the important information in the input data and enhances the model's ability to represent the data. BiLSTM adds an inverse LSTM layer to the Long Short-Term Memory (LSTM) network to improve the prediction results. For the hyperparameters in the neural network, SSA is applied for training optimization to improve the accuracy of the model.