Feasibility study of a multi-parameter probability of detection formulation for a Lamb waves–based structural health monitoring approach to light alloy aeronautical plates

Abstract

In view of an extensive literature about guided waves–based structural health monitoring of plate-like structures made of metallic and composite materials, a lack of information is pointed out regarding an effective and universally accepted approach for characterizing capability and reliability in detecting, localizing and sizing in-service damages. On the other hand, in the frame of traditional non-destructive testing systems, capability is typically expressed by means of suitable ‘probability of detection’ curves based on Berens’ model, where a linear relationship is established between probability of detection and flaw size. Although the uncertain factors are usually different between a non-destructive inspection technique and a structural health monitoring approach, it seems that a similar mathematical framework could be assumed. From this point of view, this research investigates the feasibility of application of the very recent ‘multi-parameter’ probability of detection approach, developed within the traditional non-destructive testing field, to guided waves–based structural health monitoring. In particular, numerical simulations as well as experimental responses from flawed aluminium alloy plates were combined to bring about a ‘master’ probability of detection curve. Once established, this curve can be used to study the intrinsic capability of the system in terms of probability of detection curves, overcoming the intrinsic limitation of a single predictor (like the crack size) and a statistical model typically based upon a linear behaviour between the predictor and the response.

Damage Localization in Composite Structures Using a Guided Waves Based Multi-Parameter Approach

“…Unsupervised [24,25] or supervised [26] data processing have been proposed and can be addressed to enhance the aircraft lifetime management as well [27]. Likewise, the signal response can be statistically related to the flaw size to assess the severity and size of damage from damage indicator [22,28]. Hence, a system efficiently warning hidden failures can simultaneously return the severity of damage.…”

A model-based approach for statistical assessment of detection and localization performance of guided wave–based imaging techniques

“…This requires a large number of inspection tasks for characterizing various configurations of damage that can potentially appear in the structure [17,18,19]. Many works have been done to extend this concept to SHM [20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,19,37]. Recently, Janapati et al characterized the performance of a pitch-catch SHM system for detecting a crack on an aluminum plate [19].…”

“…Nevertheless, this limitation can be overtaken by using a model-assisted approach, which produces so called ModelAssisted Probability of Detection (MAPOD) curves [21,25,26,35]. This approach, more adapted to SHM, was applied for assessing performance of various GW-based SHM systems [22,29,36,37]. However, SHM system assessment is generally limited to Technology Classification Level (TCL) 1 (i.e.…”

Modelling of the high-frequency fundamental symmetric Lamb wave using a new boundary element formulation

“…Thus, the BEM formulations based on plate theories are more suitable for the modelling of Lamb waves, especially for the cases in which the frequency is not too high. These cases are very common and of great interest in the SHM [16][17][18][19][20] because in order to simplify the wave signals and make the analysis easier, the damage detection processes are often conducted below the cut-off frequency of the first antisymmetric Lamb mode (A1) [21]. In these cases, only the fundamental symmetric Lamb mode (S0) and fundamental antisymmetric Lamb mode (A0) are excited.…”