The Application of PID Based on Optimized RBF in Thickness Control of Strip Steel

Abstract:

The strip thickness control system is difficult to establish an accurate mathematical model, and traditional PID control strategy has a poor adaptive ability, so the effect of control is always not satisfying. According to the problems above, a new control strategy of self-tuning PID controller based on RBF neural network whose parameters are optimized by PSO algorithm is proposed in the paper. The control method integrates advantages of RBF neural network as well as PID controller and good global search capability of PSO algorithm. The simulation results indicate that the method not only improves control performance and dynamic quality, but also has strong self-adapting ability and robustness. It achieved a very good control effect when used in strip thickness control system that proved the correctness and effectiveness of the control method.

• Periodical: Advanced Materials Research (Volumes 466-467)
• Pages: 52-56
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.466-467.52
• Online since: February 2012
• Authors: Yu Zhen Yu, Xin Yi Ren, Chun Yan Deng, Xiao Hui Wang
• Keywords: PSO Algorithm, RBF Neural Network, Robustness, Self-Tuning PID Control, Thickness Control

References

[1] Yang An, Ouyang Qi, Modeing and Simulation of Hydraulic AGC system of Rolling Gap Control [J], Machine tool and Hydraulic, 2008, 36(9): 243.

[2] Zhang Nai-yao. Neural Network and Fuzzy Control [M], Beijing: tsinghua university press, (2000).

[3] Xu Xin, He Han-gen, A Gradient Algorithm for Neural Network Based Reinforcement Learning [J], Chinese Journal of Computers, 2003, 26(2): 1.

[4] Zhang Ding-xue, Guan Zhi-hong, Liu Xin-zhi, RBF Neural Network Algorithm Based on PSO Algorithm and Its Application [j], Computer Engineering and Application, 2006, 42(20): 13.

[5] Shi Y, Eberhart R C, A Modified Swarm Optimizer [A]. IEEE International Conference of Evolutionary Computation [C]. Anchorage, 1998, 69.