Abstract
We present a model-order-reduction approach to simulation-based classification, with particular application to structural health monitoring. The approach exploits (1) synthetic results obtained by repeated solution of a parametrized mathematical model for different values of the parameters, (2) machine-learning algorithms to generate a classifier that monitors the damage state of the system, and (3) a reduced basis method to reduce the computational burden associated with the model evaluations. Furthermore, we propose a mathematical formulation which integrates the partial differential equation model within the classification framework and clarifies the influence of model error on classification performance. We illustrate our approach and we demonstrate its effectiveness through the vehicle of a particular physical companion experiment, a harmonically excited microtruss.
Similar content being viewed by others
Notes
Amplitudes and phases are estimated using the Matlab function fit [42], which relies on Levenberg–Marquardt algorithm.
See, e.g., [43, Chap. 3.6.2] for the Poisson’s ratio and the webpage http://pubchem.ncbi.nlm.nih.gov for the density.
We recall that standardization of data implies that we train the classifier based on the modified features \({\hat{z}}_q = \frac{ z_q - m_q^{\mathrm{bk}} }{ std_q^{\mathrm{bk}} }\) where \(m_q^{\mathrm{bk}}\), \(std_q^{\mathrm{bk}}\) are respectively the sample mean and the sample standard deviation of the training set for all classes.
We consider here the case of static data to not deal with the dependence on frequency.
References
Farrar CR, Worden K (2012) Structural health monitoring: a machine learning perspective. Wiley, New York
Amir N, Barzelay O, Yefet A, Pechter T (2010) Condenser tube examination using acoustic pulse reflectometry. J Eng Gas Turbines Power 132(1):014501
Sharp D, Campbell D (1997) Leak detection in pipes using acoustic pulse reflectometry. Acta Acust United Acust 83(3):560–566
Dawson B (1976) Vibration condition monitoring techniques for rotating machinery. Shock Vib Dig 8(12):3
Staszewski W, Boller C, Tomlinson GR (2004) Health monitoring of aerospace structures: smart sensor technologies and signal processing. Wiley, New York
Willis S (2009) Olm: a hands-on approach. In: ICAF 2009, Bridging the gap between theory and operational practice, pp 1199–1214. Springer
Benedettini O, Baines T, Lightfoot H, Greenough R (2009) State-of-the-art in integrated vehicle health management. Proc Inst Mech Eng Part G J Aerosp Eng 223(2):157–170
Ortiz EM, Clark GJ, Babbar A, Vian JL, Syrmos VL, Arita MM (2008) Multi source data integration for aircraft health management. In: Aerospace conference, 2008 IEEE, pp 1–12. IEEE
Rytter A (1993) Vibration based inspection of civil engineering structures. Ph.D. thesis, Ph.D. dissertation
Worden K, Dulieu-Barton J (2004) Damage identification in systems and structures. Int J Struct Health Monit 3:85–98
Farrar CR, Doebling SW, Nix DA (2001) Vibration-based structural damage identification. Philos Trans R Soc Lond A Math Phys Eng Sci 359(1778):131–149
Hurtado JE (2004) An examination of methods for approximating implicit limit state functions from the viewpoint of statistical learning theory. Struct Saf 26(3):271–293
Sanayei M, Onipede O (1991) Damage assessment of structures using static test data. AIAA J 29(7):1174–1179
Sanayei M, Saletnik MJ (1996) Parameter estimation of structures from static strain measurements. i: Formulation. J Struct Eng 122(5):555–562
Sanayei M, Saletnik MJ (1996) Parameter estimation of structures from static strain measurements. ii: Error sensitivity analysis. J Struct Eng 122(5):563–572
Cawley P, Adams R (1979) The location of defects in structures from measurements of natural frequencies. J Strain Anal Eng Des 14(2):49–57
Salawu O (1997) Detection of structural damage through changes in frequency: a review. Eng Struct 19(9):718–723
