Reliability of surface response to excitation method for data-driven prognostics using Gaussian process regression
In this study, the surface response to excitation method (SuRE) is investigated using a data-driven method for diagnostics and prognostics of applied load on a plate structure. The SuRE method is an emerging approach in ultrasonic wave-based structural health monitoring (SHM). In this method, high-frequency, surface-guided waves are excited on the structure using piezoceramic elements. The waves propagate and interact with internal or surface damages on the structure. State of heath is evaluated by monitoring changes in the measured frequency transfer functions. Reliability and computational efficiency of the SuRE method has been verified for several diagnostic and structural health monitoring applications. In this paper, the effectiveness of the SuRE method for prognostics and health management (PHM) of a composite plate under applied load is studied. Two piezoelectric elements are attached on the surface of a carbon fiber reinforced polymer (CFRP) composite plate. Sweep excitation-generated (150-250 kHz), surface-guided waves and the transmitted waves were monitored at the sensory position. The reference data set comprised of characteristic transfer functions was generated. SHM data using the SuRE method was captured for eleven locations of applied load between the sensor and exciter. Four data-driven prognostic models, using Gaussian Processes Regression (GPR), were qualified by interval-averaging features extracted from the spectrums and predicted the location of load. During this study, a new approach based on SuRE method is proposed for identifying the location of applied load on a composite and the optimum parameters of the study were evaluated to enhance the performance of GPR identified the optimum parameters number of SuRE method and selected features for most accurate predictions.
Fekrmandi, H., and Y. S. Gwon. “Reliability of surface response to excitation method for data-driven prognostics using Gaussian process regression.” In health monitoring of structural and biological systems XII, vol. 10600, p. 106002R. International Society for Optics and Photonics, 2018. DOI Link
Page Author: Samuel Irwin, BS student Mechanical Engineering, SDSMT