Sensor Fusion for Shape Sensing: Theory and Numerical Simulation

Research output: Contribution to conferencePosterScientific

16 Downloads (Pure)

Abstract

Computing global strain distributions in complex aerospace and wind energy structures in quasi-real time is an important challenge for the aerospace and wind energy industry. Shape sensing is being considered as a potential means for obtaining global strain fields of complex structures. This is achieved through the use of inverse finite element methods (iFEM). Traditional shape sensing techniques use a variety of sensors in isolation. These isolated approaches lead to a number of drawbacks, including significant errors in the strain and/or displacement distributions that they determine. This study focuses on reducing these errors by combining data from various sensor types.
In order to achieve this objective, a new iFEM approach was developed to improve how shape sensing determines a structure’s deformation from distributed sensors. The method utilizes a procedure that minimizes a quadratic error functional, based on the difference between the theoretical and the measured strains and displacements. The analytical iFEM equations are discretized for use in a numerical model, which is used to analyze simulated sensor data. The simulated data is obtained from a FEM analysis of a structure loaded in bending, torsion and shear, in addition to experiencing combined loads. As with previous inverse methods, the error functional weights were employed to manage missing measurements and Tikhonov regularization was applied to guarantee smoothness of the numerical solution.
In this study, it has been shown that the new iFEM methodology can determine the structural deformations to within 1% of the FEM results for each of the load cases. In addition, the methodology is capable of achieving the same accuracy as single sensor type iFEM methods with a smaller total number of sensors. The single sensor type iFEM methods that were considered made use of only strain data or only displacement data. Finally, an experimental setup consisting of a simple cantilever plate structure was constructed in order to experimentally verify the newly developed iFEM algorithm. The experimental strain and displacement data were obtained utilizing a Rayleigh backscattering fiber optic distributed sensing system and MEMS sensors, respectively. Both of these sensor types were mounted on the structure and subjected to the load cases that were simulated in the FEM analysis.
Original languageEnglish
Number of pages1
Publication statusPublished - 2016
Event27th International Conference on Adaptive Structures and Technologies - The Sagamore Resort, Lake George, United States
Duration: 3 Oct 20165 Oct 2016
Conference number: 27
http://www.icast2016.com/

Conference

Conference27th International Conference on Adaptive Structures and Technologies
Abbreviated titleICAST2016
CountryUnited States
CityLake George
Period3/10/165/10/16
Internet address

Fingerprint Dive into the research topics of 'Sensor Fusion for Shape Sensing: Theory and Numerical Simulation'. Together they form a unique fingerprint.

  • Cite this

    de Mooij, C., Martinez, M., & Benedictus, R. (2016). Sensor Fusion for Shape Sensing: Theory and Numerical Simulation. Poster session presented at 27th International Conference on Adaptive Structures and Technologies, Lake George, United States.