| 1. |
- Abdeljaber, Osama, et al.
(författare)
-
1-D CNNs for structural damage detection : verification on a structural health monitoring benchmark data
- 2018
-
Ingår i: Neurocomputing. - : Elsevier. - 0925-2312 .- 1872-8286. ; 275, s. 1308-1317
-
Tidskriftsartikel (refereegranskat)abstract
- Structural damage detection has been an interdisciplinary area of interest for various engineering fields. While the available damage detection methods have been in the process of adapting machine learning concepts, most machine learning based methods extract “hand-crafted” features which are fixed and manually selected in advance. Their performance varies significantly among various patterns of data depending on the particular structure under analysis. Convolutional neural networks (CNNs), on the other hand, can fuse and simultaneously optimize two major sets of an assessment task (feature extraction and classification) into a single learning block during the training phase. This ability not only provides an improved classification performance but also yields a superior computational efficiency. 1D CNNs have recently achieved state-of-the-art performance in vibration-based structural damage detection; however, it has been reported that the training of the CNNs requires significant amount of measurements especially in large structures. In order to overcome this limitation, this paper presents an enhanced CNN-based approach that requires only two measurement sets regardless of the size of the structure. This approach is verified using the experimental data of the Phase II benchmark problem of structural health monitoring which had been introduced by IASC-ASCE Structural Health Monitoring Task Group. As a result, it is shown that the enhanced CNN-based approach successfully estimated the actual amount of damage for the nine damage scenarios of the benchmark study.
|
|
| 2. |
- Abdeljaber, Osama, et al.
(författare)
-
Fault Detection and Severity Identification of Ball Bearings by Online Condition Monitoring
- 2019
-
Ingår i: IEEE Transactions on Industrial Electronics. - : IEEE. - 0278-0046 .- 1557-9948. ; 66:10, s. 8136-8147
-
Tidskriftsartikel (refereegranskat)abstract
- This paper presents a fast, accurate, and simple systematic approach for online condition monitoring and severity identification of ball bearings. This approach utilizes compact one-dimensional (1-D) convolutional neural networks (CNNs) to identify, quantify, and localize bearing damage. The proposed approach is verified experimentally under several single and multiple damage scenarios. The experimental results demonstrated that the proposed approach can achieve a high level of accuracy for damage detection, localization, and quantification. Besides its real-time processing ability and superior robustness against the high-level noise presence, the compact and minimally trained 1-D CNNs in the core of the proposed approach can handle new damage scenarios with utmost accuracy.
|
|
| 3. |
- Abdeljaber, Osama, et al.
(författare)
-
Optimization of linear zigzag insert metastructures for low-frequency vibration attenuation using genetic algorithms
- 2017
-
Ingår i: Mechanical systems and signal processing. - : Elsevier. - 0888-3270 .- 1096-1216. ; 84:Part A, s. 625-641
-
Tidskriftsartikel (refereegranskat)abstract
- Vibration suppression remains a crucial issue in the design of structures and machines. Recent studies have shown that with the use of metamaterial inspired structures (or metastructures), considerable vibration attenuation can be achieved. Optimization of the internal geometry of metastructures maximizes the suppression performance. Zigzag inserts have been reported to be efficient for vibration attenuation. It has also been reported that the geometric parameters of the inserts affect the vibration suppression performance in a complex manner. In an attempt to find out the most efficient parameters, an optimization study has been conducted on the linear zigzag inserts and is presented here. The research reported in this paper aims at developing an automated method for determining the geometry of zigzag inserts through optimization. This genetic algorithm based optimization process searches for optimal zigzag designs which are properly tuned to suppress vibrations when inserted in a specific host structure (cantilever beam). The inserts adopted in this study consist of a cantilever zigzag structure with a mass attached to its unsupported tip. Numerical simulations are carried out to demonstrate the efficiency of the proposed zigzag optimization approach.
|
|
| 4. |
- Abdeljaber, Osama, et al.
