神经网络与遗传算法应用于离子镀膜性能预报与工艺优化

The Application of Artificial Neural Network and Genatic Algorithms to the Prediction of Procedure Effects and Optimization of Procedure Parameter

摘要: 以多弧离子镀(TiAl) N膜为例,建立了模层性能预报神经网络模型;将神经网络模型与遗传算法结合,建立了实现工艺参数优化(设计)的遗传算法模块.实验结果验证了性能预报神经网络模型与工艺优化遗传算法模块的可靠性,为解决离子镀膜性能预报与工艺优化设计问题提供了一条先进、合理的途径.

Abstract: An Artficial Interlligence Neural Network Module (ANNM) has been built on the base of experimental data of(TiAI)N film which is obtained device of Multi-arc ion-plating, to predict the effects of ion plating procedure. In our system, a Genetic Algorithms Module(GAM) is also designed to provide us optimal values of paramrters of the procedure. The reliability of our sysytem is verified through additional experiments. Our work presents us an advanced and reasonable approach to the prediction of effects and optimization of paremeters of procedure, in the area of ion plating.