This paper develops a model-assisted approach for determining predictive probability of detection curves. The approach is “model-assisted,” as the damage-sensitive features are evaluated in combination with a numerical model of the examined structure. It is “predictive” in the sense that probability of detection (POD) curves can be constructed based on measurement records from the undamaged structure, avoiding any destructive tests. The approach can be applied to a wide range of damage-sensitive features in structural health monitoring and non-destructive testing, provided the statistical distribution of the features can be approximated by a normal distribution. In particular, it is suitable for global vibration-based features, such as modal parameters, and evaluates changes in local structural components, for example, changes in material properties, cross-sectional values, prestressing forces, and support conditions. The approach explicitly considers the statistical uncertainties of the features due to measurement noise, unknown excitation, or other noise sources. Moreover, through confidence intervals, it considers model-based uncertainties due to uncertain structural parameters and a possible mismatch between the modeled and the real structure. Experimental studies based on a laboratory beam structure demonstrate that the approach can predict the POD before damage occurs. Ultimately, several ways to utilize predictive POD curves are discussed, for example, for the evaluation of the most suitable measurement equipment, for quality control, for feature selection, or sensor placement optimization.
CarboniMCantiniS. Advanced ultrasonic probability of detection curves for designing in-service inspection intervals. Int J Fatigue2016; 86: 77–87.
2.
MIL-HDBK. Department of defense handbook, nondestructive evaluation system reliability assessment. Technical report, Department of Defense, United States, 2009.
3.
GandossiLAnnisC. Probability of detection curves: statistical best-practices. Technical report no. 41, 2010, European Network for Inspection and Qualification.
4.
RummelWDChristnerBKMullenSJ, et al. Characterization of structural assessment testing. Technical report, SAALC/MMEI/1/86, January1986.
5.
BerensAPHoveyPW. Evaluation of NDE reliability characterization. Technical report volume I, University of Dayton Research Institute, Dayton, OH, 1981.
6.
BerensAP. NDE reliability data analysis—metals handbook. Detroit, MI: ASM International, 1989.
7.
ASTME-2862. Standard practice for probability of detection analysis for hit/miss data. Technical report, ASTM International, West Conshohocken, PA, 2018.
8.
BerensAHoveyP. Flaw detection reliability criteria. Volume 1. Methods and results. Technical report, University of Dayton University Research Institute, Dayton, OH, 1984.
9.
VirkkunenIKoskinenTPapulaS, et al. Comparison of â versus a and hit/miss POD-estimation methods: a European viewpoint. J Nondestruct Eval2019; 38(4): 1–13.
10.
FalcetelliFYueNDi SanteR, et al. Probability of detection, localization, and sizing: the evolution of reliability metrics in structural health monitoring. Struct Health Monit2021; 21(6): 2990–3017.
11.
Schubert KabbanCMGreenwellBMDeSimioMP, et al. The probability of detection for structural health monitoring systems: repeated measures data. Struct Health Monit2015; 14(3): 252–264.
12.
KesslerSSDunnCTSwindellP, et al. Detection sensitivity analysis for a potential drop (PD) structural health monitoring (SHM) system. In: Proceedings of the 12th international workshop on structural health monitoring. Stanford, CA: DEStech Publications, Inc.
13.
RoachDPDelongWAWhiteS, et al. Use of composite materials, health monitoring and self-healing concepts to refurbish our civil and military infrastructure. Technical report, SAND2007-5547, Sandia National Laboratories, Albuquerque, CA, 2007.
14.
FalcetelliFCristianiDYueN, et al. Qualification of distributed optical fiber sensors using probability of detection curves for delamination in composite laminates. Struct Health Monit2023. Epub ahead of print 26 December 2022. DOI: 10.1177/14759217221141159.
15.
ThompsonRBBrascheLJLindgrenE, et al. Recent advances in model-assisted probability of detection. In: 4th European-American workshop on reliability of NDE, Berlin, Germany.
16.
HardingCHugoGBowlesS. Application of model-assisted POD using a transfer function approach. In: AIP conference proceedings, vol. 1096, pp.1792–1799. Chicago, IL: American Institute of Physics, 2009.
17.
DemeyerSJensonFDominguezN, et al. Transfer function approach based on simulation results for the determination of POD curves. In: AIP conference proceedings, Burlington, VT, USA, 17–22 July 2011, vol. 1430, pp.1757–1764.
18.
CarboniMCantiniS. A model assisted probability of detection approach for ultrasonic inspection of railway axles. In: 18th world conference on nondestructive testing, Durban, South Africa, 2012, pp.16–20.
19.
RosellAPerssonG. Model based capability assessment of an automated eddy current inspection procedure on flat surfaces. Res Nondestruct Eval2013; 24(3): 154–176.
20.
FoucherFFernandezRLeberreS, et al. New tools in CIVA for model assisted probability of detection (MAPOD) to support NDE reliability studies. In: Nondestructive evaluation of aerospace materials and structures 2018, St. Louis, MI, 05–06 June 2018, pp.32–43.
21.
FoucherFLonnèSToullelanG, et al. An overview of validation campaigns of the CIVA simulation software. In 12th European conference on non-destructive testing. Gothenburg, Sweden, 2018, pp.11–15.
22.
