DESIGN AND FEM ANALYSIS OF PERMANENT MAGNET SYNCHRONOUS MOTOR FOR ELECTRIC VEHICLE APPLICATIONS

Abstract

The depletion of fossil fuels and growing concerns relating to environmental pollution have triggered extensive research for improving the performance of electric vehicle powertrain. High performance electric vehicles are replacing conventional Internal Combustion Engine based vehicles leading to reduction in consumption of fossil fuel. This is a big breakthrough for the vehicle industry. Even though, preservation of fossil fuel can be achieved through adoption of electric vehicles, additional complexity and challenges faced in developing an efficient electric vehicle powertrain with high performance and low maintenance needs to be addressed. In this context, design of an efficient electric vehicle power train is essential. Selection of high performance traction motor, with simple and accurate control schemes, is crucial for different electric vehicle applications. Among the various kinds of electric motors that are being researched for use in electric vehicle applications, Permanent Magnet Synchronous Motor has prominent advantages such as fast response, compact size, light weight, high efficiency, brushless configuration and good dynamic performance. In this research work, conventional internal combustion engine and electric powertrain are modelled in MATLAB/Simulink for analysing the performance of these two distinct propulsion configurations. The conventional powertrain using spark ignition type of generic engine is modelled in Simulink and the information obtained from the dynamic analysis of conventional powertrain is set as the benchmark parameters for evaluating the performance of electric powertrain. The electric powertrain is modelled and simulation studies are carried out through the implementation of Space Vector Pulse Width Modulation and Field Oriented Control based control strategies. The dynamic performance of traction motor is analysed, for both traction and braking modes of operation, under different operating conditions. The modelled electric motors proposed for electric vehicle powertrain are found to be robust, with fast dynamic response in achieving the required torque and speed demand and their perturbations. The results obtained from the proposed electric vehicle powertrain models are in agreement with their suitability for EV application. The permanent magnet synchronous motor for the power train is designed using Finite Element analysis to improve the electromagnetic performance for different electric vehicle applications. Simplification of the structure of the electric motor used in electric vehicles can result in enhanced and improved performance. Accordingly, structural optimization of the motor v for electric vehicle application is carried out to enhance performance, through reduction in losses and increase in efficiency. This is achieved by designing a surface inset type permanent magnet synchronous motor and improving its performance parameters through finite element analysis, surrogate modelling technique and metaheuristic algorithms. Since the electromagnetic performance of the motor shows high dependence on the variation of its design parameters, the response surface models for various performance factors of motor are developed in terms of the specific design parameters. For this, metamodeling technique is implemented in MATLAB. The obtained meta-models are utilized as objective functions in the implementation of metaheuristics techniques to achieve improvement in the electromagnetic performance of the motor. A novel optimization technique called Integrated Taguchi method assisted polynomial Metamodeling & Genetic Algorithm is developed to optimize the motor geometry for electric compressor application, like environment control, in electric vehicles. The optimal results obtained from proposed approach are compared with those obtained from the traditional design optimization technique and found to be giving better response of the motor. In addition, a novel design optimization algorithm called Aquila Grasshopper Optimization is developed to achieve minimum core losses through improved motor geometry of the traction motor. Trench cuts are introduced in this optimized geometry to obtain light-weight and high power density. Sensitivity analysis of the motor design parameters is also carried out to predict the effect of variation in geometrical parameters on its electromagnetic performance. Further, thermal modelling of permanent magnet synchronous motor is performed via Lumped Parameter Thermal Network. The change in the temperature in different parts of the motor under different operating conditions are observed and used to detect thermal faults in the motor. To identify which part of the motor is faulty, statistical approach of Linear Discriminant Analysis method is used. Different nodes in Lumped Parameter Thermal Network are treated as differentiable classes (or groups). Finite Element analysis is used for evaluating the loss information in motor parts (nodes). The change in temperature is computed for specific operating conditions and the corresponding discriminant functions are obtained for each class feature. For particular change in temperature at any operating condition, the specific classes are identified to diagnose the faulty part in the motor.