基于多源数据的光伏电站在线热斑检测性能评价系统研究

Online Hot Spot Detection Performance Evaluation System for Photovoltaic Power Station Based on Multi-source Data

传统的热斑检测运维方式存在巡检效率低、多因素影响下的热斑成因不易判别等问题。为此,基于无人机巡检、红外图像等多源数据技术,结合基于斜率约束的红外图像与可见光图像配准的热斑定位方法、基于改进的鱼群灰色组合预测方法,设计热斑信息判别流程。在此基础上构建基于无人机红外与可见光检测的光伏电站在线热斑检测性能评价运维系统,对其进行实例应用,并通过试验验证了该系统热斑检测结果的准确性。该系统热斑定位精度高,可以主动筛除光伏组件热斑的外部影响因素,实现复杂热斑自动告警与定位排查。

There are many problems in the traditional hot spot detection operation and maintenance methods,such as low inspection efficiency and difficulty in identifying the causes of hot spots under the influence of multiple factors.Based on the multi-source data technologies such as unmanned aerial vehicle(UAV)inspection technology and infrared image,the hot spot information discrimination process was designed,with the hot spot location method based on the slope constraint of infrared image and visible image registration and the improved fish swarm grey combination prediction method combined.Then,an online hot spot detection performance evaluation operation and maintenance system for photovoltaic power plants based on UAV infrared and visible light detection was built.The accuracy of the hot spot detection results of the system was verified by practical application and experiments.The test data shows that the system has high hot spot positioning accuracy,and can actively screen out the external influencing factors of photovoltaic module hot spots,which realize automatic alarm and positioning inspection of complex hot spots.