143. modal modeling via fiber optic strain sensing for applications in structural health monitoring
Name: Benjamin Levi Martins
Grad Year: 2017
Fiber optic strain sensors make ideal instruments for in-situ structural health monitoring and damage prognosis (SHM-SP) of aerospace vehicles due to their lightweight and easily multiplexed nature. In order for fiber optic strain sensors to successfully achieve SHM-DP capabilities it is imperative that proper techniques are developed which provide quality frequency domain models generated by response only dynamic strain time histories. For such technologies to be developed, analytical models are needed to study the behavior of the transformation methodology between the time and frequency domains. Additionally, these same analytical models may be used to provide estimations of a structures modal domain behavior under varying geometric and material parameters. In this study a two degree of freedom lumped-mass, spring, damper, system was studied which utilized the superposition of two harmonic base excitation forces to generate a broad spectrum impulse excitation. The transient response of the system was studied by calculating the exact displacement time histories and transforming these to the frequency domain using the frequency response function (FRF), sensor-to-sensor transfer function (TF) and the cross-power spectra (CPS). Comparisons were made between the data quality obtained by each of the transformation methods where it was shown that the CPS is well suited to develop quality modal models using response only transient strain data. A Timoshenko beam model was introduced which relied on the Rayleigh-Ritz (R-R) method to generate the equations of motion for the beam. The equations of motion were decoupled through transformation into modal coordinates such that the uncoupled equations of motion were solved and analytical strain time histories obtained. These time histories were transformed into the frequency domain using the FRF and CPS where the quality of their modal models was assessed by comparing their extracted modal parameters to the exact values. The results of these comparisons validated the ability of the CPS to accurately model a structures frequency domain response using response only strain time histories. Lastly, a modification to the R-R/Timoshenko beam was proposed which allows for varying geometric and material properties making possible the study of any number of damage scenarios. A damage case was run and the results compared to the healthy beam case validating the models ability to capture the effects of the damage.
Industry Application Area(s)
Aerospace, Defense, Security | Civil/Structural Engineering | Materials