Fuzzy Neural Network Controller Based on Improved Genetic Algorithm for Temperature of Glass Tempering and Annealing Process

Abstract:

The temperature control of the glass tempering and annealing process has characteristics of time-varying parameters，nonlinear and big lag. It is difficult to meet the expected control effect with the common control method. To solve this problem，this paper puts forward a kind of fuzzy neural network controller optimized by genetic algorithm. First，it uses neural network to construct fuzzy logic system according to the structure equivalence rule，thus the optimization of fuzzy control rules and membership functions can be realized by finding the weight value of the neural network. Then，it uses the improved genetic algorithm to find the global optimum weighted factors with a high speed so to improve the performance of the controller. The simulation results show that the optimized fuzzy neural network controller can obtain an excellent control performance for the nonlinearity system with time- varying parameters and lag.

• Periodical: Advanced Materials Research (Volumes 154-155)
• Pages: 214-219
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.154-155.214
• Online since: October 2010
• Authors: Xiao Kan Wang, Zhong Liang Sun, Sanci Guo, Chao Qun Shen
• Keywords: Fuzzy Neural Network (FNN), Glass Annealing, Glass Tempering, Improved Genetic Algorithm (IGA), Lag, Optimation

References

[1] Jinjie Zhu, Shuting Tong, Fengjiao Guo. Journal of Building Materials, 2006，9(3)：377-380. ( In Chinese).

[2] Shiyong Li. Fuzzy control, neural netwok control and intelligent control theory [D]. Harbin Institute of Technology Press, (1998).

[3] Lingjiao Dong，Dongqing Feng. Computing Technology and Automation, 2005(04): 17-19.

[4] M. R .M. Rizk，etc. An Algorithmfor Optimum Stability Re2gion of Fuzzy Control Systems Using Genetic Algorithms [C] . Proceedings of the American Control Conference. Arlington ，VA 25 - 27 June. 2001 ：192 - 197.

[5] Xiaokan Wang，Zhongliang Sun，Lei Wang，Dongqing Feng. Design and Research Based on Fuzzy PID-parameters Self-tuning Controller with MATLAB [C]. 2008 International Conference on Advanced Computer Theory and Engineering (ICACTE 2008), Phuket Thailand: IEEE CPS, 996-999.

DOI: 10.1109/icacte.2008.9

[6] Xiaokan Wang, Lei Wang, Zhongliang Sun, Dongqing Feng. Short-Term Load Forecasting Based on RBF Adaptive Neural Fuzzy Inference[C]. 14th National Youth Communication Conference of China, USA: Scientific Reseach Publishing, 220-224.

[7] Xiaokan Wang, Zhongliang Sun, Lei Wang, Shouping Huang. Simulation and Optimization of Parameters on DC Motor Double Closed-Loop Control System Based on Simulink. 2009 International Conference on Intelligent Human-Machine Systems and Cybernetics（IHMSC09）Hangzhou: IEEE CPS, 153-156.

DOI: 10.1109/ihmsc.2009.46

[8] Xiaokan Wang，Shouxiang Zai, Zhongliang Sun. New Load Density Forecasting Method for Objective Network Planning. 2009 5th International Conference on MEMS NANO, and Smart Systems (ICMENS 2009) Dubai: IEEE CPS, 224-227.

DOI: 10.1109/icmens.2009.25