(författare)
-
Real-time vibration-based structural damage detection using one-dimensional convolutional neural networks
- 2017
-
Ingår i: Journal of Sound and Vibration. - : Elsevier. - 0022-460X .- 1095-8568. ; 388, s. 154-170
-
Tidskriftsartikel (refereegranskat)abstract
- Structural health monitoring (SHM) and vibration-based structural damage detection have been a continuous interest for civil, mechanical and aerospace engineers over the decades. Early and meticulous damage detection has always been one of the principal objectives of SHM applications. The performance of a classical damage detection system predominantly depends on the choice of the features and the classifier. While the fixed and hand-crafted features may either be a sub-optimal choice for a particular structure or fail to achieve the same level of performance on another structure, they usually require a large computation power which may hinder their usage for real-time structural damage detection. This paper presents a novel, fast and accurate structural damage detection system using 1D Convolutional Neural Networks (CNNs) that has an inherent adaptive design to fuse both feature extraction and classification blocks into a single and compact learning body. The proposed method performs vibration-based damage detection and localization of the damage in real-time. The advantage of this approach is its ability to extract optimal damage-sensitive features automatically from the raw acceleration signals. Large-scale experiments conducted on a grandstand simulator revealed an outstanding performance and verified the computational efficiency of the proposed real-time damage detection method.
|
|
| 5. |
- Avci, Onur, et al.
(författare)
-
A New Benchmark Problem for Structural Damage Detection : Bolt Loosening Tests on a Large-Scale Laboratory Structure
- 2022
-
Ingår i: Dynamics of Civil Structures, Volume 2. - Cham : Springer. - 9783030771423 - 9783030771430 ; , s. 15-22
-
Konferensbidrag (refereegranskat)abstract
- Monitoring the structural performance of engineering structures has always been pertinent for maintaining structural health and assessing the life cycle of structures. Structural Health Monitoring (SHM) and Structural Damage Detection (SDD) fields have been topics of ongoing research over the years to explore and verify different monitoring techniques and damage detection and localization procedures. In an attempt to compare performances of different methods, benchmark datasets are valuable resources since the data is made available to researchers enabling side-by-side comparisons. This paper presents a new experimental benchmark dataset generated from tests on a large-scale laboratory structure. The primary goal of the authors was to explore brand-new damage detection and quantification methodologies for efficient monitoring of structures. For this purpose, a large-scale steel grid structure with footprint dimensions of 4.2 m × 4.2 m was constructed in laboratory environment and it has been used as a test bed by the authors. The structural members of the structure are all IPE120 hot-rolled steel cross sections. The simulation of structural damage was simply loosening the bolts at one of the beam-to-girder connections, which is a slight change of rotational stiffness at the joint of the steel grid structure. The authors shared the dataset for 1 undamaged and 30 damaged conditions and published it on a public website as a new benchmark problem for structural damage detection at http://www.structuralvibration.com/benchmark/ so that other researchers can use the data and test algorithms. The authors also shared one of the damage detection tools they used, One-Dimensional Convolutional Neural Networks (1D-CNNs). The application codes, configuration files, and accompanied components of the 1D-CNNs package are available for viewers at http://www.structuralvibration.com/cnns/. © 2022, The Society for Experimental Mechanics, Inc.
|
|
| 6. |
- Avci, Onur, et al.
(författare)
-
A review of vibration-based damage detection in civil structures : from traditional methods to Machine Learning and Deep Learning applications
- 2021
-
Ingår i: Mechanical systems and signal processing. - : Elsevier. - 0888-3270 .- 1096-1216. ; 147
-
Tidskriftsartikel (refereegranskat)abstract
- Monitoring structural damage is extremely important for sustaining and preserving the service life of civil structures. While successful monitoring provides resolute and staunch information on the health, serviceability, integrity and safety of structures; maintaining continuous performance of a structure depends highly on monitoring the occurrence, formation and propagation of damage. Damage may accumulate on structures due to different environmental and human-induced factors. Numerous monitoring and detection approaches have been developed to provide practical means for early warning against structural damage or any type of anomaly. Considerable effort has been put into vibration-based methods, which utilize the vibration response of the monitored structure to assess its condition and identify structural damage. Meanwhile, with emerging computing power and sensing technology in the last decade, Machine Learning (ML) and especially Deep Learning (DL) algorithms have become more feasible and extensively used in vibration-based structural damage detection with elegant performance and often with rigorous accuracy. While there have been multiple review studies published on vibration-based structural damage detection, there has not been a study where the transition from traditional methods to ML and DL methods are described and discussed. This paper aims to fulfill this gap by presenting the highlights of the traditional methods and provide a comprehensive review of the most recent applications of ML and DL algorithms utilized for vibration-based structural damage detection in civil structures.
|
|
| 7. |
- Avci, Onur, et al.
