Related references
Note: Only part of the references are listed.
Article
Engineering, Mechanical
Jinfeng Huang et al.
Summary: This paper proposes an improved dynamic model of bearings that accurately simulates the impact characteristics of defective bearings and improves computational efficiency. A new index, MHVS-RMS, is constructed based on the dynamic model and the mathematical principle of morphology. The function relationship between MHVS-RMS and the location dimension, as well as its principle of noise reduction and feature enhancement, are derived and proven. Furthermore, a new morphological scale-difference filter and feature enhancement algorithm are proposed to improve diagnosis accuracy.
MECHANISM AND MACHINE THEORY
(2023)
Article
Energy & Fuels
Ling Xiang et al.
Summary: A new method is proposed in this study to extract multidirectional spatio-temporal features of SCADA data for wind turbine condition monitoring, using convolutional neural network and bidirectional gated recurrent unit with attention mechanism. This method can effectively detect early abnormal operation and identify failed components of wind turbine, showing better feasibility for practical wind energy application.
Article
Computer Science, Hardware & Architecture
Min Qian et al.
Summary: This article proposes a weakly supervised oversampling method to address the class imbalance problem in fault diagnosis of industrial systems. It develops a graph semisupervised learning algorithm to select high-quality synthetic samples and combines a cost-sensitive dimensionality reduction method and a boosting-based ensemble framework. The experimental results show that the proposed framework outperforms benchmark methods.
IEEE TRANSACTIONS ON RELIABILITY
(2022)
Article
Green & Sustainable Science & Technology
Panagiotis Korkos et al.
Summary: The performance of wind turbines can be improved by processing SCADA data, which also enhances decisions about maintenance schedules. The pitch system plays a critical role in analyzing relevant data for improving wind turbine operation. This study demonstrates the potential of the adaptive neuro fuzzy inference system technique for detecting pitch faults by gathering significant pitch faults and implementing the ANFIS technique. The proposed approach includes detailed preprocessing of SCADA data and emphasizes the labeling process.
Article
Green & Sustainable Science & Technology
Phong B. Dao
Summary: This paper presents a new approach for condition monitoring and fault diagnosis of wind turbines based on structural break detection in SCADA data. The method uses the Chow test to assess the stability of regression coefficients in a multiple linear regression model, enabling the detection of faults in the wind turbine.
Article
Green & Sustainable Science & Technology
Zhen Guo et al.
Summary: Sufficient samples are crucial for improving fault diagnosis performance in wind turbine gearboxes. However, obtaining fault data is challenging and time-consuming. To address this issue, an improved adversarial learning approach is proposed to generate fault features for wind turbine gearbox diagnosis. Experimental results demonstrate the effectiveness of this method in fault diagnosis.
Article
Green & Sustainable Science & Technology
Reihane Rahimilarki et al.
Summary: Fault detection and classification are crucial techniques in industrial monitoring, and deep learning approaches using convolutional neural networks have shown promise in solving the problem of fault detection in wind turbines. This article presents a novel deep learning method based on time-series analysis and CNN for fault detection and classification in wind turbines, and the simulation results demonstrate the accuracy and effectiveness of the proposed method.
Article
Green & Sustainable Science & Technology
Shilin Sun et al.
Summary: An adaptive method for blade damage identification in wind turbine blades is proposed in this paper, which can achieve global and remote health assessment, high spatial resolutions, and improved convergence rates.
Article
Green & Sustainable Science & Technology
Yongchao Zhu et al.
Summary: This study presents a novel method for real-time operational state prediction of wind turbines based on deep learning and fuzzy synthetic assessment, which can effectively reduce operation & maintenance costs and improve reliability. By analyzing monitoring data, establishing prediction dimensions, and distinguishing normal and fault states, the proposed method shows promise in detecting potential failures of WTGs.
Review
Green & Sustainable Science & Technology
Shilin Sun et al.
Summary: This paper investigates in-situ condition monitoring (CM) techniques for operational wind turbine (WT) blades and summarizes typical causes and types of blade damage during operation. Furthermore, the study provides a first-time summary of the investigation and development timeline of in-situ CM methods and types of damage that have been verified to be recognizable in field tests.
