PhD defence by Simon Vestergaard Johansen
06.12.2019 kl. 12.30 - 16.00
Automated Optimization of Broiler Production
Broiler (chicken for meat production) growth optimization reduces the amount of feed, water and electricity required to produce a mature broiler. The main objective of state-of-the-art industrial broiler production is to facilitate efficient maturation of broilers under ethically acceptable conditions at scale, e.g., 40.000 broilers per batch. The primary expense in broiler production is feed, and subsequently the primary key performance measure is feed conversion ratio (FCR), i.e., feed efficiency. Existing scientific control literature is incompatible with state-of-the-art industrial broiler production, as it exclusively focuses on active feed control despite the fact that unrestricted feeding regimes unanimously are used in practice. Climate conditions, e.g., temperature and light intensity, can be controlled and are known to significantly influence broiler growth, effectively providing a potential avenue for growth optimization, and will thus be the subject of this study.
Data driven modelling techniques are used given the absence of first-principle mathematical broiler growth models explaining the temporal relationship between broiler climate conditions and broiler growth. Dynamic neural network (DNN) models are successfully trained on industrial scale production data. Iterative learning control (ILC) provides a potential solution given the repeated nature of the production process, as it has been especially developed for systems that make repeated executions of the same finite duration task. Using DNN models in combination with ILC a combined FCR improvement of 3.8% was obtained throughout this study - corresponding to a yearly saving of 8.500€ and 26.7 tones of broiler feed per broiler house.
Associate Professor Anders la Cour-Harbo, Aalborg University, Denmark (Chairman)
Professor Daniel Berckmans, KU Leuven, Belgium
Head of Group R&D Martin Simonsen, SKIOLD A/S, Denmark
Associate Professor Jan Dimon Bendtsen, Aalborg University, Denmark
CTO Jesper Mogensen, SKOV A/S, Denmark
Associate Professor Tom Pedersen, Aalborg University, Denmark
After the defence there will be a reception at Fredrik Bajers Vej 7A1-112
Free of charge
Automation and Control, Department of Electronic Systems
Aalborg University, Fredrik Bajers Vej 7B3-104