钛合金显微组织与性能定量关系的模型

Modelling the Quantitative Correlation between the Microstructure and Mechanical Properties in Titanium Alloys

摘要:

提出利用软计算方法定量分析钛合金显微组织与性能的关系。在实现钛合金显微组织的定量分析和参数化表征的基础上，探讨以钛合金显微组织特征参数为输入变量、以性能为预测输出变量的组织与性能定量关系模型的构建方法。依据长期实践中得到的组织影响性能的定性规律及热加工工艺对组织特征的影响程度，提出定量关系模型显微组织信息预处理过程中组织特征参数评价取舍原则。运用人工神经网络建立Ti-17合金组织与室温拉伸力学性能的定量关系模型，研究建模的具体方法。该模型对各项性能指标具有较高的预测精度，说明提出的建模思想合理和模型构建方法可行

Abstract:

Modelling has now become a routine part of materials science. A model is developed for quantitative analysis and prediction of the correlation between the microstructure and mechanical properties in titanium alloys through the use of soft computing, based on quantifying microstructural features in titanium alloys. The input parameters of the model are the microstructural feature parameters, and the output parameters are some mechanical properties of the alloys. To achieve a good performance of the model, the input microstructural feature parameters must be chosen when designing model in terms of three rules: the qualitative effect of the microstructure on mechanical properties, correlative work processing and the operational efficiency of the model. A model for the analysis and prediction of the correlation between the microstructure and room-temperature tensile properties in Ti-17 titanium alloy is developed by applying artificial neural network, which is given as an example to discuss the modelling methods and approach. The model achieves the desirable predicting performance, so the modelling principle is correct and its implementation is feasible.