aiqbench analysis
This research addresses the critical challenge of solar PV forecasting in data-scarce regions caused by load shedding by integrating transfer learning with Conformalized Quantile Regression. The study demonstrates that leveraging source-domain data significantly improves both point prediction accuracy and uncertainty quantification reliability for target domains with limited historical observations. By reducing RMSE and narrowing prediction intervals, this approach offers a robust methodological advance for deploying AI-driven energy models in unstable grid environments.
Transfer Learning with Conformalized Quantile Regression for Solar PV Forecasting Under Load-Shedding-Driven Data Scarcity
arXiv:2609.26959v1 Announce Type: new Abstract: Solar photovoltaic (PV) forecasting in regions affected by load shedding is challenging because reliable historical observations are scarce. This study proposes a transfer learning framework combined with Conformalized Quantile Regression (CQR) to improve PV power forecasting and provide reliable uncertainty estimates under severe data scarcity. A source-domain PV dataset from Alice Springs, Australia, is used to pretrain a temporal forecasting model, which is then adapted to simulated Bangladesh PV data representing different levels of historical availability. Experimental results show that transfer learning reduces RMSE by up to 23.7% when only one month of target-domain data is available and by 13.7% with three months of data. The proposed Transfer Learning plus CQR framework achieves 94.3% empirical coverage with three months of target data while producing prediction intervals that are 14% narrower than those obtained without transfer learning. These results demonstrate that combining transfer learning with conformal uncertainty quantification can improve both point forecasting accuracy and uncertainty reliability when target-domain PV data are severely limited.
Abstract:Solar photovoltaic (PV) forecasting in regions affected by load shedding is challenging because reliable historical observations are scarce. This study proposes a transfer learning framework combined with Conformalized Quantile Regression (CQR) to improve PV power forecasting and provide reliable uncertainty estimates under severe data scarcity. A source-domain PV dataset from Alice Springs, Australia, is used to pretrain a temporal forecasting model, which is then adapted to simulated Bangladesh PV data representing different levels of historical availability. Experimental results show that transfer learning reduces RMSE by up to 23.7% when only one month of target-domain data is available and by 13.7% with three months of data. The proposed Transfer Learning plus CQR framework achieves 94.3% empirical coverage with three months of target data while producing prediction intervals that are 14% narrower than those obtained without transfer learning. These results demonstrate that combining transfer learning with conformal uncertainty quantification can improve both point forecasting accuracy and uncertainty reliability when target-domain PV data are severely limited.
Machine Learning (cs.LG)
arXiv:2609.26959 [cs.LG]
(or arXiv:2609.26959v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2609.26959 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 257 words
Related coverage
Report the Floor: A Training-Free Conformal Interval Is a Mandatory Baseline for Probabilistic Time-Series Forecasting
arXiv · Jun 9, 2026
PROSWIN: Probabilistic Solar Wind Speed Forecasting Using Deep Distributional Regression From Solar Images
arXiv · Sep 23, 2026
An Uncertainty-Aware Hybrid Mathematical-Machine-Learning Model for Smart Irrigation Decision Support
arXiv · Sep 15, 2026
Beyond Point Estimates: Benchmarking Uncertainty Quantification Methods on the AION-1 Astronomical Foundation Model
arXiv · Jun 9, 2026