基于人工智能钛合金热变形工艺参数优化.pdf

基于人工智能的钛合金热变形工艺参数优化 李萍 薛克敏 （合肥工业大学，材料科学与工程学院，安徽合肥，230009 ） 摘要 在深入分析热变形工艺参数对 Ti-15-3 合金显微组织及成形载荷影响的基础上，以热变形工艺参 数为设计变量，以显微组织和成形力的最佳综合为目标，建立了该合金热塑性成形工艺参数的多目标优 化数学模型。将神经网络与修正的遗传算法相结合，对 Ti-15-3 合金的热塑性成形工艺参数进行了优化， 从而为制定合金最佳的热成形工艺参数提供了一条先进、合理的途径。 关键词 优化；遗传算法；神经网络；热变形；Ti-15-3 合金 中图分类号：TG302 文献标识码：A Optimization of Hot Deformation Process for Titanium Based on Artificial Intelligence LI Ping XUE Ke-min School of Materials Science and Engineering，Hefei University of Technology，Anhui Hefei 230009 Abstract A systematic analysis of the effect of hot deformation process parameters on microstructure and load of Ti-15-3 alloy have been conducted in the previous studies ．Based on the results，multi-objection optimization model has been established for hot deformation process of Ti-15-3 alloy ．In the model，the objective is to determine the process parameters that produce a specified microstructure in the final product under the lesser load，Artificial neural network and modified genetic algorithm have been adopted to optimize the parameters of hot deformation process for Ti-15-3 alloy ．The results indicate that the optimization method presented in this paper offers a new effective means for determining the best hot deformation process parameters ． Key words Optimization ；Genetic algorithm；Neural network ；Hot deformation ；Ti-15-3 alloy CLC number：TG302 Document code ：A Fe0.13 ，C0.03 ，N0 .02 ，Si0.07 ，O0.13 ，其余为 1 引言 Ti，相变点约为 760℃。由于Ti-15-3 合金成形时 变形抗