Prediction of Cement Combined Grinding Mill Bucket Elevator Current Based on Hybrid Neural Network

Cement combined grinding is one of the key links in the cement production process, and accurate prediction of mill bucket elevator current has an important impact on its control and stable operation. However, cement combined grinding is a nonlinear system with complex working conditions. In order to bring the predicted and actual values of the mill bucket elevator current closer together. In this paper, a hybrid neural network that combines Temporal Convolutional Network with Bi-directional Long Short-Term Memory Network and introduces the Attention mechanism is utilized to predict the mill bucket elevator current. The model captures local features and long-range dependencies on the data and updates the model parameters using the Adaptive Moment Estimation optimization algorithm as the optimizer and the Mean Square Error as the loss function. Using the model to obtain the new current value, taking the standard error and the coefficient of determination as the evaluation metrics, and comparing the Convolutional Neural Network prediction model, the Long Short-Term Memory Network prediction model, and the Bi-directional Gated Recurrent Unit prediction model, it is verified that the present model outperforms the other several models, and exhibits a good performance in the current prediction of the mill bucket elevator.