Abstract
Recent legislations have necessitated policies for carbon print reduction. Buildings and in particular space heating are major energy consumers and responsible for over 34% of carbon print. This work presents a method of heating only certain parts of the building using far infrared (FIR) heating. This study gives an overview on the application of infrared radiation in heating by modern methods in tune and compatibility with climate developments for the public spaces in this decade. The case study is on a university lecture theatre and the space is split up into varyingly sized zones which enable different parts of the room to be heated depending on the time and occupation of the zone. The potential to heat each zone with FIR is implemented which runs according to the machine learning algorithm (MLA) through a practical study of real CO2 data collection and validation. This allows heating to start running before the zone(s) is occupied to optimise thermal comfort. Results show that occupation zone FIR heating saves an average of 11.175kWh through various occupations compared to the currently equipped convection wall mounted radiators. Occupation forecasting of the room using random forest machine learning has an accuracy of 97.75% for 15-minutes intervals of a day. Cost analysis for the proposed occupation heating show savings of up to 76% and 14.6% compared to convection electric and gas heating respectively. FIR provides a more efficient method of heating with the capacity for zonal implementation. The results in this research demonstrate the feasibility of FIR zonal heating for non-domestic applications.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have