摘要
提出了一种基于非线性核判别分析的悬跨管道状态识别方法。首先分析了悬跨管道状态识别问题的主要特点,根据悬跨管道状态变化频率结构的内部特征,构造由自然频率、归一化频率、频率变化率等特征参数组成的状态矢量。然后基于统计模式识别中的核判别分析原理,建立了悬跨管道的状态识别方法。最后通过实例分析研究了该方法的有效性,并讨论了悬跨管道状态矢量选取对识别效果的影响。研究结果表明:该方法能有效识别悬跨管道的多种状态,具有较强的抗测试误差能力。
An efficient approach to identify the structural condition of free spanning pipeline based on nonlinear discriminant analysis （KDA） is presented. The main characteristics of condition identification for free span （CIFS） are discussed firstly. Then, condition vectors consisting of natural frequencies, normalized frequencies and frequency change ratios are constructed based on internal relationship of frequency construction. Afterwards, an effective algorithm for CIFS is established based on the statistical pattern recognition of KDA. Finally, the validity of the proposed approach is evaluated by a case study. The effects of selection of condition vectors on the identification results are also studied. The results show that the proposed approach can identify effectively the structural condition of free span even with the presence of the measurement error.
出处
《工程力学》
EI
CSCD
北大核心
2009年第1期238-243,共6页
Engineering Mechanics
基金
国家自然科学基金重点项目(50439010)
教育部科学技术研究重大项目(305003)
关键词
水工结构
悬跨管道
状态识别
核判别分析
状态矢量
hydraulic structure
free span
condition identification
kernel discriminant analysis
condition vector
