Rotary kiln process is a strongly coupled multivariable nonlinear system with large lag. It has complex control target with less manipulate variables than controlled variables, and the boundary conditions vary frequently and severely during operation. Most rotary kilns are still under manual control such that the burning status cannot be maintained stable, the product quality is hard to be kept consistent and energy consumption remains at high level. To deal with this problem, this paper presents a pattern-based hybrid intelligent control strategy. Because of the vital importance of the burning zone temperature, several patterns representing kiln operation conditions are distinguished through burning zone temperature pattern recognition. A MIMO supervisory fuzzy controller is designed to address qualitative decoupling among controlled variables in the normal regulation phase, and a human mimic controller is employed to avoid overshoot when process large disturbance happens in the abnormal regulation phase. The two controllers are switched by the pattern recognition model, with their parameters tuned also by the pattern recognition model. This pattern-based hybrid intelligent control approach has been implemented in DCS and successfully applied in an alumina rotary kiln. Satisfactory results have shown that the adaptability and performances of the control system have been improved effectively, and remarkable benefit has been obtained.
[1]
L. P. Holmblad,et al.
The FLS application of fuzzy logic
,
1995
.
[2]
Stephen Yurkovich,et al.
Fuzzy Control
,
1997
.
[3]
Raghunathan Rengaswamy,et al.
A review of process fault detection and diagnosis: Part III: Process history based methods
,
2003,
Comput. Chem. Eng..
[4]
Mika Järvensivu,et al.
Intelligent control system of an industrial lime kiln process
,
2001
.
[5]
Hector Budman,et al.
Model predictive control with soft constraints with application to lime kiln control
,
1999
.
[6]
Esko Juuso,et al.
Intelligent control of a rotary kiln fired with producer gas generated from biomass
,
2001
.
[7]
Gerardo Beni,et al.
A Validity Measure for Fuzzy Clustering
,
1991,
IEEE Trans. Pattern Anal. Mach. Intell..
[8]
James C. Bezdek,et al.
Pattern Recognition with Fuzzy Objective Function Algorithms
,
1981,
Advanced Applications in Pattern Recognition.
[9]
S Marsili-Libelli.
Control of SBR switching by fuzzy pattern recognition.
,
2006,
Water research.
[10]
Heng Yue,et al.
Supervisory Control for Rotary Kiln Temperature Based on Reinforcement Learning
,
2006
.