Mind-evolution-based machine learning for dynamic quality control of raw materials in a cement plant

Author

Fangqing, Ma ; Yunxia, Liu

Author_Institution

Coll. of Meteorol., Mil. Univ. of Technol., Nanjing, China

Volume

3

fYear

2004

Firstpage

2409

Abstract

Mind-evolution-based machine learning (MEBML) is a new kind of evolutionary computing algorithm. MEBML substitutes similar taxis and dissimilation for crossover and mutation operators used in GA. It possesses rapider convergence and higher calculation accuracy. MEBML, used as an approach for dynamic quality control of raw material in a cement plant is presented in this paper. The dynamic quality control system based on MEBML for raw materials can control the modulus of raw materials and the coal of raw materials, the system can raise the raw materials´ quality effectively. This dynamic quality control system can also realize expert control for the load of raw mill and neuro-PID control for raw materials feeding. Results presented clearly demonstrate the feasibility of the proposed scheme.

Keywords

cement industry; control system synthesis; convergence; expert systems; genetic algorithms; learning (artificial intelligence); neurocontrollers; quality control; raw materials; three-term control; GA; cement plant; coal; convergence; crossover operator; dynamic quality control system; evolutionary computing algorithm; genetic algorithm; load expert control; mind evolution based machine learning; mutation operator; neuro PID control; raw materials; raw mill; Belts; Control systems; Educational institutions; Machine learning; Meteorology; Military computing; Milling machines; Quality control; Raw materials; Samarium;

fLanguage

English

Publisher

ieee

Conference_Titel

Intelligent Control and Automation, 2004. WCICA 2004. Fifth World Congress on

Print_ISBN

0-7803-8273-0

Type

conf

DOI

10.1109/WCICA.2004.1342026

Filename

1342026