Wind Turbine Bird Strike Avoidance System (WITURBISA)
Purpose and goal: The goal of the project is to design an automatic species identification algorithm to support the operation of wind turbines worldwide. Through the integration with the Supervisory Control and Data Acquisition (SCADA), the system will perform an automatic sensor-assisted shutdown of wind turbines based on the prediction of harmful interactions with endangered birdlife.
Expected results and effects: WITURBISA is expected to improve the quality of wildlife monitoring systems, hence it has significant environmental impacts. The system is expected to help avoid 40% of the shutdown period due to bird collisions. When completed, this system can also be used in the development phase of wind turbine projects to perform more accurate Environmental Impact Assessment to minimize the overall impact on the wildlife.
Approach and implementation: In this project, Chalmers will be involved in the development of algorithms for 3D feature extraction and classification of bird species. While existing bird classification algorithms rely on 2D features, this project takes advantage of the available 3D camera systems to develop novel 3D features in order to enhance the camera-based classification algorithms. We aim to utilize state-of-the-art machine learning techniques for efficient classification of the species.
Huu Le (contact)
Researcher at Chalmers, Electrical Engineering, Signal Processing and Biomedical Engineering, Imaging and Image Analysis
Project ID: 2019-01175
Funding Chalmers participation during 2019–2021