Author
Abstract
The rapid expansion of solar and wind power presents challenges for grid reliability and economic operation due to inherent resource variability, potentially leading to inefficient storage utilization, increased curtailment, and elevated system costs. This study proposes a cross-framework hybrid artificial intelligence approach that combines autoregressive integrated moving average (ARIMA) models for solar forecasting with long short-term memory (LSTM) networks for wind prediction, integrated through a multi-regional forecasting hub that informs reinforcement learning-based storage dispatch and region-specific policy parameters. Applied across Chile, Canada, and Germany using 8760 hourly synthetic simulations generated through validated physical models, the AI-enhanced framework achieved 8.1 % global normalized root-mean-square error, showing improved forecast accuracy compared to conventional baselines. Regional performance demonstrated consistent solar forecasting with nRMSE of 10.5–10.6 % across regions, and wind prediction accuracy of 1.8 % nRMSE in Chile and Germany and 2.0 % nRMSE in Canada. The reinforcement learning controller maintained an average battery state-of-charge around 0.80, achieved 100 % voltage compliance with IEEE 1547–2018 standards, and managed high oversupply events reaching up to 2217.9 % of load in Chile, 4580.3 % in Canada, and 2068.6 % in Germany. Economic analysis estimated a levelized cost of electricity (LCOE) reduction of approximately 18 %, with regional values of $260.6/MWh for Chile, $189.5/MWh for Canada, and $173.8/MWh for Germany. Average renewable penetration exceeded demand levels, reaching 148.8 % in Chile, 294.7 % in Canada, and 134.9 % in Germany, with minimal zero-renewable hours of 65 and 41 per year in the respective regions. Suggested policy measures include targeted storage incentives for Chile, dynamic pricing for Canada, and battery subsidies and pilot projects for Germany
Suggested Citation
Download full text from publisher
As the access to this document is restricted, you may want to
for a different version of it.
Corrections
All material on this site has been provided by the respective publishers and authors. You can help correct errors and omissions. When requesting a correction, please mention this item's handle: RePEc:eee:appene:v:401:y:2025:i:pc:s0306261925015648. See general information about how to correct material in RePEc.
If you have authored this item and are not yet registered with RePEc, we encourage you to do it here. This allows to link your profile to this item. It also allows you to accept potential citations to this item that we are uncertain about.
We have no bibliographic references for this item. You can help adding them by using this form .
If you know of missing items citing this one, you can help us creating those links by adding the relevant references in the same way as above, for each refering item. If you are a registered author of this item, you may also want to check the "citations" tab in your RePEc Author Service profile, as there may be some citations waiting for confirmation.
For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: Catherine Liu (email available below). General contact details of provider: http://www.elsevier.com/wps/find/journaldescription.cws_home/405891/description#description .
Please note that corrections may take a couple of weeks to filter through
the various RePEc services.