A systematic evaluation of indoor overheating interactions with outdoor heat conditions

Full text

The 16th Conference of the International Society of
Indoor Air Quality & Climate (Indoor Air 2020)
COEX, Seoul, Korea | July 20 - 24, 2020
Paper ID ABS-0802 ISBN: …………………….
A systematic evaluation of indoor overheating interactions with outdoor heat
conditions
Lili Ji1, Abdelaziz Laouadi2, Chang Shu1, Abhishek Gaur2, Michael Lacasse2, Liangzhu (Leon)
Wang1,*
1Department of Building, Civil and Environmental Engineering, Concordia University,
Montreal, 1455 De Maisonneuve, H3G 1M8, Montreal, Quebec, Canada
2 Construction Research Centre, National Research Council Canada
1200 Montreal Road, K1A 0R6, Ottawa, Ontario, Canada
*Corresponding email: leon.wang@concordia.ca
SUMMARY
Overheating in buildings has been a major health concern for vulnerable occupants during
extreme heatwaves. Even though indoor overheating is highly dependent on outdoor climate, it
may also be affected by other factors including building envelope characteristics, HVAC
operation and internal heat gains. This study is to investigate the synchronisation of indoor
overheating with outdoor heatwave. Archetype residential buildings were created and used to
conduct EnergyPlus simulations of 31 years climate data of three major Canadian cities. A
heatwave evaluation method developed by National Research Council of Canada was used to
define heat events and extreme summer weather years. The method uses the transient standard
effective temperature (t-SET) to rank heat events in terms of duration, intensity and severity.
The results show that in most building configurations, outdoor extreme heatwaves are
synchronised with indoor extreme overheating events. Outdoor-based extreme summer weather
years are therefore suitable to study indoor overheating risks.
KEYWORDS
Building simulation, Overheating, Heatwave, Extreme weather data, Thermal comfort
1 INTRODUCTION
Overheating problem in buildings has been a significant health concern during extreme heat
events and seems becoming more frequent as a result of climate change (Chang et al. 2020).
Overheating is found in free-running buildings, buildings with intermittent use or limited
capacity of air-conditioning, or fully air-conditioned buildings during HVAC failures or
blackouts. Indoor overheating is not solely dependent on outdoor climate, but also on building
characteristics and operation, internal heat gains, and occupant behaviour. However, it is not
clear whether indoor extreme overheating events are synchronized with extreme outdoor
heatwaves. The National Research Council (NRC) of Canada has recently proposed a method
to define outdoor heatwaves, patterns and reference extreme summer weather years based on
the transient standard effective temperature (t-SET) metric ( Laouadi et al., 2020). In this study,
the NRC method is applied to indoor overheating events with some modifications to the
threshold values of t-SET for daytime and nighttime exposures. The goal is to compare the
extreme heat events for outdoor and indoor environments.

