基于最大熵阈值分割的电气二次回路故障三维可视化识别模型

为了更直观地、实时监测了解二次回路运行工况，减少故障事后排查工作量，研究了基于最大熵阈值分割的电气二次回路故障三维可视化识别模型。采集二次回路各项运行数据，生成电气二次回路三维虚拟场景，标定电压传感器的位置以及红外成像仪的监控角度和距离；基于最大熵阈值分割处理图像，提取识别目标和背景；建立二次回路故障三维可视化识别模型，实现二维图像和三维场景的匹配，完成电气二次回路故障识别。经实验论证分析，最大熵阈值分割处理后的三维可视化图像清晰，不存在像素点丢失情况；故障识别准确率均在95%以上，误检率和漏检率均低于5%,具有更好的三维可视化识别效果和质量。

3D Visual Identification Model of Electrical Secondary Circuit Fault Based on Maximum Entropy Threshold Segmentation

In order to monitor and understand the operating conditions of the secondary circuit more intuitively and in real time, and reduce the workload of fault post troubleshooting, a three-dimensional visual identification model of electrical secondary circuit fault based on maximum entropy threshold segmentation is studied. Collecting the operation data of the secondary circuit, generating the three-dimensional virtual scene of the electrical secondary circuit, calibrating the position of the voltage sensor and the monitoring angle and distance of the infrared imager; The image is segmented based on the maximum entropy threshold, and the target and background are extracted and recognized; Establishing a three-dimensional visual recognition model of secondary circuit fault, realizing the matching of two-dimensional image and three-dimensional scene, and completing the electrical secondary circuit fault recognition. Through experimental demonstration and analysis, the 3D visualization image after maximum entropy threshold segmentation is clear, and there is no loss of pixels; The accuracy of fault identification is more than 95%, the false detection rate and missed detection rate are less than 5%. It has better 3D visual identification effect and quality.