Machine Learning based Parametric Fault diagnosis of Cuk converter

With the advent of renewable energy systems, Electric Vehicles and smart grid, there is a growing need for high performance DC-DC converters in various applications. Inorder to improve reliability and ensure sustained operation, fault diagnosis of power converters needs to be done. Parametric Fault diagnosis mainly deals with monitoring the deviation of components’ parameters from its nominal operating range present in a converter circuit. This work focusses on predicting the parametric faults in a Cuk converter using Machine Learning. Different Fault scenarios are simulated in MATLAB Simulink and the results are used in training the machine learning model. The trained ML model can sense single component fault of Cuk converter. Performance of Machine learning classifiers in terms of accuracy, along with feature extraction is discussed in this paper.