Wenzel H (2008) Health monitoring of bridges. Wiley, New York
Fan W, Qiao P (2011) Vibration-based damage identification methods: a review and comparative study. Struct Health Monit 10(1):83–111
Au SK, Zhang FL, Ni YC (2013) Bayesian operational modal analysis: theory, computation, practice. Comput Struct 126:3–14
Zang C, Imregun M (2001) Structural damage detection using artificial neural networks and measured FRF data reduced via principal component projection. J Sound Vib 242(5):813–827
Basudhar A, Missoum S (2009) A sampling-based approach for probabilistic design with random fields. Comput Meth Appl Mech Eng 198(47):3647–3655
Mainini L, Willcox K (2015) Surrogate modeling approach to support real-time structural assessment and decision making. AIAA J 53(6):1612–1626
Friswell M (1995) Finite element model updating in structural dynamics, vol 38. Springer Science & Business Media, Berlin
Hurtado JE, Alvarez DA (2003) Classification approach for reliability analysis with stochastic finite-element modeling. J Struct Eng 129(8):1141–1149
Lecerf M, Allaire D, Willcox K (2015) Methodology for dynamic data-driven online flight capability estimation. AIAA J 53(10):3073–3087
Chandola V, Banerjee A, Kumar V (2009) Anomaly detection: a survey. ACM Comput Surv (CSUR) 41(3):15
Helton JC (1993) Uncertainty and sensitivity analysis techniques for use in performance assessment for radioactive waste disposal. Reliab Eng Syst Saf 42(2):327–367
Mourelatos ZP, Kuczera R, Latcha M (2006) An efficient monte carlo reliability analysis using global and local metamodels. In: Proceedings of 11th AIAA/ISSMO multidisciplinary analysis and optimization conference, September, Portsmouth, VA
Basudhar A, Missoum S (2008) Adaptive explicit decision functions for probabilistic design and optimization using support vector machines. Comput Struct 86(19):1904–1917
Basudhar A, Missoum S (2010) An improved adaptive sampling scheme for the construction of explicit boundaries. Struct Multidiscip Optim 42(4):517–529
Basudhar A, Missoum S, Sanchez AH (2008) Limit state function identification using support vector machines for discontinuous responses and disjoint failure domains. Probab Eng Mech 23(1):1–11
Hesthaven JS, Rozza G, Stamm B (2015) Certified reduced basis methods for parametrized partial differential equations. SpringerBriefs in Mathematics, Bilbao
Quarteroni A, Manzoni A, Negri F (2015) Reduced basis methods for partial differential equations: an introduction, vol 92. Springer, Berlin
Rozza G, Huynh DP, Patera AT (2008) Reduced basis approximation and a posteriori error estimation for affinely parametrized elliptic coercive partial differential equations. Arch Comput Methods Eng 15(3):229–275
Ben-Tal A, El Ghaoui L, Nemirovski A (2009) Robust optimization. Princeton University Press, Princeton
Bertsimas D, Brown DB, Caramanis C (2011) Theory and applications of robust optimization. SIAM Rev 53(3):464–501
Bertsimas D, Sim M (2004) The price of robustness. Oper Res 52(1):35–53
Hinrichsen D, Pritchard AJ (1986) Stability radii of linear systems. Syst Control Lett 7(1):1–10
Zlobec S (1988) Characterizing optimality in mathematical programming models. Acta Appl Math 12(2):113–180
Young WC, Budynas RG (2002) Roark’s formulas for stress and strain, vol 7. McGraw-Hill, New York
MATLAB: version 8.5 (R2015a). The MathWorks Inc., Natick, Massachusetts (2015)
Harris HG, Sabnis G (1999) Structural modeling and experimental techniques. CRC Press, Boca Raton
Prudhomme S, Oden JT (1999) On goal-oriented error estimation for elliptic problems: application to the control of pointwise errors. Comput Meth Appl Mech Eng 176(1):313–331
Rubinstein RY, Kroese DP (2011) Simulation and the Monte Carlo method, vol 707. Wiley, New York
Fishman G (2013) Monte Carlo: concepts, algorithms, and applications. Springer Science & Business Media, Berlin
Hastie T, Tibshirani R, Friedman J (2009) The elements of statistical learning, vol 2. Springer, Berlin
