ResGRU: A Novel Hybrid Deep Learning Model for Compound Fault Diagnosis in Photovoltaic Arrays Considering Dust Impact

With the widespread deployment of photovoltaic (PV) power stations, timely identification and rectification of module defects are crucial for extending service life and preserving efficiency. PV arrays, subjected to severe outside circumstances, are prone to defects exacerbated by dust accumulation,...

Full description

Saved in:
Bibliographic Details
Main Authors: Xi Liu, Hui Hwang Goh, Haonan Xie, Tingting He, Weng Kean Yew, Dongdong Zhang, Wei Dai, Tonni Agustiono Kurniawan
Format: Article
Language:English
Published: MDPI AG 2025-02-01
Series:Sensors
Subjects:
Online Access:https://www.mdpi.com/1424-8220/25/4/1035
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850238633605332992
author Xi Liu
Hui Hwang Goh
Haonan Xie
Tingting He
Weng Kean Yew
Dongdong Zhang
Wei Dai
Tonni Agustiono Kurniawan
author_facet Xi Liu
Hui Hwang Goh
Haonan Xie
Tingting He
Weng Kean Yew
Dongdong Zhang
Wei Dai
Tonni Agustiono Kurniawan
author_sort Xi Liu
collection DOAJ
description With the widespread deployment of photovoltaic (PV) power stations, timely identification and rectification of module defects are crucial for extending service life and preserving efficiency. PV arrays, subjected to severe outside circumstances, are prone to defects exacerbated by dust accumulation, potentially leading to complex compound faults. The resemblance between individual and compound faults sometimes leads to misclassification. To address this challenge, this paper presents a novel hybrid deep learning model, ResGRU, which integrates a residual network (ResNet) with bidirectional gated recurrent units (BiGRU) to improve fault diagnostic accuracy. Additionally, a Squeeze-and-Excitation (SE) module is incorporated to enhance relevant features while suppressing irrelevant ones, hence improving performance. To further optimize inter-class separability and intra-class compactness, a center loss function is employed as an auxiliary loss to enhance the model’s discriminative capacity. This proposed method facilitates the automated extraction of fault features from I-V curves and accurate diagnosis of individual faults, partial shading scenarios, and compound faults under varying levels of dust accumulation, hence aiding in the formulation of efficient cleaning schedules. Experimental findings indicate that the suggested model achieves 99.94% accuracy on pristine data and 98.21% accuracy on noisy data, markedly surpassing established techniques such as artificial neural networks (ANN), ResNet, random forests (RF), multi-scale SE-ResNet, and other ResNet-based approaches. Thus, the model offers a reliable solution for accurate PV array fault diagnosis.
format Article
id doaj-art-02f09b4996a84cf5871edee794b72d44
institution OA Journals
issn 1424-8220
language English
publishDate 2025-02-01
publisher MDPI AG
record_format Article
series Sensors
spelling doaj-art-02f09b4996a84cf5871edee794b72d442025-08-20T02:01:24ZengMDPI AGSensors1424-82202025-02-01254103510.3390/s25041035ResGRU: A Novel Hybrid Deep Learning Model for Compound Fault Diagnosis in Photovoltaic Arrays Considering Dust ImpactXi Liu0Hui Hwang Goh1Haonan Xie2Tingting He3Weng Kean Yew4Dongdong Zhang5Wei Dai6Tonni Agustiono Kurniawan7School of Electrical Engineering, Guangxi University, Nanning 530004, ChinaSchool of Engineering, Taylor’s University Lakeside Campus, Subang Jaya 47500, MalaysiaSchool of Electrical Engineering, Guangxi University, Nanning 530004, ChinaSchool of Electrical Engineering, Guangxi University, Nanning 530004, ChinaDepartment of Electrical Engineering, School of Engineering & Physical Sciences, Heriot-Watt University Malaysia Campus, Putrajaya 62200, MalaysiaSchool of Electrical Engineering, Guangxi University, Nanning 530004, ChinaSchool of Electrical Engineering, Guangxi University, Nanning 530004, ChinaCollege of the Environment and Ecology, Xiamen University, Xiamen 361102, ChinaWith the widespread deployment of photovoltaic (PV) power stations, timely