Current Efficiency Prediction of Aluminum Reduction Production Based on TCGWO-KELM
Chenhua Xu, Jinzhi Zhang, Qingbao Huang, Mingkun Huang, Chun Lin Xie, Xin Feng Yu · 2019 IEEE 8th Data Driven Control and Learning Systems Conference (DDCLS) · 2019
Because the method of the current efficiency has not been properly determined in the process of aluminum reduction so far, it is difficult to obtain real-time effective current efficiency data to guide aluminum reduction production. In this paper, a current efficiency prediction model based on kernel extreme learning machine (KELM) is established, which is improved by the tent chaotic grey wolf optimization algorithm. Firstly, aluminum reduction process is analyzed and obtained the parameters affecting the current efficiency of the aluminum production. In order to avoid the adverse effect of data redundancy on the prediction model, the principal component analysis method is used to reduce the dimension of the data. Secondly, the model of current efficiency based on KELM is established. In order to improve the training precision and increase the robustness of the algorithm, the improved grey wolf optimization algorithm is proposed to optimize the key parameters of the model. Finally, the current efficiency prediction model was trained and tested using the on-site production data. The results demonstrate the effectiveness of the prediction model and can provide decision-making reference for aluminum reduction production.