James G, Witten D, Hastie T, Tibshirani R (2013) An introduction to statistical learning. Springer, Berlin
Murphy KP (2012) Machine learning: a probabilistic perspective. MIT Press, Cambridge
Vapnik VN (1998) Statistical learning theory, vol 1. Wiley, New York
Porsching T (1985) Estimation of the error in the reduced basis method solution of nonlinear equations. Math Comput 45(172):487–496
Binev P, Cohen A, Dahmen W, DeVore R, Petrova G, Wojtaszczyk P (2011) Convergence rates for greedy algorithms in reduced basis methods. SIAM J Math Anal 43(3):1457–1472
Cohen A, DeVore R (2015) Approximation of high-dimensional parametric pdes. Acta Numer 24:1–159
Maday Y, Patera AT, Rovas DV (2001) A blackbox reduced-basis output bound method for noncoercive linear problems. In: Cioranescu D, Lions JL (eds) Studies in mathematics and its applications. Elsevier Science B.V., New York, pp 533–569
Casenave F, Ern A, Lelièvre T (2014) Accurate and online-efficient evaluation of the a posteriori error bound in the reduced basis method. ESAIM Math Model Numer Anal 48(01):207–229
Cortes C, Vapnik V (1995) Support-vector networks. Mach Learn 20(3):273–297
Cristianini N, Shawe-Taylor J (2000) An introduction to support vector machines and other kernel-based learning methods. Cambridge University Press, Cambridge
Rifkin R, Klautau A (2004) In defense of one-vs-all classification. J Mach Learn Res 5:101–141
Breiman L, Friedman J, Stone CJ, Olshen RA (1984) Classification and regression trees. CRC Press, Boca Raton
Bishop CP (1995) Neural networks for pattern recognition. Oxford university press, Oxford
Tsymbal A (2004) The problem of concept drift: definitions and related work. Computer Science Department, Trinity College Dublin, vol 106, p 2
Widmer G, Kubat M (1996) Learning in the presence of concept drift and hidden contexts. Mach Learn 23(1):69–101
Moreno-Torres JG, Raeder T, Alaiz-RodríGuez R, Chawla NV, Herrera F (2012) A unifying view on dataset shift in classification. Pattern Recognit 45(1):521–530
Steinwart I (2005) Consistency of support vector machines and other regularized kernel classifiers. IEEE Trans Inf Theory 51(1):128–142
Sniedovich M (2010) A bird’s view of info-gap decision theory. J Risk Financ 11(3):268–283
Wald A (1945) Statistical decision functions which minimize the maximum risk. Ann Math 46(2):265–280
Franceschini G, Macchietto S (2008) Model-based design of experiments for parameter precision: state of the art. Chem Eng Sci 63(19):4846–4872
Pukelsheim F (1993) Optimal design of experiments, vol 50. SIAM, Philadelphia
Joshi S, Boyd S (2009) Sensor selection via convex optimization. IEEE Trans Signal Process 57(2):451–462
Boyd S, Vandenberghe L (2004) Convex optimization. Cambridge University Press, Cambridge
Eftang JL, Patera AT (2013) Port reduction in parametrized component static condensation: approximation and a posteriori error estimation. Int J Numer Methods Eng 96(5):269–302
Huynh DBP, Knezevic DJ, Patera AT (2013) A static condensation reduced basis element method: approximation and a posteriori error estimation. ESAIM Math Model Numer Anal 47(01):213–251
Smetana K (2015) A new certification framework for the port reduced static condensation reduced basis element method. Comput Meth Appl Mech Eng 283:352–383
Devroye L, Györfi L, Lugosi G (2013) A probabilistic theory of pattern recognition, vol 31. Springer Science & Business Media, Berlin
Acknowledgments
The authors thank Prof. Bernard Haasdonk (University of Stuttgart) for fruitful discussions. This work was supported by OSD/AFOSR/MURI Grant FA9550-09-1-0613, ONR Grant N00014-11-1-0713, and the MIT-Singapore International Design Center.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Ethical Standard
This work was supported by OSD/AFOSR/MURI Grant FA9550-09-1-0613, ONR Grant N00014-11-1-0713, and the MIT-Singapore International Design Center.