2 MATERIALS AND METHODS
Definition of heat events
Heat may affect thermal comfort and health of human subjects. Heat stress indices that are based
on a complete heat balance of human subjects are more suitable to study the effect of heat events
on subject comfort and health. In this study, the Standard Effective Temperature (SET) index
is used for this purpose. SET is the temperature of an environment with 50% relative humidity
(RH), mean radiant temperature equal to air temperature and airspeed < 0.1 m/s, in which the
total heat loss from the skin of a dummy subject at the actual exertion condition and with a
standard clothing insulation ( 0.6 clo for one met) is the same as that from a subject in the actual
environment with actual clothing and activity level (ASHRAE, 2017). The SET is modified to
take into account the transient nature of heat events and activity levels during daytime and
nighttime exposure, and thus is named Transient Standard Effective Temperature (t-SET)
(Laouadi et al., 2020). The magnitude of a heat event of the given duration (∆t) is defined as
follows:
  󰇛󰥝  󰥝
󰇜
  (1)
Where SETH (°C·h) is the magnitude of the heat event (zero if negative),  is the calculation
time step (h), and 󰥝
 is the reference value of 󰥝 above the comfort level, which triggers
a physiological response to cool a subject body and a subject action to restore the thermal
comfort. The reference value of 󰥝 is fixed at 30°C (corresponding to the initiation of
sweating or slight warm sensation) for un-acclimatized people and 31.2°C for acclimatised
people (Laouadi et al., 2020) during daytime exposure. During night time, the reference value
of 󰥝 is set to be equal to the lower temperature limit of the adaptive thermal comfort range
for the location under consideration or fixed at 26°C (the upper limit value of thermal comfort)
for subjects under outdoor or indoor environment exposure, respectively.
A countable (or “meaningful”) heat event is declared if its daytime magnitude () is
higher than a minimum value ( ). In this study, this minimum value is fixed at
     corresponding to a body water loss lower than 1.2% when a subject is
exposed to thermal conditions of one degree SET above the reference value of 󰥝
(exposure time of four hours). A heatwave is defined as continuous countable heat events
occurring over at least two days. To define exterior heatwaves, 󰥝 is calculated considering
air temperature, relative humidity, mean radiant temperature (MRT), wind speed and solar
radiation of the outdoor climate, and a subject is assumed walking outdoors during daytime and
sleeping during night time. To define an indoor overheating event, 󰥝 is calculated
considering indoor air temperature, relative humidity, mean radiant temperature (MRT) and
airspeed, and a subject is assumed seating quietly indoors during daytime and sleeping during
night time. An overheating event is then defined as continuous meaningful heat event occurring
over at least two days. Heatwaves and overheating events both could be characterised by three
features: duration, intensity, and severity. The duration D (days) is measured in terms of the
number of days of sustained heat events. The severity ( denoted by SETH) is calculated as the
summation of magnitudes of daily heat events:
    󰇛󰥝  󰥝
󰇜
   (2)

Where SETH (°C·h) is the severity of a heatwave or overheating event of a duration of two or
more days. The intensity I (°C) is calculated as the ratio of severity to duration (expressed in
hours):
  
 (3)
Accordingly, one can distinguish three major types of heatwaves or overheating events, namely:
long, intense and severe or a combination of long and intense, long and severe, or severe and
intense.
Building models and configurations
Natural Resources Canada (NRCan) has generated archetype models of residential buildings
(Parekh, 2012). 500,000 houses were rated across Canada to understand the characteristics of
existing and new houses for predicting the effects of construction changes. Among those
houses, there were 438,746 single-detached houses. According to the archetype characteristics,
we have created two archetype models for the single-detached house and row house to study
space overheating using the EnergyPlus software (v9.2; DOE, 2020). Each house model has
two floors above ground, attic space and a full basement, as shown in Figure 1. The first floor
is assigned to living room where occupants spend their time during daytime and the second
floor is assigned to bedrooms where occupants sleep at night. The house construction
characteristics are summarised in Table 1. The single-detached house has windows on each
facade of the first and second floors, whereas the row house has windows on the south and north
(or east and west) facades. Internal heat gains from occupancy, lights and equipment are taken
from the National Building Code of Canada (NRC, 2015): 3 people per house, lighting gains
=5 W/m2 of the heated area; and equipment gains = 5 W/m2.
A B
Figure 1. The geometry of building models: a) single-detached house and b) row house
Table 1. Building properties of the archetypes - baseline buildings
Property
Single-detached house
Row house
Old
Construction
New
Construction
Old
Construction
New
Construction
Orientation of windows
N-S-E-W
N-S-E-W
N-S (or E-W)
N-S (or E-W)
Heated area (m2)
160.20
160.20
220.70
220.70
Window/Wall ratio (%)
15.27
15.27
23.95
23.95
Exterior Wall
Effective R (m2 KW−1)
1.80
3.20
1.80
3.20
Attic insulation
Effective R (m2 KW−1)
3.60
8.20
3.60
8.20
Window (wooden frame)
U-Value (W m−2 K−1)
2.58
1.58
2.58
1.58
Window SHGC
0.702
0.67
0.702
0.67
Design Infiltration Rate (ACH)
6.86
2.32
9.32
2.80

