Optimal sensor placement through Bayesian experimental design: effect of measurement noise and number of sensors Giovanni Capellari Eleni Chatzi Stefano Mariani 3 rd International Electronic Conference on Sensors and Aplications, 10-15 November 2016
Motivation Structural Health Monitoring can be conceptually divided in three stages: in our work, we will focus on the design of the sensor network SHM system design ๐ Data collection ๐ Parameters estimation ๐พ Decision making
Motivation Optimal The usefulness of the sensor network depends on the SHM system number, type and location of the sensors. Therefore, design we need a method to quantify the information obtained by the acquisition system. measurement error Estimates SHM system Uncertainty Identifiability cost configuration # sensors
Optimal sensor placement: deterministic methods The existing approaches does not take into account the measurement noise, i.e. the sensors accuracy. EFI EVP KE Sensitivity to damage M. Meo , G. Zumpano , (2005), M. Bruggi , S. Mariani , (2013), Leyder , C., Ntertimanis , V., Chatzi , E., Frangi , A. (2015).
Optimal sensor placement: Bayesian framework In a Bayesian sense, the optimal spatial configuration ๐ โ of the sensor network can be found by maximizing the Shannon information gain. In order to compute it, we use a Monte Carlo approximation. Expected gain in Shannon information ๐ โ = arg max ๐โ๐ฌ ๐(๐) ๐ ๐ = เถฑ เถฑ ๐ฃ ๐, ๐, ๐พ ๐ ๐พ, ๐ ๐ ๐๐พ๐๐ ๐ ฮ Monte Carlo sampling Prior: ๐พ~๐ ๐พ Likelihood: ๐~๐(๐|๐พ, ๐) ๐ ๐๐ฃ๐ข ๐ ๐๐ 1 1 ln ๐ ๐ ๐ ๐พ ๐ , ๐ ๐ ๐ ๐ ๐พ ๐ , ๐ ๐ ๐ โ เท โ ln เท ๐ ๐๐ฃ๐ข ๐ ๐๐ ๐=1 ๐=1 X. Huan , Y. M. Marzouk , (2013).
Model evaluation The measurements are related to the mechanical parameters to be estimated through a FEM-based forward model. The sensor accuracy is taken into account through a fictitious measurement noise. โข Evaluation of the likelihood ๐ ๐ ๐ ๐พ ๐ , ๐ = ๐ ๐ ๐ ๐ โ ๐ฏ ๐พ ๐ , ๐ โข Forward model ๐ = ๐ฏ ๐พ, ๐ + ๐ Measurement noise ๐ฏ ๐พ, ๐ = ๐ด ๐ ๐ณ(๐พ) โ1 ๐ฎ ๐ ๐พ ๐น ๐ ๐ณ ๐พ = เท ๐น โ 1 ๐ณ ๐ฃ๐๐ โ ๐ณ ๐ Observation matrix ๐=1
Optimization In order to reduce the computational cost of the forward model, a cheaper surrogate model is built. โข Surrogate model: polynomial chaos expansion ๐พ ๐ ~๐ ๐พ , ๐ ๐ ~๐ฑ ๐ ๐ ๐ ๐ ๐ ๐ ๐ = ๐พ ๐ ๐ ๐๐ท๐น = ๐ ๐ = ฯ ๐ทโโ ๐ ๐ง ๐ฝ ๐ ๐ฝ ๐ ๐ ๐ ๐บ๐น = ๐ฏ ๐พ ๐ , ๐ ๐ โข Optimization: Covariance Matrix Adaptation Evolution Strategy (CMA-ES) ๐ โ โ ๐ ๐ , ๐ซ โ โ ๐ ๐ร ๐ ๐ 1 . ๐ ๐ ~๐ + ๐๐ช ๐ ๐, ๐ซ 2. ๐ and ๐ซ are updated through cumulation 3. Check the tolerance on ๐ ๐ N. Hansen , S.D. Mรผller , P. Koumoutsakos , (2003).
