Study On Accuracy Stability Testing Of Servo Feed System On CNC Machine Tool

With continuous improvement of equipment manufacturing requirements on advanced manufacturing technology, the low stability of accuracy on domestic CNC machine tools has become a key restriction for the machine tool industry to develop towards the field of high speed precision machine tools. So stability of accuracy is an urgent problem to be solved.This paper is based on the National Science and Technology Major project of "Exemplary Applications of Domestic Rolling Functional Components on High-End CNC Machine Tools". The accuracy stability of servo feed system and its testing methods were studied. The main contents include the following aspects:Servo feed system and its key transmission components’failures locations, modes and causes were analyzed via the method of FMEA. On this basis, the composition of feed system positioning errors was classified. And then a method, which was based on failures formation process, was introduced to analyze the stability of accuracy, and a prediction equation of accuracy stability was deduced in this process.A coupling mathematical model of AC servo system and mechanical transmission was established. By modeling the tricyclic structure of servo control system and tuning its parameters, the steady and dynamic characteristics of control system under different servo parameters was analyzed. And a mathematical model of mechanical transmission system in hybrid conditions was achieved through dynamic analysis, and then the influence of the stiffness on positioning accuracy was formulated.Designed a test concerning the accuracy stability of servo feed system, including the design and selection of test content, the methods of measuring accuracy, and testing equipment. Then initial processing methods of experimental data were explored.Designed appropriate reliability statistical methods to analyze test data from different stages. First, achieved reliability statistics of field test data by using two-parameter Weibull distribution model and probability plots methods. Second, the concept of accuracy spectrum was proposed to integrate the test model with reliability theory. Then multiple regression methods based on least squares was utilized to estimate the parameters of accuracy spectrum distribution model. Finally, in the process of analyzing the whole life data, a non-homogeneous Poisson process model based on AMSAA was introduced to describe the variation of positioning accuracy along with running distance, and thus the forecast of accuracy stability was achieved.