Product Overview

Product Overview
Wind power generation is proportional to the cube of wind speed, meaning even small changes in wind speed can lead to significant fluctuations in power output.Solar power generation is also influenced by multiple weather factors such as solar irradiance, temperature, and humidity. To achieve accurate generation forecasting, high-resolution weather forecasts tailored to the exact location of each power plant are essential.
Limitations of Conventional Weather Forecasts
Traditional weather forecasting typically provides spatial resolution of 5 km or more, making it difficult to deliver accurate predictions for the exact location of power plants. In addition, hourly wind speed forecasts are insufficient for wind power prediction—detailed forecasts at 10-minute intervals are required.
Solving the Problem with Proprietary Weather Forecasting
REcast utilizes the WRF (Weather Research and Forecasting) regional model to generate high-resolution weather forecasts tailored to power plant locations. With spatial resolution below 1 km and temporal resolution at 10-minute intervals, the system significantly improves generation forecasting accuracy.

Key Technologies

Generation Forecasting Based on Proprietary Weather Models
We generate numerical forecast data at 10-minute intervals with a 1 km × 1 km resolution and process it in real time. By combining this data with satellite imagery through ensemble techniques, we further improve prediction accuracy. Our solution delivers one of the most advanced solar and wind generation forecasting capabilities globally, while continuously improving forecasting models based on customer requirements.
Cloud-based High-Performance Computing
Using HPC powered by Amazon Web Services, computing environments can be configured within an hour. Large-scale parallel computing accelerates big data processing, achieving the same performance with only 45% of the computing resources and reducing costs by more than 50%.
Improved Generation Forecast Accuracy
When applying our proprietary forecast data, wind farms across six locations showed improved prediction accuracy compared to national meteorological forecasts, and similar or better performance compared to international forecasts. For solar power plants nationwide, sampling results from 41 sites also demonstrated significantly improved performance compared to domestic forecasts.