Vibration-based assessment of delaminations in FRP composite plates
Research output: Contribution to journal › Article › Scientific › peer-review
|Number of pages||13|
|Journal||Composites Part B : Engineering|
|Publication status||Published - 1 Jul 2018|
|Publication type||A1 Journal article-refereed|
Delamination is a frequently occurring type of damage in laminated fibre reinforced polymer (FRP) composites and causes substantial loss in structural stiffness and usable service life. The detection of delaminations in FRP composites is critical for the safe and reliable use of these materials in aeronautical and other industries. Structural Health Monitoring (SHM) techniques based on vibration measurements have proven to be promising towards this end. There have been comprehensive studies of FRP beams with through-width delaminations, but the damage assessment of FRP plates with embedded delaminations using frequency-based detection has not been extensively studied. To solve the inverse problem of determining size and location of delamination from changes in the natural frequencies, this paper presents a new surrogate assisted optimisation (SAO) method for predicting the location and size of delaminations in fibre reinforced composite plates using natural frequency shifts as indicative parameters. The proposed frequency-based delamination assessment method is validated using finite element models of FRP plates with embedded delaminations and by experimental modal analysis. Modal testing was conducted using scanning laser vibrometer on carbon/epoxy and glass/epoxy FRP plates that were manufactured with artificially induced delaminations. The proposed SAO algorithm was compared to an Artificial Neural Network (ANN) method in terms of database size, prediction accuracy and sensitivity to noisy data. The results show that the proposed inverse algorithm can predict the delamination parameters of location and size with good accuracy for numerically simulated frequency shift data but the prediction accuracy was reduced with experimental data. A comparison of the two inverse algorithms show that the SAO method has significant advantages compared to the ANN algorithm for delamination prediction.