Wind generation forecasting

Energy Sector

Customer

Spanish energy company with a global presence.

Description

Prediction of wind generation based on meteorological information (three-dimensional grids of wind speed, pressure, etc.).

Results

Improvement of 10 percentage points compared to traditional forecasting models. Higher profitability for our client by participating in the US electricity auction.

Technology

Data ingestion with Spark (Scala) in Big Data cluster based on Hortonworks distribution. Modelling with scikit-learn and LightGBM. Orchestration and monitoring with Apache Airflow.