Method of AC Servo System Based on Genetic Algorithm and Neural Network PID Control

Abstract:

BP network model has become one of the important neural network model, is used in many fields, but it has some defects. As from a mathematical perspective, it is a nonlinear optimization problem, which inevitably has the local minima problem; BP neural network learning algorithm has slow convergence rate, and the convergence speed and the initial weights of choice; network structure, namely the hidden layer nodes selection is still no theory until, but according to the experience. Based on the BP algorithm the local extreme values, considering the genetic algorithm and BP algorithm is combined with, on the BP neural network optimization. Neural network using genetic algorithm optimization mainly includes three aspects: the connection weights of evolution, evolutionary network structure, learning the rules of evolution.

• Periodical: Advanced Materials Research (Volume 717)
• Pages: 563-567
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.717.563
• Online since: July 2013
• Authors: Wen Chun Chang, Cheng Chen
• Keywords: GA-BP Neural Network, Genetic Algorithm (GA), Learning Rules

References

[1] Yunxing Shu, Zhang Yongsheng, Yu. Based on genetic algorithm and BP network optimization research. Journal of Shandong Institute of building materials, 2000,14 ( 1): 22~24

[2] Tan Yonghong. Adaptive control based on BP neural networks. Control theory and applications, 1994, 11 ( 1) :84~87

[3] Shen Qing, Hu Dewen, spring. Neural network application technology. National University of Defense Technology press, 1993,2~50

[4] He Yubin, Li Xinzhong. Neural network control technology and its application. Science Press, (2000)

[5] Wang Fengqin, Gao Ying, Zhao Jun. Neural network optimization based on genetic algorithm. Journal of Yanshan University, 2001,25 ( 3) :234~238

[6] Luo Ronghai robot automatic guidance, tracking control research. Nanjing Aerospace University Master thesis, 2002:12~13

[7] K.S. Narendra, K.Parthasarathy.Identification and Control for Dynamical Systems Using Neural Networks.IEEE Trans Neural Networks, 1990,1 ( 1): 4~27

DOI: 10.1109/72.80202

[8] Hang Yong . Diesel engine control model and control algorithm design and simulation research . Jiangsu Polytechnic University doctoral thesis, (2002)