Airing with side-hung and pivot-hung windows - Simulations and heat loss

Abstract

Efforts to reduce energy consumption in buildings have increased with the pending impact of climate change. Heat losses through airing behaviours remain a largely unknown factor due to the variety of reasons for opening a window, making generalization difficult. Currently, an estimation of 4 kWh/m2 heated area for airing heat loss is assumed in Sweden. Additionally, the resulting airflow from opening a window needs to be quantified under various conditions to find a resulting heat loss, which has been done using on-site measurements and computational fluid dynamics (CFD). The aim of this study was firstly establishing air flow data for different opening angles and temperature differences for two rooms with two different window types, the first being side-hung windows and the second pivot-hung windows. For this, transient simulations in the CFD software Simcenter Flovent were performed, with additional measurements using CO2 as a tracer gas taken in the room with pivot-hung windows to allow for a very limited validation. Secondly, several selected data sets of window opening angles throughout the year taken from apartments in southern Sweden were analysed. Thirdly, yearly airing heat loss was calculated by pairing the air flow data with angles and temperature differences from the user data. As the selected user data sets were not deemed representative, additional parametric schedules were created for the yearly heat loss calculation which depended on various parameters such as opening angle, duration, frequency, and time of day of opening. From the CFD simulations, the pivot-hung windows resulted in higher airflow, but it was unclear whether this was due to the larger window size or the different construction. It also seemed to allow for more efficient air exchange when cross-ventilating, although this is inconclusive due to the lack of comparability between the rooms. Adding wind increased the air flow significantly. There was some deviation between measurements and simulation results, but they were shown to be in a similar order of magnitude when accounting for errors. The heat loss calculations with user data showed a lot of variation with some scenarios comparable to the recommended value from regulations, while others were significantly off. The parametric study revealed that angle, duration and frequency as contained in the data affect the results significantly, while window size was also identified as a large factor. Future studies should concentrate on evaluating additional user data in order to identify further different patterns and driving factors on window opening which could be integrated into simulations.

Popular Abstract

Lately, efforts have been made to reduce the energy used in buildings in various ways. To meet requirements, energy simulations have become a popular method to estimate the energy use in new and existing buildings. Many factors are included in these simulations, however, when it comes to opening windows, there are many variables that make an accurate representation of the heat that is lost by it difficult. The purpose of this study was to investigate these variables.
The study was done by applying a specific type of simulation (CFD) used for representing airflow patterns around geometric structures. Two rooms with different window types were created in the software and the airflow through them was measured for different variations of opening the windows (for example, different opening angles and different window constellations) and under different outdoor conditions such as temperature and wind. The airflow simulations showed little difference between the window types beyond what was assumed to be caused by the different sizes of the windows. Adding wind increased the airflow manyfold, although different wind speeds and directions were not tested.
For some of these scenarios, the values from the simulation were compared to real life airflow measurement in the actual room that was modelled. The simulation and measurement results were similar, but there were some differences. As the measurements can be affected by varying conditions, it is hard to say how accurate the simulation really is, but it is likely that there is some room for improvement by adding more detail to the model.
The second part of the study involved investigating data of window opening angles measured over a longer period of time in several different apartments, representing window opening behaviour of the building’s inhabitants. Different patterns according to the type of window and person in charge could be observed, and outdoor temperature proved a main influence in some of the datasets, but not all. The data was generally not extensive enough to make any clear conclusions.
Lastly, the two previous parts of the study were combined in a parametric study covering many different scenarios. For this, the airflows from the simulation were matched to the window opening behaviour data and indoor/outdoor temperatures, and the corresponding heat loss that is carried through the opened window was calculated and summed up for a whole year. The result was also compared to a given value that is recommended to use for the airing heat loss by the Swedish building code. In addition to the available behaviour data, further schedules were created by varying parameters such as opening angle, duration and frequency. The calculations showed that the results are very different per scenario which in turn shows how the heat loss from opening windows is affected by many different things and a simple estimation most likely will not give an accurate result. Generally, the duration of opening and the opening angle had the most impact on the airing heat loss, followed by how often the window is opened. The size of the window should also not be neglected. Wind was not included in the heat loss calculation, but the simulation results show that it can have a huge impact as well.
In the future, similar studies could include a more extensive analysis of different wind conditions to be included in the calculations. Additionally, a larger user behaviour database could give some better conclusions as to how long and how much the windows are generally opened under different conditions.

• author: Driehaus, Kathrin
• supervisor: Dennis Johansson
• organization: Department of Building and Environmental Technology; Division of Energy and Building Design
• course: AEBM01 20241
• year: 2024
• type: H2 - Master's Degree (Two Years)
• subject: Technology and Engineering
• keywords: Airing, Window opening, Computational fluid dynamics, Heat loss, Occupant behaviour
• language: English
• id: 9163214
• date added to LUP: 2024-06-24 11:11:01
• date last changed: 2024-06-24 11:11:01