RENEWABLE & SUSTAINABLE ENERGY REVIEWS
(2022)
Article
Green & Sustainable Science & Technology
Juchuan Dai et al.
Summary: This paper comprehensively reviews the development of the blade industry, load theory, load measurement, and load control of large-scale Horizontal Axis Wind Turbines (HAWTs). It discusses the key issues and corresponding progress in blade research and application, and outlines the main areas of attention in future blade research and application, providing effective countermeasures.
Article
Automation & Control Systems
Shilin Sun et al.
Summary: In this article, an environmentally adaptive and contrastive representation learning method is proposed to improve the accuracy of asset condition monitoring. The method utilizes regression models and contrastive learning to obtain valid data representations. Moreover, the method establishes a reference statue for healthy assets and employs control charts for online monitoring. The remarkable performance of the method is demonstrated with real-world measurements and comparative analysis.
IEEE TRANSACTIONS ON CYBERNETICS
(2022)
Article
Green & Sustainable Science & Technology
Wanqiu Chen et al.
Summary: This study presents a framework for fault diagnosis of wind turbine faults using transfer learning algorithms Inception V3 and TrAdaBoost, and introduces a new evaluation index 'Comprehensive Index'. Traditional machine learning algorithms perform poorly for unbalanced and differently distributed datasets, while the novel transfer learning algorithm TrAdaBoost shows superior performance in dealing with such challenges.
Article
Automation & Control Systems
Qun He et al.
Summary: The paper proposes a new spatio-temporal multiscale neural network (STMNN) to capture spatio-temporal correlations in SCADA data from wind turbines. It contains two parallel feature extraction modules for temporal multiscale and spatial multiscale features, and adopts focal loss instead of cross-entropy loss to address data imbalance. The experimental results demonstrate the effectiveness of the STMNN model in practical applications.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2021)
Article
Green & Sustainable Science & Technology
Seongpil Cho et al.
Summary: The paper develops a fault detection and diagnosis method for a hydraulic blade pitch system in a spar-type floating wind turbine, using a Kalman filter for fault detection and an artificial neural network for fault diagnosis. The neural network model shows effective performance after training, validation, and test procedures.
Article
Automation & Control Systems
Luoxiao Yang et al.
Summary: The article presents a monitoring method based on conditional convolutional autoencoder for identifying wind turbine blade breakages, which has been validated with datasets from multiple commercial wind farms. The method derives reconstruction errors reflecting system dynamics changes caused by impending blade breakages and uses statistical process control to develop boundaries for triggering blade breakage alarms. Comparison with classical autoencoder and conditional autoencoder-based methods demonstrates advantages of the proposed method.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2021)
Article
Automation & Control Systems
Huaikuan Yi et al.
Summary: The article proposed a minority clustering SMOTE (MC-SMOTE) method to improve the imbalance classification performance and verified its superiority through experiments on benchmark datasets and real industrial data.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2021)
Article
Green & Sustainable Science & Technology
Yanting Li et al.
Summary: This paper proposes a fault diagnosis method for wind turbines based on parameter-based transfer learning and convolutional autoencoder, suitable for small-scale data. The method can transfer knowledge from similar wind turbines and shows advantages in fault diagnosis.
Article
Green & Sustainable Science & Technology
Xiaoan Yan et al.
Summary: This paper proposes a new approach based on MSSD and IKC to effectively extract multichannel fault information of WTDS and automatically discriminate different fault patterns through PLS. Practical engineering data demonstrate the effectiveness of this method, which shows superiority in fault information extraction and health condition identification compared to other multichannel methods and traditional single-channel approaches.
Article
Green & Sustainable Science & Technology
Yonghao Miao et al.
Article
Green & Sustainable Science & Technology
Yingning Qiu et al.
Article
Green & Sustainable Science & Technology
Cancan Yi et al.
Article
Green & Sustainable Science & Technology
Yanhua Pang et al.
Article
Automation & Control Systems
Jinhai Liu et al.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2019)
Review
Energy & Fuels
Estefania Artigao et al.
Article
Automation & Control Systems
Wei Qiao et al.
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS
(2015)