Date of Graduation


Document Type


Degree Name

Bachelor of Science in Civil Engineering

Degree Level



Civil Engineering


Grimmelsman, Kirk A.


Selvam, R. Panneer

Second Reader

Heymsfield, Ernest


The bridge population in the US is currently aging and deteriorating rapidly. More than 30% of the bridges across the country have already exceeded their expected design life. Therefore, it is important to develop more timely, reliable and quantitative alternatives to the qualitative visual inspection approach that is currently used to evaluate these structures. One experimental approach that has been researched extensively for quantitatively characterize bridges is Ambient Vibration Testing (AVT) also known as Operational Modal Analysis (OMA). In this approach, the vibration responses of a structure due to unmeasured and uncontrolled ambient dynamic excitation are measured and analyzed to identify the modal parameters of the structure. Modal parameters (which generally include the natural frequencies, mode shapes, and damping ratios) are system properties that are directly related to the mass and stiffness characteristics of a structure. Changes in the modal parameters will reflect changes in the mass and stiffness of a structure due to damage or deterioration. Although there are many advantages in obtaining a quantitative description of a structure’s in-situ condition, especially for supporting more rational and reliable management decisions, there are many potential sources of uncertainty associated with AVT that can limit the utility of the characterization for such purposes. Furthermore, since the dynamic excitation used in AVT testing is unmeasured, it is difficult to quantify and evaluate these uncertainties. This paper presents a research study that was designed to evaluate the effects of uncertainty in the unmeasured ambient dynamic excitation on the identified modal parameters of a multi-girder bridge. A novel dynamic excitation system was used in this study to provide controlled dynamic excitation to the bridge that was consistent with the assumed nature of ambient dynamic excitation. A number of controlled variations to the assumed nature were also evaluated. The modal parameters for these excitation cases are compared to results from uncontrolled ambient vibration and those obtained from traffic crossing the bridge. The experimental results clearly indicated that the characteristics of the dynamic excitation had a significant impact on the identified modal parameters for the bridge. The dynamic excitation of this relatively short span bridge due to natural environmental sources was the least effective for identifying the modal parameters. Traffic related excitation permitted more modes to be identified, but the results reflected some degree of interaction between the vehicles and the structure. The dynamic excitation provided by the tactile transducers that most closely matched the assumed characteristics of uncontrolled and unmeasured dynamic excitation provided the most reasonable modal parameter results of the cases evaluated. Although traffic excitation on top of the full band controlled white noise excitation yielded the most modal parameters of any case evaluated, these modal parameters exhibit more uncertainty due to the interactions between the vehicles and structure than the same parameters identified from the full band controlled excitation case without traffic crossing the structure. Considering the results obtained from each dynamic excitation case, the modal parameters identified for full band white noise excitation (with no traffic crossing the bridge) provided the most reasonable results.