(författare)
-
An Overview of Deep Learning Methods Used in Vibration-Based Damage Detection in Civil Engineering
- 2022
-
Ingår i: Dynamics of Civil Structures, Volume 2. Conference Proceedings of the Society for Experimental Mechanics Series.. - Cham : Springer. - 9783030771430 - 9783030771423 ; , s. 93-98
-
Konferensbidrag (refereegranskat)abstract
- This paper presents a brief overview of vibration-based damage identification studies based on Deep Learning (DL) in civil engineering structures. The presence, type, size, and propagation of structural damage on civil infrastructure have always been a topic of research. In the last couple of decades, there has been a significant shift in the damage detection paradigm when the advancements in sensing and computing technologies met with the ever-expanding use of artificial neural network algorithms. Machine-Learning (ML) tools enabled researchers to implement more feasible and faster tools in damage detection applications. When an artificial neural network has more than three layers, it is typically considered as a “deep†learning network. Being an important accomplishment of the ML era, DL tools enable complex systems which are made of several layers to learn implementations of data with outstanding categorization and compartmentalization capability. In fact, with proper training, a DL tool can operate directly with the unprocessed raw data and help the algorithm produce output data. Competitive capabilities like this led DL algorithms perform very well in complicated problems by dividing a relatively large problem into much smaller and more manageable portions. Specifically for damage identification and localization on civil infrastructure, Convolutional Neural Networks (CNNs) and Unsupervised Pretrained Networks (UPNs) are the known DL tools published in the literature. This paper presents an overview of these studies. © 2022, The Society for Experimental Mechanics, Inc.
|
|
| 8. |
- Avci, Onur, et al.
(författare)
-
Control of Plate Vibrations with Artificial Neural Networks and Piezoelectricity
- 2020
-
Ingår i: Sensors and Instrumentation, Aircraft/Aerospace, Energy Harvesting & Dynamic Environments Testing. - Cham : Springer. - 9783030126759 - 9783030126766 ; , s. 293-301
-
Konferensbidrag (övrigt vetenskapligt/konstnärligt)abstract
- This paper presents a method for active vibration control of smart thin cantilever plates. For model formulation needed for controller design and simulations, finite difference technique is used on the cantilever plate response calculations. Piezoelectric patches are used on the plate, for which a neural network based control algorithm is formed and a neurocontroller is produced to calculate the required voltage to be applied on the actuator patch. The neurocontroller is trained and run with a Kalman Filter for controlling the structural response. The neurocontroller performance is assessed by comparing the controlled and uncontrolled structural responses when the plate is subjected to various excitations. It is shown that the acceleration response of the cantilever plate is suppressed considerably validating the efficacy of the neurocontroller and the success of the proposed methodology.
|
|
| 9. |
|
|
| 10. |
- Avci, Onur, et al.
(författare)
-
Convolutional Neural Networks for Real-Time and Wireless Damage Detection
- 2020
-
Ingår i: Dynamics of Civil Structures. - Cham : Springer. - 9783030121143 - 9783030121150 ; , s. 129-136
-
Konferensbidrag (övrigt vetenskapligt/konstnärligt)abstract
- Structural damage detection methods available for structural health monitoring applications are based on data preprocessing, feature extraction, and feature classification. The feature classification task requires considerable computational power which makes the utilization of centralized techniques relatively infeasible for wireless sensor networks. In this paper, the authors present a novel Wireless Sensor Network (WSN) based on One Dimensional Convolutional Neural Networks (1D CNNs) for real-time and wireless structural health monitoring (SHM). In this method, each CNN is assigned to its local sensor data only and a corresponding 1D CNN is trained for each sensor unit without any synchronization or data transmission. This results in a decentralized system for structural damage detection under ambient environment. The performance of this method is tested and validated on a steel grid laboratory structure.
|
|
| 11. |
- Avci, Onur, et al.