identification and rectification of module defects are crucial for extending service life and preserving efficiency. PV arrays, subjected to severe outside circumstances, are prone to defects exacerbated by dust accumulation, potentially leading to complex compound faults. The resemblance between individual and compound faults sometimes leads to misclassification. To address this challenge, this paper presents a novel hybrid deep learning model, ResGRU, which integrates a residual network (ResNet) with bidirectional gated recurrent units (BiGRU) to improve fault diagnostic accuracy. Additionally, a Squeeze-and-Excitation (SE) module is incorporated to enhance relevant features while suppressing irrelevant ones, hence improving performance. To further optimize inter-class separability and intra-class compactness, a center loss function is employed as an auxiliary loss to enhance the model’s discriminative capacity. This proposed method facilitates the automated extraction of fault features from I-V curves and accurate diagnosis of individual faults, partial shading scenarios, and compound faults under varying levels of dust accumulation, hence aiding in the formulation of efficient cleaning schedules. Experimental findings indicate that the suggested model achieves 99.94% accuracy on pristine data and 98.21% accuracy on noisy data, markedly surpassing established techniques such as artificial neural networks (ANN), ResNet, random forests (RF), multi-scale SE-ResNet, and other ResNet-based approaches. Thus, the model offers a reliable solution for accurate PV array fault diagnosis.https://www.mdpi.com/1424-8220/25/4/1035photovoltaic (PV) systemfault diagnosisdust impactI-V curveresidual network (ResNet)bidirectional gated recurrent unit (BiGRU)
spellingShingle Xi Liu
Hui Hwang Goh
Haonan Xie
Tingting He
Weng Kean Yew
Dongdong Zhang
Wei Dai
Tonni Agustiono Kurniawan
ResGRU: A Novel Hybrid Deep Learning Model for Compound Fault Diagnosis in Photovoltaic Arrays Considering Dust Impact
Sensors
photovoltaic (PV) system
fault diagnosis
dust impact
I-V curve
residual network (ResNet)
bidirectional gated recurrent unit (BiGRU)
title ResGRU: A Novel Hybrid Deep Learning Model for Compound Fault Diagnosis in Photovoltaic Arrays Considering Dust Impact
title_full ResGRU: A Novel Hybrid Deep Learning Model for Compound Fault Diagnosis in Photovoltaic Arrays Considering Dust Impact
title_fullStr ResGRU: A Novel Hybrid Deep Learning Model for Compound Fault Diagnosis in Photovoltaic Arrays Considering Dust Impact
title_full_unstemmed ResGRU: A Novel Hybrid Deep Learning Model for Compound Fault Diagnosis in Photovoltaic Arrays Considering Dust Impact
title_short ResGRU: A Novel Hybrid Deep Learning Model for Compound Fault Diagnosis in Photovoltaic Arrays Considering Dust Impact
title_sort resgru a novel hybrid deep learning model for compound fault diagnosis in photovoltaic arrays considering dust impact
topic photovoltaic (PV) system
fault diagnosis
dust impact
I-V curve
residual network (ResNet)
bidirectional gated recurrent unit (BiGRU)
url https://www.mdpi.com/1424-8220/25/4/1035
work_keys_str_mv AT xiliu resgruanovelhybriddeeplearningmodelforcompoundfaultdiagnosisinphotovoltaicarraysconsideringdustimpact
AT huihwanggoh resgruanovelhybriddeeplearningmodelforcompoundfaultdiagnosisinphotovoltaicarraysconsideringdustimpact
AT haonanxie resgruanovelhybriddeeplearningmodelforcompoundfaultdiagnosisinphotovoltaicarraysconsideringdustimpact
AT tingtinghe resgruanovelhybriddeeplearningmodelforcompoundfaultdiagnosisinphotovoltaicarraysconsideringdustimpact
AT wengkeanyew resgruanovelhybriddeeplearningmodelforcompoundfaultdiagnosisinphotovoltaicarraysconsideringdustimpact
AT dongdongzhang resgruanovelhybriddeeplearningmodelforcompoundfaultdiagnosisinphotovoltaicarraysconsideringdustimpact
AT weidai resgruanovelhybriddeeplearningmodelforcompoundfaultdiagnosisinphotovoltaicarraysconsideringdustimpact
AT tonniagustionokurniawan resgruanovelhybriddeeplearningmodelforcompoundfaultdiagnosisinphotovoltaicarraysconsideringdustimpact