Long-term comparative analysis of machine learning models: A deep dive into applications of artificial intelligence for enhancing photovoltaic performance prediction

This study tackles a key research gap by applying comparative analysis on several well-known classic machine learning models to predict photovoltaic performance under variable environmental conditions and dust levels. It uses a year-long dataset that includes environmental factors such as irradiance...

Full description

Saved in:
Bibliographic Details
Main Authors: Ali Akbar Yaghoubi, Mahdi Gandomzadeh, Aslan Gholami, Roghayeh Gavagsaz-Ghoachani, Majid Zandi
Format: Article
Language:English
Published: Elsevier 2025-09-01
Series:International Journal of Electrical Power & Energy Systems
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S0142061525004144
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849235955575160832
author Ali Akbar Yaghoubi
Mahdi Gandomzadeh
Aslan Gholami
Roghayeh Gavagsaz-Ghoachani
Majid Zandi
author_facet Ali Akbar Yaghoubi
Mahdi Gandomzadeh
Aslan Gholami
Roghayeh Gavagsaz-Ghoachani
Majid Zandi
author_sort Ali Akbar Yaghoubi
collection DOAJ
description This study tackles a key research gap by applying comparative analysis on several well-known classic machine learning models to predict photovoltaic performance under variable environmental conditions and dust levels. It uses a year-long dataset that includes environmental factors such as irradiance, ambient and module temperatures, wind speed, humidity, precipitation, and dust accumulation, along with electrical outputs like voltage, current, and power. Data were collected from two PV systems, one regularly cleaned and one left naturally soiled, at the same operating conditions. Six supervised machine learning models were trained: K-Nearest Neighbors (KNN), Cross-Decomposition Regression, Decision Trees, Support Vector Regression (SVR), Multi-Layer Perceptron (MLP), and Ensemble Learning. The models were trained to predict differences in voltage, current, and power between the two systems, achieving R-squared scores over 85% in most cases. KNN and Decision Trees performed best with fewer neighbors and shallow tree depths though they might suffer from computational costs in terms of large input data. SVR and MLP showed high sensitivity to hyperparameters such as activation functions and iteration limits, respectively. Although most ensemble methods performed well, Adaboost’s accuracy declined as the number of estimators increased. The outcomes of this study offer practical insights for real-world applications, including predictive maintenance and energy optimization. The predictive models developed herein can support enhanced PV system design, the prediction of soiling effects using environmental data, optimized dust removal methods and cleaning schedules, and data-driven energy policy planning.
format Article
id doaj-art-8f42787c2d0943138a2bf79f0ec66caf
institution Kabale University
issn 0142-0615
language English
publishDate 2025-09-01
publisher Elsevier
record_format Article
series International Journal of Electrical Power & Energy Systems
spelling doaj-art-8f42787c2d0943138a2bf79f0ec66caf2025-08-20T04:02:32ZengElsevierInternational Journal of Electrical Power & Energy Systems0142-06152025-09-0117011086610.1016/j.ijepes.2025.110866Long-term comparative analysis of machine learning models: A deep dive into applications of artificial intelligence for enhancing photovoltaic performance predictionAli Akbar Yaghoubi0Mahdi Gandomzadeh1Aslan Gholami2Roghayeh Gavagsaz-Ghoachani3Majid Zandi4Department of Renewable Energy Engineering, Shahid Beheshti University, IranDepartment of Renewable Energy Engineering, Shahid Beheshti University, IranDepartment of Renewable Energy Engineering, Shahid Beheshti University, IranDepartment of Renewable Energy Engineering, Shahid Beheshti University, IranCorresponding author.; Department of Renewable Energy Engineering, Shahid Beheshti University, IranThis study tackles a key research gap by applying comparative analysis on several well-known classic machine learning models to predict photovoltaic performance under variable environmental conditions and dust levels. It uses a year-long dataset that includes environmental factors such as irradiance, ambient and module temperatures, wind speed, humidity, precipitation, and dust accumulation, along with electrical outputs like voltage, current, and power. Data were collected from two PV systems, one regularly cleaned and one left naturally soiled, at the same operating conditions. Six supervised machine learning models were trained: K-Nearest Neighbors (KNN), Cross-Decomposition Regression, Decision Trees, Support Vector Regression (SVR), Multi-Layer Perceptron (MLP), and Ensemble Learning. The models were trained to predict differences in voltage, current, and power between the two systems, achieving R-squared scores over 85% in most cases. KNN and Decision Trees performed best with fewer neighbors and shallow tree depths though they might suffer from computational costs in terms of large input data. SVR and MLP showed high sensitivity to hyperparameters such as activation functions and iteration limits, respectively. Although most ensemble methods performed well, Adaboost’s accuracy declined as the number of estimators increased. The outcomes of this study offer practical insights for real-world applications, including predictive maintenance and energy optimization. The predictive models developed herein can support enhanced PV system design, the prediction of soiling effects using environmental data, optimized dust removal methods and cleaning schedules, and data-driven energy policy planning.http://www.sciencedirect.com/science/article/pii/S0142061525004144Solar energyPhotovoltaic systemsPerformanceDustSoilingMachine learning
spellingShingle Ali Akbar Yaghoubi
Mahdi Gandomzadeh
Aslan Gholami
Roghayeh Gavagsaz-Ghoachani
Majid Zandi
Long-term comparative analysis of machine learning models: A deep dive into applications of artificial intelligence for enhancing photovoltaic performance prediction
International Journal of Electrical Power & Energy Systems
Solar energy
Photovoltaic systems
Performance
Dust
Soiling
Machine learning
title Long-term comparative analysis of machine learning models: A deep dive into applications of artificial intelligence for enhancing photovoltaic performance prediction
title_full Long-term comparative analysis of machine learning models: A deep dive into applications of artificial intelligence for enhancing photovoltaic performance prediction
title_fullStr Long-term comparative analysis of machine learning models: A deep dive into applications of artificial intelligence for enhancing photovoltaic performance prediction
title_full_unstemmed Long-term comparative analysis of machine learning models: A deep dive into applications of artificial intelligence for enhancing photovoltaic performance prediction
title_short Long-term comparative analysis of machine learning models: A deep dive into applications of artificial intelligence for enhancing photovoltaic performance prediction
title_sort long term comparative analysis of machine learning models a deep dive into applications of artificial intelligence for enhancing photovoltaic performance prediction
topic Solar energy
Photovoltaic systems
Performance
Dust
Soiling
Machine learning
url http://www.sciencedirect.com/science/article/pii/S0142061525004144
work_keys_str_mv AT aliakbaryaghoubi longtermcomparativeanalysisofmachinelearningmodelsadeepdiveintoapplicationsofartificialintelligenceforenhancingphotovoltaicperformanceprediction
AT mahdigandomzadeh longtermcomparativeanalysisofmachinelearningmodelsadeepdiveintoapplicationsofartificialintelligenceforenhancingphotovoltaicperformanceprediction
AT aslangholami longtermcomparativeanalysisofmachinelearningmodelsadeepdiveintoapplicationsofartificialintelligenceforenhancingphotovoltaicperformanceprediction
AT roghayehgavagsazghoachani longtermcomparativeanalysisofmachinelearningmodelsadeepdiveintoapplicationsofartificialintelligenceforenhancingphotovoltaicperformanceprediction
AT majidzandi longtermcomparativeanalysisofmachinelearningmodelsadeepdiveintoapplicationsofartificialintelligenceforenhancingphotovoltaicperformanceprediction