Tripathi et al. (2026) Performance Evaluation of a Machine Learning Based Framework for Solar Irradiance Prediction

Identification

• Journal: Lecture notes in networks and systems
• Year: 2026
• Date: 2026-01-01
• Authors: Arpit Tripathi, Aabhya Jain, Oshin Sharma
• DOI: 10.1007/978-3-032-15401-9_11

Research Groups

Department of Computer Science and Engineering, SRM Institute of Science and Technology, Delhi NCR Campus, Modinagar, Ghaziabad, Uttar Pradesh, India

Short Summary

This study evaluates machine learning models (XGBoost, MLP) for solar irradiance prediction using meteorological data from the HI-SEAS weather station, demonstrating XGBoost's superior performance for precise forecasts.

Objective

• To analyze and predict solar irradiance using machine learning algorithms based on meteorological data from the HI-SEAS weather station.

Study Configuration

• Spatial Scale: HI-SEAS weather station
• Temporal Scale: September 2016 to December 2016

Methodology and Data

• Models used: XGBoost, Multi-Layer Perceptron (MLP). Feature selection techniques: SelectKBest, Extra Tree Classifier.
• Data sources: Meteorological data from the HI-SEAS weather station, including temperature, humidity, pressure, wind speed, and cloud cover.

Main Results

• XGBoost outperformed MLP in solar irradiance prediction.
• XGBoost achieved a Root Mean Squared Error (RMSE) of 81.45 W/m², a Mean Absolute Error (MAE) of 32.43 W/m², and an R-squared score of 0.93.
• MLP achieved an RMSE of 100.06 W/m², an MAE of 43.75 W/m², and an R-squared score of 0.89.
• Cloud cover and temperature were identified as crucial factors for predicting solar irradiance.

Contributions

• Demonstrated the superior performance of XGBoost over Multi-Layer Perceptron for solar irradiance prediction using real-world meteorological data.
• Identified key meteorological features (cloud cover and temperature) that are crucial for accurate solar irradiance forecasting.
• Provided a machine learning-based framework capable of producing precise solar irradiance forecasts, which can enhance renewable energy systems.

Funding

• Not explicitly mentioned in the paper.

Citation

@article{Tripathi2026Performance,
  author = {Tripathi, Arpit and Jain, Aabhya and Sharma, Oshin},
  title = {Performance Evaluation of a Machine Learning Based Framework for Solar Irradiance Prediction},
  journal = {Lecture notes in networks and systems},
  year = {2026},
  doi = {10.1007/978-3-032-15401-9_11},
  url = {https://doi.org/10.1007/978-3-032-15401-9_11}
}