Surrogate Modeling of Random Electromagnetic Field Coupling to Transmission Lines Using Support Vector Machine Regression

This paper addresses the challenges of analyzing random electromagnetic field coupling to transmission lines, which are critical for electromagnetic compatibility (EMC). Traditional deterministic methods struggle with stochastic variations. To overcome this, the study employs machine learning, specifically Support Vector Machine (SVM) regression, to create an efficient surrogate model. Trained on simulation data, the model accurately predicts system behavior under random conditions, reducing computational complexity compared to Monte Carlo simulations. The approach enhances EMC studies by providing a scalable and robust method for analyzing random field coupling effects on transmission lines.