摘要
针对动态系统仿真模型有效性验证中输出响应间存在强相关性的问题,基于误差评分法和主成分分析法,提出了一种既能全面分析动态系统输出响应间曲线特征差异,又能解决输出响应间相关性问题的多元响应分析(MRA)方法,并结合具体实例对MRA方法进行了验证。结果表明,该方法能够实现多元数据的降维并对具有强相关性的数据进行评分,提高了验证效率和鲁棒性。
To solve the problem of strong correlation between output responses in the validation of dynamic system simulation model,a Multivariate Response Analysis(MRA)method is proposed,which can comprehensively analyze the difference of curve characteristics between output responses of dynamic systems and solve the correlation between output responses based on principal component analysis and error scoring.The MRA method is verified by practical cases.The results show that this method can effectively achieve dimensionality reduction of multivariate data,and score the data with strong correlation,and improve the verification efficiency and robustness.
作者
雷正保
邢欢
陈小勇
Lei Zhengbao;Xing Huan;Chen Xiaoyong(Large Structural Impact Laboratory,Changsha University of Science and Technology,Changsha 410114;Key Laboratory of Highway Engineering of Ministry of Education,Changsha University of Science and Technology,Changsha 410114)
出处
《汽车技术》
CSCD
北大核心
2020年第8期41-45,共5页
Automobile Technology
基金
国家自然科学基金项目(51675059)。
关键词
车辆安全技术
动态系统
模型验证
主成分分析
Vehicle safety technology
Dynamic system
Model verification
Principal component analysis
