A new method of processing laser scanning data of radial section dimensions for ring forgings. (1st March 2020)

• Record Type: Journal Article
• Authors: Fu, Xian-bin; Miao, Shu; Zhang, Yu-cun; Li, Qun; Kong, Tao; Zhang, Gui-ru
• Is Part Of: Measurement. Volume 153(2020)
• Publication Date: 2020-03-01
• Subjects: Data processing algorithm -- Laser scanning data -- Ring forgings -- Topological metric space -- Gradient descent method; Weights and measures -- Periodicals; Measurement -- Periodicals; Measurement; Weights and measures; Periodicals; 530.8
• DOI: 10.1016/j.measurement.2019.107430
• Languages: English
• ISSNs: 0263-2241

Abstract

Highlights: A new method of processing laser scanning data for ring forgings is proposed. The information space sets of laser scanning data are established. The uniform continuity between the various areas of information space is verified. The undetermined coefficients are solved by gradient descent and uniform isomorphism factor. Abstract: Ring forgings are widely used in petrochemical, power generation, aerospace, nuclear industry and other industrial fields. In the forging process, the radial section size data greatly affect the forging accuracy of ring forgings. Therefore, a new method of processing laser scanning data of radial section dimensions for ring forgings is proposed. Firstly, the information space set of laser scanning data is established. This information space is divided by continuous concentric circles. Secondly, according to the information relationship between the various areas in the information space, the uniform isomorphism factor is defined. By combining the gradient descent method with the uniform isomorphic factor, the undetermined coefficients are solved. Thirdly, the reduction ratio is set according to the density of scanning data points. By combining the reduction ratio with the undetermined coefficients, the number of points needed to be reserved in each area of a concentric circle is determined. The remaining points and the repetitive points in the intersection of two concentric circles are deleted. Finally, the method is repeated until the last point of the laser scanning data. In this way, the laser scanning data of radial section for ring forging is processed. When the method is applied in the experiment, the size data of radial section is extracted from the processing data of laser scanning radial section line. Two groups of dimension data measurement experiment for standard and non-standard ring forgings are carried out respectively. The dimensions measured from the unprocessed and processed data points are compared with true values. The dimensions measured from the processed data have higher accuracy and smaller error. The feasibility of this method is proved.