基于ICEEMDAN的连续梁桥车致振动信号的HHT分析

改进的带有自适应噪声的完备集合经验模式分解（improved complete ensemble empirical mode decomposition with adaptive noise，ICEEMDAN）是传统经验模式分解（empirical mode decomposition，EMD）方法的发展，在桥梁结构损伤识别领域具有较好的应用前景. 首先，以数值模拟信号为对象，采用ICEEMDAN方法进行桥梁车致动信号的数据分解和Hilbert谱分析，提取损伤引起的频谱特征变化和建立损伤识别方法；然后，利用该方法对实测振动信号的振型分量进行识别；最后，以实测信号的一阶振型分量为对象，对其Hilbert瞬时频率谱的特征进行了分析和讨论. 研究结果表明:模拟信号中的振型振动分量数比实测信号中多，其中模拟信号中不显著的高阶竖弯振动分量在实测信号中没有发现； 一阶振型振动分量的瞬时频率可作为桥梁损伤识别的特征参数，用于进行损伤有无、损伤定位甚至损伤定量的判断； 损伤识别效果受测点位置影响很小； 该方法不依赖有限元模型即可完成桥梁损伤有无的识别和损伤定位，且数据采集简单，具有实际工程中应用可行性. 

Hilbert-Huang Transfer Analysis on Vehicle-Induced Vibration Signal of Continuous Bridges Based on ICEEMDAN

As an improvement of the empirical mode decomposition (EMD), the improved complete ensemble empirical mode decomposition with adaptive noise (ICEEMDAN) has a good application potential in the field of structure damage identification of bridges. In this study, ICEEMDAN and Hilbert transforms were firstly performed on the simulated vehicle-induced vibrations of bridge, and the spectral feature changes caused by the damage were extracted with the established damage identification method. The established method was then applied on the measured vibration data of a real bridge, and the mode components of the vibration signals were identified. The characteristics of Hilbert instantaneous frequency spectrum were finally analyzed and discussed based on the first-order mode components of the measured signal. The results are shown as follows: There are more vibration components in the simulated signal than in the measured signal, and the insignificant mode components in the simulated signal do not appear in the measured signal; The instantaneous frequency of the first-order mode vibration component can be used as the characteristic parameter of bridge damage identification for the determination of the occurrence of damage, the damage location or even the quantification of the damage; The damage recognition capability of the established method is insensitive to the position of the measurement point; The proposed method has a feasibility in engineering practice, which completes the identification of bridge damage and damage location with a simple data acquisition procedure, and without relying on finite element model calculations.