The excitation control of the generator has a great impact on the stability of the terminal voltage and the power grid, but traditional excitation control schemes have shortcomings of the dependency on the accurately mathematical model of the generator and weak robustness. In order to improve the control performance of the system, an online learning ANN-inversion(OLANNI) excitation controller of the multi-machine power system is designed in the paper. Firstly, the reversibility of the multi-machine excitation system is analyzed for the model of synchronous generator sets. Then, an online learning ANNI-inversion excitation controller is designed based on the ANN-inversion excitation controller oftline trained and an online learning algorithm of the ANN-inversion excitation controller is proposed with the idea of basis functions. Finally, simulations are conducted for the typical two-area four-machine power system. Simulation results show that the proposed OLANNI excitation controller is superior to both AVR/PSS and the offline learning ANNI excitation controller in the control performance when the controlled system suffers disturbance.
[1]
Wenxin Liu,et al.
Design of an adaptive neural network based power system stabilizer
,
2003,
Neural Networks.
[2]
George W. Irwin,et al.
Neural control of turbogenerator systems
,
1997,
Autom..
[3]
Om P. Malik,et al.
Application of an inverse input/output mapped ANN as a power system stabilizer
,
1994
.
[4]
Frank L. Lewis,et al.
Neural net robot controller with guaranteed tracking performance
,
1993,
Proceedings of 8th IEEE International Symposium on Intelligent Control.
[5]
Yuan Zhu-zhi.
Observer-based neural networks robust control for a class of nonlinear system
,
2004
.
[6]
Sun Zheng-shun,et al.
APPLICATION OF INVERSE SYSTEM METHOD FOR POWER SYSTEM INTEGRATED CONTROL
,
2001
.