Reseach On Characteristic Curve Reconstruction Of Special Ring Parts Based On 2D Laser Scanning

Special ring is the use of expensive titanium alloy and high temperature alloy forging two materials,with excellent high temperature strength,good resistance to oxidation,thermal corrosion resistance,good fatigue properties and other comprehensive performance,with the modernization The progress of industrial technology,China's power,shipbuilding,nuclear power,energy and chemical industry to achieve rapid development of the production of special ring parts quality requirements are also getting higher and higher.Accurate control and control of the forging process in the size of special ring parts to improve the quality of forging production plays an important role.However,in the process of forging of special ring parts,two-dimensional laser scanner is often affected by factors such as emissivity,high temperature background radiation,oxide scale covering,environmental stray light and so on,so the measurement error is difficult to meet the measurement requirements.In this paper,the characteristic curve reconstruction method is proposed for the on-line inspection of the special ring forging and the reconstruction of the characteristic curve.Two-dimensional point cloud data to eliminate noise,two-dimensional point cloud data splicing and two-dimensional point cloud data feature point extraction are studied.Firstly,a two-dimensional histogram of the original point cloud data is established based on the prior probabilistic algorithm,and the noise points and non-noise points are classified by region.Using the curve probability threshold segmentation algorithm,the histogram data is processed,model to eliminate the noise data in the two-dimensional point cloud data and lay a foundation for improving the accuracy of the characteristic curve reconstruction of special ring parts.Secondly,according to the different points of the different locations,these points are divided into target point and moving points.The target points as reference point match point cloud.Binary array method instead of coordinate transformation calculates the rotation coefficient and the translation coefficient.Through the objective functions minimization processing,to achieve point cloud registration.The least squares fitting method is used to obtain the characteristic parameters.According to the characteristics of the parameters,the point cloud rotates to achieve point cloud data stitching.Thirdly,according to the point cloud data of the two-dimensional laser scanning,the B-spline interpolation algorithm is used to fit the point cloud data to obtain the target curve.The target curve is smoothed.Then the curve points are defined with as the center to curvature angle and get candidate points.Accord to the suppression conditions and the template matching technology get the real feature points.The real feature points calculate a special ring piece of the reliable characteristics of the size of the curve.Finally,the experiments are carried out to eliminate the noise,splicing and feature point extraction in this experiment.The accuracy,integrity and reliability of the proposed algorithm are verified by the analysis of the experimental results.