Search

To achieve the goal of high-speed and highly accurate machining, this paper proposes a real-time control algorithm based on Taylor΄s expansion for implementing variable feed rate non-uniform rational B-spline (NURBS) curve. To efficiently compute the NURBS curve and its derivatives in real-time, some effective methods are proposed. In contrast to the existing linear and circular interpolators, the proposed interpolator can maintain small contour errors and feedrate fluctuations. It also can shorten the size of machining code effectively compared with the linear and circular interpolators. At the same time, the NURBS-format milling codes adapting the dual NURBS Curve interpolation algorithm for 5 axis high-speed machining are adopted in this study. Besides, an effective algorithm is adopted to minimize the kinematics error by means of the shortest path scheme. [ABSTRACT FROM AUTHOR]

Precision positioning stage is the key part of precision feed system of a machine. To compensate for the effect of processing precision affected by the vibration of wheelhead of precision grinder, a micro-feed worktable using piezoelectric actuators, which serves as an auxiliary feeding device, has been introduced for common cylindrical grinders. The stiffness of its support-oriented mechanism is modeled and analyzed by PRO/E and FEA. A close-loop control system based on special algorithm is presented and the prototype of a macro-micro driven ultra-precision positioning stage on grinder is developed. The experimental results show that the positioning accuracy of the stage is 0.1±0.02 μm, its average stiffness is 207 N/μm and the cylindricity of machined parts is less than 2 μm. [ABSTRACT FROM AUTHOR]

The basic principle and method of grey system model and its application in thermal error modeling on machine tools has been presented. How to utilize thermal error sequence surveyed to model and predict is discussed by adopting total data GM (1,1) model, new information GM(1,1) model and metabolic GM(1,1) model after function transform is brought forward to improve degree of smoothness of primary data sequence. Finally prediction precision of metabolic GM (1,1) model is 7.2% and 15.46% higher than those of new information GM (1,1) model and total data GM (1,1) model respectively. The comparison of the experiment’s results indicates the accuracy and robustness advantages of metabolic GM (1,1) model over other two models. [ABSTRACT FROM AUTHOR]