Ai-based smart real-time pv panels soiling recognizing system using deep neural network framework on nvidia jetson nano embedded gpu




HIGHLIGHTS
  • What: This study provides a clear direction for future research interested in applying machine_learning techniques to improve the reliability and maintenance of PV systems. The aim of the object detection target is to identify and classify visual objects in images or videos. In the context of this research paper, the object detection methods employed are based on the YOLOv8 algorithm proposed in a previous research work . The authors propose several architectural reforms and a series of enhancements, collectively known as a bag-of-freebies, to improve model performance.
  • Who: YOLOv and colleagues from the Laboratory of Advanced Systems Engineering National School of Applied Sciences, Ibn Tofail, have published the research work: AI-Based Smart Real-Time PV Panels Soiling Recognizing System Using Deep Neural Network Framework on NVIDIA Jetson Nano Embedded GPU, in the Journal: (JOURNAL) of 27/Dec/2023
  • How: This research paper introduces a novel solution to this issue by integrating the Deep Neural Network (DNN) with advanced image processing techniques for dust and soiling classification to suggest the appropriate cleaning intervention strategy. This research paper presents an innovative real-time approach that integrates YOLOv8 deep learning-based object detection techniques with image processing methods. For training the authors utilized Google Colab an Integrated To evaluate the performance of the model based on the enhanced YOLOv8 version the authors conducted comparative experiments with two widely used object detection models.
  • Future: This decision is relayed to the autonomous cleaning system to provide a choice of cleaning operations more details in future work. Future work will focus on exploring the implementation of the proposed system for autonomous Photovoltaic cleaning robots. In future research many improvement ideas can be discussed.
SUMMARY

    The proposed methodology is based on a novel approach that uses Denoising Convolutional Neural_Networks (DnCNN) for image analysis and evaluation. Three types of Neural_Networks for PV Fault Detection and Diagnosis were discussed: Sallow Neural_Networks, Deep Neural_Networks, and hybrid models, which combine ANNs with other machine_learning methods. This study provides a clear direction for future research interested in applying machine_learning techniques to improve the reliability and maintenance of PV systems. It introduces the status evaluation method based on the YOLOv8 algorithm and image processing approaches for soiling and dust detection. In the last several years, there has been substantial growth in the adoption of photovoltaic technologies and development in utilizing techniques of deep learning, especially Convolutional Neural_Networks (CNNs) and artificial_intelligence (AI), for evaluating the state of photovoltaic (PV) systems. The input image containing the objects to be detected (a) is preprocessed for the neural_network algorithm through the preprocessing step (b). The preprocessed image is passed through a Convolutional Neural_Network (CNN) architecture as its backbone (c). A prototype has been developed to assess and validate a tools and services for managing and annotating datasets for machine_learning. Development Environment (IDE) known for its support of machine_learning and deep learning tasks. Examples of dust image detection for two PV panels, PV1 and PV2: (a) input image, (b) HSV image, (c) Thresholded image for a range of dust colors, (d) Grayscale image, (e) Dust edge boundary image (f)(g) Eroding and Dilating image (h) % covered dust area object classes, exhibits a consistent decline, stabilizing around 3.0 for training and 4.0 for validation. The authors proposed a YOLOv8 framework to detect soiling on photovoltaic PV panels and perform operations on visible images and image processing techniques using the computer vision library OpenCV to extract the dust layer on the PV panel for the state PV panel evaluation. @@

LAY DEFINITIONS
  • Artificial neural networks: Artificial neural networks, usually simply called neural networks, are computing systems vaguely inspired by the biological neural networks that constitute animal brains. An ANN is based on a collection of connected units or nodes called artificial neurons, which loosely model the neurons in a biological brain
  • Convolutional neural network: In deep learning, a convolutional neural network (CNN, or ConvNet) is a class of deep neural networks, most commonly applied to analyzing visual imagery. They are also known as shift invariant or space invariant artificial neural networks (SIANN), based on their shared-weights architecture and translation invariance characteristics
  • Development environment: In software deployment an environment or tier is a computer system in which a computer program or software component is deployed and executed. In simple cases, such as developing and immediately executing a program on the same machine, there may be a single environment, but in industrial use the development environment and production environment are separated; often with several stages in between
  • Graphical user interface: The graphical user interface is a form of user interface that allows users to interact with electronic devices through graphical icons and audio indicator such as primary notation, instead of text-based user interfaces, typed command labels or text navigation. GUIs were introduced in reaction to the perceived steep learning curve of command-line interfaces, which require commands to be typed on a computer keyboard
  • Artificial intelligence: Theory and development of COMPUTER SYSTEMS which perform tasks that normally require human intelligence. Such tasks may include speech recognition, LEARNING; VISUAL PERCEPTION; MATHEMATICAL COMPUTING; reasoning, PROBLEM SOLVING, DECISION-MAKING, and translation of language

