PRF Technologies Files Patent Application Covering Proprietary Methods For Solar Plant–Level Micro-Climate Modeling Designed To Significantly Improve The Accuracy Of Solar Energy Production Forecasts In Competitive Electric Utility Markets

Patent filing follows acceptance into the NVIDIA Connect Program 

Proprietary software targets weather forecast-driven revenue optimization, reduced penalty risk, and improved market performance for utility-scale solar assets

TEL AVIV, Israel, Jan. 20, 2026 (GLOBE NEWSWIRE) -- PRF Technologies Ltd. (NASDAQ:PRFX) ("PRF" or the "Company") (formerly "PainReform") today announced that it has filed a patent application covering proprietary methods for solar plant–level micro-climate modeling designed to significantly improve the accuracy of solar energy production forecasts in competitive electric utility markets.

The patent follows acceptance into the prestigious NVIDIA Connect Program and aims to protect core intellectual property underlying DeepSolar Predict, PRF Technologies' solar forecasting software platform. The technology is designed to address a critical limitation in conventional solar forecasting: the reliance on regional or broad-area weather models that fail to capture the highly localized conditions that directly impact power output at individual solar plants.

By modeling micro-climate behavior unique to each solar installation, the patented approach aims to enable more precise short-term and intraday production forecasts that reflect real operating conditions rather than broad regional averages. The system continuously adapts to localized environmental dynamics, allowing forecasts to evolve in real time as conditions change.

Improved forecast accuracy has direct commercial implications for solar asset owners, operators, and energy market participants. More reliable production estimates can potentially support stronger positioning in day-ahead and intraday electricity markets, reduce financial penalties from forecast errors, enhance revenue capture during favorable pricing conditions, and improve operational decision-making across plant control and energy management systems.