To investigate the impact of different passive measures in buildings on the indoor overheating
condition, we have applied six different configurations to both old and current constructions.
Three primary passive measures are considered, including interior blinds, exterior shades and
natural ventilation through windows. Interior blinds are opened by setting the slat angle as
vertical (slat angle 90°) and closed by setting the angle at 175° (almost horizontal). Exterior
shades are always closed with 5% solar transmittance. The above three configurations could be
combined with or without natural ventilation. Natural ventilation is realised through opening
windows for 25% when the indoor temperature exceeds the setpoint 26°C and the outdoor
temperature. Using the above set of parameters, 1920 simulations (2 constructions, two
orientations for the row house, six configurations, 31 historical weather files and one typical
meteorological year (TMY) weather file, and three cities including Ottawa, Toronto, and
Montreal) have been conducted.
Selection and comparison of extreme years
Thirty-one years (1986-2016) of historical climate data have been used in the simulations to
capture all types of outdoor heatwaves and indoor overheating events. The climate data were
generated for selected Canadian cities using the methodology of Gaur et al. (2019). The
summer period is fixed from May to September. Heatwaves and overheating events for each
year are identified and sorted by maximum duration, intensity, and severity. The maximum
values are assigned to each year. For each simulation case, the top two extreme years among 31
years are listed and compared in terms of heatwaves and overheating events. The software
CumFreq (Oosterbaan, 2019) is used to extract the extreme years that have a return period of
15.5 years (second rank out of 31 years). The synchronisation of extreme overheating events
with heatwaves is accepted if both are in the first two ranks.
3 RESULTS AND DISCUSSION
Comparison of indoor and outdoor-based extreme years
Table 2 lists the first two extreme years of heatwaves and overheating events for the single-
detached house with old and current constructions in Ottawa, Ontario, Canada. The duration,
intensity and severity of heatwaves and overheating events are also shown in the table. Outdoor-
based extreme years that have a return period closest to 15.5 years (calculated using CumFreq
software) are coloured in red. The extreme year 2010 is selected as the representative year for
Ottawa that could be considered as the reference summer weather year (RSWY) (Laouadi et
al., 2020) since it combines two features of heat waves - long and severe . The year 2010 also
ranks in the top two indoor-based extreme years for all house configurations (coloured in red).
Similarly, the extreme years of 2006 and 2010 for Toronto and Montreal (not presented here)
rank in the top two overheating extreme years for 75% and 58% of house configurations,
respectively. For a row house with North-South and East-West orientations in Ottawa, 2010
ranks in the top two overheating extreme years for 80% and 78% of house configurations,
respectively. Overall, for most building configurations, exterior heatwaves also result in
extreme interior overheating events. Therefore, under cold climates of Canada, outdoor-based
extreme summer weather years are suitable for assessing overheating risk in free-running
buildings. However, further testings of the NRC model for generating extreme summer years
that are independent of buildings are needed to assess their suitability for other climates such
as warm, hot, hot and humid or tropical climates where heatwaves may be present all year long.
Table 2. Top two extreme years of exterior heatwave and interior overheating events for the
single-detached house in Ottawa, Ontario, Canada. Bolded years are outdoor-based extreme

