M.Sc Thesis


M.Sc StudentElbaz Yair
SubjectHybrid Approach to Predict and Optimize Flaw Detection
Capabilities using Laser Shearography
DepartmentDepartment of Aerospace Engineering
Supervisor ASSOCIATE PROF. Haim Abramovich
Full Thesis textFull thesis text - English Version


Abstract

In recent years, there is significant integration of composite materials in the aircraft structure due to their unique benefits. Flaws in composite structures may be formed already during the manufacturing process or during service. These flaws may develop in different areas of the aircraft and cause structure failure and even a flight safety incident. Effective and reliable NDT (non-destructive testing) capability is the guarantee for early structural flaws detection in order to maintain the integrity of the aircraft structure and ensure flight safety.

NDT  of composite structures is a global challenge. There are different NDT methods, each of which has advantages and limitations and which provides a certain solution.

One of the main methods, which has been gaining momentum in recent years, is the Laser Shearography Testing (LST).

LST is an NDT technology based on the analysis of the optical derivative of the perpendicular displacement of the surface being examined under excitation. This method uses laser and interferometric imaging, which can be used to detect subsurface flaws.

Today, due to dependence on several parameters, the detection capabilities of the various NDT methods for composite structures, especially in the LST, are not known in advance and individual experiments using dedicated test specimens are required - each case on its own.

The manner in which the NDT procedures are currently being developed is based on accumulated experience, trial and error, and a number of "rules of thumb" yielding only a general assessment, if any, of the manner in which the flaws are modulated and stimulated.

In the present work, we present a theoretical, analytical and experimental study of the engineering prediction of shearography fringe pattern. In our study, we first present the mathematical and the physical models of the LST. Then, based on these models, we had developed a methodology, which was implemented as a simulation that can evaluate the LST fringe pattern in advance. This simulation focuses on a circular delamination flaw that is stimulated by vacuum excitation. The simulation was verified by comparing its results to representative cases from the literature and by conducting a practical experiment using the LST system. The simulation results showed a good agreement with the experimental results.

Subsequently, "Detection Capability Envelopes" were constructed for the LST method and making it possible to assess the flaws detection capability in advance, depending on a variety of parameters. Investigation of the detection capability and construction of these envelopes was applied for the three most common types of skin structures in aviation - aluminum, composite solid laminate and composite sandwich.

This methodology and technique will allow a more correct and accurate design of the test specimens and lead to optimal test parameters defining process by a hybrid approach.

In addition, the hybrid method can be used as a guide and as a reference point for all those involved in the testing process using the LST method, developers and technicians alike, for the more efficient and accurate flaws detection in aviation structures - which will lead to improvements of  the flight safety.