遗传算法及其在平面度误差求解中的应用

来源期刊：中南大学学报(自然科学版)1999年第1期

论文作者：喻寿益 廖平

文章页码：92 - 95

关键词：遗传算法;平面度误差;误差分析

Key words：genetic algorithms; plane error; error analysis

摘    要：对标准遗传算法提出了一些改进,并应用于计算满足最小区域法的平面度误差.采用实数值编码,其计算结果的精确度非常高,理论上可以获得全局最优解.改进的遗传算法简单明了,收敛速度快,在计算机上容易实现,可用于三坐标测量机等测量平面度误差的数据处理.

Abstract: Genetic algorithms can effectively find out the optimal solution.But the traditional genetic algorithms has some defects, for example, the existence of precocity, stagnancy and blind mutation. Besides, its calculating precision is limited by the length of binary encoding, and sometimes it cannot obtain the global optimal solution. Some improving methods for the traditional genetic algorithms have been discussed in this paper. Simulation shows that numerical encoding and save the best individual into the next generation can overcome the limitation of precision, find out the global optimal solution and obtain fast convergence velocity. Improved genetic algorithms can be effectively calculate the plane error, realized easily using computer and applied to the three coordinate measuring machine.