Research On Dam Deformation Monitoring Method Based On 3D Laser Scanning Technology

The safety of the dam is of great significance to the protection of people’s lives and social benefits.Deformation monitoring realizes deformation monitoring and analysis through the comparison of measured data in different periods,which can reflect the safe operation status of the dam in advance and intuitively.It plays an important role in the construction and operation management stage.Dam surface deformation is the main monitoring content,while traditional monitoring methods mainly adopt the"single point"measurement mode,which requires monitoring points to be set on the building,and the number is limited.3D laser scanning（3D laser scanning）technology can scan the spatial data of the surface of the ground object in all directions,large area and high efficiency,and obtain high-density,high-resolution massive three-dimensional point clouds^[1].Dam deformation monitoring based on 3D laser scanning technology requires data processing procedures such as data collection,point cloud splicing,point cloud registration,deformation surface extraction,and point cloud comparison.Any process will affect the results of dam deformation analysis.In order to improve the accuracy of laser data acquisition,this project adopts the method of multi-site scanning and spherical target layout to realize the complete dam crest and point cloud collection of the back water surface.The spherical target method is suitable for large-scale operations,and has high splicing efficiency and accuracy,which meets the accuracy requirements of dam deformation monitoring.In addition,the registration of point clouds at different periods is an important processing link to realize dam deformation monitoring,and the accuracy of registration directly affects the results of deformation analysis.During the experiment,the stable rock mass in front of and on both sides of the dam was selected as the registration element,and the two-phase coordinate conversion parameters were solved by the ICP algorithm.This paper takes the downstream surface of the dam as the experimental object,and uses 4 point cloud data features（normal vector,roughness,surface change rate,curvature）to obtain a relatively smooth deformed surface.Through experimental comparison,it is found that the roughness characteristics can be very good.It is easy to distinguish between surface points and non-surface points of the dam.Based on the analysis and comparison of deformation comparison methods based on proximity（C2C,Cloud to Cloud）,point-to-model（C2M,Cloud to Mesh）and multi-scale model to model（M3C2,Multi scale Model to Model Cloud Comparison）,Taking into account the deformation characteristics of the arch dam,a method based on the comparison of the arch ring model to the model deformation is proposed.The more robust moving least square method is used in the arch ring modeling method.Finally,the difference between the two arch rings is compared to indicate the dam is deformed.In the experiment,the downstream face of Xiahuikeng dam is taken as the research object,and the results show that the calculation results of this method（-11.98mm）are similar to those of c2m（-12.01mm）and m3c2（-11.92mm）.The C2C method（-30.21mm）is not suitable for dam deformation analysis because of its large calculation result.Therefore,the method proposed in this paper is feasible and meets the requirements of dam deformation analysis.