Genetic algorithm-based output power optimisation of fault tolerant five-phase brushless direct current drives applicable for electrical and hybrid electrical vehicles

Abstract

This study presents a method to improve the output power of five-phase brushless direct current (BLDC) motors under different faulty conditions. Different machine connections are considered while having open-circuit fault in one and two stator phases, and both fundamental and third harmonic component of stator currents are controlled to improve the amplitude and quality of generated torque under faulty conditions. Rated root-mean-square value of stator phase currents is considered as the main limiting factor of generated electrical torque. Genetic algorithm is used in the optimisation procedure of stator reference currents to gain more output power under the fault. Automotive applications are kept in mind, and to verify the theoretical developments, experimental tests are conducted on a five-phase BLDC motor with in-wheel outer-rotor configuration.