Research on waveform classification and loss optimization using machine learning

This paper discusses how the existing core material loss model can be improved through machine learning in the field of power electronics technology. With the rapid development of the national economy and advancements in science and technology, the application of power electronics technology is becoming increasingly popular across various fields, particularly in communication power supplies, new energy power conversion, and electric vehicles. The focus of this paper is on magnetic components, which play a crucial role in the design of high-frequency, high-power-density, and high-reliability converters. In this study, a machine learning method that combines data preprocessing and feature extraction is proposed. Principal component analysis (PCA) and various machine learning algorithms (such as logistic regression, random forest, and gradient boosting decision trees) were employed to classify and verify different excitation waveforms, thereby improving the model’s accuracy in complex practical situations.