Design Of Patrol Inspection Device For Abnormal Dam Surface And Research On Crack Detection Algorithms

Pumped-storage power stations can store electricity by pumping water to high places when the power is sufficient,and release water during peak hours to ease the energy shortage.On the one hand,pumped storage power stations can convert excess electrical energy when the electrical load is low to high-value electrical energy at peak power consumption.On the other hand,it can provide backup power for the power grid,which can reduce grid fluctuations and improve grid efficiency.It has high economic value and good environmental adaptability.However,over time,the reservoir panel of a pumped storage power station may be cracked due to external factors such as wind and sun,and structural aging.In order to prevent this hazard from happening or to minimize its consequences,regular inspections of reservoir panels are required.Traditional manual on-site inspections are time-consuming and labor-intensive,and have certain risks.So,in order to achieve an efficient and reliable automatic inspection of the reservoir panel,a dam surface abnormal inspection device needs to be developed.For the automatic inspection of the surface anomalies of reservoirs,there is currently no complete solution in China.This article designs and implements an automatic inspection device for dam surface abnormality monitoring.The device is designed in a hierarchical way and is divided into four parts:mechanical layer,electrical layer,image acquisition layer and control calculation layer.First of all,through the calculation of the load of the dam wave wall,this paper determines the automatic walking mode of the dam.Based on this,the mechanical layer frame is designed and implemented.Then,select the drive wheels,winches,sensors,and power supply units to build the electrical layer as required.After that,design and implement the camera bracket and putter and complete the selection of camera and gimbal to build the image acquisition layer.Finally,the PC 104 embedded computer and the control carrier board are used to realize the walking and image acquisition control of the dam surface abnormality inspection device.This paper not only completes the design and implementation of the software and hardware framework of the automatic inspection device,but also studies the image stitching,image enhancement,and image segmentation algorithms for dam panel images.Select improved SURF and RANSAC image registration algorithms to achieve image stitching.By comparing several classic image denoising algorithms,a median filtering algorithm is selected to complete the image denoising.The image defogging algorithm is implemented by the image defogging algorithm based on the dark channel prior and guided filtering.Finally,the image segmentation algorithm is used to monitor the cracks of the dam panel.Based on the requirements of dam panel automatic inspection,this paper designs and implements dam surface abnormal inspection devices,and realizes dam panel abnormal detection.The dam inspection automatic walking device developed in this paper has been put into use and has stable operation,indicating that the research design in this paper meets the requirements of scientific rationality and can meet the requirements of automatic inspection of reservoir panels.