Bayesian OSP framework Training surrogate model Maximizing information Sample input variables Sample design variable ๐พ ๐ ~๐ ๐พ , ๐ ๐ ~๐ฑ ๐ ๐ ๐ ๐ ๐ ๐ ๐ ๐ ๐ = ๐พ ๐ MC approximation ๐(๐ ๐ ) System response ๐ฏ ๐บ๐น ๐พ ๐ , ๐ ๐ Update ๐ ๐ โ ๐ ๐+1 (CMA-ES) PCE surrogate ๐ฏ ๐บ๐น ๐พ ๐ , ๐ ๐ โ ๐ฏ ๐๐ท๐น ๐พ ๐ , ๐ ๐ Check tolerance on ๐ ๐ ๐ โ ๐ ๐ ๐+1 Optimal configuration ๐ โ
Application: simply supported plate 10x10 mesh: 726 d.o.f. Displacement measurements 4 zones: ๐พ = ๐น 1 , ๐น 2 , ๐น 3 , ๐น 4
Application: simply supported plate Choice of prior distribution ๐ ๐พ Optimal position of ๐ ๐ก = 4 sensors, results of 10 algorithm runs ๐พ = ๐น 1 ๐น 2 ๐น 3 ๐น 4 ๐ ๐ก : # sensors ๐ ๐๐ท๐น : # PCE samples ๐ ๐ก = 4, ๐ ๐๐ท๐น = 10 4 , ๐ : PCE polynomial degree ๐ = 10 , ๐ ๐๐ท = 5 ยท 10 3 ๐ ๐๐ท : # MC samples ๐ ๐พ ~๐ฑ 2 ๐น ๐ ๐พ ~๐ฑ 0, ๐น 3 , ๐น
Application: simply supported plate Effect of ฯ ๐ Contour of the objective function with one sensor for each possible location on the plate with different standard deviations of the measurement noise. 2 ๐~๐ช 0, ๐ ๐ ๐ ๐ก : # sensors ๐พ = ๐น 2 ๐ ๐๐ท๐น : # PCE samples ๐ ๐ก = 1, ๐ ๐๐ท๐น = 10 4 , ๐ : PCE polynomial degree ๐ = 10 , ๐ ๐๐ท = 5 ยท 10 3 ๐ ๐๐ท : # MC samples 0.6936 2.9 0.697 0.6935 0.6965 0.6934 2.8 0.696 0.6933 2.7 0.6932 0.6955 0.6931 0.695 2.6 0.693 0.6945 0.6929 2.5 0.694 0.6928 0.6935 2.4 0.6927 0.693 0.6926 ฯ ๐ = 10 โ3 m ฯ ๐ = 10 โ4 m ฯ ๐ = 10 โ5 m
Application: simply supported plate Effect of ฯ ๐ and number of sensors Contour of the objective function with one sensor for different standard deviations and number of sensors. 2 ๐~๐ช 0, ๐ ๐ ๐ ๐ก : # sensors ๐พ = ๐น 2 ๐ ๐๐ท๐น : # PCE samples ๐ ๐๐ท๐น = 10 4 , ๐ : PCE polynomial degree ๐ = 10 , ๐ ๐๐ท = 5 ยท 10 3 ๐ ๐๐ท : # MC samples
Conclusions โข Optimal sensor placement and SHM system design โข Take into account: Bayesian optimal experimental design - Measurements uncertainties - Number of sensors โข Maximization of expected information gain between prior and posterior โข Use of surrogate model (PCE) for MC approximation and stochastic optimization (CMA-ES) methods for computational speed-up โข Future developments: larger number of sensors, larger number of parameters, application to complex cases
References Bruggi , M., and Mariani , S. (2013). โOptimization of sensor placement to detect damage in flexible plates.โ Engineering Optimization , 45(6), 659 โ 676. Capellari , G., Eftekhar Azam , S., Mariani , S. (2016). โTowards real -time health monitoring of structural systems via recursive Bayesian filtering and reduced order modelling.โ International Journal of Sustainable Materials and Structural Systems , In Press. Hansen , N., Mรผller , S. D., Koumoutsakos , P. (2003). โReducing the time complexity of the derandomized evolution strategy with Covariance Matrix Adaptation (CMA- ES).โ Evolutionary Computation , 11(1), 1-18. Huan , X., and Marzouk , Y. M. (2013). โSimulation -based optimal Bayesian experimental design for nonlinear systems.โ Journal of Computational Physics , 232(1), 288 โ 317. Leyder , C., Ntertimanis , V., Chatzi , E., Frangi , A. (2015). โOptimal sensors placement for the modal identification of an innovative timber structure.โ Proceedings of the 1 st International Conference on Uncertainty Quantification in Computational Sciences and Engineering , 467-476. Lindley , D. V. (1972). Bayesian Statistics, A Review , Society for Industrial and Applied Mathematics, SIAM. Marelli , S., and Sudret , B. (2015). UQLab User Manual , Chair of Risk, Safety & Uncertainty Quantification, ETH Zรผrich. Capellari , G., Chatzi , C., Mariani , S. (2016). An optimal sensor placement method for SHM based on Bayesian experimental design and Polynomial Chaos Expansion Proceedings of the VII European Congress on Computational Methods in Applied Sciences and Engineering . Meo , M., and Zumpano , G. (2005). โOn the optimal sensor placement techniques for a bridge structure.โ Engineering Structures , 27, 1488-1497. Ryan , K. J. (2003). โEstimating expected information gains for experimental designs with application to the random fatigue-limit model .โ Journal of Computational and Graphical Statistics , 12(3), 585-603. Papadimitriou , C., (2004). โOptimal sensor placement methodology for parametric identification of structural systems.โ Journal of Sound and Vibration, 278 (4), 923-947.
Recommend
More recommend