摘要
为了改进风险识别手段,提出了风险空间的概念。首先通过样本学习,使风险空间具备风险评估能力;然后针对高维空间下分类难度大,分类结果可解释性、可表达性差的缺点,采用主成分分析算法对风险空间进行降维,以利于后续处理。针对航天产品的实例验证表明,该方法操作简便、效果明显。
In order to improve the risk identification method,it proposes the concept of risk space.Based on sample learning,the risk space has the ability of risk classification.Aiming at the difficulty of classification in high dimensional space and the poor interpretability and expressiveness of classification results,it uses the principal component analysis(PCA)to reduce the dimension of the risk space.The application example shows that the method is reliable,easy to operate,effective and promising.
作者
陈刚
齐海雁
曹晓
徐雪萍
陈华
Chen Gang;Qi Haiyan;Cao Xiao;Xu Xueping;Chen Hua(Shanghai Aerospace Equipment Manufacturer Co.,Ltd.,Shanghai,200245,China)
出处
《机械设计与制造工程》
2019年第12期95-98,共4页
Machine Design and Manufacturing Engineering
基金
国家智能制造专项(2016ZXFM03002)
关键词
航天产品
风险识别
风险空间
样本学习
分类
主成分分析
aerospace products
risk identification
risk space
sample learning
classification
PCA
