Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Improving indoor air quality and occupant health through smart control of windows and portable air purifiers in residential buildings

Improving indoor air quality and occupant health through smart control of windows and portable... Indoor exposure to PM (particulate matter with aerodynamic diameter less than 2.5 μm) has a substantial 2.5 negative impact on people’s health. However, indoor PM can be controlled through effective ventilation and 2.5 filtration. This study aimed to develop a smart control framework that (1) combines a portable home air purifier (HAP) and window control system to reduce indoor PM concentrations whilst maintaining thermal comfort; (2) 2.5 evaluates the associated health impacts and additional energy use. The proposed framework was demonstrated through a simulation-based case study of a low-energy apartment. The simulation results showed that joint control of HAP and window openings has great potential to not only maintain thermal comfort but also achieve effective PM removal which, consequently, can lead to considerable health benefits at a low additional energy cost. 2.5 Compared to similar previous studies, the strength of the proposed control framework lies in combining window operations and HAPs in the same system and including both thermal comfort and indoor PM as the control 2.5 targets. This work also introduces a novel concept of linking a building control system with a health impact assessment, an important and innovative step in the creation of holistic and responsive building controls. Practical application: This study proposes a novel control framework that jointly controls portable home air purifiers (HAPs) and windows to maintain thermal comfort and achieve effective PM removal. The 2.5 simulation results suggest that such a hybrid control strategy can result in considerable health benefits at low additional energy costs. Keywords Window operation, indoor air quality, thermal comfort, health impact assessment, air purifier, smart building control Received 12 January 2022; Revised 22 April 2022; Accepted 22 April 2022 UCL Institute for Environmental Design and Engineering, London, UK Department of Civil Engineering, Tampere University, Tampere, Finland Corresponding author: Yan Wang, The Bartlett School for Environment, Energy and Resources, UCL Institute for Environmental Design and Engineering, 14 Upper Woburn Place, London WC1H 0NN, UK. Email: y.wang.18@ucl.ac.uk 572 Building Services Engineering Research & Technology 43(5) modern 1-bedroom apartment in London, UK. Com- Introduction pared to similar previous studies, the strength of the Considerable research efforts have been made in re- control framework proposed lies in combining window cent decades to improve indoor air quality (IAQ) to operations and HAPs in the same system and including provide healthy indoor environments for occupants. both thermal comfort and indoor PM as the control 2.5 Currently, carbon dioxide (CO ) sensors are com- targets. This work also introduces a novel concept of monly used in building control systems, but CO is linking a building control system with a health impact not representative of all indoor air pollutants including assessment, an important and innovative step in the particulate matter (PM). PM refers to a mixture of creation of holistic and responsive building controls. airborne liquid droplets and solid particles and is categorised as PM (≤1 μm), PM (≤2.5 μm) and 1 2.5 Material and methods PM (≤10 μm) based on aerodynamic diameter. PM is of particular concern because it can infiltrate 2.5 Description of case study deeply into the respiratory system, causing severe health problems including cardiovascular diseases and The case study residence is a 1-bedroom flat, ap- 2,3 2 asthma. A link has been established between ex- proximately51m , located on the ninth floor of a 13- posure to PM and an increase in all-cause mor- storey residential building built in 2015. The building is 2.5 tality. Thus, reductions in PM are estimated to have sited in a busy urban area in London, UK, adjacent to 2.5 major health benefits. two heavily trafficked roads. The Energy Performance Due to the important role window opening plays Certificate (EPC) for the flat is band B, with band A in shaping the indoor environment, implementing being the highest and band D being the average rating automatic window control systems has been deemed for dwellings in England and Wales. The monitored flat a promising building control strategy. Several papers was located within a building equipped with decen- reported the findings of deploying automated win- tralised mechanical ventilation and heat recovery 5–7 dow systems to facilitate ventilative cooling, or (MVHR) without mechanical cooling in each dwelling. The operation of the MVHR system was, therefore, minimise the amount of time with high indoor CO individually controlled by the occupants of each flat. concentration. In comparison, very limited studies The filtration of the MVHR system in the case study of window control systems considered indoor PM . 