Optimization of Night Cooling in Commercial Buildings - Using Genetic Algorithms and Neural Networks

Examensarbete för masterexamen

Please use this identifier to cite or link to this item: https://hdl.handle.net/20.500.12380/253023
Download file(s):
File Description SizeFormat 
253023.pdfFulltext4.82 MBAdobe PDFView/Open
Type: Examensarbete för masterexamen
Master Thesis
Title: Optimization of Night Cooling in Commercial Buildings - Using Genetic Algorithms and Neural Networks
Authors: Dahlström, Emmy
Rönn, Linus
Abstract: Periodically since the two oil crisis in the 1970s, there has been a focus in Sweden on reducing the energy use in buildings. This focus has evolved into the building regulations used today. With the stricter energy requirements and the interest from society in environmental issues there is a need to use more optimized control systems for cooling and ventilation. Night cooling is an example of this. The purpose of night cooling is to decrease the cooling need in buildings by ventilating at night with cold outdoor air. The thesis uses case studies to examine if it is possible to optimize night cooling set points and time schedules regarding energy consumption and indoor climate for two retail stores in Göteborg. The optimizations were done with genetic algorithms, neural networks and building energy models, based on logged control data for the two stores. The study suggest that the energy consumption could be reduced with 15% for both facilities with the optimized control settings compared to the original. The project also shows that even unoptimized night cooling has benefits to energy consumption. The sensitivity analysis shows that a reduction around 10% for similar buildings are plausible with the optimized settings from the case studies. The projects concludes that the use of logged control data in combination with genetic algorithms and neural networks is an efficient way for both calibration and optimization of building energy models. The industry moves towards an increase of available logged control data. As such, it is important to be able to properly utilize the data for improving the accuracy of building energy simulations. The method used in this project is an example of this.
Keywords: Materialvetenskap;Byggnadsteknik;Materials Science;Building engineering
Issue Date: 2017
Publisher: Chalmers tekniska högskola / Institutionen för bygg- och miljöteknik
Chalmers University of Technology / Department of Civil and Environmental Engineering
Series/Report no.: Examensarbete - Institutionen för bygg- och miljöteknik, Chalmers tekniska högskola : BOMX02-17-50
URI: https://hdl.handle.net/20.500.12380/253023
Collection:Examensarbeten för masterexamen // Master Theses



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.