Open Collections

Abstract

In recent years, as induction motor drive systems are increasingly applied in industrial automation, such as in renewable energy and electric vehicles, the demands for performance analysis and validation of motor drive systems have grown significantly. A newly proposed simulation approach, exponential integrator algorithm, has been introduced due to its great numerical accuracy and computational efficiency across various circuit configurations when simulating power electronic systems in prior-art works. However, the research gap is that the application of this state-of-the-art numerical algorithm for induction machine drive system simulation is unexplored. Under this background, this work proposes a high-order, variable-step Exponential Integrator-Voltage Behind Reactance (EI-VBR) machine model which retains the desirable efficiency and stability of EI algorithm and the structural simplicity of the VBR machine model. Additionally, several numerical efficiency optimization techniques are incorporated to enable model precomputation, which is one of the key features of EI algorithm to further enhance the overall simulation efficiency. By applying EI-VBR model to typically induction machine drive case studies, the simulation results showcase its important numerical advantages. The comparison with other widely used Ordinary Differential Equations (ODE) solvers from popular commercial software packages, such as Simulink/SimPowerSystems and PLECS, illustrates the proposed EI-VBR model outperforms all the prior-art models in terms of computational efficiency while maintaining great numerical accuracy. Under non-stiffness conditions, the EI-VBR model achieves up to a 20-fold speedup compared to the fastest solvers from commercial software packages. As the system stiffness increases, its advantages become more pronounced, achieving at least a 50-fold simulation speedup.