Performance Evaluation of Structural Health Monitoring System Applied to Full-Size Composite Wing Spar via Probability of Detection Techniques

Abstract

Probability of detection (POD) is an acknowledged mean of evaluation for many investigations aiming at detecting some specific property of a subject of interest. For instance, it has had many applications for Non-Destructive Evaluation (NDE), aimed at identifying defects within structural architectures, and can easily be used for structural health monitoring (SHM) systems, meant as a compact and more integrated evolution of the former technology. In this paper, a probability of detection analysis is performed to estimate the reliability of an SHM system, applied to a wing box composite spar for bonding line quality assessment. Such a system is based on distributed fiber optics deployed on the reference component at specific locations for detecting strains; the attained data are then processed by a proprietary algorithm whose capability was already tested and reported in previous works, even at full-scale level. A finite element (FE) model, previously validated by experimental results, is used to simulate the presence of damage areas, whose effect is to modify strain transfer between adjacent parts. Numerical data are used to verify the capability of the SHM system in revealing the presence of the modeled physical discontinuities with respect to a specific set of loads, running along the beam up to cover its complete extension. The POD is then estimated through the analysis of the collected data sets, wide enough to assess the global SHM system performance. The results of this study eventually aim at improving the current strategies adopted for SHM for bonding analysis by identifying the intimate behavior of the system assessed at the date. The activities herein reported have been carried out within the RESUME project.

Keywords: FE simulations; aeronautic composite structures; de-bonding; probability of detection; structural health monitoring.

Figure 1

Schematic view of the spar used for the POD activities.

Figure 2

Detail of the C-spars modeling, with the position of the imposed damage regions highlighted and the fiber optics (dotted lines). Upper and bottom skin panels are not visualized. Caps are depicted in blue, while the webs are plotted in red.

Figure 3

Mesh refinement in proximity to the damaged areas.

Figure 4

Application of the Segment method for modeling optical fibers (white line): detail of the curved path.

Figure 5

Application of the Merge method for modeling optical fibers (white line): detail of the curved path.

Figure 6

Application of Virtual Contact method for modeling optical fibers (white line): a detail of the curved path.

Figure 7

Finite element simulation of fiber optic strain measures by Segment, Merge, and Virtual Contact methods.

Figure 8

Aeronautical full-scale composite spar.

Figure 9

SHM methodology flow-chart based on cross-correlation analysis.

Figure 10

Simplified sketch of the beam with position of damage (black rectangles), fiber optics (yellow lines), and load position (red arrow).

Figure 11

Experimental strain map during quasi-static loading.

Figure 12

SHM feature extraction: time domain cross-correlation (left); space domain cross-correlation (right). Threshold limit (TL) (blue line).

Figure 13

SHM readout of the baseline structure for the healthy spar cap.

Figure 14

SHM readout of the damaged structure.

Figure 15

Example of the relation between measured response by SHM and de-bonding length by C-scan.

Figure 16

Examples of numerical estimation of strain responses by different loading position: symmetric loading (left); asymmetric loading (right).

Figure 17

Detail of the composite beam. The top skin has been removed to provide a better view of the 4 fibers, embedded at the interface between the adhesive layer and the top skin. The 4 installed fibers are represented by 4 fine lines, while the yellow dots indicate the sensitive points, where the strain values are retrieved.

Figure 18

Relation between measured response by SHM (black distribution) and real de-bonding length (red distribution). Relation with threshold values by B-basis one-side limit (yellow and blue dotted circles) by using k_B = 1.456 numerically tabulated for a data set of 187 elements, according to one-side tolerance limit of the normal distribution.

Figure 19

Outcome of the SHM algorithm, in terms of the probability of detecting the damage edges, for the considered configuration (given constraints and damage areas; running point vertical load). The x-axis shows the 201 sensor IDs, arranged along the fiber with a constant 8 mm step, while the y-axis shows the number of occurrences, normalized with respect to the performed runs, which refer to the damage edge detection. The red bands represent the extension of the three damage areas; the green lines represent the tapering lines (thickness variations); and the yellow lines represent the location of the two supports. The blue rectangles indicate three arbitrary regions of structural healthy conditions.

Figure 20

Number of identified sensors: (a) the number of sensors identified by the SHM algorithm as indicators of damage occurrence, for each of the 3 damage zones; (b) the number of sensors identified by the SHM algorithm for the 3 healthy zones (see Figure 19). REMARK: Top and bottom vertical scales are different.

Figure 21

Normalization of the values reported in Figure 20, with respect to the total number of considered sensors in detail: (a) normalization by 14 sensors (D3), 25 sensors (D2), and 21 sensors (D5); (b) 3 arbitrary healthy zones normalized by 15 sensors. REMARK: Top and bottom vertical scales are different.

Figure 21

Normalization of the values reported in Figure 20, with respect to the total number of considered sensors in detail: (a) normalization by 14 sensors (D3), 25 sensors (D2), and 21 sensors (D5); (b) 3 arbitrary healthy zones normalized by 15 sensors. REMARK: Top and bottom vertical scales are different.