MODEL PREDICTIVE CONTROLLER BASED AC-DC BIDIRECTIONAL CONVERTER FOR PHEV

Conversion from combustion based vehicles to environmental friendly electrical vehicle has become a trend among automobile industries. This thesis brings out a new idea to support power requirement of vehicle local loads through the power harvesting from piezoelectric generation associated with the vehicle's vibratory parts. Second priority of the paper is to develop a model predictive control algorithm of three-phase bidirectional AC-DC converter that can be used for V2G and G2V applications. This algorithm utilizes the discrete behavior of the converter to determine appropriate switching states that minimize the cost function. The proposed predictive control scheme allows bidirectional power transfer with instantaneous mode changing capability and fast dynamic response. This paper also presents the results obtained for the study conducted using MATLAB. The test was carried out with a 8kW induction motor to confirm the feasibility of the proposed control scheme. Both the simulation and experimental results prove that the converter is operated with sinusoidal grid current with acceptable THD level.