Aimed at the low quality of the product and low recovery rate of coke calcinations in rotary kiln, an intelligent optimization operations system for coke calcinations rotary kiln is presented to stabilize the temperature in calcination zone in this paper. The key operation parameters in rotary kiln are optimized using improved case-based reasoning (CBR). The case searching and matching adopted the weighted k-nearest neighbor (KNN) approach. An adaptive particle swarm optimization (PSO) with mutation is applied to optimize the weighted vector of features and the number of nearest neighbors of the weighted k-nearest neighbor approach in case searching and matching. Moreover, the control strategies in feeding rate of the disc feeder and interlock protection strategy are introduced briefly. The running results show that the intelligent optimization operation system can efficiently stabilize the temperature at calcination zone and can enhance the feeding rate control accuracy. Moreover, it can improve the coke calcination ratio and the mean productivity of the calcined coke per day. It brings significant economic benefit and social benefits to the carbon plant and have a bright future and a wide market.
[1]
Xiao Yun-mao.
Application of Case Based Reasoning in Injection Moulding
,
2008
.
[2]
T. Krink,et al.
Particle swarm optimisation with spatial particle extension
,
2002,
Proceedings of the 2002 Congress on Evolutionary Computation. CEC'02 (Cat. No.02TH8600).
[3]
Pak Kin Wong,et al.
Case-based Reasoning and Adaptation in Hydraulic Production Machine Design
,
2002
.
[4]
Ambrose Sunny Ochi-Okorie,et al.
Disease diagnosis validation in TROPIX using CBR
,
1998,
Artif. Intell. Medicine.
[5]
E.B Reategui,et al.
Combining a neural network with case-based reasoning in a diagnostic system
,
1997,
Artif. Intell. Medicine.
[6]
Hojjat Adeli,et al.
Case-based reasoning in steel bridge engineering
,
2005,
Knowl. Based Syst..
[7]
S. Wesley Changchien,et al.
Design and implementation of a case-based reasoning system for marketing plans
,
2005,
Expert systems with applications.