The Direct Inverse-Model Control Based on Neural Networks for Inverts

This paper proposed the inverse model control strategy of the inverter in the field of active power filter. Since the inverse model of the inverter can not be easily as well as accurately obtained, BP neural network is introduced to approach the inverse model. After the training of the neural network, we can connect it with the inverter to get a nearly ideal linear system. Various simulations are done under different conditions to prove this method to be effective. Finally an active power filter system is simulated, the result proved to be favorable.