Optimizing Wind Turbine Management Through AI-Driven Load and Placement Predictions

Authors

Laurene Robinson, St Mary's University

Files

Publication Date

Summer 2024

Digital Publisher

Digital Commons at St. Mary's University

Collection

McNair Scholars Symposium

Keywords

Machine Learning; Renewable Energy; Wind Turbines; Real-Time Data Integration; Power Production

Description

The maintenance and operation of wind turbines create challenges when it comes to maximizing energy efficiency while also producing a minimal amount of wear. It is necessary to determine when is the optimal time to stop a turbine, that being if it's due to mechanical issues, lack of power demand, or environmental factors. Stopping the turbines can save companies resources and improve their system's longevity. In this study, we explore the feasibility of using machine learning to predict optimal loads for these conditions. By modeling variables such as wind availability and current power demand, the model aims to predict what factors lead to the most efficiency. Online simulated datasets might not involve environmental conditions, simulating the data with perfect laminar flow. So we simulated our own data in order to see how turbulence affects airflow patterns over time. To address ⁤⁤these challenges, this research proposes developing an ⁤⁤AI model capable of continuously optimizing turbine ⁤⁤load and placement decisions. Such a model would be able to integrate real-time data on wind conditions and power demand, and then adapt accordingly to environmental changes along with operational constraints. By using AI’s predictive capabilities, the study aims to increase energy production efficiency by making the systems that control the turbines dynamic in the way they continuously adapt to the conditions.

Disciplines

Computer and Systems Architecture | Other Electrical and Computer Engineering | Power and Energy

Format

MOV

Medium

video

Size or Duration

18:21 minutes

City

San Antonio

Creative Commons License

This work is licensed under a Creative Commons Attribution-NonCommercial-No Derivative Works 4.0 International License.