Comparative Analysis of AI-Driven Machine Learning Models for Fault Detection and Maintenance Optimization in Photovoltaic Systems DOI Creative Commons

Abdellahi Moulaye Rchid,

Moussa Attia, Mohamed Basyony

et al.

Solar Energy and Sustainable Development, Journal Year: 2025, Volume and Issue: 14(1), P. 361 - 378

Published: April 28, 2025

With the increasing adoption of solar photovoltaic (PV) systems, ensuring their reliability and efficiency is crucial for sustainable energy production. However, traditional fault detection methods rely on expensive manual inspections or sensor-based monitoring, often slow inefficient. This study aims to bridge this gap by leveraging machine learning techniques enhance maintenance optimization in PV systems. We evaluate five advanced models—Random Forest, XGBoost, Artificial Neural Networks (ANN), Convolutional (CNN), Support Vector Machines (SVM)—using accurate operational data from a 250-kW power station. The dataset includes key parameters such as current, voltage, output, temperature, irradiance. Data preprocessing included outlier removal, feature selection via Pearson correlation, normalization improve model performance. models were trained tested using an 80-20 split evaluated based classification accuracy, precision, recall, F1-score. Our results show that XGBoost achieved highest accuracy (88%), making it best candidate real-time predictive maintenance. Random Forest also performed well (87% accuracy), particularly handling noisy data. ANN CNN effectively detected long-term degradation patterns, supporting preventive strategies. Based these findings, we propose dual strategy: detection, while monitor gradual system deterioration. research provides practical framework integrating into management, offering scalable solution reliability, reduce costs, optimize efficiency.

Language: Английский

Langzeitbeobachtungen des Bodenwasserhaushalts in Österreich und ihr Wert in Gegenwart und Zukunft DOI
Thomas Weninger,

Verena Jagersberger,

Valentina Pelzmann

et al.

Österreichische Wasser- und Abfallwirtschaft, Journal Year: 2025, Volume and Issue: unknown

Published: March 17, 2025

Citations

0
Deep learning for efficient high-resolution image processing: A systematic review DOI Creative Commons
Albert Dede, Henry Nunoo‐Mensah, Eric Tutu Tchao

et al.

Intelligent Systems with Applications, Journal Year: 2025, Volume and Issue: unknown, P. 200505 - 200505

Published: March 1, 2025

Language: Английский

Citations

0
Development of Mathematical and Computational Models for Predicting Agricultural Soil–Water Management Properties (ASWMPs) to Optimize Intelligent Irrigation Systems and Enhance Crop Resilience DOI Creative Commons
Brigitta Tóth, Oswaldo Guerrero-Bustamante, Michel Murillo

et al.

Agronomy, Journal Year: 2025, Volume and Issue: 15(4), P. 942 - 942

Published: April 12, 2025

Soil–water management is fundamental to plant ecophysiology, directly affecting resilience under both anthropogenic and natural stresses. Understanding Agricultural Soil–Water Management Properties (ASWMPs) therefore essential for optimizing water availability, enhancing harvest resilience, enabling informed decision-making in intelligent irrigation systems, particularly the face of climate variability soil degradation. In this regard, present research develops predictive models ASWMPs based on grain size distribution dry bulk density soils, integrating traditional mathematical approaches advanced computational techniques. By examining 900 samples from NaneSoil database, spanning diverse crop species (Avena sativa L., Daucus carota Hordeum vulgare Medicago Phaseolus vulgaris Sorghum Pers., Triticum aestivum Zea mays L.), several are proposed three key ASWMPs: soil-saturated hydraulic conductivity, field capacity, permanent wilting point. Mathematical demonstrate high accuracy (71.7–96.4%) serve as practical agronomic tools but limited capturing complex soil–plant-water interactions. Meanwhile, a Deep Neural Network (DNN)-based model significantly enhances performance (91.4–99.7% accuracy) by uncovering nonlinear relationships that govern moisture retention availability. These findings contribute precision agriculture providing robust soil–water management, ultimately supporting against environmental challenges such drought, salinization, compaction.

Language: Английский

Citations

0
A Daily Reference Crop Evapotranspiration Forecasting Model Based on Improved Informer DOI Creative Commons

Jiangjie Pan,

Long Yu, Bo Zhou

et al.

Agriculture, Journal Year: 2025, Volume and Issue: 15(9), P. 933 - 933

Published: April 25, 2025

Daily reference crop evapotranspiration (ET0) is crucial for precision irrigation management, yet traditional prediction methods struggle to capture its dynamic variations due the complexity and nonlinearity of meteorological conditions. To address this, we propose an Improved Informer model enhance ET0 accuracy, providing a scientific basis agricultural water management. Using soil data from Yingde region, employed Maximal Information Coefficient (MIC) identify key influencing factors integrated Residual Cycle Forecasting (RCF), Star Aggregate Redistribute (STAR), Fully Adaptive Normalization (FAN) techniques into model. MIC analysis identified total shortwave radiation, sunshine duration, maximum temperature at 2 m, 28–100 cm depth, surface pressure as optimal features. Under five-feature scenario (S3), improved achieved superior performance compared Long Short-Term Memory (LSTM) original models, with MAE reduced 0.065 (LSTM: 0.637, Informer: 0.171) MSE 0.007 0.678, 0.060). The inference time was also by 31%, highlighting enhanced computational efficiency. effectively captures periodic nonlinear characteristics ET0, offering novel solution management significant practical implications.

Language: Английский

Citations

0
Comparative Analysis of AI-Driven Machine Learning Models for Fault Detection and Maintenance Optimization in Photovoltaic Systems DOI Creative Commons

Abdellahi Moulaye Rchid,

Moussa Attia, Mohamed Basyony

et al.

Solar Energy and Sustainable Development, Journal Year: 2025, Volume and Issue: 14(1), P. 361 - 378

Published: April 28, 2025

With the increasing adoption of solar photovoltaic (PV) systems, ensuring their reliability and efficiency is crucial for sustainable energy production. However, traditional fault detection methods rely on expensive manual inspections or sensor-based monitoring, often slow inefficient. This study aims to bridge this gap by leveraging machine learning techniques enhance maintenance optimization in PV systems. We evaluate five advanced models—Random Forest, XGBoost, Artificial Neural Networks (ANN), Convolutional (CNN), Support Vector Machines (SVM)—using accurate operational data from a 250-kW power station. The dataset includes key parameters such as current, voltage, output, temperature, irradiance. Data preprocessing included outlier removal, feature selection via Pearson correlation, normalization improve model performance. models were trained tested using an 80-20 split evaluated based classification accuracy, precision, recall, F1-score. Our results show that XGBoost achieved highest accuracy (88%), making it best candidate real-time predictive maintenance. Random Forest also performed well (87% accuracy), particularly handling noisy data. ANN CNN effectively detected long-term degradation patterns, supporting preventive strategies. Based these findings, we propose dual strategy: detection, while monitor gradual system deterioration. research provides practical framework integrating into management, offering scalable solution reliability, reduce costs, optimize efficiency.

Language: Английский

Citations

0