Conflict of interest
The authors declare that they have no conflict of interest.
Appendices
Appendix 1: Existence and Uniqueness of the Solutions to the Monitoring Problems
In this appendix, we discuss existence and uniqueness of solutions to the monitoring problems and indeed we develop an explicit—through not readily evaluated—expression for \(g^\mathrm{opt,bk}\). For the sake of clarity, we recap the definition of \(P_{w^{\mathrm{bk}}}\) in (28),
and of \(P_{(\mu ,\xi )}^{\mathrm{exp}}\) in (44),
We assume here that \(P_{w^{\mathrm{bk}}}\) and \(P_{(\mu ,\xi )}^{\mathrm{exp}}\) are Borel-measurable, that is they are defined on all open sets of \({\mathcal {P}}^{\mathrm{bk}}\) and \({\mathcal {P}}^{\mathrm{exp}}\), respectively. A sufficient condition for which \(P_{w^{\mathrm{bk}}}\) and \(P_{(\mu ,\xi )}^{\mathrm{exp}}\) are Borel-measurable is that \(w^{\mathrm{bk}} \in L^1({\mathcal {P}}^{\mathrm{bk}})\) and \(p_{\xi } \in L^1({\mathcal {V}})\). We further recall the random pair
with probability distribution \(P_{(Z^{\mathrm{bk}},Y)}\), and
with probability distribution \(P_{(Z^{\mathrm{exp}},Y)}\).
Next Lemma shows that if \({\mathbf {z}}^{\mathrm{bk}}\) and \({\mathbf {z}}^{\mathrm{exp}}\) are continuous, \(f^{\mathrm{dam}}\) is Borel-measurable, and \(P_{w^{\mathrm{bk}}}\) and \(P_{(\mu ,\xi )}^{\mathrm{exp}}\) are Borel-measurable, then \(P_{(Z^{\mathrm{bk}},Y)}\) and \(P_{(Z^{\mathrm{exp}},Y)}\) are also Borel-measurable.
Lemma 1
Suppose that
-
1.
the probability measures \(P_{w^{\mathrm{bk}}}\) in (28) and \(P_{(\mu ,\xi )}^{\mathrm{exp}}\) in (44) are Borel-measurable;
-
2.
the bk features \({\mathbf {z}}^{\mathrm{bk}}: {\mathcal {P}}^{\mathrm{bk}} \rightarrow {\mathbb {R}}^Q\) and the experimental features \({\mathbf {z}}^{\mathrm{exp}}: {\mathcal {P}}^{\mathrm{exp}} \rightarrow {\mathbb {R}}^Q\) are continuous;
-
3.
the discrete function \(f^{\mathrm{dam}}: {\mathcal {P}}^{\mathrm{bk}} \rightarrow \{1,\ldots ,K\}\) is Borel-measurable.
Then, the probability measures \(P_{(Z^{\mathrm{bk}},Y)}\) and \(P_{(Z^{\mathrm{exp}},Y)}\) are Borel-measurable on \({\mathbb {R}}^Q \times \{1,\ldots ,K\}\).
