Operation State Evaluation of High-voltage Power Bushing Based on Artificial Intelligence Image Processing Algorithm and Broadband Dielectric Spectrum Analysis Technology

In this paper, during operation of UHV capacitive bushing, there is the direct relationship between its external operation state and its internal insulation state. Now, the external image of UHV capacitive bushing is acquired through the artificial intelligent manipulator. In the process of patrol inspection, the robot continuously collects visual information such as external contour photos and videos of typical power equipment in the substation, and uses the immune genetic algorithm to segment key components of power equipment for specific visual image, and then carries out the subsequent processing. At the same time, it is found that the complex permittivity of the sample is fitted with the three-relaxation Cole-Cole model function, and the ideal fitting effect is obtained. There is a strong nonlinear relationship between the micro-structure of the bushing and the temperature. In the temperature range [-40℃,160℃], the real part of the dielectric constant, the tangent value of the loss angle and the AC conductivity have a complex change process. Further, through theoretical analysis of the internal causes of the temperature gradient of UHV capacitive bushing, theoretical and practical analysis of the artificial intelligence algorithm for processing infrared and ultraviolet image database is given. The artificial intelligence image processing algorithm proposed in this paper can effectively output two-dimensional numerical images of the RBF network. Infrared image database processing algorithm and image noise elimination algorithm of substation power equipment are proposed. The original two-dimensional data of the field image is input into Bayesian topological network for the feature elimination after NHMC parameter learning. Take the UHV post insulator as an example. The seed point is automatically selected near the red point, and then the black and white binary growth of the image is started in the post insulator area, followed by morphological processing, and finally the edge mark is made for the UHV post insulator. This paper innovatively proposes the high voltage power bushing operation state evaluation method combining the artificial intelligence image processing algorithm and the broadband dielectric spectrum analysis technology, and proposes artificial intelligence algorithm image processing technology for safe and stable operation of high voltage power equipment.