Dynamic Positioning Error Analysis And Modeling Of CNC XY Working Table

In order to improve the measuring accuracy of CNC on-machine measuring system,error compensation is an effective way.The CNC on-machine measuring system mainly consists of CNC machine tool system,probe system,measurement macro program and measurement software.XY working table is an important part of CNC machine tools,and its positioning accuracy has significant influence on the measuring accuracy of CNC on-machine measuring system.At present,high speed,high efficiency and high precision are developing in the direction of CNC on-machine measuring system.The influence is increasingly significant to the measuring accuracy of CNC on-machine measuring system under different motion conditions,such as load,speed and X/Y position.There is great significance for carrying out error compensation and improving the measuring accuracy of CNC on-machine measuring system to analysis and modeling of the error correlation of CNC on-machine measuring system and the dynamic positioning error of XY working table.In this paper,Error sources and influencing factors of CNC on-machine measuring system are analyzed.The two-dimensional stacking table was selected as an example to consider error correlation influence of the lower Y-axis guide rail system to the upper X-axis guide rail system,and a more comprehensive Abbe error compensation model was established.In order to study the error generation mechanism of XY working table,several error models such as friction model,ball screw system model,grating measurement system model and guide rail system model were established respectively.Finally,a mathematical model of the dynamic positioning error of XY working table was established.In order to establish a high-precision dynamic positioning error prediction model for CNC machine tool XY working table,PSO-SVM algorithm modeling and BP neural network algorithm modeling were adopted respectively.The mean square error of error modeling of PSO-SVM algorithm is 2.4108 mm.The mean square error of BP neural network algorithm error modeling is 7.8936 mm.The results show that the PSO-SVM algorithm can establish a more accurate prediction model for the dynamic positioning error of CNC machine tool XY working table.In order to simulate and verify mathematical model of the dynamic positioning error of XY working table,the relevant experimental device was built by using Renishaw XL80 laser interferometer and the relevant parameters in the model were measured.The simulation results show that the XY working table has the optimal measuring speed of 12mm/s.The HP laser interferometer is triggered in real time by the output signal of the grating measuring system.And the dynamic positioning error of XY working table at different motion conditions was measured.The experimental results show that XY working table has the optimal measuring speed of 14mm/s,and the experimental results are almost the same as the theoretical results.In this paper,the error correlation of CNC on-machine measuring system is expounded.Taking the self-developed CNC XY working table as a prototype,the theoretical model of positioning error and high precision prediction model are established.And the optimal measuring speed of XY working table was obtained.At the same time,the dynamic positioning error of the CNC XY working table under different conditions was measured and analyzed,and which lays a theoretical and practical foundation for the development of error compensator.Figure [45] table [2] reference [55]...