Research On The Status Detection And Evaluation Technology Of Existing Railway Based On Point Cloud Information

With the demand of railway speeding and passenger dedicated line construction, the application of new technologies in track dynamic and static detection is more widely, which has shifted from the traditional static detection methods of "gauge ruler and artificial visual" to track inspection vehicle detection instruments, added the regular track inspection vehicle inspection mode. Due to the fact that railway mileage in our country is long, the line structure type and traffic density are different, especially the traffic density of existing railway is large, while the skylight operation time is short, which requires the non-contact field measuring method. As well as to adapt to the BIM of railway construction requirements, collection of three-dimensional information along the line is also a very important research topic. When train speed increased, there are always conflicts between the track static and dynamic management values of irregularity. The versine of static detection curve is still subject to the "Repair Regulation", while the horizontal acceleration of dynamic detection curve occurs deduction phenomenon in the Ⅱ, Ⅲ grade. Which mainly because the increased detection indexes of track irregularity in the new Track Dynamic Management Interim Test Standard do not match with the conventional ones. In the current, artificial static detection is still the main mode of the maintenance and repair work, how to introduce new track static detection technologies to meet the high ride comfort requirements of speed line is needed to be solved.The application of three-dimensional laser scanning technology in railway can effectively solve the problems, especially its point cloud data can provide detailed information needed in line detection, and provide support for three-dimensional modeling. In view of this, the paper takes the application of three-dimensional laser scanning technology in existing railway status detection and evaluation as the research topic, mainly complete the following work, such as equipment selection, operation mode, data processing, track feature extraction, point cloud data curvature analysis, the curve adjusting, and the surveying program analysis of turnout, etc. Which have important theoretical and practical value. The works have been complete are mainly as following:1) Research on the operation mode of terrestrial scanning technology in existing railway surveyAccording to the working principle of three-dimensional laser scanner, select Surphaser25HSX and Z+F IMAGER 5010C, which are scanners with high precision, to separately complete the initial experiments both in road and rail engineering, analyze the accuracy of data collection and the application feasibility. Analyze the factors that affect the accuracy and improving measures, and provide a referencing operation mode for the field applications of three-dimensional laser scanner, through field testing and data analysis. Which use the phase-shift laser scanning equipment, set stations as "之", use the spherical target, and the general scanning radius is not more than 30m.2) Research on the line point cloud data processing and application methods that suit the characteristics of the railway lineCombined with the line major’s technical features, research the line point cloud data post-processing methods, analyze the line point cloud data registration and fitting algorithm of spherical target center. Based on the completion of point cloud data noise removal and redundancy elimination, realized using three-dimensional software to complete 3D modeling of existing railway, to provide basic data for the three-dimensional data management of existing railway equipment. Use the point cloud as the CAD base model, and take track direction and bed section as the example, to carry out the preliminary extraction of the design features from line, plane, and body. Conduct the acquisition of bed volume and size based on the surface modeling, which can be used as the base model of railway reconstruction project, and the detection and evaluation of specific projects have higher degree of matching. Based on the continuity of point cloud data, propose the line static detection method of rail direction irregularity based on curvature changing rate of unit based long (5m distance), which make up the operation technology of existing railway where the curvature changing rate is exceeded, and can also be used in track construction quality control of new railway, to help to improve the track initial irregularity.3) Research on the curve adjusting method of existing curve based on point cloud dataIn order to meet the high ride comfort and comfort needs after train speeding, conduct in-depth data mining of the continuous point cloud data obtained by laser scanning technology. From the idea from coordinate to curvature, and to curvature changing rate, has described the coordinated curve adjusting method based on point cloud data, analyzed the curve adjusting algorithms based on the inspection data of the track inspection vehicle, proposed the adjusting method of curvature changing rate being overrun of existing railway. The method is different from the traditional method that seek the optional radius through taking the minimum combinations of moving amount as the objective function, to get the perfect smooth position of curve, so that to get the moving amount, whose character is directly object-oriented, takes the gauge indicators of curvature changing rate as the red line, and return the overrun to the feasible region, which matches the maintenance and requirement requirements of integration of rail and vehicle. And the adjusted line has a better track geometry, but also is more consistent with the night train in motion characteristics.4) Form the application scheme of three-dimensional laser scanning technology in turnout detectionAgainst the features of high requirements and more test items of turnout geometric detection, introducing the laser scanning technology to turnout track geometric detection is the useful exploration,which is adapted to the modern work operations, less interference, easy to operate, and full data acquisition. The paper mainly carried the experimental study from data collection, data processing, data recognition and data use. Take full consideration of the particularity of turnout structure geometry detection in the test (multi-point and multi-index). By inversion, carry out the instruments leveling during the single-station measuring, to improve the operability and accuracy of single-station data, and complete the data extraction of geometric line position through the construction of section or feature surface. However, because of the large volume of point cloud data, it requires further optimization field data collection program in its specialized application process, Strengthen the model analysis of line and plane reconstruction process, and the CAD base model reconstruction problem.