Gaussian Process Regression based Model for Prediction of Discharge Voltage of Air Gaps under Positive Polarity Lightning Impulse Voltages

Abstract

Discharge voltage of insulation is pivotal in the design of High Voltage systems. In this work, a machine learning algorithm is used to develop a model to predict the discharge characteristics of air. Finite Element Method (FEM) simulations have been performed to extract different electric field and energy features of air gaps in the range 5mm-40mm under lightning impulses of positive polarity. While developing the model, these features along with gap lengths are considered. The features have been used for training a machine learning algorithm based on Gaussian Process Regression (GPR) to develop the model. The results obtained from the model are validated with measured experimental data. A good comparison between the predicted data and the measured data establishes the accuracy of the predicted model. The proposed methodology is compared using different kernel functions.