• Smart Structures and Systems
• /
• v.14 no.3
• /
• pp.377-395
• /
• 2014

Vibration-based fault detection and condition monitoring of rotating machinery, using statistical process control (SPC) combined with statistical pattern recognition methodology, has been widely investigated by many researchers. In particular, the discrete wavelet transform (DWT) is considered as a powerful tool for feature extraction in detecting fault on rotating machinery. Although DWT significantly reduces the dimensionality of the data, the number of retained wavelet features can still be significantly large. Then, the use of standard multivariate SPC techniques is not advised, because the sample covariance matrix is likely to be singular, so that the common multivariate statistics cannot be calculated. Even though many feature-based SPC methods have been introduced to tackle this deficiency, most methods require a parametric distributional assumption that restricts their feasibility to specific problems of process control, and thus limit their application. This study proposes a nonparametric multivariate control chart method, based on multiscale wavelet scalogram (MWS) features, that overcomes the limitation posed by the parametric assumption in existing SPC methods. The presented approach takes advantage of multi-resolution analysis using DWT, and obtains MWS features with significantly low dimensionality. We calculate Hotelling's $T^2$-type monitoring statistic using MWS, which has enough damage-discrimination ability. A bootstrap approach is used to determine the upper control limit of the monitoring statistic, without any distributional assumption. Numerical simulations demonstrate the performance of the proposed control charting method, under various damage-level scenarios for a bearing system.

• Journal of the Korean Society of Hazard Mitigation
• /
• v.9 no.5
• /
• pp.9-13
• /
• 2009

Modal parameters including resonant frequencies and mode shapes are heavily utililized in most damage identification throries for structural health monitoring. However, extracting modal parameters from dynamic responses needs postprocessing which inevitably involves errors in curve-fitting resonants as well as transforming the domain of responses. In this paper, the applicability of a damage identification method based on free vibration responses to the in-sevice responses is experimentally verified. The experiment is performed via applying periodic and nonperiodic moving loads to a simply supported beam and displacement responses are measured. The moving load is simulated using steel balls and a downhill device. The damage identification results show that the in-service response may be applicable to identifying damage in the beam.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.28 no.1A
• /
• pp.135-146
• /
• 2008

In this paper, a hybrid damage monitoring scheme for prestressed concrete (PSC) girder bridges by using sequential acceleration and impedance signatures is newly proposed. Damage types of interest include prestress-loss in tendon and flexural stiffness-loss in a concrete girder. The hybrid scheme mainly consists of three sequential phases: damage alarming, damage classification, and damage estimation. In the first phase, the global occurrence of damage is alarmed by monitoring changes in acceleration features. In the second phase, the type of damage is classified into either prestress-loss or flexural stiffness-loss by recognizing patterns of impedance features. In the third phase, the location and the extent of damage are estimated by using two different ways: a mode shape-based damage detection to detect flexural stiffness-loss and a natural frequency-based prestress prediction to identify prestress-loss. The feasibility of the proposed scheme is evaluated on a laboratory-scaled PSC girder model for which hybrid vibration-impedance signatures were measured for several damage scenarios of prestress-loss and flexural stiffness-loss.

• Structural Engineering and Mechanics
• /
• v.77 no.1
• /
• pp.47-56
• /
• 2021

This paper deals with damage detection using a Gapped Smoothing Method (GSM) combined with deep learning. Convolutional Neural Network (CNN) is a model of deep learning. CNN has an input layer, an output layer, and a number of hidden layers that consist of convolutional layers. The input layer is a tensor with shape (number of images) × (image width) × (image height) × (image depth). An activation function is applied each time to this tensor passing through a hidden layer and the last layer is the fully connected layer. After the fully connected layer, the output layer, which is the final layer, is predicted by CNN. In this paper, a complete machine learning system is introduced. The training data was taken from a Finite Element (FE) model. The input images are the contour plots of curvature gapped smooth damage index. A free-free beam is used as a case study. In the first step, the FE model of the beam was used to generate data. The collected data were then divided into two parts, i.e. 70% for training and 30% for validation. In the second step, the proposed CNN was trained using training data and then validated using available data. Furthermore, a vibration experiment on steel damaged beam in free-free support condition was carried out in the laboratory to test the method. A total number of 15 accelerometers were set up to measure the mode shapes and calculate the curvature gapped smooth of the damaged beam. Two scenarios were introduced with different severities of the damage. The results showed that the trained CNN was successful in detecting the location as well as the severity of the damage in the experimental damaged beam.

• Proceedings of the KIEE Conference
• /
• 2005.05a
• /
• pp.243-245
• /
• 2005

Condition monitoring is desirable for increasing machinery availability, reducing consequential damage, and improving operational efficiency. In this paper, a model-based method using neural network modeling of induction noter in vibration spectra is proposed for machine fault detection and diagnosis. The short-time Fourier transform (STFT) is used to process the quasi-steady vibration signals to continuous spectra so that the neural network model can be trained with vibration spectra. And the faults are detected from changes in the expectation of vibration spectra modeling error. The effectiveness of the proposed method is demonstrated through experimental results.

