Innovative application of point cloud algorithm in power grid
There is no doubt that from the initial entry of the digital twin into the power field until now, the demand for the digital intelligent transformation of the power grid is in a constantly changing state.
In the past, the “digital twin of visual mode” is essentially a combination of various types of data, and such a combination seems to be together on the surface, once the use of some proprietary analysis, it will be found that these different sources, different levels of data are not connected, difficult to flexibly integrate applications.
The above is also the reason why many customers believe that the value of digital twins is not high.
Today, the digital twin has truly entered the stage of creating value from good-looking, and behind the release of the value of power data by Youli Technology is its gradually in-depth understanding of the power grid field.
Zhang Xuebing pointed out that the field of power grid has high requirements for accuracy, especially for substation facilities, substations and other closed accuracy requirements of more than 99.99%, hoping to use millimeter-level accuracy data to accurately analyze and judge the state of power inspection.
Specific to practical applications, the use of visible light cameras during the day can easily monitor and track targets. However, at night, the visible light camera has certain limitations, and is greatly affected by the weather, and the visible light imaging effect of the environment such as rain, cloud and night is greatly disturbed.
Because of this, unlike most companies on the market to use artificial modeling models to build three-dimensional applications of the technical path, Youlitech created a set of substation three-dimensional ultra-high precision laser point cloud as the data basis, through high-precision three-dimensional color point cloud and camera linkage control to establish a three-dimensional intelligent inspection system.
Compared with the ordinary laser point cloud data, the ultra-high precision laser point cloud data of Youli Technology comprehensively exceeds the three aspects of restoration degree, aesthetics and practicality. Based on ultra-high precision laser point cloud data, the full-station model can achieve millimeter-level accuracy, point density of 200,000 points/square meter, the average error is within 2MM, and the maximum error is not more than 6MM.
The system uses high-precision point cloud for point marking, mobilizes cameras, inspection robots, drones, auxiliary control equipment and other detection equipment to obtain video and photo data, analyzes hidden table and defect information through background image recognition, and displays it on the three-dimensional model.
Based on the results of liDAR point cloud and the integration of visible light image data, Youlitech conducts remote customized inspection tasks, video surveillance at the control station and inspection by robots, intelligent analysis of acquired visual images based on 3D real-time perception inspection, automatic tracking of abnormal devices, alarm monitoring and fault linkage combined with real-time signals of devices, etc. Achieve remote intelligent three-dimensional inspection.
In the words of Zhang Xuebing, Youli Technology is currently the first domestic LiDAR this type of new terminal equipment used in large substations, DC stations, and provide corresponding operation and maintenance services.