Study On Geometry Control Of Short-Line-Match Precast Segmental Bridge Based On Photogrammetry

Short-line-match precast segmental bridge is one of the main forms of bridge industrialized construction,and the geometric control method is used to precisely control the unstress geometry of segments in the precast process to achieve design status.The current geometry control of short-line-match precast segmental bridges mainly use total station measurement following six-point principle,while the limitations and deficiencies of traditional method have begun to appear with the development of industrialized construction,on one hand,the geometry control method with total station measurement is labor and time intensive,and there are some limitations and problems such as manpower and financial resources cost,low efficiency and high risk of human error.On the other hand,the six-point principle is a simplified mode to restore as-built model of the object with the traditional local point sampling total station measurement,which cannot represent the segmental shape perfectly.Photogrammetry acquires multi-view 2D images and then quickly reconstruct geometric data of the object,with the advantages of non-contact,faceted measurement,high accuracy,and fast speed.Relevant research shows that photogrammetry has better capabilities than traditional technical methods in structural digital reconstruction and shape measurement.Photogrammetry combined with computer vision technology has become one of the core technologies of automatic measurement and digital control,which not only improves the accuracy of measurement control,but also gradually increases the automation of industrial construction.In order to improve the automation and efficiency in geometry control of shortline-match precast segmental bridge,based on the review and summary of the existing interdisciplinary researches on bridge engineering and photogrammetry,this paper introduce photogrammetry into the geometry control of short-line-match method,and an innovative and automatic geometry control method of short-line-match precast segmental bridge based on photogrammetry is established taking main research route that "overview summary-solution-data acquisition-data processing-data application".The main work of this paper includes:(1)Combining with construction process and geometry control of short-line-match precast segmental bridge,as well as the characteristics of photogrammetric reconstruction,the adaptability and feasibility to apply photogrammetry on the geometry control of short-line-match process are discussed.According to the data structure from photogrammetry,a point cloud with wireframe model and a six-coded point model of the segment are proposed respectively to solve geometry control of short-line-match precast segmental bridge.(2)In order to acquire high-precision photogrammetry data of precast segment,the standards for constructing the data quality driven photogrammetric scanning system in the scene of segment precast is proposed.The factor affecting accuracy is analyzed based on the photogrammetric digital projection mathematical model,then the optimal parameters interval of photogrammetric scanning system in the scene of segment precast is proposed through the field factor comparison test of real bridge.According to the scanning visibility,lighting conditions and geometric size in short-line-match process,the prototype and implementation workflow of data quality driven photogrammetric scanning system are established from the aspects of equipment selection,system parameters,acquisition scheme and system integration.(3)The initial Sf M point cloud from photogrammetric reconstruction is a set of disordered spatial points without topology,including surface points of objects in the scene,which cannot be analyzed and applied directly.The automatic geometry control of short-line-match method based on 3D point cloud is established from three aspects:point cloud processing,model transformation and control model application.In point cloud processing,an automatic point cloud processing is proposed,which realizes the conversion from initial point cloud to final application point cloud without affecting the accuracy.Firstly,a method for identifying effective point cloud based on encoded points with multi-layer bounding boxes is proposed to realize automatic recognition and extraction of segment effective point cloud in the initial point cloud;Secondly,a concave convex discrimination mode considering the accuracy and spatial distribution characteristics of point cloud data is proposed to realize the high-precision identification and segmentation of shear key areas on end surface of precast segment,which can solve the problem that the shear key areas have influence on the plane recognition and fitting of the end surface in the precast segment point cloud.In model transformation,the hybrid model of point cloud with wireframe is used as basic model of linear control,which can improve the accuracy and efficiency of the model calculation and iteration in the precast segment assembly simulation process,and also can accurately characterize the 3D shape of segment.A method for building the wireframe model accurately from segment point cloud data using 3D straight line detection is proposed.In control model application,to realize the geometry analysis and adjust of segment,and automatic setting out of matched segment,based on the hybrid model of point cloud with wireframe,the geometry control method of short-line-match precast segmental bridge is established from three aspects: segment assembly simulation,parameter control method for intelligent adjustment template and multilevel geometry adjustment strategy.Relevant conclusions show that the geometry control method proposed in this paper has high accuracy,stability,and applicability.Compared with the traditional geometry control method,on one hand,the photogrammetric scanning system can measure the segment shape automatically,which can improve efficiency and reduce risk of human error;On the other hand,the hybrid model can represent the shape and assembly process of segments accurately,which has a more direct and clear physical meaning in linear control,and the method presented in this paper can avoid the decline of control accuracy caused by the change of individual measuring points in traditional method.(4)The automatic geometry control of short-line-match method based on six encoded points is established from data processing and control model application.In data processing,firstly,the photogrammetric reconstruction mode is optimized considering recognition accuracy of coded marker,then a coordinate transform method for six encoded points from initial reconstructed coordinate system to final precast geometry control coordinate system is proposed.In control model application,to realize the geometry analysis and adjust of segment,and automatic setting out of matched segment,based on the model of six encoded points,the geometry control method of short-line-match precast segmental bridge is established from three aspects: segment geometry analysis,parameter control method for intelligent adjustment template and multilevel geometry adjustment strategy.Relevant conclusions show that the geometry control method proposed in this paper has high accuracy,stability,and applicability.Compared with the traditional geometry control method,the photogrammetric scanning system can measure the encoded point coordinates automatically,which can improve efficiency and reduce risk of human error,and the method presented in this paper has high practical application and popularization value.