【摘要】 针对控制复杂、难以精确描述数学模型的复杂加工过程,提出集成遗传算法和人工神经网络的多目标优化方法。把加工工艺参数和工艺指标参数分别看作神经网络的输入和输出,建立相对精确的数学模型,随后借助训练好的神经网络和遗传算法来优化加工参数。由于该方法解决了适应度函数难以获得的问题,为复杂过程多目标优化提供了强有力的工具。在电火花线切割中的应用验证了该方法的可实施性和有效性。更多还原

【Abstract】 An integrated computational intelligence method was proposed for modeling and optimization of high speed Wire Electro-Discharge Machining(WEDM) process,which was hard to develop practical mathematical models.Taking the manufacturing parameters and target parameters as the input and output values of a Neural Network(NN) respectively,the mathematical model for the process was set up with considerable precision.The manufacturing parameters then can be optimised with Genetic Algorithms(GAs) and trained NN.On account of this method did not need practical fitness function of the objects,so that it was the strong implemention for optimising complex processes.The test results prove that this method is feasible and effective.更多还原