| Operational Modal Analysis (OMA) is a technique that characterizes a structure/system on the basis of output responses only. It is an emerging field in structural dynamics and has been applied to complex structures that are often difficult to analyze using traditional Experimental Modal Analysis (EMA) techniques. However, the unavailability of input force information, in the case of OMA, makes the overall process significantly more complex as it affects every stage of modal analysis including data acquisition, data processing, parameter estimation, etc. Factors such as these have been responsible for the lack of frequency domain OMA algorithms, inconsistent damping estimation, etc. This research provides useful insights into the OMA techniques by in-depth exploration of the various assumptions under which OMA works and suggests some new signal processing approaches and algorithms to aid in modal parameter estimation using OMA techniques. The dissertation starts with a general literature review of OMA which is followed by presentation of the Unified Matrix Polynomial Approach (UMPA) to the OMA problem. In subsequent chapters, a frequency domain, lower order algorithm and a spatial domain algorithm are presented. Next, new signal processing techniques like Blind Source Separation/Independent Component Analysis (BSS/ICA) are adapted for OMA purposes. The performance of these algorithms is verified by conducting studies on analytical and experimental systems. Intense analytical studies are conducted to understand the effect of the violation of OMA assumptions on OMA modal parameter estimation, especially in view of inconsistent damping estimation. The algorithms developed in this dissertation are then applied to two newly built cable-stayed bridges for assessing their performance in real-life situations. Finally, the research conclusions are presented and recommendations for the future research in this area are given. |
