Title :
Quality control of coal jigging in a coal-preparing plant using MEBML method
Author :
Ma, Fangqing ; Liu, Yunxia
Author_Institution :
Coll. of Inf. & Electr. Eng., China Univ. of Ming & Technol., Jiangsu, China
Abstract :
Mind-evolution-based machine learning (MEBML) is a new kind of evolutionary computing algorithm. MEBML substitutes similartaxis and dissimilation for crossover and mutation operators used in GA. It possesses more rapid convergence and higher calculation accuracy. In this paper, MEBML was used as an approach for dynamic quality control of coal jigging in a coal-preparing plant. The dynamic quality control system based on MEBML for jigger can control the ashes of three products and their product ratios, the system can raise the fine coal quality effectively. The system can also realize expert control and neuro-PID control for feeding, discharging, winding, watering loop control. Results presented in the paper clearly demonstrate the feasibility of the proposed scheme.
Keywords :
closed loop systems; coal; coal ash; evolutionary computation; fuel processing industries; learning (artificial intelligence); neurocontrollers; production control; pulverised fuels; quality control; three-term control; PID control; coal jigging; coal-preparing plant; convergence; crossover operators; dynamic quality control system; evolutionary computing algorithm; expert control; genetic algorithm; loop control; mind-evolution-based machine learning; mutation operators; neurocontrol; similartaxis; Accuracy; Ash; Control systems; Convergence; Genetic mutations; Machine learning; Machine learning algorithms; Neural networks; Quality control; Statistics;
Conference_Titel :
Control and Automation, 2005. ICCA '05. International Conference on
Print_ISBN :
0-7803-9137-3
DOI :
10.1109/ICCA.2005.1528155
