Adaptive Neuro Fuzzy Inference System Based controller for Electric Vehicle's hybrid ABS braking

Abstract

To improve braking safety, modern electric vehicles are equipped with both friction and regenerative braking systems. The combined use of these two modes is mainly carried out in ABS operation. This paper deals with the problem of electric braking torque control in ABS mode. The EV considered is equipped with a traction motor of the interior permanent magnet synchronous machine type (IPMSM). Using field-oriented control (FOC) strategy, two torque regulators, namely adaptive fuzzy PID (AFPID) and adaptive neuro-fuzzy inference system (ANFIS) are developed and their control performances are analyzed and compared. In order to perform a realistic evaluation of the two aforementioned regulators, the hybrid ABS system, consisting of a slip controller, a braking torque references distribution law (electrical/friction), a friction brake, and an electric brake (IPMSM) is considered. Based on several performance indices, e.g. torque ripple, flux ripple, inverter switching frequency, control accuracy, dynamic response, and algorithm complexity, a comparison is made by simulation using Matlab /Simulink.