Giorgos Petrou, Anna Mavrogianni, Anastasia Mylona, Rokia Raslan, Gurdane Virk, Michael Davies
Languages: English | Pages: 10 pp
Bibliographic info:
38th AIVC Conference "Ventilating healthy low-energy buildings", Nottingham, UK, 13-14 September 2017

According to the 2016 Household Projections report, England’s housing stock could reach 28 million households by 2039 with approximately one fifth being new constructions. A significant proportion of these newly built dwellings may face a high risk of overheating as a result of the combined effects of climate change and more stringent building thermal efficiency standards, if not appropriately designed. Reliable methods for predicting indoor overheating risk are required to avoid potentially negative impacts of excess indoor temperature exposure on occupant thermal comfort and wellbeing while simultaneously minimising the use of mechanical ventilation and cooling. Building Energy Simulation (BES) software are widely used in the building construction industry to estimate the overheating risk of new developments. CIBSE’s recently released methodology for predicting overheating in new dwellings aims to achieve consistency between existing prediction methods currently applied by building designers and engineers. BES tools are abstract representations of reality and large differences in model outputs are often observed between tools. The level of overheating risk predicted through the CIBSE method may hence depend on the choice of software and its underlying assumptions. Such an effect could directly impact CIBSE’s efforts in creating a standardised procedure across the industry. This research project utilised inter-model comparison along with sensitivity analysis to investigate the differences in overheating risk prediction between two commonly used software packages, EnergyPlus and IES VE. The sensitivity analysis resulted in a total of nine variations of the single-aspect, high-rise flat, simulated in each software. Looking at individual models, there was a general agreement between either software’s predictions and the literature’s suggestions on the factors that may be driving overheating. Measures such as increased thermal mass, external shading, north-facing direction and cross-ventilation lowered the predicted risk. However, discrepancies between software were observed with only two EnergyPlus models successfully meeting both overheating criteria, compared to all the IES VE models. This work therefore concludes that the choice of BES tool could greatly impact the predicted risk of overheating.