PV Abnormal Shading Detection Based on Convolutional Neural Network

Abnormal shading on the surface of photovoltaic panels will seriously damage its power generation efficiency and life. This paper established an abnormal shading detection system for photovoltaic panels. First, remove the photovoltaic panel and find the shading. Finally, the object detection algorithm is used to identify and classify the shading. Experimental results show that the system’s abnormal shading classification accuracy is 94.5%, which is 23% higher than the old method.

[1]  Efstratios I. Batzelis,et al.  Energy models for photovoltaic systems under partial shading conditions: a comprehensive review , 2015 .

[2]  L. Seguier,et al.  A proposed graphical electrical signatures supervision method to study PV module failures , 2015 .

[3]  Shengxian Cao,et al.  Cleaning cycle optimization and cost evaluation of module dust for photovoltaic power plants in China , 2019, Clean Technologies and Environmental Policy.

[4]  Ali Farhadi,et al.  YOLO9000: Better, Faster, Stronger , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[5]  Mohd Amran Mohd Radzi,et al.  Power loss due to soiling on solar panel: A review , 2016 .

[6]  Kaiming He,et al.  Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks , 2015, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[7]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[8]  Shaocheng Tong,et al.  Neural Networks-Based Adaptive Finite-Time Fault-Tolerant Control for a Class of Strict-Feedback Switched Nonlinear Systems , 2019, IEEE Transactions on Cybernetics.

[9]  Qiang Yang,et al.  Transfer Learning based Photovoltaic Module Defect Diagnosis using Aerial Images , 2018, 2018 International Conference on Power System Technology (POWERCON).

[10]  Ali Farhadi,et al.  You Only Look Once: Unified, Real-Time Object Detection , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[11]  Huaguang Zhang,et al.  Identification method for a class of periodic discrete-time dynamic nonlinear systems based on Sinusoidal ESN , 2018, Neurocomputing.

[12]  Shivkumar Kalyanaraman,et al.  DeepSolarEye: Power Loss Prediction and Weakly Supervised Soiling Localization via Fully Convolutional Networks for Solar Panels , 2017, 2018 IEEE Winter Conference on Applications of Computer Vision (WACV).

[13]  Claudia Buerhop,et al.  Faults and infrared thermographic diagnosis in operating c-Si photovoltaic modules: A review of research and future challenges , 2016 .

[14]  Kaiming He,et al.  Feature Pyramid Networks for Object Detection , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[15]  Kun Liu,et al.  Solar cell surface defect inspection based on multispectral convolutional neural network , 2018, Journal of Intelligent Manufacturing.

[16]  Y. El Basri,et al.  A shadow fault detection method based on the standard error analysis of I-V curves , 2016 .

[17]  Yan-Jun Liu,et al.  Adaptive neural network-based control for a class of nonlinear pure-feedback systems with time-varying full state constraints , 2018, IEEE/CAA Journal of Automatica Sinica.

[18]  D. Cormode,et al.  PV system power loss and module damage due to partial shade and bypass diode failure depend on cell behavior in reverse bias , 2015, 2015 IEEE 42nd Photovoltaic Specialist Conference (PVSC).

[19]  Bill Marion,et al.  Performance and Economic Analysis of Distributed Power Electronics in Photovoltaic Systems , 2011 .

[20]  Ali Farhadi,et al.  YOLOv3: An Incremental Improvement , 2018, ArXiv.