Study On Spatial Structure Of AC Corona Discharge Based On Digital Image Processing Technology

In the past,the state detection of corona discharge mainly focused on pulsed electrical signals,ultra-high frequency electromagnetic signals,ultrasonic signals,gas components.However,there are few studies on the most direct optical images for discharge.As the most commonly used information carrier,image processing technology is an important branch of state recognition.As a real-time online non-contact detection method,image detection is bound to be intelligently developed in the electric power field and process control in the low-temperature plasma field.In this paper,a digital camera was used to take visible light images of corona discharge for a long time scale,and the corona discharge spatial structure was studied based on digital image technology from grayscale image morphology and color image chromaticity,which supports the assessment of the state of corona discharge.Study on discharge spatial structure distribution based on gray image,the spatial development information of corona discharge was studied from the two-dimensional distribution of morphological information,including corona onset diagnosis and discharge intensity recognition: the feasibility and effectiveness of visible light optical detection for corona discharge are analyzed,and a method for diagnosing corona discharge optical image is proposed based on adaptive threshold segmentation;the gray-level probability distribution of the discharge space is calculated,and the corona discharge intensity diagnosis method based on the gray level histogram is given;the spatial texture descriptors such as smoothness,consistent values,and entropy are used to achieve quantitative analysis of corona discharge intensity;the spatial boundary features such as the perimeter,diameter,area,and circumscribed area of the digital image processing field are introduced to achieve a simple division of the discharge intensity of the corona discharge in an extremely non-uniform field.Study on discharge spatial structure distribution based on color image,the spatial development information of corona discharge was studied from the two-dimensional distribution of chromaticity information,including the RGB and HSV chroma index division of discharge stage,spatial structure distribution description of the chromaticity angles in different discharge gas environment,and the feasibility exploration of color temperature: the sum gray levels of the image quantization index is given and the proportion of the chroma components that can be interpreted using the discharge emission spectrum is obtained;the difference between B and G component is extracted as the characteristic chroma index of the discharge image,the chromaticity division of the discharge stage is achieved,and the associated changes of the characteristic chroma and electrical indicators at different discharge stage are obtained;the average gray level and chroma component proportion of the statistical indicators distributed along the axial direction are used to achieve the spatial structure distribution description of chromaticity angles in different discharge gas environment;the non-equal interval division and grading method is used to obtain the typical distribution maps of HSV components in different discharge stages and the distribution characteristics of corresponding pixel points for each component magnitude,and the effective division of the corona discharge stage is also achieved;the color temperature index in the field of chromaticity is introduced,the relationship between color temperature and voltage is nonlinear,and the color temperature description of discharge intensity is achieved.Through the study of the discharge spatial structure distribution based on the gray image,the judgment of the discharge corona onset and the diagnosis of the corona discharge intensity are realized.Through the study of the spatial structure distribution of color images,the chromaticity division of the corona discharge stage is given,and the correspondence between the optical image characteristics and the division of traditional electrical methods is achieved,the description of the spatial structure distribution of chromaticity angles in different discharge gas environment is also completed,besides,the correlation between the characteristic chroma index and the micro-indicators such as electrical and spectral indicators are further realized.