Optimization of the Hydraulic Control System Utilizing BP Neural Network Control Strategy Based on Genetic Algorithm

Abstract:

For the characteristic of the MMS series Thermo-Mechanical Simulator hydraulic control system, using traditional PID control method can not achieve the desired control effect. Basing on genetic algorithm, BP neural network, which has the arbitrary non-linear approximation ability, self-learning ability and generalization ability, has been used into the hydraulic control system to achieve the online adjustment of the weighting coefficients and the adaptive adjustment of PID control parameters. The results of simulation and online tests show that the control effect of hydraulic system has been improved significantly, and the accurate control of hydraulic system hammer displacement has been realized.

• Periodical: Applied Mechanics and Materials (Volumes 397-400)
• Pages: 1245-1252
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.397-400.1245
• Online since: September 2013
• Authors: Ying Ying Feng, Nan Mu Hui, Zong An Luo, Dian Hua Zhang
• Keywords: BP Neural Network (BPNN), Genetic Algorithm (GA), Hydraulic Control System, Online Test, Simulation

References

[1] Fuji Dempa Kori Co., Ltd, in: Specification of Themecmastor-Z, High Temperature Deformation Simulation and Mechanical Testing Machine Tokyo, Japan: Fuji Dempa Kori Co., Ltd. (1996).

[2] Nippes E F, Savage W F: Welding Journal Vol. 28 (1949), p.599.

[3] Zong-An LUO: A new generation of thermal simulation technology and equipment-The development of MMS-100 Thermo-Mechanical Simulator, Shenyang, Northeastern University, 2006: 116-122. (In Chinese).

[4] Zhi-Jun Chen: The Research of Computer Measuring and Controlling System for Hydraulic Universal Testing Machine , Hangzhou, Zhejiang University of Technology, 2003. (In Chinese).

[5] Xin YAO: Proceedings of the IEEE Vol. 87 (1999), p.1423.

[6] Phua P K H, MING Dao-hua: IEEE Transactions on Neural Networks, Vol. 14 (2003), p.1460.

[7] Qi-quan LIN, Da-shu PENG and Yuan-zhi ZHU: Journal of Central South University of Technology Vol. 12 (2005), p.380.

[8] Lippmann R P: IEEE Communication Magzine Vol. 27 (1989), p.47.

[9] Lang K J, Waibel A H and Hinton G E: Neural Networks Vol. 3 (1990), p.23.

[10] Pan H: Computer Application Vol. 25 (2005), p.2777.

[11] Gua R S: International Journal of Pattern Recognition and Artificial Intelligence Vol. 18 (2004), p.79.

[12] Holland J H: SIAM J of Computing, Vol. 2(1973), p.88.

[13] Brockwell P J, Davis R A: Time series Theory and Methods [M]. New York: Springer-Verlag New York Inc, (1987).

[14] Srinivas M, Patnaik L M: IEEE Trans on SMC Vol. 24 (1994), p.656.

[15] Yong-Tang Li, Bu-Fang Lei and Yu-Zhuang Gao: Metallurgical Industry Press Vol. 1 (2003), p.9 (In Chinese).

[16] Juan-Ping Zhao, Yu-Xia Zhang: Journal of Shenyang Institute of Chemical Technology Vol. 21 (2007), p.134(In Chinese).