基于遗传算法的几何产品形位误差计算

Form and Position Errors Computation of Geometrical Products Using Genetic Algorithm

几何产品形位误差的计算是机械制造领域中的关键问题,属于非线性优化问题,应用遗传算法的全局优化能力实现了对误差问题的精确求解。重点研究了算法的实现技术,包括统一计算模型的建立、适应度函数的确定、种群繁衍方式的设计等。采用谐波技术和实测实例对算法进行了验算。结果显示算法合理,计算精度高,对实现形位误差的精确计算具有重要意义。

The evaluation of form and position errors is the key technique in the field of manufacturing which belongs to non -linear optimization problems. The genetic algorithm （GA） -based method is adopted to precisely, solve the problem by using its global optimiza- tion advantage. The key techniques of GA implementation have been studied, including founding the evaluation model, constructing the fitness function, devising the reproduction strategies, etc.. The data with given errors generated through harmonic analysis and those sampled by a Brown＆Sharpe CMM （Coordinate Measurement Machine） from a part are both used to verify the GA -based method. All the results show that the method is correct and more precise than other methods.