Study On The Fitting Of The 3D Curved Surface And Fitting Error Evaluation Of The Locomotive Of High-Speed Trains

High-speed train technology is key to the development of China's railway transportation technology,and streamlined locomotive design is vital to the aerodynamic performance of high-speed trains.Therefore,the three-dimensional standard measurement on the locomotive of high-speed trains is a very important step in the process of locomotive production.Current researches are focused on the use of point cloud data in testing the shape of locomotives,and the three-dimensional fitting of locomotive surface based on point cloud data is key to the three-dimensional standard measurement on the locomotive of high-speed trains.In light of the current researches,this paper will focus on the method of point cloud surface fitting of locomotives and the corresponding error analysis.Specific research content include:1.Analysis on the fitting methods for curved surface with point cloud data.Several classical fitting methods using point cloud data are introduced to explain the algorithm principles and main process of the curved surface fitting algorithm.Features of each algorithm are also explained in details.2.Poisson surface fitting algorithm.An in-depth research is conducted on the classical Poisson surface fitting algorithm,with detailed explanations given on the principles and specific steps of this particular algorithm.Based on the point cloud features of the locomotives of high-speed trains,a more targeted analysis is conducted on the Poisson surface fitting algorithm.3.Locomotive surface fitting based on the Poisson algorithm.In regard to issues such as data gaps and outliers that came up when using Poisson surface fitting algorithm for the restructuring of the three-dimensional surface of the locomotives of high-speed trains,a pre-processing method is proposed to guarantee the quality of patches following the Poisson algorithm.To reduce the amount of computing when using the Poisson algorithm for surface fitting,a minimized algorithm is brought up to reduce the complexity of Poisson algorithm in both time and space.Regarding the ambiguity issues that came up when extracting equivalent contour surface with Poisson surface fitting algorithm,an improved marching cubes algorithm is proposed.With such method,the holes on the surface fitting caused by the uncertainties when connecting the triangular patches are eradicated.4.Analysis on the error evaluation of the curved surface of the locomotives of highspeed trains.Evaluation indexes are proposed for the curved surface of the locomotives of high-speed trains,which focuses on the accuracy of size,error of surface profile and fairness of the surface.5.Experiment results and analysis.Following the experiment conducted on the locomotive surface fitting method based on the Poisson algorithm,results are provided,and error analysis is conducted,which proves the feasibility of the algorithm brought up in this paper on the curved surface fitting of the locomotives of high-speed trains.