Multivariate Cooperative Internal Mode Control of RBF Neural Network for Power System Chaos Suppression

In this paper, a multivariate cooperative internal mode control method based on RBF neural network (RBF-NN) inverse system is proposed to suppress the chaotic behavior in the power system. Firstly, a seven-dimensional model of the controlled power system including energy storage (ES) and static var compensator (SVC) is constructed, and the chaotic dynamics of the system is analyzed by local bifurcation diagram, attractor phase diagram and timing diagram, and the merging crisis and coexistence of chaotic attractors are found in the power system under the action of the low-frequency power disturbance. Secondly, considering the parameter uncertainties of the power system, an RBF-NN inverse system model of the controlled power system is established based on inverse system theory and neural network theory to realize its pseudo-linearization, and a multivariable cooperative internal mode controller is designed to suppress the chaotic behavior in the power system by combining the ES and the SVC. Finally, the effectiveness and robustness of the proposed chaos suppression control strategy are verified by simulation.