Vibration‐based structural condition assessment using convolution neural networks

Khodabandehlou, Hamid; Pekcan, Gökhan; Fadali, M. Sami

doi:10.1002/stc.2308

Cited by 103 publications

(82 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Thus, CNNs, which are suitable for handling a large volume of data with a sequence characteristic in the input layer, are also appropriate to handle time-history measurement signals such as dynamic structural responses. Recently, they are being applied in studies on dynamic structural responses, such as investigations of damage detection in bridge structures, 25 seismic damage identification, 26 and dynamic response estimation. 27 In previous studies on seismic performance evaluation in building structures, time-history measurement signals were manually processed and handled in the input layer of a conventional NN.…”

Section: Introductionmentioning

confidence: 99%

Seismic response prediction method for building structures using convolutional neural network

Park

2020

Struct Control Health Monit

View full text Add to dashboard Cite

Summary In this study, a method of predicting the seismic responses of building structures based on a convolutional neural network (CNN) is proposed. In the method, the time histories of acceleration responses previously measured in a building during earthquakes are used in the CNN input layer, with the corresponding time histories of the displacement responses being used in the CNN output layer. The correlations between the features automatically extracted from the acceleration responses by the convolution and pooling operations in the various CNN layers and the displacement responses in the CNN output layer are established through CNN training. The trained CNN predicts the displacement responses for a future earthquake event using only the measured acceleration responses. To verify the validity of the proposed method, a numerical study on the ASCE benchmark model and an experimental study on a reinforced concrete frame structure are conducted. In the numerical study, the structural responses of the ASCE model subject to artificial earthquakes are utilized for CNN training, and the performance of the trained CNN is verified through seismic response prediction. In the experimental study, shaking table tests are performed for various earthquakes to measure the seismic responses of the reinforced concrete test structure. The seismic response prediction performance of the proposed method is examined using CNN trained with the measured seismic responses. Furthermore, the performances of CNNs trained with different numbers of datasets generated by the proposed data generation method based on data overlapping with the same data pool are discussed with related to the number of CNN training iteration number.

show abstract

Section: Introductionmentioning

confidence: 99%

Seismic response prediction method for building structures using convolutional neural network

Park

2020

Struct Control Health Monit

View full text Add to dashboard Cite

show abstract

“…The change in this health indicator value can be used to identify the crack severity level using the binary segmentation methodology. In different works, different health indicators are used for structural health monitoring. In these works, if it is not possible to visually inspect the structure, then binary segmentation methodology can be used for damage state identification based on the extracted health indicators.…”

Section: Resultsmentioning

confidence: 99%

Gear pitting severity level identification using binary segmentation methodology

Kundu

Darpe

Kulkarni

2019

Struct Control Health Monit

View full text Add to dashboard Cite

With growth of a defect on the gear tooth surface, vibration response of the geared rotor system changes, which can be quantified using a health indicator. Based on the health indicator, identification of an accurate health stage or categorization is crucial so that the pitting severity classification can be done precisely. In all the prior reported works, the fault severity classification approaches are applied to the seeded pitting fault, and hence, exact state change point is known in advance. However, in the real life, the gear tooth surface is subjected to a natural pitting progression, and exact state change point of the gear is not known a priori. This study presents a binary segmentation methodology for identification of multiple degradation stages such as initial pitting, medium pitting, and severe pitting in a spur gear subjected to natural pitting progression. The gear damage progression is represented by a correlation coefficient-based degradation parameter extracted from the residual vibration signal. Based on the change in distributional properties such as mean and variance of the degradation parameter with time progression, three degradation stages of the gear are identified. A flexible miniature inspection camera is used to access and assess the actual physical damage on the gear tooth surface. Based on the visual inspection pictures, actual damage area at the identified degradation stages is quantified. Performance of the binary segmentation methodology is validated through six run-to-failure accelerated gear pitting experiments.

show abstract

“…Khodabandehlou et al [67] developed a similar CNN-based structural damage detection technique. A one-fourth-scale laboratory structure of a reinforced concrete bridge was used to experimentally demonstrate the proposed method.…”

Section: Vibration-based Structural Damage Detection In Civil Infrastmentioning

confidence: 99%

1-D Convolutional Neural Networks for Signal Processing Applications

Kıranyaz

İnce

Abdeljaber

et al. 2019

ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

252

View full text Add to dashboard Cite

During the last decade, Convolutional Neural Networks (CNNs) have become the de facto standard for various Computer Vision and Machine Learning operations. CNNs are feed-forward Artificial Neural Networks (ANNs) with alternating convolutional and subsampling layers. Deep 2D CNNs with many hidden layers and millions of parameters have the ability to learn complex objects and patterns providing that they can be trained on a massive size visual database with ground-truth labels. With a proper training, this unique ability makes them the primary tool for various engineering applications for 2D signals such as images and video frames. Yet, this may not be a viable option in numerous applications over 1D signals especially when the training data is scarce or application-specific. To address this issue, 1D CNNs have recently been proposed and immediately achieved the state-of-the-art performance levels in several applications such as personalized biomedical data classification and early diagnosis, structural health monitoring, anomaly detection and identification in power electronics and motor-fault detection. Another major advantage is that a real-time and low-cost hardware implementation is feasible due to the simple and compact configuration of 1D CNNs that perform only 1D convolutions (scalar multiplications and additions). This paper presents a comprehensive review of the general architecture and principals of 1D CNNs along with their major engineering applications, especially focused on the recent progress in this field. Their state-of-the-art performance is highlighted concluding with their unique properties. The benchmark datasets and the principal 1D CNN software used in those applications are also publically shared in a dedicated website.

show abstract

Vibration‐based structural condition assessment using convolution neural networks

Cited by 103 publications

References 32 publications

Seismic response prediction method for building structures using convolutional neural network

Seismic response prediction method for building structures using convolutional neural network

Gear pitting severity level identification using binary segmentation methodology

1-D Convolutional Neural Networks for Signal Processing Applications

Contact Info

Product

Resources

About