Eefficient switching of three-phase PFC converters based on a sensorless current control neuro-computing space vector modulation

A sensorless current control scheme based on a neuro-computing SVM is proposed in this paper. The neuron indexes in the proposed scheme are used to identify the highest current carrying phase and the switching vectors. In addition, the proposed scheme is simple, accurate, and easy to implement. Moreover, four different switching patterns are introduced, and a novel switching pattern that reduces the switching events by a factor of 2/3 and maintains the advantages of the 60deg-clamping is proposed. Simulation results confirm the validity of the analytical work