Licence: cc-by

Site reference: https://www.iieta.org/download/file/fid/145255

DOI reference: https://www.doi.org/10.18280/isi.290503

Summary powered by SciencePOD SUMMSci Version version 5.5 (C) 2023 Context powered by www.wikipedia.org

This summary is a productivity tool designed to help quickly identify suitable research studies. Always refer to the full-text article for further information.




source https://magazine.sciencepod.net/ai-based-smart-real-time-pv-panels-soiling-recognizing-system-using-deep-neural-network-framework-on-nvidia-jetson-nano-embedded-gpu/

Comments

Post a Comment

Popular posts from this blog

Exploring the bilayer approach to analyzing lead-free perovskite solar cells` numerical aspects for high efficiency

HIGHLIGHTS What: According to this paper a volume lead-free perovskite structure is quantitatively investigated using 1D solar cell capacitance simulator (SCAPS) software. Developing the proposed device structure in an experimental environment is the aim of this preliminary study. In this work, the amphoteric defect with uniform energy distribution above the absorber layer`s Ev was investigated. Who: Shahid Farhan et al. from the Department of Electrical Engineering, Government College University, Pakistan have published the research: Exploring the Bilayer Approach to Analyzing Lead-Free Perovskite Solar Cells` Numerical Aspects for High Efficiency, in the Journal: (JOURNAL) Future: The expected results of this research are expected to provide significant perspectives for the future research and manufacturing of lead-free PSCs with an emphasis on their commercial feasibility. SUMMARY Transport layer (ETL), a perovskite absorber layer, and a hole transport layer (HTL) a...
Read more

Can europe recover its photovoltaic manufacturing capacity?

HIGHLIGHTS What: Figure 2 below illustrates the changing dynamics of the global PV market, and the influence of the Chinese PV market, 2 For the purpose of this report, IEA PVPS countries are those that are either member in their own right or through the adhesion of the EC. but also the fast pace of growth in India and emerging countries. Who: Melodie de l`Epine from the Japan Japan have published the Article: Snapshot of Global PV Markets -, in the Journal: (JOURNAL) of 23/04/2023 Future: Repowering is still relatively unusual given the age of the oldest installations but it is expected to increase in the near future – serial defects with backsheets manufactured in the period 2009 – 2011 is a good example as the past 2 years have seen a few hundred MW replaced. As penetration rates increase new governance models compatible with market and climate policy driven deployment targets will need to be established to ensure PV can be smoothly deployed. SUMMARY Fc PVPS Task 1 ...
Read more

A panorama of brazilian social housing research: scope, gaps and intersections

HIGHLIGHTS What: The SR process was developed following four steps (Barat et_al, 2017; Cahyo, 2021) presented in Figure 1: (a) research motivation: the state of the art of Brazilian SHs was the research motivation. The results of this analysis are presented in Results and discussion item; and_(d) analysis and synthesis: existing studies developed in Brazil were identified and contextualised. Who: Social Housing. MCMV. Housing et al. from the Universidade Estadualvulnerability of low-income populationsHowever, studies developed in Brazil have published the research work: Ambiente Construído: resiliente e sustentável, in the Journal: (JOURNAL) How: Strings and sample analysis are presented in Figure 2. Future: In the absence of changes in the production of SH in which public authorities and users could lead the main decisions on design and location many of the recurrent problems will be repeated in future projects. SUMMARY The aim of SLRs is to gather empirical studies...
Read more