(författare)
-
Efficiency Validation of One Dimensional Convolutional Neural Networks for Structural Damage Detection Using A SHM Benchmark Data
- 2018
-
Ingår i: 25th International Congress on Sound and Vibration 2018, ICSV 2018: Hiroshima Calling. - : International Institute of Acoustics and Vibration (IIAV). - 9781510868458 ; , s. 4600-4607
-
Konferensbidrag (övrigt vetenskapligt/konstnärligt)abstract
- In this paper, a novel one dimensional convolution neural network (1D-CNN) based structural damage assessment technique is validated with a benchmark study published by IASC-ASCE Structural Health Monitoring Task Group in 2003. In contrast with predominant machine learning based structural damage detection techniques of the literature, the technique shown in this paper runs without manual feature extraction or preprocessing stages. It runs directly on the raw vibration data. In CNNs, the stages of feature extraction and feature classification are merged into one stage; therefore, the proposed technique is efficient, feasible and economical. Utilizing the optimal features learned by 1D CNNs, the proposed CNN-based technique considerably improves the classification efficiency and accuracy. The performance improvement of the proposed technique is assessed by calculating the “Probability of Damage” values for damage estimations. The unseen structural damage cases between the two extreme end structural cases (zero damage and total damage) were successfully identified. Consequently, it is validated that the improved CNN-based technique is efficient since it predicted the level of damage consistently with the structural damage cases defined in the existing benchmark.
|
|
| 12. |
- Avci, Onur, et al.
(författare)
-
Monitoring framework development for a network of multiple laboratory structures
- 2024
-
Ingår i: Journal of Building Engineering. - 2352-7102. ; 92
-
Tidskriftsartikel (refereegranskat)abstract
- The stadium structures have unique structural features increasing the significance of structural monitoring systems specifically designed for them. Aside from vibrations serviceability concerns and human -induced excitations, the development and propagation of structural damage under all possible atmospheric and seismic conditions need to be closely monitored for structural resiliency and integrity of the stadia. As such, Structural Health Monitoring (SHM) methods combined with effective data evaluation methodologies need to be deployed to monitor the structural performance of stadiums. Even though stadia monitoring has been performed at multiple locations in the world, a web based and real-time SHM network of stadia is not known to authors. As a preliminary study for the network implementation of stadia monitoring with acceleration measurements, the presented work focuses on the fundamental steps to accomplish this goal, with a collaborative research effort between Qatar University, the University of Central Florida, and University of Alberta. The authors performed analytical investigations and experimental testing on stadium -type structures built in laboratory environments for the development of the SHM framework. Specialized signal processing algorithms, sensing suites and approaches considering multi -scale monitoring were used on collected acceleration measurements. The novelty of the work presented in this manuscript are the following items which exist simultaneously in the developed SHM framework. The developed framework is a web -based monitoring application where structural damage is detected in real-time. The proposed methodology operates directly on raw acceleration signals and runs at a network level. With that, the damage detection, damage localization, and damage quantification tasks are performed simultaneously, while the feature extraction and classification stages are combined in one learning body.
|
|
| 13. |
- Avci, Onur, et al.
(författare)
-
One-Dimensional Convolutional Neural Networks for Real-Time Damage Detection of Rotating Machinery
- 2022
-
Ingår i: Rotating Machinery, Optical Methods & Scanning LDV Methods, Volume 6. Conference Proceedings of the Society for Experimental Mechanics Series. - Cham : Springer. ; , s. 73-83
-
Konferensbidrag (refereegranskat)abstract
- This paper presents a novel real-time rotating machinery damage monitoring system. The system detects, quantifies, and localizes damage in ball bearings in a fast and accurate way using one-dimensional convolutional neural networks (1D-CNNs). The proposed method has been validated with experimental work not only for single damage but also for multiple damage cases introduced onto ball bearings in laboratory environment. The two 1D-CNNs (one set for the interior bearing ring and another set for the exterior bearing ring) were trained and tested under the same conditions for torque and speed. It is observed that the proposed system showed excellent performance even with the severe additive noise. The proposed method can be implemented in practical use for online defect detection, monitoring, and condition assessment of ball bearings and other rotatory machine elements. © 2022, The Society for Experimental Mechanics, Inc.
|
|
| 14. |
- Avci, Onur, et al.
(författare)
-
Structural Damage Detection in Civil Engineering with Machine Learning : Current State of the Art
- 2022
-
Ingår i: Sensors and Instrumentation, Aircraft/Aerospace, Energy Harvesting & Dynamic Environments Testing. - Cham : Springer. - 9783030759872 ; , s. 223-229
-
Konferensbidrag (refereegranskat)abstract
- This paper presents a brief overview of vibration-based structural damage detection studies that are based on machine learning (ML) in civil engineering structures. The review includes both parametric and nonparametric applications of ML accompanied with analytical and/or experimental studies. While the ML tools help the system learn from the data fed into, the computer enhances the task with the learned information without any programming on how to process the relevant data. As such, the performance level of ML-based damage identification methodologies depends on the feature extraction and classification steps, especially on the classifier choices for which the characteristic nature of the acceleration signals is recorded in a feasible way. Yet, there are several issues to be discussed about the existing ML procedures for both parametric and nonparametric applications, which are presented in this paper.
|
|
| 15. |
- Avci, Onur, et al.
