A Predictive Energy Management Strategy for Hybrid Electric Vehicle Using Fast Iteration Algorithm

This study proposes a predictive energy management strategy for hybrid electric vehicle using fast iteration algorithm. First, a sectional power preconditioning method based on demand power prediction is proposed to increase the generated output current throttle opening and power prediction sequence. Second, the fast iteration sequential clustering quadratic programming (FSCQP) algorithm is proposed in the receding horizon of model predictive control. The clustering algorithm is applied to gather up the points with less iteration step size in the iteration process, and to design the termination criterion based on the distance between the cluster centers of various iteration domains. Moreover, projection matrix is introduced to modify the search direction of the optimization process in subproblems. The direction of gradient projection is applied as the steepest descent direction to replace the original descent direction. Finally, the performance of the proposed strategy is validated in simulation tests. The results reveal that the proposed strategy achieves 5.21% less fuel consumption and 12.46% less iterations while maintaining the stable power output.