Top content

Anh-Duy Pham

PhD student

E-Mail

Coppenrath Innovation Centre
Hamburger Straße 24
LOK 15, Raum 01.13.03A
49084 Osnabrück
Tel.:  +49 541 969-6341

Profiles

LinkedIn
Researchgate
ORCID
GitHub

 

 

Informed Machine Learning on Sparse Data and Information in the context of barn climate and emissions

Livestock plays an important role for the supply of high quality food and has a high economic andsocial relevance. However, it is also a significant contributor of pollutants, which negatively affectour environment and health (e.g. greenhouse gases, ammonia, airborne pathogens). There is anurgent need to mitigate these pollutant emissions. To do so, accurate measurements of emissionsand barn climate are a fundamental requirement.Due to extensive costs, actual measurements are usually limited to only few sensors inside andaround livestock housing systems. The measured variables are non-linear and show a hightemporalspatial variability. This leads to large information gaps between the sensors and thereforeto high uncertainties in the measurement results.The goal of this project is to overcome these limitations by combining different approaches andsources of information. Computational fluid dynamics (CFD) will be applied for a variety ofboundary conditions to generate ground truth information. This information will be used in a hybridArtificial Intelligence (AI) approach, where data-driven as well as informed machine learning will beapplied, making use of the provided domain knowledge, e.g., via CFD and the respectivesimulations. The AI will then be coupled with the sensor data with the goal of generating mostaccurate data on emission and barn climate in real time.After validation, the combined approach will be applied on a larger number of housing systems,thus further enabling the discovery of new knowledge and previously unknown correlations usingdata science and machine learning approaches

Project team: Anh-Duy Pham (Ph.D. Student), Prof. Dr. Martin Atzmüller (UOS), Prof. Dr. Tim Römer (UOS), Dr.-Ing. David Janke (ATB), Prof. Dr. Thomas Amon (ATB)

GIL Tagung 2024

Find the digital version of the conference poster here: GIL 2024 POSTER