Cement Grinding Power Consumption Prediction Based on AO-TCN-GRU

In order to solve the problem that the power consumption of cement grinding is difficult to predict, this paper proposes a prediction model. This model is based on the Aquila optimization algorithm and combines the Time Convolutional Network (TCN) and the Gated Recurrent Unit(GRU). Firstly, this paper analyses the cement grinding process and chooses the variables related to the power consumption of cement grinding. Then the mutual information method is used to verify it. Secondly, the data is filtered using mean filtering to smooth it, and then normalized. Finally, with the help of the Aquila optimization method, the hyperparameters of the established TCN-GRU model are optimized. So the cement power consumption prediction model based on AO-TCN-GRU is established. The experimental results show that the R² of the proposed model is 0.971 and the MAPE is 0.17%. It indicates that the proposed model can accurately predict the power consumption of cement.