Enhanced Fault Detection in Photovoltaic Panels Using CNN-Based Classification with PyQt5 Implementation DOI Creative Commons
Younes Ledmaoui, Adila El Maghraoui, Mohamed El Aroussi

et al.

Sensors, Journal Year: 2024, Volume and Issue: 24(22), P. 7407 - 7407

Published: Nov. 20, 2024

Solar photovoltaic systems have increasingly become essential for harvesting renewable energy. However, as these grow in prevalence, the issue of end life modules is also increasing. Regular maintenance and inspection are vital to extend lifespan systems, minimize energy losses, protect environment. This paper presents an innovative explainable AI model detecting anomalies solar panels using enhanced convolutional neural network (CNN) VGG16 architecture. The effectively identifies physical electrical changes, such dust bird droppings, implemented PyQt5 Python tool create a user-friendly interface that facilitates decision-making users. Key processes included dataset balancing through oversampling data augmentation expand dataset. achieved impressive performance metrics: 91.46% accuracy, 98.29% specificity, F1 score 91.67%. Overall, it enhances power generation efficiency prolongs while minimizing environmental risks.

Language: Английский

DropletCoin DropletCoin: Pioneering Sustainable AI and Emerging Technologies through Blockchain Innovation DOI Creative Commons

Tyrone Moodley

Published: Jan. 5, 2025

DropletCoin represents an innovative fusion of blockchain technology and renewable energy solutions, targeting the substantial demands AI emerging technologies. This paper presents UMD v3.0 IoT device, designed for logging solar production, its seamless integration with Dandelion Blockchain efficient data capture processing. By utilizing tokenized credits IoT-based monitoring, enables decentralized, carbon-neutral computing networks. Findings reveal a 30% reduction in costs 40% decrease carbon emissions smart city applications.

Language: Английский

Citations

0
Predicting the Temperature of a Permanent Magnet Synchronous Motor: A Comparative Study of Artificial Neural Network Algorithms DOI Creative Commons
Nabil El Bazi, Nasr Guennouni, Mohcin Mekhfioui

et al.

Technologies, Journal Year: 2025, Volume and Issue: 13(3), P. 120 - 120

Published: March 17, 2025

The accurate prediction of temperature in Permanent Magnet Synchronous Motors (PMSMs) has always been essential for monitoring performance and enabling predictive maintenance the industrial sector. This study examines efficiency a set artificial neural network (ANN) models, namely Multilayer Perceptron (MLP), Long Short-Term Memory (LSTM), Recurrent Neural Network (RNN), Convolutional (CNN), predicting Temperature. A comparative evaluation is conducted using common indicators, including root mean square error (RMSE), absolute (MAE), coefficient determination (R2), to assess accuracy each model. intent identify most favorable model that balances high with low computational cost.

Language: Английский

Citations

0
Faults Detection and Diagnosis of a Large-Scale PV System by Analyzing Power Losses and Electric Indicators Computed Using Random Forest and KNN-Based Prediction Models DOI Creative Commons

Yasmine Gaaloul,

Olfa Bel Hadj Brahim Kechiche, Houcine Oudira

et al.

Energies, Journal Year: 2025, Volume and Issue: 18(10), P. 2482 - 2482

Published: May 12, 2025

Accurate and reliable fault detection in photovoltaic (PV) systems is essential for optimizing their performance durability. This paper introduces a novel approach diagnosis large-scale PV systems, utilizing power loss analysis predictive models based on Random Forest (RF) K-Nearest Neighbors (KNN) algorithms. The proposed methodology establishes baseline model of the system’s healthy behavior under normal operating conditions, enabling real-time deviations between expected actual performance. Faults such as string disconnections, module short-circuits, shading effects have been identified using two key indicators: current error (Ec) voltage (Ev). By focusing losses indicator, this method provides high-accuracy without requiring extensive labeled data, significant advantage where data acquisition can be challenging. Additionally, contribution work identification correction faulty sensors, specifically pyranometer misalignment, which leads to inaccurate irradiation measurements disrupts diagnosis. ensures input models, RF achieved an R2 0.99657 prediction 0.99459 prediction, while KNN reached 0.99674 estimation, improving both accuracy overall outlined was experimentally validated real-world from 500 kWp grid-connected system Ain El Melh, Algeria. results demonstrate that innovative offers efficient, scalable solution detection, enhancing reliability large reducing maintenance costs.

Language: Английский

Citations

0
Enhanced Fault Detection in Photovoltaic Panels Using CNN-Based Classification with PyQt5 Implementation DOI Creative Commons
Younes Ledmaoui, Adila El Maghraoui, Mohamed El Aroussi

et al.

Sensors, Journal Year: 2024, Volume and Issue: 24(22), P. 7407 - 7407

Published: Nov. 20, 2024

Solar photovoltaic systems have increasingly become essential for harvesting renewable energy. However, as these grow in prevalence, the issue of end life modules is also increasing. Regular maintenance and inspection are vital to extend lifespan systems, minimize energy losses, protect environment. This paper presents an innovative explainable AI model detecting anomalies solar panels using enhanced convolutional neural network (CNN) VGG16 architecture. The effectively identifies physical electrical changes, such dust bird droppings, implemented PyQt5 Python tool create a user-friendly interface that facilitates decision-making users. Key processes included dataset balancing through oversampling data augmentation expand dataset. achieved impressive performance metrics: 91.46% accuracy, 98.29% specificity, F1 score 91.67%. Overall, it enhances power generation efficiency prolongs while minimizing environmental risks.

Language: Английский

Citations

0