Contour Error Compensation Algorithm Based On Spatial Iterative Learning

Contour accuracy is an important indicator of the quality of CNC machined products.It is mainly affected by contour errors.Contour control is mainly divided into two aspects:contour error estimation and contour error control.Concerning the digital curves collected by the CNC system,the contour error estimation still has the problem of insufficient accuracy.Iterative learning control has a better effect when dealing with repetitive errors.Traditional iterative learning control acts on the inner loop of CNC system,which is not easy to implement,and its stability lacks corresponding theoretical proof.For a type of motion control system with repetitive characteristics in the space cycle,the iterative learning algorithm based on the time cycle is not applicable.In order to better solve the problems in mass repetitive machining,a set of contour error compensation mechanism with complete theoretical analysis is established in this paper.Aiming at the current situation of large-scale repeated processing in the manufacturing industry,a contour error compensation algorithm based on spatial iterative learning is proposed,which mainly includes a spatial iterative learning module,a contour error estimation module,and a feed rate adjustment module.Aiming at a type of motion control system with repetitive characteristics in the space cycle,an iterative learning control algorithm based on the space cycle is proposed,and the contour error is compensated by modifying the reference trajectory.This method does not involve the underlying controller,and there is no open controller in commercial products.In this case,the method has better applicability and is more convenient to implement.The stability of spatial iterative learning algorithm is proved by analyzing the dynamics model of the CNC machine tool mechanical system and the Lyapunov function,that is,the contour error can be continuously reduced with the iteration increases.Aiming at the planned position and actual position in the form of discrete data points collected from the numerical control system,a contour error estimation method based on Hermite spline is proposed.The estimation effect of this estimation method is verified in experiments and compared with common contour error estimation methods,which shows that the proposed method has higher estimation accuracy and estimation efficiency.Based on the iterative modification of the geometric trajectory algorithm,a feed rate adjustment method based on the trajectory curvature feature is proposed,which can greatly reduce the peak contour error peakof the trajectory in the large curvature area on the basis of bottomless operation efficiency,and solve peak of the local contour error in the trajectory is large where there is the large curvature,which does not meet the processing accuracy requirements.The method in this paper is not only suitable for planar trajectories,but also for three-dimensional trajectories.This method can be easily extended to three-dimensional trajectories.Finally,a three-axis CNC machine tool was used to verify the contour error compensation effect of the method in the fourth-order elliptical trajectory and flower curve.The peak of the contour error was reduced by about 80%.The peak value of contour error has been further reduced.In three-dimensional trajectory,contour error compensation also has a significant effect.