Effect of multiple state characteristics on damage and wave propagation and compensation methods for damage detection

点击翻译按钮获取中文摘要

A guided ultrasonic wave (GUW) based structural health monitoring (SHM) system must be able to detect and distinguish changes in GUW propagation resulting from various state characteristics, such as structural damage and external environmental conditions, to be reliable. In the first Funding Period we were able to develop a simple but powerful compensation for temperature influence on GUW propagation. In addition to the temperature in a realistic test scenario, external static loads, e.g. caused by different fuel levels in the tank of aircraft wings, also determine the test conditions under which a SHM system must operate. Since external loads have a measurable influence on the GUW propagation and thus could undermine the developed temperature compensation method. In this subproject (SP), the simultaneous influence of temperature and static load on GUW propagation is investigated experimentally in order to develop methods that compensate for both state characteristics simultaneously. The influence of damage on GUW propagation will also be investigated. For this purpose, two types of fibre-metal laminates (FML), namely Glass Laminate Aluminium Reinforced Epoxy (GLARE) and Carbon Fibre Reinforced Polymer (CFRP) steel laminates, are considered to map impedance and metal volume content differences. For efficient damage diagnostics based on GUW, establishing the validated damage classes (DC) is an essential prerequisite. These DC will be correlated with GUW signals and thus enable comprehensive diagnostics. Since impact damage makes up a large portion of common damage, it is one of the most important DC. Impact damage characteristics include matrix cracking, delamination between the metal and prepreg plies, fibre breakage, metal cracks etc. These characteristics can be similar in different affected samples, making further differential classification based on the presence and absence of damage characteristics difficult. In order to verify, whether samples of the same DC can be further differentiated, they are examined after the application of mechanical loads by means of X-ray computed tomography (CT). It is assumed that this procedure makes differences in the damage patterns more prominent. Especially in the case of impact damages which have already had to endure loads compared to newly emerged damage, these investigations provide further subclasses, enabling an extension of the developed taxonomy. These new subclasses are correlated with GUW measurements, which enables an enlarged database for damage diagnostics using ultrasonic monitoring. The damage characteristics will be used in SP3 for model validation while the X-ray CT and GUW-data labelled according to DCs are used in SP 4 for damage diagnosis with machine learning methods. Since the condition of integrated sensors and electronics after application of mechanical loads is also investigated, the results form the basis for optimizing the functional design and integration methods in SP2.