Application of Data Analysis Techniques and Machine Learning Methods to the Development of High Voltage Circuit Breakers

This study aims to develop a systematic methodology utilizing data analysis and machine learning (ML) techniques to investigate the influence of various design parameters on the thermal clearing capability of high-voltage circuit breakers (HVCBs). By analyzing design parameters, test circuit parameters, and test result parameters, the objective is to identify correlations and interdependencies, and develop thermal clearing criteria that improve prediction of thermal clearing capability in high-current test duties using a CO<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf> + C<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">4</inf>F<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">7</inf>N-based gas mixture for current interruption.ML analysis demonstrates that design parameters can be used to predict circuit breaker (CB) thermal clearing performance, thereby revealing patterns that cannot be obtained from correlation methods alone. Correlation analysis indicates that arc resistance is closely linked to thermal clearing. Simple regression models effectively predict compression volume (CV) pressure peak using design parameters, arcing time, and applied current. Successful classification of thermal clearing performance through ML methods with an accuracy of up to 91% requires the incorporation of test result parameters such as arc resistance. Support Vector Machines (SVM) can be used to define a hyperplane that separates successful tests from failures based on pressure, arcing time, and R<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">200</inf>. The results provide a conservative guideline for the nozzle design regarding pressure at current zero (CZ).