• Structural Engineering and Mechanics
• /
• v.54 no.1
• /
• pp.105-119
• /
• 2015

Structural damage detection using modal strain energy (MSE) is one of the most efficient and reliable structural health monitoring techniques. However, some of the existing MSE methods have been validated for special types of structures such as beams or steel truss bridges which demands improving the available methods. The purpose of this study is to improve an efficient modal strain energy method to detect and quantify the damage in complex structures at early stage of formation. In this paper, a modal strain energy method was mathematically developed and then numerically applied to a fixed-end beam and a three-story frame including single and multiple damage scenarios in absence and presence of up to five per cent noise. For each damage scenario, all mode shapes and natural frequencies of intact structures and the first five mode shapes of assumed damaged structures were obtained using STRAND7. The derived mode shapes of each intact and damaged structure at any damage scenario were then separately used in the improved formulation using MATLAB to detect the location and quantify the severity of damage as compared to those obtained from previous method. It was found that the improved method is more accurate, efficient and convergent than its predecessors. The outcomes of this study can be safely and inexpensively used for structural health monitoring to minimize the loss of lives and property by identifying the unforeseen structural damages.

• Smart Structures and Systems
• /
• v.12 no.3_4
• /
• pp.441-463
• /
• 2013

In this paper, a simplified planar model is developed for damage estimation of interlocked caisson systems. Firstly, a conceptual dynamic model of the interlocked caisson system is designed on the basis of the characteristics of existing harbor caisson structures. A mass-spring-dashpot model allowing only the sway motion is formulated. To represent the condition of interlocking mechanisms, each caisson unit is connected to adjacent ones via springs and dashpots. Secondly, the accuracy of the planar model's vibration analysis is numerically evaluated on a 3-D FE model of the interlocked caisson system. Finally, the simplified planar model is employed for damage estimation in the interlocked caisson system. For localizing damaged caissons, a damage detection method based on modal strain energy is formulated for the caisson system.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.22 no.10
• /
• pp.1010-1019
• /
• 2012

Non-contact detection of internal defects in a rail using ultrasonic waves is discussed in this paper. Cracks in a rail may be the cause of a serious railway accident such as derailment if left undetected. Concurrent rail inspection method based on ultrasonic testing using piezoelectric transducers has several limitations as it should keep physical contact with the rail. This work suggests a non-contact detection of internal defects in a rail using ElectroMagnetic Acoustic Transducers (EMAT) which can produce and measure ultrasonic waves in a rail without any couplant. The EMATs for rail inspection are designed and fabricated and their working performance is verified through a series of experiments on various types of internal defects in test rails. The effect of lift-off between the transducers and the rail on the generated signals is also discussed.

• Structural Monitoring and Maintenance
• /
• v.6 no.2
• /
• pp.125-145
• /
• 2019

Bridge scour is one of the predominant causes of bridge failure. Current climate deterioration leads to increase of flooding frequency and severity and thus poses a higher risk of bridge scour failure than before. Recent studies have explored extensively the vibration-based scour monitoring technique by analyzing the structural modal properties before and after damage. However, the state-of-art of this area lacks a systematic approach with sufficient robustness and credibility for practical decision making. This paper attempts to develop a data-driven methodology for bridge scour monitoring using support vector machines. This study extracts features from the bridge dynamic responses based on a generic sensitivity study on the bridge's modal properties and selects the features that are significantly contributive to bridge scour detection. Results indicate that the proposed data-driven method can quantify the bridge scour damage with satisfactory accuracy for most cases. This paper provides an alternative methodology for bridge scour evaluation using the machine learning method. It has the potential to be practically applied for bridge safety assessment in case that scour happens.

• Smart Structures and Systems
• /
• v.6 no.9
• /
• pp.1057-1077
• /
• 2010

An early detection of structural damages is critical for the decision making of repair and replacement maintenance in order to guarantee a specified structural reliability. Consequently, the structural damage detection, based on vibration data measured from the structural health monitoring (SHM) system, has received considerable attention recently. The traditional time-domain analysis techniques, such as the least square estimation (LSE) method and the extended Kalman filter (EKF) approach, require that all the external excitations (inputs) be available, which may not be the case for some SHM systems. Recently, these two approaches have been extended to cover the general case where some of the external excitations (inputs) are not measured, referred to as the adaptive LSE with unknown inputs (ALSE-UI) and the adaptive EKF with unknown inputs (AEKF-UI). Also, new analysis methods, referred to as the adaptive sequential non-linear least-square estimation with unknown inputs and unknown outputs (ASNLSE-UI-UO) and the adaptive quadratic sum-squares error with unknown inputs (AQSSE-UI), have been proposed for the damage tracking of structures when some of the acceleration responses are not measured and the external excitations are not available. In this paper, these newly proposed analysis methods will be compared in terms of accuracy, convergence and efficiency, for damage identification of structures based on experimental data obtained through a series of laboratory tests using a scaled 3-story building model with white noise excitations. The capability of the ALSE-UI, AEKF-UI, ASNLSE-UI-UO and AQSSE-UI approaches in tracking the structural damages will be demonstrated and compared.