The Decision Algorithm of Cement Mill Operation Index Based on Improved Differential Evolution Algorithm

• Published in: Automatic control and computer sciences Vol. 56; no. 6; pp. 533 - 545
• Main Authors: Chong Liu, Yang, Xunian, Zheng, Lizhao, Hao, Xiaochen
• Format: Journal Article
• Language: English
• Published: Moscow Pleiades Publishing 01.12.2022; Springer Nature B.V
• Subjects: Algorithms; Cement; Computer Science; Constraints; Control Structures and Microprogramming; Evolutionary algorithms; Evolutionary computation; Power consumption; Prediction models; Specific surface; Surface area; cement grinding mill; differential evolutionary algorithm; LSTM; XGBoost; PCA
• ISSN: 0146-4116; 1558-108X
• DOI: 10.3103/S0146411622060049

An improved differential evolutionary cement mill operation index decision algorithm based on constraint control and selection strategy is proposed to address the problem that the operation index is usually decided by manual experience in the cement mill operation process, which causes unqualified cement specific surface area and excessive cement mill power consumption. The algorithm uses an improved differential evolutionary algorithm to solve the long short-term memory network (LSTM) power consumption prediction model and specific surface area prediction model based on PCA and XGBoost. Constraints are set in the solution process to obtain the optimal solution that satisfies the quality index and the power consumption index. The optimal solution is used to guide the scheduling of each piece of equipment in the production process and to make decisions on the operating index of the cement mill grinding process so that the specific surface area is qualified and the power consumption is reduced.