2.5 building was found to be minimal (ISO Coarse 45%) in As one rare example, An et al. recently used the a previous study. There was a cooking extract hood reinforcement learning approach to develop an au- available in the open plan kitchen-living room. During tomatic window control system to mitigate indoor the semi-structured interviews, residents from the case PM . However, when outdoor air quality is poor, 2.5 study flat reported that they turned on the MVHR this approach cannot reduce indoor PM concen- 2.5 system only occasionally, although the design intent trations. In this regard, it is worthwhile to consider was to provide continuous background ventilation. In alternative strategies such as portable home air pu- regards to cooking, they reported preparing simple and rifiers (HAPs) that use high-efficiency particulate air (HEPA) filters. Notably, the new generation of HAPs quick breakfasts without using the oven or cooktop, (such as those used in a recent study ) has built-in and cooked dinner about twice a week using the front PM sensors and can be connected to the internet to burner of the cooktop with the extract hood turned on. 2.5 realise instant remote control, showing great po- Temperature, relative humidity, CO and PM 2 2.5 tential to be part of an advanced building automation were measured by air quality sensors (Eltek AQ 110) in system. the living room of the flat and outside the building. As This paper presents a novel control framework that shown in Figure 1, the outdoor sensor was placed on integrates HAPs and automatic window systems to the ground floor at the left façade of building A directly reduce indoor PM concentrations and maintain facing a road. The monitored flat is situated in Building 2.5 thermal comfort. The proposed framework was dem- B, with only the balcony side having external walls and onstrated through a simulation-based case study of a the other boundary walls adjacent to neighbouring flats Wang et al. 573 Figure 1. Indoor and outdoor monitoring locations (left) and the floor plan of the case study flat (right). Note that to protect the residents’ privacy, schematic drawings were used for illustration. or the inner corridor. The indoor sensor was placed on simulation software because it has been previously an internal wall of the living room (about 1.6 m above validated in simulations of indoor pollutants and the floor), while the status (open or closed) of the thermal environment. The key input parameters double-glazed balcony door in the living room was and assumptions for the EP model are detailed in monitored by magnetic reed switches sensors (Eltek Table 2. The U-values for the building envelope GS34). This balcony door is referred to as ‘window’ in and windows were determined in a previous case the following text. The sampling frequency for all study of another flat from the same building. The sensors was every 5 min. The equipment specifications key parameters (e.g. discharge coefficient) related are detailed in Table 1. More details about the envi- to the airflow model were estimated based on a ronmental monitoring and participant interviews and recent calibration effort focusing on the indoor surveys (including sleep and wellbeing surveys, IAQ CO concentration of this flat. Note that because opinions, and occupant behaviours) can be found in the MVHR system was rarely used, it was not 10,12 previous publications. considered during the EP model development phase. Building model development Indoor PM modelling. Cooking schedules, deposi- 2.5 Model inputs and assumptions. A building physics- tion rate and penetration factor were determined to based model was developed in EnergyPlus 9.4 (EP) model indoor PM concentration. The following 2.5 to simulate indoor temperature and PM in the provides a description of the process of determina- 2.5 living room of the flat. EP was chosen as the tion for these factors. 