人工神经网络在钢铁材料力学性能预测方面的应用

人工神经网络模型特别适用于非线性系统。具有较好的学习精度和概括能力。已成功应用于钢铁材料力学性能的预测。使用人工神经网络模型，通过输入合金元素、组织、生产工艺参数可预测钢铁材料的抗拉强度、延伸率、韧性、疲劳和蠕变性能。概要叙述了人工神经网络在预测板材、球墨铸铁的常温力学性能，合金结构钢的淬透性。高速钢、不锈耐热钢的热强度以及微合金钢热扭转性能方面的应用。

Application of Artificial Neural Networks in Prediction of Mechanical Properties of Iron and Steel Materials

The artificial neural networks model is available especially to apply in non-linearity system, has better leaning precision and summarizing ability, and has been successfully applied to prediction of mechanical properties of iron and steel materials. The tensile strength, elongation, toughness, fatigue and creep properties of iron and steel materials can be predicted using artificial neural networks model by input alloy elements, structure, and production parameters. The applications of artificial neural networks in prediction of ambient temperature mechanical properties of strip and spheroidal graphite cast iron, hardenability of alloy steel, hot strength of high speed steel and heat resistant - stainless steel and hot torsion properties of microalloying steel are summarily presented in this article.