Neural-network-based inverse control method for active power filter system

A new type of active power filter (APF) is described. A multi-layered neural network based inverse control method for this APF system is proposed. The functioning of the APF system is based on a switching network whose characteristics are nonlinear. The characteristic of the switching on-off time of the switching network and the output current of the APF was demonstrated. The switching on-off time can be instantaneously calculated by using the Neural-Network-Based Inverse Control algorithm proposed. The neural network was designed. An all digital control way may be realized by using the algorithm. The validation of the results of Inverse Control for APF is proposed as well.