节点文献
基于人工神经网络的微合金钢力学性能预报
Prediction of properties of microalloy steel based ibartificial neural network model
【摘要】 用人工神经元方法分析微合金钢性能、组织成份及铁素体晶粒尺寸之间关系,对某种微合金钢的力学性能、各相的分数和铁素体的晶粒尺寸之间的关系进行了研究,并通过实验数据训练了一套BP网络.结果表明,该方法是用于组织性能预报的有效方法之一,误差可以控制在5%以内.
【Abstract】 A method for design of artificial neural network to predict mechanical properties of microalloyed steel is studied. The relationships between mechanical properties and volume of each phase and ferrite grain size are analyzed. Based on the experimental data, a neural network is established. Results show that artificial neural network is an effective method for the prediction of properties, errors can be controlled within 5%.
【关键词】 人工神经元;
BP;
微合金钢;
性能预报;
【Key words】 artificial neural network(ANNs); back propagation; microalloyed steel; property prediction;
【Key words】 artificial neural network(ANNs); back propagation; microalloyed steel; property prediction;
【基金】 国家高技术研究发展计划资助项目(2001AA779030).
- 【文献出处】 哈尔滨工业大学学报 ,Journal of Harbin Institute of Technology , 编辑部邮箱 ,2004年12期
- 【分类号】TG113.25
- 【被引频次】6
- 【下载频次】177