Refining the Forecast: Advanced Data Preprocessing for Accurate Wind Prediction

Overview

Accurate wind forecasting is challenging due to wind patterns being affected by weather, geography and seasons. Traditional methods of predicting wind are often incapable of capturing all the patterns and so are unable to effectively predict future behaviour. AI-driven wind forecasting provides a solution, since machine learning models perform well in scenarios with large amounts of past data and the need for accurate predictions of the future. This project focused on the data preprocessing step, when the initial dataset, containing past measurements such as atmospheric pressure or precipitation amount, is transformed in various ways to make it easier for the model to make accurate predictions. Preprocessing has major effects on the final predictions and accuracy, and so carries immense potential for improving wind forecasting. Such predictions are used, for example, in selecting sites for building wind farms and when scheduling maintenance of the existing ones.

Awards (1)

• Selected for CWSF 2026

Competition history

• CWSF 2026 Energy

Related projects

Closest projects by meaning, across every fair and year in the corpus.