A study on a new algorithm to optimize ball mill system based on modeling and GA

• Published in: Energy conversion and management Vol. 51; no. 4; pp. 846 - 850
• Main Authors: Wang, Heng, Jia, Min-ping, Huang, Peng, Chen, Zuo-liang
• Format: Journal Article
• Language: English
• Published: Kidlington Elsevier Ltd 01.04.2010; Elsevier
• Subjects: Applied sciences; Ball mill; Ball mills; Coal and derived products; Electric energy; Energy; Exact sciences and technology; Fuels; Genetic algorithm; Genetic algorithms; Mathematical models; Mechanical preparation; Neural network; Neural networks; Nonlinearity; Optimization; Power consumption; Processing; Pulverizing; Electric energy; Neural network; Ball mill; Genetic algorithm; Energy conservation; Spraying; Coal; Electric power consumption; Energy savings; Optimization; Operating conditions
• ISSN: 0196-8904; 1879-2227
• DOI: 10.1016/j.enconman.2009.11.020

Aiming at the disadvantage of conventional optimization method for ball mill pulverizing system, a novel approach based on RBF neural network and genetic algorithm was proposed in the present paper. Firstly, the experiments and measurement for fill level based on vibration signals of mill shell was introduced. Then, main factors which affected the power consumption of ball mill pulverizing system were analyzed, and the input variables of RBF neural network were determined. RBF neural network was used to map the complex non-linear relationship between the electric consumption and process parameters and the non-linear model of power consumption was built. Finally, the model was optimized by genetic algorithm and the optimal work conditions of ball mill pulverizing system were determined. The results demonstrate that the method is reliable and practical, and can reduce the electric consumption obviously and effectively.