As insulators play an important role in power transmission lines, we propose an intelligent method based on the convolution neural network (CNN) to evaluate the corona discharge of an insulator. In this method, the imaging device is a dual-spectra camera with a visible channel and an ultraviolet (UV) channel. The CNN is adopted to identify the detection distance of the insulator with the visible channel. To train the network, the dataset of the insulator is obtained by the experimental setup and deep convolutional generative adversarial networks. Through adjusting the training parameters and optimizing the network structure, an optimal trained model is achieved. Then the image pixel ratio method is adopted to measure the UV signal strength of the images captured by the UV channel. Meanwhile, the relationship between the detection distance and the UV signal strength is discussed. The critical value for the corona discharge of the insulator is obtained via experiments at the standard detection distance. Finally, the corona discharge of the insulator is evaluated by combining the detection distance with the UV signal strength. The experimental results show the method has the advantages of high accuracy and robustness and can effectively evaluate the corona discharge of the insulator.
We present a method based on image pixel value to evaluate the characteristics of the corona discharge of polymer insulators. To capture the images of the solar-blind ultraviolet (SBUV) band, an ultraviolet (UV) imaging detector based on an image intensifier is designed. First, through the experiments’ results with the UV light source, the proposed method based on the image pixel value is proved to detect solar-blind signals more effectively than the photon counting method. Then, by conducting experiments with the insulator at a fixed observation distance, the corona characteristics of the insulator can be described through analyzing the images captured by the detector. Finally, the experiments at different observation distances are conducted, and the image pixel value ratios of the recorded images with certain observation distance are computed. The results reveal that the developing stages of different corona discharges can be uniquely described. Based on the method, the characteristics of a corona discharge in the SBUV band can be described accurately and objectively.
Ultraviolet detection technology has been widely focused and adopted in the fields of ultraviolet warning and corona detection for its significant value and practical meaning. The component structure of ultraviolet ICMOS, imaging driving and the photon counting algorithm are studied in this paper. Firstly, the one-inch and wide dynamic range CMOS chip with the coupling optical fiber panel is coupled to the ultraviolet image intensifier. The photocathode material in ultraviolet image intensifier is Te-Cs, which contributes to the solar blind characteristic, and the dual micro-channel plates (MCP) structure ensures the sufficient gain to achieve the single photon counting. Then, in consideration of the ultraviolet detection demand, the drive circuit of the CMOS chip is designed and the corresponding program based on Verilog language is written. According to the characteristics of ultraviolet imaging, the histogram equalization method is applied to enhance the ultraviolet image and the connected components labeling way is utilized for the ultraviolet single photon counting. Moreover, one visible light video channel is reserved in the ultraviolet ICOMS camera, which can be used for the fusion of ultraviolet and visible images. Based upon the module, the ultraviolet optical lens and the deep cut-off solar blind filter are adopted to construct the ultraviolet detector. At last, the detection experiment of the single photon signal is carried out, and the test results are given and analyzed.
In order to measure spectral transmittance of solar-blind filter ranging from ultraviolet to visible light accurately, a high-precision filter transmittance measuring system based on the ultraviolet photomultiplier is developed. The calibration method is mainly used to measure transmittance in this system, which mainly consists of an ultraviolet photomultiplier as core of the system and a lock-in amplifier combined with an optical modulator as the aided measurement for the system. The ultraviolet photomultiplier can amplify the current signal through the filter and have the characteristics of low dark current and high luminance gain. The optical modulator and the lock-in amplifier can obtain the signal from the photomultiplier and inhibit dark noise and spurious signal effectively. Through these two parts, the low light passing through the filters can be detected and we can calculate the transmittance by the optical power detected. Based on the proposed system, the limit detection of the transmittance can reach 10-12, while the result of the conventional approach is merely 10-6. Therefore, the system can make an effective assessment of solar blind ultraviolet filters.