Investigation of harmonic resonances in the power system of an electric vehicle

Sammanfattning

Heavy road transport manufacturers such as Scania are developing new electric vehicles in order to reduce greenhouse gas emissions. These vehicles have many systems which are connected through power electronic converters that introduces ripples into the power system. These ripples excite harmonic resonances in the power system’s passive circuit posing new design challenges. This project addresses the challenge by developing a time-frequency analysis method of measurement data and creating a model for the purposes of ripple estimation. Measurement data is gathered in accordance with standard measurement procedures at Scania. The Short-Time Fourier Transform is then used to perform time-frequency analysis. By multiplying smaller segments of the signal with a window function and applying the Fast Fourier Transform, the frequency-domain components of the signal can be obtained for various timesteps. Series resonances that amplify current ripples were determined to be the main problem. Therefore different subsystems can be modeled as simple series RLC circuits whose component values can be determined from data sheets or estimated. The data analysis indicates that lowering the switching frequency increases the amplitude of any ripples which is worse for the condition of the power system in the long run. It also indicates at what frequencies significant ripples occur which could inform filter design to avoid the issue of harmonic resonances. The model captures the general trends in the frequency and magnitude of the largest ripples absorbed by each subsystem although it does not match the measurement perfectly. In the long term, this model could be further developed to estimate how changes in various components caused by heating while driving may affect the ripples absorbed by the subsystems.