采用自适应遗传算法的机床公差分配研究
Research on the Tolerance Allocation of Machine Tools Based on Adaptive Genetic Algorithm

针对当前机床几何精度建模忽视装配过程中的调整量,以及机床公差分配时缺乏科学可行的方法问题,利用状态空间模型描述机床实际装配过程,考虑装配过程中的调整控制量,建立了更加准确的机床装配精度模型,并引入种群多样性指标,构建了用于机床公差分配的自适应遗传算法。以TGK46100精密卧式坐标镗床为研究对象,建立了装配精度要求与基础大件角度误差之间的映射关系,以零件加工总成本最小为目标函数,采用构建的自适应遗传算法,完成了该机床基础大件角度公差的分配。结果表明:与不考虑装配调整量的偏差累积方法相比,该方法放宽了零件加工精度,最大放宽幅度达到了36？？4％,平均放宽幅度为12.0％,从而在满足最终装配精度要求的前提下,降低了零件加工制造成本,为机床公差分配提供了更加准确的精度建模方法和可行合适的公差分配方法。
The current modeling methods of machine tool’s geometric precision often ignore the effects of adjustment, lacking of scientific and feasible approaches to allocate the tolerance of machine tools. This paper adopts the state space model to describe the actual assemble process of a machine tool, and the adjustments in the assemble process are considered. Thus a more accurate model is set up. By introducing the index of the species diversity, an adaptive genetic algorithm for tolerance allocation is built up. The method was applied in the tolerance distribution of the angular tolerance in TGK46100. Firstly, the relationship of assembly accuracy requirements and angle error of large parts is constructed. Then taking the total cost of parts machining as the objective function and using the adaptive genetic algorithm, the optimal angle tolerance is obtained. The results show that, compared with the deviation accumulation method without considering the assembly adjustment, the machining accuracy of parts is relaxed and the maximum relaxation rate reaches 12％, with an average of 36.4％. Hence the parts manufacturing cost is reduced when the final assembly accuracy requirements are met. This research provides an accurate modeling method for machine tools’ tolerance distribution