摘要
采用BP人工神经网络建立AZ91D镁合金力学性能与挤压铸造工艺参数的关系模型,研究并分析了工艺参数对合金力学性能的影响规律。结果表明:比压对合金力学性能影响最强,浇注温度次之,保压时间最弱。在浇注温度680℃、比压200 MPa、保压时间25 s、模具温度280℃条件下可使合金获得良好综合性能。BP网络模型预测的准确率最高为96%,具有良好的可靠性和推广价值,对挤压铸造AZ91D镁合金生产具有实践指导和理论借鉴意义。
The modeling was established by BP neural network to build the relationship between the mechanical properties of AZ91D alloy and squeezing casting processing parameters. The regular rule of parameter on alloy properties was analyzed. The results show that effect of pressure on the mechanical properties of the alloy is strongest, secondly pouring temperature, the effect results of pressure holding time is weakest. It can make the alloy to obtain good comprehensive performance in the condition of pouring temperature 680℃, pressure 200MPa, holding pressure time 25 s, mold temperature 280℃. The prediction by BP network model has good reliability and spreading value, and the highest accuracy rate of the prediction is 96%. Therefore, BP network model has the practical and theoretical significance on the squeeze casting AZ91D alloy oroduction.
出处
《热加工工艺》
CSCD
北大核心
2013年第7期46-48,共3页
Hot Working Technology
基金
华南理工大学国家金属材料近净成形工程技术研究中心开放基金项目(2011010)
河北省教育厅青年基金项目(2011252)
河北省教育厅基金项目(Z2010203)
河北省科技支撑计划项目(11212143)