(författare)
-
Structural Damage Detection in Real Time: Implementation of 1D Convolutional Neural Networks for SHM Applications
- 2017
-
Ingår i: Structural Health Monitoring & Damage Detection, Volume 7. - : Springer. - 9783319541082 ; , s. 49-54
-
Konferensbidrag (övrigt vetenskapligt/konstnärligt)abstract
- Most of the classical structural damage detection systems involve two processes, feature extraction and feature classification. Usually, the feature extraction process requires large computational effort which prevent the application of the classical methods in real-time structural health monitoring applications. Furthermore, in many cases, the hand-crafted features extracted by the classical methods fail to accurately characterize the acquired signal, resulting in poor classification performance. In an attempt to overcome these issues, this paper presents a novel, fast and accurate structural damage detection and localization system utilizing one dimensional convolutional neural networks (CNNs) arguably for the first time in SHM applications. The proposed method is capable of extracting optimal damage-sensitive features automatically from the raw acceleration signals, allowing it to be used for real-time damage detection. This paper presents the preliminary experiments conducted to verify the proposed CNN-based approach.
|
|
| 16. |
- Avci, Onur, et al.
(författare)
-
Structural Health Monitoring with Self-Organizing Maps and Artificial Neural Networks
- 2020
-
Ingår i: Topics in Modal Analysis & Testing. - Cham : Springer. - 9783030126834 - 9783030126841 ; , s. 237-246
-
Konferensbidrag (övrigt vetenskapligt/konstnärligt)abstract
- The use of self-organizing maps and artificial neural networks for structural health monitoring is presented in this paper. The authors recently developed a nonparametric structural damage detection algorithm for extracting damage indices from the ambient vibration response of a structure. The algorithm is based on self-organizing maps with a multilayer feedforward pattern recognition neural network. After the training of the self-organizing maps, the algorithm was tested analytically under various damage scenarios based on stiffness reduction of beam members and boundary condition changes of a grid structure. The results indicated that proposed algorithm can successfully locate and quantify damage on the structure.
|
|
| 17. |
- Avci, Onur, et al.
(författare)
-
Vibration suppression in metastructures using zigzag inserts optimized by genetic algorithms
- 2017
-
Ingår i: Shock & Vibration, Aircraft/Aerospace, Energy Harvesting, Acoustics & Optics. - Cham : Springer. - 9783319547343 - 9783319547350 ; , s. 275-283
-
Konferensbidrag (övrigt vetenskapligt/konstnärligt)abstract
- Metastructures are known to provide considerable vibration attenuation for mechanical systems. With the optimization of the internal geometry of metastructures, the suppression performance of the host structure increases. While the zigzag inserts have been shown to be efficient for vibration attenuation, the geometric properties of the inserts affect the suppression performance in a complex manner when attached to the host structure. This paper presents a genetic algorithm based optimization study conducted to come up with the most efficient geometric properties of the zigzag inserts. The inserts studied in this paper are simply cantilever zigzag structures with a mass attached to the unsupported tips. Numerical simulations are run to show the efficiency of the optimization process.
|
|
| 18. |
- Avci, Onur, et al.
(författare)
-
Wireless and real-time structural damage detection : a novel decentralized method for wireless sensor networks
- 2018
-
Ingår i: Journal of Sound and Vibration. - : Elsevier. - 0022-460X .- 1095-8568. ; 424, s. 158-172
-
Tidskriftsartikel (refereegranskat)abstract
- Being an alternative to conventional wired sensors, wireless sensor networks (WSNs) are extensively used in Structural Health Monitoring (SHM) applications. Most of the Structural Damage Detection (SDD) approaches available in the SHM literature are centralized as they require transferring data from all sensors within the network to a single processing unit to evaluate the structural condition. These methods are found predominantly feasible for wired SHM systems; however, transmission and synchronization of huge data sets in WSNs has been found to be arduous. As such, the application of centralized methods with WSNs has been a challenge for engineers. In this paper, the authors are presenting a novel application of 1D Convolutional Neural Networks (1D CNNs) on WSNs for SDD purposes. The SDD is successfully performed completely wireless and real-time under ambient conditions. As a result of this, a decentralized damage detection method suitable for wireless SHM systems is proposed. The proposed method is based on 1D CNNs and it involves training an individual 1D CNN for each wireless sensor in the network in a format where each CNN is assigned to process the locally-available data only, eliminating the need for data transmission and synchronization. The proposed damage detection method operates directly on the raw ambient vibration condition signals without any filtering or preprocessing. Moreover, the proposed approach requires minimal computational time and power since 1D CNNs merge both feature extraction and classification tasks into a single learning block. This ability is prevailingly cost-effective and evidently practical in WSNs considering the hardware systems have been occasionally reported to suffer from limited power supply in these networks. To display the capability and verify the success of the proposed method, large-scale experiments conducted on a laboratory structure equipped with a state-of-the-art WSN are reported.
