The on-site detection technique for silicone oil level in high voltage cable terminal using infrared image processing techniques

Oil-filled high voltage cable terminals are widely used in urban power transmission networks. It is a common issue that the oil leakage poses troubles on the reliability of the cable lines. This paper proposes a technique to evaluate the oil level in the cable terminal using infrared imaging technique and corresponding image processing algorithm. In this paper, Canny operator and the threshold value of the sum of pixels are utilized to extract the region of interest. The peak envelop curve and its gradient curve are obtained to identity the oil level in the terminal. An on-site detection case of an oil-leaked terminal verifies the proposed technique.