A Real-time Data Analysis Method Based On Interpolation Point Surface Reconstruction

With the advancement of CNC and information technology,it becomes feasible to record the massive real-time data of the CNC system at the frequency of interpolation cycle during the entire machining process.The massive real-time data includes but not limit to the following types: command data and feedback signals.These real-time data give a full record of the machining process,and are closely related to the quality of surface precision machining.However,few scholars have studied the relationship between real-time data and surface quality.To address the problem,a method for evaluating the discontinuity of real-time data is proposed to study the relationship between real-time data and surface quality.To achieve this goal,the interpolation points need to be reconstructed as surface model.Because the distribution of interpolation point is extremely dense,the interpolation points need to be compressed firstly to improve the efficiency of the surface reconstruction.Therefore,a bidirectional interpolation feature points recognition method is proposed,which is efficient and has high feature point recognition rate.In order that the triangular surface can be reconstructed efficiently,the interpolation point should be compressed with feature reserved.A compression method with adaptive compress ratio is proposed for interpolation points based on recognized feature points.It achieves the reservation of geometric features and adaptive compression rate,which lays foundations for the efficient surface reconstruction and accurate quantitation of real-time data discontinuity.Then the triangle mesh surface is constructed with simplified data,which provides the topological relation between interpolation points.Given the interpolation point constructed as triangulated surface,a quantification approach is proposed to analyze the relationship between real-time data and surface quality.The discontinuity of real-time data can be quantified accurately,and the positions where the discontinuous real-time data occurs can be located automatically.Accordingly,whether the discontinuity of real-time data is corresponding to the surface defect can be determined.Additionally,the method brought up in the paper has been verified in actual processing.The experimental results showed that the discontinuity of real-time data can be quantified accurately.And the proposed method provides evidences to the analysis of defect causes and the improvement of machining quality.