The Fault-Tolerant Generator for Vehicles with a Hybrid Power Plant Design and Optimization

Abstract

The aim of this work is the development of the algorithm and the permanent magnets starter-generator topology selecting process for a 500-kW hybrid power plant with a rotor speed of 2500 rpm, taking into account the fault-tolerance requirements for maximum energy density and maximum efficiency. The novelty of the algorithm is that numerical optimization is performed for each topology by using a genetic algorithm before the estimation of its main advantages. Thus, the optimal geometric dimensions of topologies are first determined and then compared. This feature distinguishes the proposed algorithm from those known from literature. The algorithm improves accuracy and provides an effective selection of starter-generator topologies for the transport industry. The topology selection is based on the methods of two-dimensional and three-dimensional computer simulations in the Ansys Maxwell software package. The 3D-simulation is used to estimate the distribution of eddy-current losses in the rotor. The obtained results are useful to those engineers and researchers who are engaged in the development of electrical machines for hybrid power plants.