SpencerFW. Nonparametric POD estimation for hit/miss data: a goodness of fit comparison for parametric models. In: AIP conference proceedings, San Diego, CL, USA, 18–23 July 2010, vol. 1335, pp.1557–1564.
23.
DominguezNReboudCDuboisA, et al. A new approach of confidence in POD determination using simulation. In: AIP conference proceedings, Denver, CO, USA, 15–20 July 2012, vol. 1511, pp.1749–1756.
24.
KnoppJGrandhiRZengL, et al. Considerations for statistical analysis of nondestructive evaluation data: hit/miss analysis. E J Adv Maint2012; 4(3): 105–115.
25.
AbdessalemABJensonFCalmonP. Quantifying uncertainty in parameter estimates of ultrasonic inspection system using Bayesian computational framework. Mech Syst Signal Process2018; 109: 89–110.
26.
AldrinJCKnoppJSSabbaghHA. Bayesian methods in probability of detection estimation and model-assisted probability of detection evaluation. In: AIP conference proceedings, Denver, CO, USA, 15–20 July 2012, vol. 1511, pp.1733–1740.
27.
JensonFDominguezNWillaumeP, et al. A Bayesian approach for the determination of POD curves from empirical data merged with simulation results. In: AIP conference proceedings, Denver, CO, USA, 15–20 July 2012, vol. 1511, pp.1741–1748.
28.
HoveyPW. Using the information in field service inspections to assess the damage state of the aircraft and the detection capability of the inspection system. In: AIP conference proceedings, Chicago, IL, USA, vol. 1096, pp.1855–1861.
29.
AnnisCGandossiLMartinO. Optimal sample size for probability of detection curves. Nucl Eng Des2013; 262: 98–105.
30.
BenvenisteABassevilleMMoustakidesG. The asymptotic local approach to change detection and model validation. IEEE Trans Autom Control1987; 32(7): 583–592.
31.
DöhlerMMevelLZhangQ. Fault detection, isolation and quantification from Gaussian residuals with application to structural damage diagnosis. Ann Rev Control2016; 42: 244–256.
32.
MendlerADöhlerMVenturaCE. A reliability-based approach to determine the minimum detectable damage for statistical damage detection. Mech Syst Signal Process2021; 154: 107561.
33.
FarrarCRWordenK. Structural health monitoring: a machine learning perspective. Oxford, UK: Wiley, 2012.
34.
CasellaGBergerRL. Statistical inference. Pacific Grove, CA: Duxbury, 2002.
35.
ReyndersEPintelonRDe RoeckG. Uncertainty bounds on modal parameters obtained from stochastic subspace identification. Mech Syst Signal Process2008; 22(4): 948–969.
36.
DöhlerMMevelL. Efficient multi-order uncertainty computation for stochastic subspace identification. Mech Syst Signal Process2013; 38(2): 346–366.
37.
MellingerPDöhlerMMevelL. Variance estimation of modal parameters from output-only and input/output subspace-based system identification. J Sound Vib2016; 379: 1–27.
38.
GrèsSDöhlerMAndersenP, et al. Subspace-based Mahalanobis damage detection robust to changes in excitation covariance. Struct Control Health Monit2021; 28: e2760.
39.
BalmèsEBassevilleMBourquinF, et al. Merging sensor data from multiple temperature scenarios for vibration-based monitoring of civil structures. Struct Health Monit2008; 7(2): 129–142.
40.
MagalhãesFCunhaÁCaetanoE. Vibration based structural health monitoring of an arch bridge: from automated OMA to damage detection. Mech Syst Signal Process2012; 28: 212–228.
41.
SohnHWordenKFarrarCR. Statistical damage classification under changing environmental and operational conditions. J Intell Mat Syst Struct2002; 13(9): 561–574.
42.
MontgomeryDC. Introduction to statistical quality control. 6th ed.Jefferson City, MO: John Wiley & Sons, 2007.
43.
SchoefsFClèmentANouyA. Assessment of ROC curves for inspection of random fields. Struct Safety2009; 31(5): 409–419.
44.
PapadopoulosVGiovanisDG. Stochastic finite element methods: an introduction. 2018 Springer International (ebook).
45.
LiuWKBelytschkoTManiA. Random field finite elements. Int J Num Methods Eng1986; 23(10): 1831–1845.
46.
GhanemRGSpanosPD. Stochastic finite elements: a spectral approach. North Chelmsford, MA: Courier Corporation, 2003.
47.
AstillCImosseirSShinozukaM. Impact loading on structures with random properties. J Struct Mech1972; 1(1): 63–77.
48.
MendlerADöhlerMVenturaCE. Sensor placement with optimal damage detectability for statistical damage detection. Mech Syst Signal Process2022; 170: 108767.
49.
NelsonRB. Simplified calculation of eigenvector derivatives. AIAA J1976; 14(9): 1201–1205.
50.
HeylenWSasP. Modal analysis theory and testing. Leuven, Belgium: Katholieke Universiteit Leuven, 1997.
51.
MottersheadJELinkMFriswellMI. The sensitivity method in finite element model updating: a tutorial. Mech Syst Signal Process2011; 25(7): 2275–2296.
52.
SimoenEde RoeckGLombaertG. Dealing with uncertainty in model updating for damage assessment: a review. Mech Syst Signal Process2015; 56–57: 123–149.
