Photovoltaic system fault diagnosis based on binary salp swarm and optimized support vector machine DOI Creative Commons
Tawfik Thelaidjia, Nabil Chetih, Zouhir Boumous

et al.

South Florida Journal of Development, Journal Year: 2024, Volume and Issue: 5(12), P. e4863 - e4863

Published: Dec. 31, 2024

In this study, we develop a pattern recognition method that utilizes dimensionality reduction and an optimized support vector machine (SVM) for fault diagnosis in photovoltaic systems, based on three-phase currents data. Initially, eleven (11) statistical descriptors are calculated from each phase currents. As result, thirty-three (33) included the feature vector. However, not all equally sensitive to faults. Because of this, use binary salp swarm optimisation algorithm (BSSA) application counter-propagation artificial neural networks classification error as fitness function choose most exclude those with low sensitivity. Finally, optimal is adopted ensure task. The suggested approach evaluated by using real dataset. obtained results demonstrate BSSA has high convergence speed can effectively select pertinent features. Furthermore, rate indicates be employed system diagnosis.

Language: Английский

Fault detection in photovoltaic systems using unmanned aerial vehicle-captured images and rough set theory DOI

C. V. Prasshanth,

Sujatha Narayanan,

Naveen Venkatesh Sridharan

et al.

Solar Energy, Journal Year: 2025, Volume and Issue: 290, P. 113348 - 113348

Published: Feb. 16, 2025

Language: Английский

Citations

1
A Particle Swarm Optimization-Based Ensemble Broad Learning System for Intelligent Fault Diagnosis in Safety-Critical Energy Systems with High-Dimensional Small Samples DOI Creative Commons

Jiasheng Yan,

Yang Sui,

Tao Dai

et al.

Mathematics, Journal Year: 2025, Volume and Issue: 13(5), P. 797 - 797

Published: Feb. 27, 2025

Intelligent fault diagnosis (IFD) plays a crucial role in reducing maintenance costs and enhancing the reliability of safety-critical energy systems (SCESs). In recent years, deep learning-based IFD methods have achieved high accuracy extracting implicit higher-order correlations between features. However, excessive long training time learning models conflicts with requirements real-time analysis for IFD, hindering their further application practical industrial environments. To address aforementioned challenge, this paper proposes an innovative method SCES that combines particle swarm optimization (PSO) algorithm ensemble broad system (EBLS). Specifically, (BLS), known its low complexity classification accuracy, is adopted as alternative to SCES. Furthermore, EBLS designed enhance model stability high-dimensional small samples by incorporating random forest (RF) strategy into traditional BLS framework. order reduce computational cost EBLS, which constrained selection hyperparameters, PSO employed optimize hyperparameters EBLS. Finally, validated through simulated data from complex nuclear power plant (NPP). Numerical experiments reveal proposed significantly improved diagnostic efficiency while maintaining accuracy. summary, approach shows great promise boosting capabilities

Language: Английский

Citations

0
Intelligent real-time status identification for anti-roll tank via solid-liquid triboelectric nanogenerators DOI
Xingang Xu, Hao Wu, Zhongjie Li

et al.

Ocean Engineering, Journal Year: 2025, Volume and Issue: 327, P. 120987 - 120987

Published: March 17, 2025

Citations

0
Enhancing industrial machinery maintenance through advanced fault and novelty detection using variational autoencoder and hybrid transformer model DOI
H. Hamdaoui, Looh Augustine Ngiejungbwen, Jinan Gu

et al.

Structural Health Monitoring, Journal Year: 2025, Volume and Issue: unknown

Published: March 26, 2025

In the context of Industry 4.0, new sensing and communication technologies have unlocked vast amounts process data, offering significant potential for its transformation into actionable insights to support manufacturing decisions. The reliable detection diagnosis faults in rolling element bearings pose a challenge condition-based maintenance fault (FDD), which are critical strategies enhancing equipment reliability reducing operational costs. Deep learning methods, such as convolutional neural networks (CNNs), can extract features from vibration signals compared traditional signal processing. However, these methods isolation insufficient reliably detect novel conditions variable working environments. Also, existing novelty anomaly criteria not accurate enough correctly distinguish or unseen faults. This study introduces multi-fault framework leveraging variational autoencoder with Mahalanobis distance (MD) scores unknown condition hybrid CNN-Swin transformer (Swin-T) model incremental classification. Using frequency-domain image-based representation signals, CNN-based feature extractor after projecting patch embedding layer simplified Swin-T is trained incrementally allow continuous adaptation. Extensive validation three separate datasets simulation test rigs demonstrates superior performance method over cutting-edge models FDD (ND), achieving near-perfect accuracy (99.7%), precision (99.8%), recall (99.6%), F1 score (99.7%). ND outperformed approaches an MD threshold yielding true-positive rate 98.9% false-positive 1.2%. Additionally, improved classification by up 5.4% newly introduced types, highlighting adaptability. These results demonstrate framework’s ability enhance efficiency industrial machinery identifying both known high precision.

Language: Английский

Citations

0
Practical implementation based on Histogram of Oriented Gradient descriptor combined with Deep Learning: Towards intelligent monitoring of a photovoltaic power plant with robust faults predictions DOI
Nadji Hadroug,

Amel Sabrine Amari,

Walaa Alayed

et al.

Journal of Industrial Information Integration, Journal Year: 2024, Volume and Issue: unknown, P. 100760 - 100760

Published: Dec. 1, 2024

Language: Английский

Citations

0
Photovoltaic system fault diagnosis based on binary salp swarm and optimized support vector machine DOI Creative Commons
Tawfik Thelaidjia, Nabil Chetih, Zouhir Boumous

et al.

South Florida Journal of Development, Journal Year: 2024, Volume and Issue: 5(12), P. e4863 - e4863

Published: Dec. 31, 2024

In this study, we develop a pattern recognition method that utilizes dimensionality reduction and an optimized support vector machine (SVM) for fault diagnosis in photovoltaic systems, based on three-phase currents data. Initially, eleven (11) statistical descriptors are calculated from each phase currents. As result, thirty-three (33) included the feature vector. However, not all equally sensitive to faults. Because of this, use binary salp swarm optimisation algorithm (BSSA) application counter-propagation artificial neural networks classification error as fitness function choose most exclude those with low sensitivity. Finally, optimal is adopted ensure task. The suggested approach evaluated by using real dataset. obtained results demonstrate BSSA has high convergence speed can effectively select pertinent features. Furthermore, rate indicates be employed system diagnosis.

Language: Английский

Citations

0