years with return periods closest to 15.5 years. Representative years are selected when they
combine at least two features of heat events.
Exterior
Year
Duration
Year
Intensity
Year
Severity
Rep Year
2010
7
2002
3.83
2010
437
2010
1987
7
2006
2.89
1987
415
Case1: Single-detached house with old constructions
Configuration
Year
Duration
Year
Intensity
Year
Severity
Rep Year
Closed windows + open
internal blinds
2005
23
2002
4.82
2005
1922
2005
2003
17
1987
4.67
2010
1613
Closed windows + closed
internal blinds
2003
15
2006
5.11
2010
1251
2010
2010
14
2002
4.52
2005
1153
Open windows + open
internal blinds
2010
4
2002
2.42
2010
224
2010
1988
4
2010
2.33
2002
175
2002
Open windows + closed
internal blinds
2010
4
2010
2.13
2010
204
2010
2002
3
2002
2.13
2002
154
2002
Closed windows + closed
external shading
2013
5
2002
4.47
2010
414
2010
2006
5
2010
4.31
2013
400
2013
Open windows + closed
external shading
2010
3
2010
2.27
2010
164
2010
2002
3
2002
1.98
2002
143
2002
Case2: Single-detached house with current constructions
Configuration
Year
Duration
Year
Intensity
Year
Severity
Rep Year
Closed windows + open
internal blinds
1987
62
2005
7.71
1987
7890
1987
1988
50
2010
7.40
1988
7737
1988
Closed windows + closed
internal blinds
1989
47
2005
6.94
1988
6469
1989
1988
45
2010
6.61
1989
6189
1988
Open windows + open
internal blinds
2010
4
2002
2.60
2010
247
2010
1988
4
2010
2.58
1988
188
1988
Open windows + closed
internal blinds
2010
4
2010
228.00
2010
228
2010
2002
3
2002
171.31
2002
171
2002
Closed windows + closed
external shading
2010
14
2002
5.92
2010
1625
2010
2003
13
1994
5.19
2005
1380
Open windows + closed
external shading
2010
4
2010
2.15
2010
207
2010
2002
3
2002
2.09
2002
151
2002
Indoor heat stress levels during exterior heat waves
As mentioned above, there are three types of heatwaves and overheating events: long, severe
and intense. According to the cumulative frequency analysis for Ottawa, 2010 (Jul. 5-10) and
2006 (Jul. 31- Aug. 03) are selected as the extreme years with long/severe and intense heat
waves, respectively (Laouadi et al., 2020). Figure 2 shows the indoor temperature and t-SET of
single-detached house and row house (N-S) with old construction, closed interior blinds and
without natural ventilation. As shown in Figure 2 (A1) (B1), the difference between indoor and
outdoor temperature is more significant during night time than during daytime. As shown in
Figure 2 (A2) (B2), during the long/severe and intense heatwaves, the t-SET in both single-
detached house and row house (N-S) is higher than 26°C for at least two days during night for
several hours, which would disturb sleeping, and higher than 31.2°C when occupants are
assumed awake (7:00 am to 10:00 pm). The t-SET is above 34.5°C for several hours during the
daytime of the whole heatwave, in which condition the thermal sensation is very uncomfortable
and the people’s physiological state is profuse sweating according to Parson (2003); the t-SET
is even up to above 37.5°C for several hours, which indicates people would feel very
uncomfortable, and might lead to failure of thermoregulation.

Figure 2. Comparison of outdoor temperature, indoor temperature and t-SET
4 CONCLUSIONS
For most typical configurations of free-running residential buildings, indoor extreme
overheating events are synchronized with extreme outdoor heatwaves. Outdoor-based extreme
summer weather years as developed by NRC (Laouadi et al., 2020) are therefore suitable to
assess overheating risk in buildings. The results for the indoor temperatures and t-SET during
the periods of heatwaves showed that building occupants would be under extreme heat stress
(excessive sweating during daytime; t-SET > 34.5°C) and their sleep would be disturbed during
night time (t-SET> 26°C). Sleep disturbance weakens the physiological response of occupants
in the next hot days, which may consequently affect the health of vulnerable occupants.
5 ACKNOWLEDGEMENT
The research was supported by the Natural Sciences and Engineering Research Council
(NSERC) of Canada through the Discovery Grants Program and the Advancing Climate Change
Science in Canada Program led by the corresponding author of the paper, and the National
Research Council Construction Research Centre. The authors were thankful for their support.
6 REFERENCES
ASHRAE. 2017. ANSI/ASHRAE Standard 55-2017, Thermal Environmental Conditions for Human
Occupancy. Atlanta: American Society of Heating, Refrigerating, and Air-Conditioning Engineers.
Chang S. et al. 2020. Building Information Survey and Site Investigation of Summertime Overheating
Potentials in Vulnerable Buildings in Montreal City. Submitted to 2020 Building Performance
Analysis Conference and SimBuild.
Gaur, A. et al. 2019. Climate data to undertake hygrothermal and whole building simulations under
projected climate change influences for 11 Canadian cities. Data, 4(2).
Laouadi A. et al. 2020. Development of reference summer weather years for analysis of overheating risk
in buildings. Journal of Building Performance Simulation, 0(0), 1–19.
Oosterbaan R. J., 2019. CumFreq, free software for probability distribution. Available at:
https://www.waterlog.info/cumfreq.htm
National Research Council (NRC), 2015. National Building Code of Canada 2015, National Research
Council of Canada, Ottawa.
Parekh A., 2012. Representative Housing Thermal Archetypes for Energy Analysis Models. Final
Report. CanmetENERGY Ottawa, Natural Resources Canada, 275 pages.
Parsons K. 2003. Human Thermal Environments. Second edition; Taylor & Francis. New York, NY.
U.S. Department of Energy, EnergyPlus simulation software, Version 9.2.0, 2019. Available at:
https://energyplus.net/