|
|
| 19. |
- Kiranyaz, Serkan, et al.
(författare)
-
1-D Convolutional Neural Networks for Signal Processing Applications
- 2019
-
Ingår i: 2019 IEEE International Conference on Acoustics, Speech, and Signal Processing. - : IEEE. - 9781479981311 - 9781479981328 ; , s. 8360-8364
-
Konferensbidrag (refereegranskat)abstract
- 1D Convolutional Neural Networks (CNNs) have recently become the state-of-the-art technique for crucial signal processing applications such as patient-specific ECG classification, structural health monitoring, anomaly detection in power electronics circuitry and motor-fault detection. This is an expected outcome as there are numerous advantages of using an adaptive and compact 1D CNN instead of a conventional (2D) deep counterparts. First of all, compact 1D CNNs can be efficiently trained with a limited dataset of 1D signals while the 2D deep CNNs, besides requiring 1D to 2D data transformation, usually need datasets with massive size, e.g., in the "Big Data" scale in order to prevent the well-known "overfitting" problem. 1D CNNs can directly be applied to the raw signal (e.g., current, voltage, vibration, etc.) without requiring any pre- or post-processing such as feature extraction, selection, dimension reduction …
|
|
| 20. |
- Kiranyaz, Serkan, et al.
(författare)
-
1D convolutional neural networks and applications : A survey
- 2021
-
Ingår i: Mechanical systems and signal processing. - : Elsevier. - 0888-3270 .- 1096-1216. ; 151
-
Tidskriftsartikel (refereegranskat)abstract
- 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 electrical 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 publicly shared in a dedicated website. While there has not been a paper on the review of 1D CNNs and its applications in the literature, this paper fulfills this gap. (C) 2020 The Author(s). Published by Elsevier Ltd.
|
|
| 21. |
- Kiranyaz, Serkan, et al.
(författare)
-
Zero-shot motor health monitoring by blind domain transition
- 2024
-
Ingår i: Mechanical systems and signal processing. - : Elsevier. - 0888-3270 .- 1096-1216. ; 210
-
Tidskriftsartikel (refereegranskat)abstract
- Continuous long-term monitoring of motor health is crucial for the early detection of abnormalities such as bearing faults (up to 51% of motor failures are attributed to bearing faults). Despite numerous methodologies proposed for bearing fault detection, most of them require normal (healthy) and abnormal (faulty) data for training. Even with the recent deep learning (DL) methodologies trained on the labeled data from the same machine, the classification accuracy significantly deteriorates when one or few conditions are altered, e.g., a different speed or load, or for different fault types/severities with sensors placed in different locations. Furthermore, their performance suffers significantly or may entirely fail when they are tested on another machine with entirely different healthy and faulty signal patterns. To address this need, in this pilot study, we propose a zero -shot bearing fault detection method that can detect any fault on a new (target) machine regardless of the working conditions, sensor parameters, or fault characteristics. To accomplish this objective, a 1D Operational Generative Adversarial Network (Op-GAN) first characterizes the transition between normal and fault vibration signals of (a) source machine(s) under various conditions, sensor parameters, and fault types. Then for a target machine, the potential faulty signals can be generated, and over its actual healthy and synthesized faulty signals, a compact, and lightweight 1D Self-ONN fault detector can then be trained to detect the real faulty condition in real time whenever it occurs. To validate the proposed approach, a new benchmark dataset is created using two different motors working under different conditions and sensor locations. Experimental results demonstrate that this novel approach can accurately detect any bearing fault achieving an average recall rate of around 89% and 95% on two target machines regardless of its type, severity, and location.
|
|
