跳到主要导航 跳到搜索 跳到主要内容

Determination of the influence of processing parameters on the mechanical properties of the Ti-6Al-4V alloy using an artificial neural network

  • Yu Sun
  • , Weidong Zeng
  • , Yuanfei Han
  • , Xiong Ma
  • , Yongqing Zhao
  • , Ping Guo
  • , Gui Wang
  • , Matthew S. Dargusch
  • Northwestern Polytechnical University Xian
  • University of Queensland
  • Northwest Institute for Nonferrous Metal Research
  • DMTC Limited

科研成果: 期刊稿件 › 文章 › 同行评审

48 引用 (Scopus)

摘要

There are many difficulties associated with the development of a quantitative correlation model relating the thermo-mechanical processing parameters to mechanical properties due to the complexity of the problem. In this research, based on the experimental data obtained from a series of forging and heat treatment experiments, the correlation model between hot processing parameters and the mechanical properties of the Ti-6Al-4V alloy has been established using an artificial neural network (ANN) approach. In the proposed model, the input variables are forging temperature, degree of deformation, annealing temperature and annealing time. The mechanical properties are determined as the output variables, including ultimate tensile strength, yield strength, elongation and reduction in area. Subsequently, the generalization capability of the trained ANN model was tested using an unseen data sample. The combined influence of hot processing parameters on the mechanical properties is further studied using the present model. It is found that a reliable correlation between processing parameters and mechanical properties of the Ti-6Al-4V alloy can be obtained. The artificial neural network method is capable of presenting the complex nonlinear relationship including interactions associated with hot processing parameters and mechanical properties.

源语言英语
页(从-至)239-244
页数6
期刊Computational Materials Science
卷60
DOI
出版状态已出版 - 7月 2012

学术指纹

探究 'Determination of the influence of processing parameters on the mechanical properties of the Ti-6Al-4V alloy using an artificial neural network' 的科研主题。它们共同构成独一无二的学术指纹。

引用此