Automatic Detection Of Oil Level Of Electric Power Transformers Based On Infrared Image

The development of smart grid requires it can monitor on-line and carry out safety warning, detect the defects and take measures to eliminate hidden dangers. Infrared detection technology has been widely used as an effective means of monitoring the fault. The combining of infrared detection technology and image monitoring system greatly improved the level of power system fault monitoring. However, the current detection methods still require manual diagnosis, unable to meet real-time and accuracy.In order to detect the oil level of electrical power transformers in the substations accurately and automatically, a method based on the infrared image is put forward. The edge points of the oil conservator of the electrical power transformer are identified by the edge detection process on the infrared image of the transformer oil conservator. An iterative elliptic fitting approach is used to obtain the location and the shape of the oil conservator. The points far away from the obtained ellipse edge are eliminated in each iteration step until the difference of the ellipse between two adjacent iterations is quite small and meets the requirement of the convergent precision. Several temperature measurement lines are implanted inside the oil conservator. Taking the derivatives of the temperature respect to the position for each temperature measurement line, the point with the maximum temperature changing ratio is regarded as the boundary point of the oil surface. Based on the obtained boundary points and by a least squares approach, the oil surface can be determined. Finally, the oil level is obtained by normalizing the area of the oil part and the oil-free part within the ellipse into the circle shape.The result of the experimentation shows that the proposed method is feasible and it can detect the edge of the ellipse, fitting elliptical contours, and automatically detect the transformer oil level,according to the points with the greatest gray change in temperature measurement.Therefore. it can be used in the infrared monitoring and warning system in unmanned substations.