China International Youth Conference on Electrical Engineering (CIYCEE 2020)

    Partial discharge is one of the major causes of power equipment failures and operation accidents. Discharges in daylight environment are difficult to be observed due to the weak light signal. Digital images are often used to study corona discharge. However, ultraviolet cameras have limitations such as low pixels, high prices. This paper proposes a new method for diagnosing corona discharge in daylight environment. Based on the needle-plate electrode air gap discharge structure, we use a common visible light camera to take photos of the discharge. By converting the photo from RGB to HIS color space, we extract the hue and intensity information of each pixel. Analogous to the spectrogram, we propose the hue-intensity spectrum. After correlating the Hue-Intensity spectrum with the applied voltage value during discharge and the discharge images in a dark light environment, we found that the newly proposed spectrum has a go.od division effect in the discharge development stage. Meanwhile, we found that the corona inception voltage can be roughly estimated from the curve graph of relative brightness and applied voltage. This method provides a new idea for the diagnosis of discharge in daylight environment.