Co-kriging for delamination quantification in composite structures

• Grant number: 23/14432-1
• Support Opportunities: Scholarships abroad - Research Internship - Master's degree
• Start date: February 19, 2024
• End date: May 18, 2024
• Field of knowledge: Engineering - Mechanical Engineering - Mechanics of Solids
• Principal Investigator: Samuel da Silva
• Grantee: Henrique Cordeiro Novais
• Supervisor: Jean-François Deu
• Host Institution: Faculdade de Engenharia (FEIS). Universidade Estadual Paulista (UNESP). Campus de Ilha Solteira. Ilha Solteira , SP, Brazil
• Institution abroad: Conservatoire National des Arts et Métiers (Cnam), France
• Associated to the scholarship: 22/12575-7 - Co-Kriging Modeling for Delamination Quantification in Laminates., BP.MS

Abstract

Delamination is a common type of damage in laminate composites that needs to be detected and quantified early. This dissertation's central theme is applying the co-Kriging method to quantify this damage. While Ordinary Kriging has been employed for such prognoses, co-kriging enables the integration of a limited amount of high-cost data (derived from experimental tests or calibrated models) with more affordable data from basic models without prior calibration. The suggested approach correlates multiple data sets from Lamb wave simulations, both numerical and experimental, in laminates. Our numerical data is anticipated to be cheap, derived from a model based on the Kirchhoff-Love plate theory. This model integrates a rudimentary delamination damage model to examine its influence on specific signal outputs and damage indices. This inexpensive numerical data will be merged with results from experimental tests on laminates with piezoceramics that actuate and receive Lamb waves. The combined data from both sources aims to refine the surrogate model, leading to more accurate predictions regarding the extent and impact of the simulated delamination area. A primary challenge of this project is the simulation of damage conditions within the model, given the complexities associated with the effects of delamination. Our goal for the internship at CNAM is to develop this model with Prof. Jean-François Deu. In simulating this ¨damage scenario, we aim to produce co-variable data to forecast the delamination area in laminates. This forecast will be realized by merging two weighted high-cost and low-cost variables averages. (AU)