基于AEEMD和改进DATA-SSI算法的桥梁结构模态参数自动化识别

发布时间：2018-11-08 15:44

【摘要】：模态参数作为桥梁结构最重要的动力参数之一,在实际运用中,可通过监测其变化情况来辨识结构的使用性能,精确地参数识别对保障桥梁健康运营具有十分重要的意义。鉴于此,该文对现阶段常用的振动信号降噪处理算法和模态参数识别算法进行了相应的改进。一方面,提出一种新的信号自适应分解与重构算法,即自适应总体平均经验模态分解算法(AEEMD),该算法相比总体平均经验模态分解算法(EEMD)而言,能够根据信号的自身特征自动化确定添加白噪声的幅值标准差和集成平均次数;能更好地处理端点效应;同时还能够保证所得本征模态函数之间不存在模态混叠现象;最终实现有效IMF分量的自动化筛选和信号重构。另一方面,将多维数据聚类分析算法引入随机子空间算法中,并以频率值、阻尼比以及振型系数为因子建立判别矩阵,以智能化区分虚假模态和真实模态,最终实现模态参数自动化识别。文章最后分别用模拟信号和实际桥梁测试信号对所提算法的有效性进行验证,结果表明,该文所提算法能运用于实际桥梁结构的模态参数自动化识别。
[Abstract]:Modal parameters, as one of the most important dynamic parameters of bridge structures, can be used to identify the performance of bridges by monitoring their variations, and accurate identification of parameters is of great significance to ensure the healthy operation of bridges. In view of this, the noise reduction algorithms and modal parameter identification algorithms commonly used at present are improved. On the one hand, a new adaptive signal decomposition and reconstruction algorithm, adaptive population average empirical mode decomposition (AEEMD),) algorithm, is proposed. Compared with the total average empirical mode decomposition algorithm (EEMD), this algorithm is more efficient than the global average empirical mode decomposition algorithm (EEMD). The amplitude standard deviation and integration average times of adding white noise can be automatically determined according to the characteristics of the signal. It can better deal with the endpoint effect and ensure that there is no mode aliasing between the eigenmode functions. Finally, the automatic screening of effective IMF components and signal reconstruction can be realized. On the other hand, the multi-dimensional data clustering analysis algorithm is introduced into the random subspace algorithm, and the discriminant matrix is established based on frequency value, damping ratio and mode coefficient, so as to distinguish the false mode from the real mode intelligently. Finally, the automatic identification of modal parameters is realized. Finally, the validity of the proposed algorithm is verified by the analog signal and the actual bridge test signal. The results show that the proposed algorithm can be applied to the automatic identification of the modal parameters of the actual bridge structure.
【作者单位】：重庆交通大学;西南交通大学;中国中铁二院工程集团有限责任公司;
【基金】：国家自然科学基金(11627802、51478071)
【分类号】：U446

【参考文献】