Ventilation and shading to reduce overheating in UK homes: an evaluation using matched pair test houses with synthetic occupants
thesisposted on 27.11.2020, 08:56 authored by Ben M RobertsBen M Roberts
Overheating in homes is a problem because high indoor temperatures present a risk to occupant health and well-being. Overheating in the UK is already a problem and is expected to increase in the future as the climate warms, homes become better insulated, and the population ages and urbanises. Air-conditioning could be used, but would increase electricity demand at a time when climate change targets require the UK to reduce its carbon emissions.
Strategies are needed to help keep UK homes cool in summer and prevent the uptake of air-conditioning. Occupant-controlled strategies for cooling could make use of existing features of the homes: operable windows, internal shading, internal doors, or a combination of these.
Occupants influence the indoor temperature of homes, but there is a gap in knowledge about what occupants should do, and when, to reduce indoor temperatures in summer and what effect these interventions might have in typical UK homes. This thesis aims to examine the effect of occupant-controlled cooling strategies on ventilation rate and indoor temperature through measurement and modelling of a matched pair of test houses.
The two houses are a nominally identical pair of adjoining 1930s semi-detached two-storey houses located in the East Midlands of England. The houses are thermally as-built, except for the addition of double-glazing and loft insulation. Measurements assured that the houses had similar heat loss coefficients and air permeability. The houses were instrumented with sensors to measure indoor temperatures, ventilation rates, and the weather conditions. Synthetic occupancy devices operated windows, internal shading, internal doors and mimicked the internal heat gains of occupants and appliances.
Three main experiments were conducted. Experiment~1 evaluated the effect of window opening with curtains open and closed on ventilation rates. The experiment used tracer gas in side-by-side simultaneous tests in single rooms with the internal doors closed (i.e. single-sided ventilation). In one house the room being tested had windows and curtains open, in the corresponding room in the other house the windows were also open, but the curtains were closed. Other tracer gas tests were also done to measure whole house infiltration with internal doors and curtains open, but windows closed. Measured infiltration was compared to infiltration derived from calculation methods.
Results showed that keeping the curtains open in bedrooms allowed for higher ventilation rates compared to when curtains were closed. Measured infiltration was lower than predicted by all calculation methods.
Experiment~2 evaluated the effect of occupant-controlled strategies on indoor temperatures and overheating, which was assessed using the CIBSE TM59 protocol. The experiment used a series of side-by-side tests using synthetic occupancy to directly compare different strategies including windows always closed, windows open in response to indoor temperature and occupancy (the TM59 protocol), or windows open according to time of day; daytime closing of curtains; varying the use of internal doors; or a combination of these. The houses were monitored for an entire summer period (May-September 2017) which featured a heatwave in June (as defined by the UK Met Office) when outdoor temperatures reached a maximum of 30.5\,\degree C, exceeded 30\,\degree C on two consecutive days, and were above 28.5\,\degree C on two other days within a 5-day period. Analysis of temperatures and overheating in all seven tests focused on rooms pertinent to the TM59 overheating assessment: living rooms, kitchens, and bedrooms.
It was found that night ventilation of bedrooms and a ground floor room with internal doors open was the most effective occupant-controlled strategy to keep homes cool in summer. However, neither of the strategies trialled during the June heatwave prevented the houses from overheating, and so alternative strategies are needed.
Experiment~3 evaluated the ability of dynamic thermal models to accurately predict overheating. The experiment used the data gathered in Experiment~2 to empirically validate models in a two-phase multi-model exercise. This involved four experienced, industry practitioners using two different dynamic thermal modelling programs. Models were first constructed in a blind phase where modellers received information about the test houses, the occupancy profiles, and weather conditions. Models were then modified in an open phase where modellers received the test house temperature measurements and, with the other modellers, adjusted their models to try and improve the predictions. The models' predicted hours of overheating were compared with the measured hours using the BS~EN~15251 Category II threshold for living rooms during occupied hours and the CIBSE static threshold of 26\,\degree C for bedrooms during sleeping hours. The models developed in each phase were also used to predict the hours of overheating using the TM59 procedure, which allowed for inter-model comparison.
All four dynamic thermal models predicted higher maximum temperatures and lower minimum temperatures than were measured, especially during the June heatwave when predictive accuracy was most needed. Overheating hours were predicted higher than measured in living rooms and predicted lower than measured in bedrooms. Even when the modellers had access to the measured temperatures they could not eliminate the over- and under-prediction of temperatures. The inter-model variability in predictions, which is due to the differences in the models and the way they were used, was quantified as the Simulation Resolution.
This work has demonstrated the value of matched pair houses for understanding the effect that different occupant-controlled cooling strategies had on ventilation rates and indoor temperature in summer. It has also demonstrated and quantified the unreliability of the overheating predictions of dynamic thermal models. These and other results will be valuable to house builders, those concerned with assuring the health and well-being of UK citizens, and the academics, engineers, and consultants that use dynamic thermal models to assess summertime overheating.
EPSRC Centre for Doctoral Training in Energy Demand (LoLo)
Engineering and Physical Sciences Research CouncilFind out more...
- Architecture, Building and Civil Engineering