Structural Health Monitoring using deep learning with optimal finite element model generated data

Abstract Identifying damage through Structural Health Monitoring (SHM) methods is increasingly attracting attention due to multiple maintenance and failure prevention applications. In order to create reliable SHM systems for structural damage identification (DI) tasks, access to large amounts of data containing measured structural responses is usually necessary. The data acquisition is mostly based on direct experimental responses up to now and requires time consuming measurements in various working and ambient conditions of the structure. In the present work, a novel SHM method is tested where all data is solely derived from FE calculated responses, after an initial experimental cost for FE model updating on the healthy structure state. The proposed method can be especially applied in cases where specific damage types are expected or anomalies are adequately defined so they can be effectively simulated by FE models. Origin of such models may then be the healthy experimental status. To test the proposed SHM system, the optimal FE model of an experimental benchmark linear beam structure is constructed, simulating an undamaged condition. In order to check the robustness of the proposed method the damage magnitudes imposed on the benchmark are kept small and combined with random excitations. Next, the optimal FE model is used for generating labeled SHM vibration data through a repetitive load case scheme which also includes uncertainties simulation. The data derived from the optimal FE model is finally used to train a Deep Learning (DL) Convolutional Neural Network (CNN) classifier which is after experimentally validated on the benchmark structure. The optimal FE generated data proves to be able to train an accurate CNN that can predict adequately the experimental benchmark states. A comparison is also given with a CNN trained by the corresponding nominal FE model data which is found not reliable on the experimental validations. The presented combination of optimal FE and DL is a potential solution for future SHM tools and further investigation is encouraged.

[1]  Dimitrios Giagopoulos,et al.  Integrated Reverse Engineering Strategy for Large-Scale Mechanical Systems: Application to a Steam Turbine Rotor , 2018, Front. Built Environ..

[2]  Gokhan Pekcan,et al.  Vibration‐based structural condition assessment using convolution neural networks , 2018, Structural Control and Health Monitoring.

[3]  Nitish Srivastava,et al.  Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..

[4]  David P. Thambiratnam,et al.  Vibration characteristics and damage detection in a suspension bridge , 2016 .

[5]  Dmitri Tcherniak,et al.  An experimental study on the data-driven structural health monitoring of large wind turbine blades using a single accelerometer and actuator , 2019, Mechanical Systems and Signal Processing.

[6]  Khandakar M. Rashid,et al.  Times-series data augmentation and deep learning for construction equipment activity recognition , 2019, Adv. Eng. Informatics.

[7]  S. Audebert,et al.  Uncertainties in structural dynamics: overview and comparative analysis of methods , 2015 .

[8]  Joaquim Figueiras,et al.  Damage detection under environmental and operational effects using cointegration analysis – Application to experimental data from a cable-stayed bridge , 2020 .

[9]  Hao Sun,et al.  Optimal sensor placement in structural health monitoring using discrete optimization , 2015 .

[10]  Taghi M. Khoshgoftaar,et al.  A survey on Image Data Augmentation for Deep Learning , 2019, Journal of Big Data.

[11]  Seiichi Uchida,et al.  Mining the displacement of max-pooling for text recognition , 2019, Pattern Recognit..

[12]  Yan Liu,et al.  A sparsity-based stochastic pooling mechanism for deep convolutional neural networks , 2018, Neural Networks.

[13]  John Dalsgaard Sørensen,et al.  Quantification of traffic and temperature effects on the fatigue safety of a reinforced-concrete bridge deck based on monitoring data , 2019, Engineering Structures.

[14]  Hong Chang,et al.  A novel fault diagnosis technique for wind turbine gearbox , 2019, Appl. Soft Comput..

[15]  Costas Papadimitriou,et al.  Variability of updated finite element models and their predictions consistent with vibration measurements , 2012 .

[16]  Jerzy Małachowski,et al.  Finite element analysis of vehicle-bridge interaction , 2006 .

[17]  Martin T. Hagan,et al.  Neural network design , 1995 .

[18]  Gyuhae Park,et al.  Structural Health Monitoring With Autoregressive Support Vector Machines , 2009 .

[19]  Laurent Mevel,et al.  Structural health monitoring with statistical methods during progressive damage test of S101 Bridge , 2014 .

[20]  L. Cooper,et al.  When Networks Disagree: Ensemble Methods for Hybrid Neural Networks , 1992 .

[21]  Jun Li,et al.  Structural damage identification based on autoencoder neural networks and deep learning , 2018, Engineering Structures.

[22]  Dimitrios Giagopoulos,et al.  Computational framework for model updating of large scale linear and nonlinear finite element models using state of the art evolution strategy , 2017 .

[23]  Moncef Gabbouj,et al.  Real-time vibration-based structural damage detection using one-dimensional convolutional neural networks , 2017 .

[24]  E. Cross,et al.  Detection of sub-surface damage in wind turbine bearings using acoustic emissions and probabilistic modelling , 2020, Renewable Energy.

[25]  K. Krishnan Nair,et al.  Time Series Based Structural Damage Detection Algorithm Using Gaussian Mixtures , 2007 .

[26]  Costas Papadimitriou,et al.  Structural health monitoring and fatigue damage estimation using vibration measurements and finite element model updating , 2018, Structural Health Monitoring.

[27]  Yang Yu,et al.  A novel deep learning-based method for damage identification of smart building structures , 2018, Structural Health Monitoring.

[28]  Nickolas S. Sapidis,et al.  Dynamic and structural integrity analysis of a complete elevator system through a Mixed Computational-Experimental Finite Element Methodology , 2018 .

[29]  Filip Ksica,et al.  Integration and test of piezocomposite sensors for structure health monitoring in aerospace , 2019 .

[30]  Faris Elasha,et al.  Planetary bearing defect detection in a commercial helicopter main gearbox with vibration and acoustic emission , 2018 .

[31]  Charles R. Farrar,et al.  Robust structural health monitoring under environmental and operational uncertainty with switching state-space autoregressive models , 2019 .

[32]  Nazih Mechbal,et al.  A probabilistic multi-class classifier for structural health monitoring , 2015 .

[33]  Robert X. Gao,et al.  Deep learning and its applications to machine health monitoring , 2019, Mechanical Systems and Signal Processing.

[34]  Spilios D. Fassois,et al.  Statistical Time Series Methods for Vibration Based Structural Health Monitoring , 2013 .

[35]  Dmitri Tcherniak,et al.  Active vibration-based structural health monitoring system for wind turbine blade: Demonstration on an operating Vestas V27 wind turbine , 2017 .

[36]  Peter Volgyesi,et al.  Structural health monitoring of bridges with piezoelectric AE sensors , 2015 .

[37]  Michael Affenzeller,et al.  Machine learning based concept drift detection for predictive maintenance , 2019, Comput. Ind. Eng..

[38]  Claudomiro Sales,et al.  Machine learning algorithms for damage detection: Kernel-based approaches , 2016 .