摘要

A novel approach, which can be used for dynamic characteristics analysis of machine tools based on unit structure (US), is reported in this paper. The concepts of unit structures for design of machine tools are defined. In order to satisfy the dynamic characteristics requirement of high natural frequency and light-weight of US, a design method of multi-disciplinary optimization of NSGA-II about unit structures driven by natural frequency and mass is developed. Through analyzing the unit structures, key factors affecting the natural frequency and mass are extracted, and the mathematical models of natural frequency and mass about unit structures are also established by using central composite design and response surface model. The goal of high natural frequency and light-weight is reached by using the multi-objective genetic algorithms. The Pareto optimal set is also obtained. The dynamic behavior of US is investigated by the experimental modal analysis. To show the efficiency of the proposed novel method, the example of YKW51250 gear shaping machine bed is used. Through optimization of NSGA-II about US of YKW51250 machine bed, the natural frequency of YKW51250 gear shaping machine bed is increased by 30.4% and its mass decreased by 5.2% comparing with the original design. By studying the dynamic characteristics of the simplified machine tools bed, useful laws are obtained, and these laws can be used in primary design of NC machine tools structures. The optimal method based on US can be also applied to the dynamic optimal design of machine tools and other similar equipments.