Monitoring method for the posture of a double-column horizontal rotating isolation switch

A method for monitoring the opening and closing state of a double-column horizontal rotating isolation switch is proposed. The method includes a point cloud data acquisition module, a point cloud data preprocessing module, a point cloud data segmentation module, and an isolation switch posture recognition module. Firstly, a laser radar is arranged in the working area of the isolation switch to ensure comprehensive acquisition of point cloud data. The acquired point cloud data is downsampled and subjected to statistical filtering to remove outliers, obtaining high-precision point cloud data. An improved K-means clustering algorithm is designed to segment the point cloud data of the region where the conductive arm of the isolation switch is located. Through training the PointNet network, a model that can automatically recognize the key nodes between the two ends of the conductive arm is obtained. Then, by calculating the horizontal angle of the conductive arm and comparing it with a predefined threshold, the opening and closing state of the isolation switch can be determined.