Detailed analysis of various techniques used for fault detection and analysis in multilevel inverters based on renewable energy sources

With the advent of industrialization, power generation based on renewable energy sources is so prominent. This in turn leads to a drastic increase in the use of multilevel inverters to drive such high power and high voltage applications. This paper aims at discussing various fault classification techniques used to detect faults. Comparison starts from traditional techniques like fast fourier transform, wavelet transform focuses on fault classification in cascaded ${H}$ bridge nine-level inverter. Comparative fault analysis and classification are presented. Fast Fourier transform and wavelet-based fault classification are few traditional methods used to classify faults in multi-level inverters. With the advent of soft computing techniques fuzzy logic controller based fault classification and mitigation came into practice. Furthermore, machine learning based fault classification based on Support vector machine and various other techniques also contributes to fault diagnosis of multilevel inverters.ANN based fault diagnosis and detection has embarked a trend in this field. Fault classification and detection based on neural networks are widely used in research nowadays. This research study will not only focus to discuss on advantages, disadvantages and application but also will give an insight of various fault possibilities and mitigation techniques available to rectify them. A brief of various renewable energy sources used for every mitigation technique discussed, is also listed to give a clear insight of effective utilization under such energy demand and crisis existing worldwide.