A New Control Strategy for the STATCOM Based on Multi-Model Neural Network PI Controller

Aiming to dealing with the problems of power factor compensation and the limitations of the conventional PI controller in the Static Synchronous Compensator (STATCOM), a new control strategy based on multi-model and neural network PI controller is proposed. This control scheme applied the multi-model and neural network technology to the PI controller to meet the accuracy and speed of the power factor compensation under different impact loads. Meanwhile, the neural network technology is used to tune the PI controller parameters values according to an optimal control law, which can meet the requirements of full range working conditions and optimality. Simulation experiments show that compared to the traditional PI controller, PI controller based on multi-model and neural network is proved to be better capable of adapting to the changes of impact loads with a higher compensating precision, which makes the power factor maintained at about 1 after compensation.