Insulator Contamination Monitoring Using Machine Learning Methods: A Review

Insulators are one of the most important parts of power systems that can affect the overall performance of high voltage (HV) transmission lines and substations. High voltage (HV) insulators are critical to the successful operation of HV transmission lines, and failure of any because of contamination can result in flashover voltages that lead to power outages. Due to airborne pollution particles attach to the insulator surfaces, if moisture exists, the adhesive solution particles form conductive layers on the insulator surface that leading leakage current. In this paper several methods aim to monitor the contamination levels in high-voltage insulators were reviewed including thermal imaging, ultraviolet imaging, digital imaging, ultrasonic signal, sound signals, leakage current and partial discharge. Also, various machine learning methods aim to classification of insulator contamination level have been investigated as well as their advantages and disadvantages based. on the published researches.

Keywords:

machine learning , neural network , classification , insulator pollution