Laser & Optoelectronics Progress, Volume. 60, Issue 24, 2412005(2023)

Infrared Image Fault Detection of Photovoltaic Modules Based on Residual Photovoltaic Network

Mingzheng Sun and Hao Li*
Author Affiliations
  • School of Earth Science and Engineering, Hohai University, Nanjing 211100, Jiangsu, China
  • show less
    Figures & Tables(17)
    Model of ResPNet
    Model of Residual block
    Model of Respblock
    Model of Gblock
    Ensemble neural network framework for prediction of photovoltaic modules
    Training loss of each model
    Validation accuracy of each model
    Feature layer visualization information of underlying feature information enhancement module. (a) Feature maps information of ResNet-50; (b) feature maps information of ResPNet[D]
    Feature layer visualization information of Respblock module. (a) Feature maps information of ResPNet[D]; (b) feature maps information of ResPNet[P]
    Feature layer visualization information of Gblock module. (a) Feature maps information of ResPNet[P]; (b) feature maps information of ResPNet[G]
    Test results of different photovoltaic module fault detection models. (a) ResNet; (b) ResPNet
    Test results of cascaded photovoltaic module fault detection
    • Table 1. Description of photovoltaic module and dataset partitioning

      View table

      Table 1. Description of photovoltaic module and dataset partitioning

      Fault nameFault descriptionNumber of imagesNumber of images in train and validation datasetNumber of images in test dataset
      CellThere is a single square hot spot in the photovoltaic module18771689188
      Cell-multiThere are multiple square hot spots in the photovoltaic module12881160128
      CrackingCracks exist on the surface of the photovoltaic module94084595
      Hot-spotA single hot spot exists on the surface of the photovoltaic module24922425
      Hot-spot-multiThere are multiple hot spots on the surface of the photovoltaic module24622224
      ShadowingShadows created by plants,artifacts between adjacent rows1056950106
      DiodeBypass diode short circuit causes that 1/3 photovoltaic module cannot operate normally14991349150
      Diode-multiThe short circuit of the bypass diode causes that 2/3 of the photovoltaic module cannot operate normally17515817
      VegetationThe plants shade the photovoltaic module16391476163
      SoilingDust and garbage may block the surface of the photovoltaic module20418321
      Offline-moduleThe entire photovoltaic module fails to operate properly82774483
      No-anomalyTrouble-free1000090001000
    • Table 2. Type of photovoltaic module images

      View table

      Table 2. Type of photovoltaic module images

      Fault nameImage data
      Cell
      Cell-multi
      Cracking
      Hot-spot
      Hot-spot-multi
      Shadowing
      Diode
      Diode-multi
      Vegetation
      Soiling
      Offline-module
      No-anomaly
    • Table 3. Ablation experiment results of infrared image classification accuracy of photovoltaic modules

      View table

      Table 3. Ablation experiment results of infrared image classification accuracy of photovoltaic modules

      DatasetModelAc /%
      Infrared Solar ModulesResNet-5083.2
      ResPNet[D]83.8
      ResPNet[P]84.4
      ResPNet[G]84.6
    • Table 4. Model accuracy of infrared image classification in photovoltaic modules

      View table

      Table 4. Model accuracy of infrared image classification in photovoltaic modules

      ModelAc /%Ap /%Ar /%FF1-Score /%Speed /ms
      ShuffleNetV21976.859.854.156.84.9
      MobileNetV22078.363.057.259.94.0
      ResNet-501483.274.267.970.95.1
      ResNet-1011483.574.367.070.412.3
      EfficientNetV72181.468.763.666.126.9
      Alves et al.模型1178.8
      CNN-SVM1282.9
      Le et al.模型1384.1
      ResPNet[G]84.675.867.671.530.8
    • Table 5. Integrated models for infrared image classification accuracy of photovoltaic module

      View table

      Table 5. Integrated models for infrared image classification accuracy of photovoltaic module

      ModelNo. of modelAc /%
      Proposed by Le et al. 131585.9
      ResPNet[G]385.9
    Tools

    Get Citation

    Copy Citation Text

    Mingzheng Sun, Hao Li. Infrared Image Fault Detection of Photovoltaic Modules Based on Residual Photovoltaic Network[J]. Laser & Optoelectronics Progress, 2023, 60(24): 2412005

    Download Citation

    EndNote(RIS)BibTexPlain Text
    Save article for my favorites
    Paper Information

    Category: Instrumentation, Measurement and Metrology

    Received: Mar. 22, 2023

    Accepted: Apr. 20, 2023

    Published Online: Nov. 27, 2023

    The Author Email: Li Hao (lihao@hhu.edu.cn)

    DOI:10.3788/LOP230912

    Topics