Research On The Thermal Error Model Of Ball Screws In Servo

The ball screw is widely used in precision servo system of numerical control machine and coordinate measuring machine for its high efficiency and long longevity. As the heat generated by friction between bearing and lead screw nut will result in the temperature increase of the leading screw and thereby deformation of bass screw, thermal error must not be disregarded. Considering that the lead screw is long in numerical control machine and Computerized Numerical Control Machine, the thermal error is especially obvious. Through study of the current research on thermal error modeling and relevant compensation technology both at home and abroad, the author of the paper found that there exist some flaws either in empirical modeling or theoretical modeling. The thesis does some researches in the temperature field and thermal error model of ball screw based on a large number of experiments on single axis semiclosed-loop servo combined with relevant theoretical analysis. It establishes a temperature field model of ball screw and makes prediction on thermal error with two methods. The structure of the thesis is as below.1. The Establishment of Experiment Platform.The author establishes servo platform based on semiclosed-loop control method with ball screw driving gear and“IPC+control card”open motion control system, and then builds a temperature acquisition system with PT100 thermal resistor, temperature transmitter, NI6251 capture card and Labview. Besides, two methods are introduced for the measurement of ball screw thermal error, namely Agilent HP5518A laser interferometer and HP55280A linear components.2. Establishment of Ball Screw Temperature Field Model and Prediction of Ball Screw’s Thermal Error in Servo System.On the basis of previous researches, appropriate revisions on boundary-value conditions have been achieved combining the practical situation, a model that can predict ball screw’s temperature distribution under single thermal source is developed and a method to identify model parameters is proposed. Then, by analyzing the ball screw’s temperature distribution under multi-thermal sources, the ball screw’s temperature distribution model in servo system is developed which utilizes the superposition of thermal sources’effect on feed screw. Furthermore, ball screw’s thermal error is predicted based on the established temperature field model and the mechanical thermal deformation theory.3. Prediction of Ball Screw’s Thermal Error Based on the BP Neural Network. Firstly, the structure and the training method of BP Neural Network is selected. Secondly, a model of ball screw’s error based on the BP Neural Network is established in servo system.4. Experiment Design and Results Analysis.Correlative experiments are conducted to verify the feasibility of both methods and comparison of two thermal deformation prediction methods is made. Besides, correspondent experimental program is designed to verify this ball screw temperature field model.