针对集合经验模态分解算法存在的不足之处，提出了一种基于聚类分析的集合经验模态分解算法（CEEMD），以实现对响应信号的降噪与重构。首先对输入信号进行特征分析以确定加入白噪声的幅值标准差以及EEMD集成次数；其次进行EEMD分解；并对所得本征模态函数（IMF）利用欧式距离进行聚类分析，以检验所得本征模态函数之间是否存在模态混叠现象；然后采用模糊综合评价法计算每个IMF与实测信号之间的模糊相似系数，以便选出有效的IMF分量；再利用主成分分析和帕累托图法对保留下来的有效IMFs进行信号的重构，进而达到对实测信号的有效分解和降噪效果。为了验证该算法能运用于实际桥梁中，本文对某大型斜拉桥进行实例分析，首先对传感器所测响应信号进行重构，然后将其作为数据驱动随机子空间算法的输入，进行模态参数识别，同时为了进一步验证该算法所得结果比现有算法更为精确，对各算法结果进行了对比分析，结论是本文所提算法能对响应信号进行更好的降噪与重构，且所得结果更接近真实值，能运用于实际桥梁的模态参数识别。
Abstract
In this paper, aiming at the shortcomings of the ensemble empirical mode decomposition algorithm, an improved decomposition algorithm is proposed based on clustering analysis to achieve the goal that the noise reduction and reconstruction of response signal. Firstly, make an analysis for the input signals, and then to verify the amplitude standard deviation of added white noise and integration times of EEMD. Secondly, making EEMD decomposition and clustering analysis for the obtained intrinsic mode function (IMF) by using Euclidean distance to verify that whether there is a modal aliasing in the obtained intrinsic mode functions. Finally, it is necessary that calculating the fuzzy similarity coefficient between each IMF and measured signals by using fuzzy comprehensive evaluation method in order to select out the effective IMF components, and then using principal component analysis method and Pareto Diagram method reconstruct the signal of preserved effective IMFs, so that can achieve effective decomposition of the measured signal and the noise reduction. An empirical analysis was taken to a large cable-stayed bridge in this
paper in oder to verify that the algorithm can be applied to actual bridges. In the first place, refactoring the measured response signal of the sensor and then regarding it as the input of the data-driven stochastic sunspace algorithm,which is be used to identify the modal parameters. And at the same time,making comparison analysis for the results of the various algorithms to further validate this algorithm is much more accurate than the existing algorithms. The conclusion is that this proposed algorithm can reconstruct better and has a great noise reduction performance of the response signal, and this result is more closer to the real value, which can be applied to modal parameters identification of real bridge.
关键词
桥梁工程 /
CEEMD /
模糊综合评价法 /
主成分分析 /
帕累托图 /
Data-SSI
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1]漆景星,侯艳红. 桥梁可靠性评估综述[J]. 浙江交通职业技术学院学报,2011,01:20-22.
[1]Qi Jingxing, Hou Yanhong.Literature Review of Bridge Reliability Assessment [J].Journal of Zhejiang Vocational and Technical Institute of Transportation, 2011(01).
[2] Lin P, Zhang N, Ni B. On-line modal parameter monitoring of bridges exploiting multi-core capacity by recursive stochastic subspace identification method[C].American Control Conference, 2008. IEEE, 2008:632 - 637.
[3]周晶. 基于环境振动模态参数识别随机子空间方法与应用[D].兰州理工大学,2008.
[3]Jin Zhou. The method and application of model parameter identification,[D]. Lanzhou University of Technology, 2008.
[4]Peeters B and De Roeck G 1999 Reference -based stochastic subspace identification for output-only modal analysis Mech.Syst. Signal Process. 13855–78.
[5]郑近德,程军圣,杨宇. 改进的EEMD算法及其应用研究[J]. 振动与冲击,2013,23.
[5]Jingde Zheng, Junsheng Cheng, Yu Yang. Research of Improved EEMD Algorithm and its Application [J].Journal of Vibration and Shock, 2013（23）
[6]Joseph Lardies, Minh-Ngi Ta. Modal parameter identification of stay cables from output-only measurements. Proceeding of the Mechanical Systems and Signal Processing, 25 (2011) 133 –150.
[7]蔡艳平,李艾华,徐斌,许平,何艳萍. 集成经验模态分解中加入白噪声的自适应准则[J]. 振动.测试与诊断,2011,06:709-714+811.
[7]Cai Yanping, Li Aihua, Xu Bing, Xu Ping, He Yanping. The adaptive rule of Ensemble Empirical Mode Decomposition with added white noise [J]. Journal of Vibration Measurement & Diagnosis, 2011, 06:709-714+811.
[8]唐东明. 聚类分析及其应用研究[D].电子科技大学,2010.
[8]Dongming Tang. Cluster analysis and its application [D]. University of electronic science and technology of china, 2010.
[9]许顺国,牟瑞芳,张雪梅. 模糊数学综合评判法在水质评价中的应用——以成都市府河为例[J]. 唐山师范学院学报,2007,02:68-70.
[9]Shunguo Xu, Ruifang Mou, Xuemei Zhang. Application of Fuzzy Comprehensive Evaluation Method in Water Quality Evaluation of Chengdu Fuhe River [J].Journal of Tangshan Normal University, 2007（02）:68-70
[10]谢沙沙,李素梅,王光华,马辰阳. 基于主成分分析和支持向量机的人眼注视识别[J]. 信息技术,2014,07:163-166+170.
[10]Shasha Xie, Sumei Li, Guanghua Wang, Chenyang Ma. The human eye gaze recognition based on main component analyst and support vector machine [J].Journal of Information and technology, 2014(07).
[11] Jing Hang, Yuan Haiqing, Zhao Yi, etc. Applying Hilbert-Huang Transform to Identifying Structural Modal Parameters [C]. ettandgrs, vol. 2, pp.617-621, 2008
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
PDF(3521 KB)
