Automated health condition diagnosis of in situ wood utility poles using an intelligent non-destructive evaluation (NDE) framework
- Authors: Yu, Yang , Subhani, Mahbube , Hoshyar, Azadeh , Li, Jianchun , Li, Huan
- Date: 2020
- Type: Text , Journal article
- Relation: International Journal of Structural Stability and Dynamics Vol. 20, no. 10 (2020), p.
- Full Text:
- Reviewed:
- Description: Wood utility poles are widely applied in power transmission and telecommunication systems in Australia. Because of a variety of external influence factors, such as fungi, termite and environmental conditions, failure of poles due to the wood degradation with time is of common occurrence with high degree uncertainty. The pole failure may result in serious consequences including both economic and public safety. Therefore, accurately and timely identifying the health condition of the utility poles is of great significance for economic and safe operation of electricity and communication networks. In this paper, a novel non-destructive evaluation (NDE) framework with advanced signal processing and artificial intelligence (AI) techniques is developed to diagnose the condition of utility pole in field. To begin with, the guided waves (GWs) generated within the pole is measured using multi-sensing technique, avoiding difficult interpretation of various wave modes which cannot be detected by only one sensor. Then, empirical mode decomposition (EMD) and principal component analysis (PCA) are employed to extract and select damage-sensitive features from the captured GW signals. Additionally, the up-to-date machine learning (ML) techniques are adopted to diagnose the health condition of the pole based on selected signal patterns. Eventually, the performance of the developed NDE framework is evaluated using the field testing data from 15 new and 24 decommissioned utility poles at the pole yard in Sydney. © 2020 World Scientific Publishing Company.
- Description: This research is supported by Australian Research Council via Linkage Project (LP110200162) and Industrial Transforming Research Hub for Nanoscience Based Construction Materials Manufacturing (IH150100006) as well as Ausgrid. The authors greatly appreciate the ¯nancial and technical supports from the funding bodies.
Corrosion and coating defect assessment of Coal Handling and Preparation Plants (CHPP) using an ensemble of deep convolutional neural networks and decision-level data fusion
- Authors: Yu, Yang , Hoshyar, Azadeh , Samali, Bijan , Zhang, Guang , Rashidi, Maria , Mohammadi, Masoud
- Date: 2023
- Type: Text , Journal article
- Relation: Neural Computing and Applications Vol. 35, no. 25 (2023), p. 18697-18718
- Full Text: false
- Reviewed:
- Description: In view of the problems of ineffective feature extraction and low detection accuracy in existing detection system, this study presents a novel machine vision-based approach composed of an ensemble of deep convolutional neural networks (CNNs) and improved Dempster-Shafer (D-S) theory-based data fusion to evaluate corrosion and coating defect of coal handling and preparation plants. To start with, the structural surface image is sent to each transferred CNN for initial defect identification. Then, an improved D-S fusion algorithm is proposed to combine the identification results from different CNNs, which are vectors consisting of statistical indicators of all the potential damage severity categories. The decision-level fusion of different CNNs can effectively improve image classification. To validate the performance of the proposed method, a dataset made of 3593 surface images with different defect severities captured from mining infrastructure in field is established together with data augmentation. The validation result demonstrates that the proposed method is able to effectively improve the recognition accuracy of defect severity and reduce the wrong recognition rate. Finally, the robustness of the proposed approach is also appraised by polluting the images with different types and intensities of noise, with satisfactory results. © 2023, The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature.
Nonlinear characterization of magnetorheological elastomer-based smart device for structural seismic mitigation
- Authors: Yu, Yang , Hoshyar, Azadeh , Li, Huan , Zhang, Guang , Wang, Weiqiang
- Date: 2021
- Type: Text , Journal article
- Relation: International Journal of Smart and Nano Materials Vol. 12, no. 4 (2021), p. 390-428
- Full Text:
- Reviewed:
- Description: Magnetorheological elastomer (MRE) has been demonstrated to be effective in structural vibration control because of controllable stiffness and damping properties with the effect of external magnetic fields. To achieve a high performance of MRE device-based vibration control, a robust and accurate model is necessary to describe nonlinear dynamics of MRE device. This article aims at realising this target via nonlinear modeling of an innovative MRE device, i.e. MRE vibration isolator. First, the field-dependent properties of MRE isolator were analysed based on experimental data of the isolator in various dynamic tests. Then, a phenomenal model was developed to account for these unique characteristics of MRE-based device. Moreover, an improved PSO algorithm was designed to estimate model parameters. Based on identification results, a generalised model was proposed to clarify the field-dependent properties of the isolator due to varied currents, which was then validated by random and earthquake-excited test data. Based on the proposed model, a frequency control strategy was designed for semi-active control of MRE devices-incorporated smart structure for vibration suppression. Finally, using a three-storey frame model and four benchmark earthquakes, a numerical study was conducted to validate the performance of control strategy based on the generalised current-dependent model with satisfactory results. © 2021 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group.
Proposed machine learning techniques for bridge structural health monitoring : a laboratory study
- Authors: Noori Hoshyar, Azadeh , Rashidi, Maria , Yu, Yang , Samali, Bijan
- Date: 2023
- Type: Text , Journal article
- Relation: Remote Sensing Vol. 15, no. 8 (2023), p.
- Full Text:
- Reviewed:
- Description: Structural health monitoring for bridges is a crucial concern in engineering due to the degradation risks caused by defects, which can become worse over time. In this respect, enhancement of various models that can discriminate between healthy and non-healthy states of structures have received extensive attention. These models are concerned with implementation algorithms, which operate on the feature sets to quantify the bridge’s structural health. The functional correlation between the feature set and the health state of the bridge structure is usually difficult to define. Therefore, the models are derived from machine learning techniques. The use of machine learning approaches provides the possibility of automating the SHM procedure and intelligent damage detection. In this study, we propose four classification algorithms to SHM, which uses the concepts of support vector machine (SVM) algorithm. The laboratory experiment, which intended to validate the results, was performed at Western Sydney University (WSU). The results were compared with the basic SVM to evaluate the performance of proposed algorithms. © 2023 by the authors.
Spectrum of Variable-Random trees
- Authors: Liu, Fei , Ting, Kaiming , Yu, Yang , Zhou, Zhi-Hua
- Date: 2008
- Type: Text , Journal article
- Relation: The Journal of Artificial Intelligence Research Vol. 32, no. (2008), p. 355-384
- Full Text: false
- Reviewed:
- Description: In this paper, we show that a continuous spectrum of randomisation exists, in which most existing tree randomisations are only operating around the two ends of the spectrum. That leaves a huge part of the spectrum largely unexplored. We propose a base learner VR-Tree which generates trees with variable-randomness. VR-Trees are able to span from the conventional deterministic trees to the complete-random trees using a probabilistic parameter. Using VR-Trees as the base models, we explore the entire spectrum of randomised ensembles, together with Bagging and Random Subspace. We discover that the two halves of the spectrum have their distinct characteristics; and the understanding of which allows us to propose a new approach in building better decision tree ensembles. We name this approach Coalescence, which coalesces a number of points in the random-half of the spectrum. Coalescence acts as a committee of "experts" to cater for unforeseeable conditions presented in training data. Coalescence is found to perform better than any single operating point in the spectrum, without the need to tune to a specific level of randomness. In our empirical study, Coalescence ranks top among the benchmarking ensemble methods including Random Forests, Random Subspace and C5 Boosting; and only Coalescence is significantly better than Bagging and Max-Diverse Ensemble among all the methods in the comparison. Although Coalescence is not significantly better than Random Forests, we have identified conditions under which one will perform better than the other.