摘要
将一种新型的统计回归方法——偏最小二乘方法(partial least square,PLS)用于热作模具钢的性能预测,借助提取主元的思想,利用PLS对热作模具钢工作温度下的性能参数和材料的热疲劳性能之间的相关信息进行筛选和综合,在此基础上建立热作模具钢热疲劳性能预测模型。结果表明模型对材料的热疲劳性能有较好的预测能力。同时,提高材料的高温屈服强度、冲击韧度、热稳定性、抗氧化性及伸长率有助于改善材料的热疲劳性能。
A new approach based on partial least square was used to predict hot working steels performance of thermal-fatigue.Based on the idea of extracting principal components,correlation information between materials properties and thermal-fatigue resistance under the operating temperature was screened and synthesized.Moreover,an integrated assessment model was built based on the information of these materials.The results indicate that this model had good ability to forecast steels thermal-fatigue performance.In...
出处
《机械强度》
CAS
CSCD
北大核心
2010年第5期824-828,共5页
Journal of Mechanical Strength
关键词
热疲劳
材料性能
偏最小二乘
预测
Thermal-fatigue
Material property
Partial least-squares
Prediction