Proof
We only prove that \(P_{(Z^{\mathrm{bk}},Y)}\) is Borel-measurable. The proof of the measurability of \(P_{(Z^{\mathrm{exp}},Y)}\) is analogous. Let \(A \subset {\mathbb {R}}^Q\) be an open set and let \(k \in \{1,\ldots ,K\}\). We must show that \(P_{(Z^{\mathrm{bk}},Y)}( A \times \{ k \} )\) is well-defined. By construction, we have that
As a result, we must show that
is a Borel set. Since \({\mathbf {z}}^{\mathrm{bk}}\) is continuous and A is open, \(\left( {\mathbf {z}}^{\mathrm{bk}} \right) ^{-1} (A)\) is also open. On the other hand, by assumption, \(\left( f^{\mathrm{dam}} \right) ^{-1} (k)\) is Borel. Thesis follows by recalling that the intersection of Borel sets is also Borel. \(\square\)
Lemma 1 can be used to prove the following important result related to the existence and uniqueness of solutions to (20) and to (19). We state here only the result for (20). An analogous discussion applies also to the monitoring problem (19).
Proposition 2
The optimal solution \(g^\mathrm{opt, bk}\) to (20) is given by
Here, \(P_{(Z^{\mathrm{bk}},Y)} ( Y= k | \, Z^{\mathrm{bk}} = {\mathbf {z}} )\) denotes the conditional probability of the event \(\{Y= k\}\) given \(\{ Z^{\mathrm{bk}} = {\mathbf {z}} \}\).
Furthermore, if there exists \(\epsilon >0\) such that
then any solution g to (20) satisfies \(g^\mathrm{opt, bk}({\mathbf {z}}) = g({\mathbf {z}})\) for \(P_{Z^{\mathrm{bk}}}\)-almost every \({\mathbf {z}} \in {\mathbb {R}}^Q\).
Proof
We first show that (55) is measurable. Recalling [74, Theorem A.24 page 586], since \(P_{(Z^{\mathrm{bk}},Y)}\) is Borel-measurable, if we denote by \(P_{Z^{\mathrm{bk}}}\) the corresponding marginal distribution, there exists for \(P_{Z^{\mathrm{bk}}}\)—almost-every \({\mathbf {z}} \in {\mathbb {R}}^Q\) and for \(k=1,\ldots ,K\) a measurable function \(\xi _k^{\mathrm{bk}} : {\mathbb {R}}^Q \rightarrow {\mathbb {R}}\) such that \(\xi _k^{\mathrm{bk}}({\mathbf {z}}) = P_{(Z^{\mathrm{bk}},Y)}(Y=k | Z^{\mathrm{bk}}={\mathbf {z}})\). Recalling that the pointwise maximum of measurable functions is also measurable, this implies that, under the hypotheses of Lemma 1, the function \(g^\mathrm{opt, bk}\) in (55) is measurable.
We now observe that
Since \(g^\mathrm{opt,bk}\) is measurable, this implies that \(g^\mathrm{opt,bk}\) is a solution to (20).
Let g be a classifier such that \(R^{\mathrm{bk}}( g^\mathrm{opt,bk} ) = R^{\mathrm{bk}}( g)\). Using the same reasoning as before, it is possible to verify that
Recalling (56), we find
Then, we must have
which implies that \(g({\mathbf {z}}) =g^\mathrm{opt,bk} ({\mathbf {z}})\) for \(P_{Z^{\mathrm{bk}}}\)-almost every \({\mathbf {z}} \in {\mathbb {R}}^Q\). \(\square\)
Appendix 2: Parametric-Affine Expansion for the Microtruss Problem
Below, we report the parameter-dependent coefficients \(\{ {\varTheta }_q\}_{q=1}^{10}\) and the parameter-independent bilinear forms \(\{ a^q\}_{q=1}^{10}\) associated with the microtruss problem considered in this work.
and
Rights and permissions
About this article
Cite this article
Taddei, T., Penn, J.D., Yano, M. et al. Simulation-Based Classification; a Model-Order-Reduction Approach for Structural Health Monitoring. Arch Computat Methods Eng 25, 23–45 (2018). https://doi.org/10.1007/s11831-016-9185-0
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11831-016-9185-0