Study on Improved Neural Network PID Control for Coal Mining Power Grid Arc Suppression Coil Compensating System

Abstract:

The paper analyzed arc suppression coil with magnetic bias compensating system with linear system rules. The nonlinear and time-variable performances are considered during model building process. In order to optimize control effect, the paper adopted improved BP neural network PID controller with closed loop control method. Improve BP neural network with the combination of the two strategies, adding momentum method and adaptive learning rate adjustment, can not only effectively suppress the network appearing local minimum but also good to shorten learning time and improve stability of the network furthermore. The results of simulation and experiments indicate that arc suppression coil based on improved neural network with PID control method can quickly and accurately control the compensating capacitive current to an expected value and it has strong robustness. The paper also provided core controller with software and hardware designing scheme based on STM32 microcontroller.

• Periodical: Key Engineering Materials (Volumes 426-427)
• Pages: 427-431
• DOI: https://doi.org/10.4028/www.scientific.net/KEM.426-427.427
• Online since: January 2010
• Authors: C.Y. Ma, C.L. Wang, J.H. Liu, X.B. Li, R. Liang
• Keywords: Arc Suppression Coil Compensating System, Improved BP Neural Network, PID Control

References

[1] W.H. Zhang: Doctoral dissertation of CUMT (2008).

[2] L.H. Mou, Q.H. Meng: Power supply and distribution security technology (Mechanical Industry Press, China 2003).

[3] R. Ding, C.L. Wang: Industry and Mine Automation (2007) No. 2, pp.1-3.

[4] J.K. Liu: Advanced PID Control and its MATLAB Simulation (Electronic industry Press, China 2003).