Optimisation of fractional-order PI controller for bidirectional quasi-Z-source inverter used for electric traction system

Abstract

This study presents the optimisation of fractional-order proportional-integral (FOPI) controllers for a bidirectional quasi-Z-source inverter (QZSI) in an electric vehicle (EV) off-road application. An ant colony optimisation Nelder-Mead (ACO-NM) algorithm is used for the optimisation of the controller parameters. This optimisation method is applied to enhance the performance of FOPI control for bidirectional QZSI. Ziegler-Nichols (ZN) with relay and the pole placement tuning method are also used for the FOPI controller design for comparison purposes. The modelling and the control design of bidirectional QZSI for an electric traction system are presented and discussed. Simulations are performed to verify the efficacy of the proposed controller structure with the bidirectional QZSI for two standardised driving cycles. The result shows that the FOPI controller designed with the ACO-NM algorithm provides more suitable ageing performance index values for the battery. The ACO-NM algorithm permits to reduce the root-mean-square value and the standard deviation by 2 and 5% of the battery current compared to the ZN tuning method and direct battery supply topology, respectively. The bidirectional QZSI with this type of controller can globally enhance the performance of EVs by optimising the electric power consumption and extending its driving range.