An image-based deep transfer learning approach to classify power quality disturbances

• DOI: 10.1016/j.epsr.2022.108795
• Publication Date: 2022-09-16T05:10:48Z
• Authors: Grazia Todeschini; Karan Kheta; Cinzia Giannetti
• Keywords: Transfer of learning

Power quality disturbances (PQDs) consist in deviation of voltage and current waveforms from the ideal sinusoid at fundamental frequency, need to be monitored ensure a reliabile electrical supply. While, traditionally, power monitoring has been performed using signal processing techniques, coupled with shallow Machine Learning classifiers or wave change detection methods, more recently, new approaches, based on Deep Learning, have proposed. These methods potential achieve high classification accuracy remove extensive data pre-processing, hence being suitable for real-time deployments. However, performance only demonstrated synthetically generated data. In order address limitations related time accuracy, this paper proposes novel end-to-end framework automated PQDs Transfer Learning. The proposed approach uses small set images train model classify different types PQDs. This method leverages existing pre-trained models image shows consistent varying resolution. methodology provides pathway towards effective deployment systems applications.