574 Building Services Engineering Research & Technology 43(5) Table 1. The specifications of sensors. Sensor Parameter Range Resolution Accuracy Eltek AQ 110 Temperature 30.0 to 65.0°C 0.1°C ±0.2°C (at 20°C) ±0.4°C (5 to 40°C) ±1.0°C (20 to 65°C) RH 0.0–100.0% 0.1% ±2% RH (0–90% RH) ±4% RH (0–100% RH) CO 0–5000 ppm 1 ppm ±50 ppm 3 3 PM (≤2.5 μm) 0.00–500.00 μg/m 0.01 μg/m 2.5 PM (≤10.0 μm) Eltek GS34 Window status 0 (closed) or 1 (open) Table 2. Key inputs and assumptions for the living room in the EP model. Category Values/settings 2 1 External wall U-value: 0.18 (Wm k ) Air mass flow coefficient: 0.0011 (kgs ) Window Height: 2.0 m, width: 0.9 m 2 1 U-value: 0.92 (Wm k ) Width factor for the open state: 1 Discharge coefficient when open: 0.65 1 1 Air mass flow coefficient when opening is closed: 0.0001 (kgs m ) 1 1 Air mass flow coefficient when opening is closed: 0.0001 (kgs m ) Opening schedule: Measurement data Wall, floor, ceiling adjacent to Assumed to be adiabatic neighbouring flats Door to corridor Always fully open based on occupant survey 2 16 Internal gain Lighting: Power density (7 W/m ) and schedule as per UK NCM Equipment: 30% of the area as a kitchen and 70% of the area as a lounge; the power 2 2 density (kitchen:30.28 W/m , lounge: 3.9 W/m ) and schedules as per UK NCM Hourly external weather Including air temperature, air relative humidity, global, diffuse and direct irradiance, wind speed and direction, obtained from London City Airport station, about 4 km away from the case study building Emission rates (E) and cooking schedules. Smoking 5 μg/m ) in the morning and occasional large peaks and cooking were previously found to be the primary (usually over 50 μg/m ) in the evening. Thus, two sources of high indoor PM concentrations. Since rules were used to generate breakfast and dinner 2.5 the occupants of the case study flat were not smokers, schedules for the EP model: only cooking was modelled here. Consistent with occupant survey results, preliminary observations of 1) when there was a morning peak of measured the measured indoor PM concentrations found that PM concentrations of 5–10 μg/m at 6–9am, 2.5 2.5 there were frequent small peaks (typically around 5min’ use of microwave (E: 0.03 mg/min ) Wang et al. 575 and toasting (E: 0.11 mg/min ) were as- measured indoor PM concentration, for example, 2.5 th sumed to occur during the corresponding 6–9am on 4 August. This difference was likely due period; to the location of the outdoor sensor which was at the 2) when there was an evening peak of PM ground level, directly adjacent to a busy road. The 2.5 concentrations over 50 μg/m at 6–9 pm, measured flat, meanwhile, was located on the other cooking (E: 1.60 mg/min ) was assumed to side of the building on the ninth floor (as illustrated in happen during the corresponding rising pe- Figure 1). Another drawback was the disparity be- riod, and a 20% capture efficiency (CE) of the tween the large peaks in the simulation and measured extract hood was assumed (a midrange of the values, likely a consequence of using general as- CE for the front burner that was estimated to sumptions about cooking emission rates and cooking be 4%–39% ). schedules. Generally, the PM model underesti- 2.5 mates the emission rate, which leads to large errors Penetration factor (P) and deposition rate (K). From (especially RMSE). the literature, the values of both P and K are de- pendent on particle size. For PM , the range for the 2.5 Control strategies and simulation scenarios penetration factor (P) is 0.7–1.0 ; P is less than one when the window is closed, while it should be ap- The mean of the measured indoor PM concen- 2.5 proximately equal to one when the window is open trations during the monitoring period in this flat was 21,22 3 3 for naturally ventilated buildings. The measured 4.90 μg/m , below the WHO annual limit of 5 μg/m , outdoor PM data was used as the outdoor con- and no days exceeded the WHO 24-h limit of 15 μg/ 2.5 3 25 taminant source in the EP model. The deposition rate m . Thus, to create a case that more closely re- was reported to be more varied, e.g. 0.06–0.39 h , sembles PM concentrations modelled in other 2.5 123 1 24 0.21–0.63 h , 0.30–0.69 h . After comparing intervention studies, hypothetical scenarios were the simulated and measured indoor PM concen- developed for one summer week and one winter 2.5 trations, the best-fit values of P (ranging between 0.7 week. The modelled parameters are illustrated in and 1.0) and K (ranging between 0.06 and 0.69) were Table 4, and represent scenarios with both high in- found using the assumed cooking emission rate door and outdoor PM concentrations. A 15-min 2.5 (from above), the inferred cooking schedule, and the breakfast and a 30-min dinner were set to repeat measured outdoor PM data. every day, based upon the cooking schedule adopted 2.5 in a previous study. The outdoor PM data was 2.5 Model tests. The outcomes of test simulations sourced from an outdoor air quality station about showed that the combination of K = 0.69 and p = 0.7 2.2 km away from the case study building. The (when the window is closed), 1 (when the window is chosen weeks saw higher-than-average levels of open) gave the best fit of the simulated PM con- outdoor PM concentrations in both the summer 2.5 2.5 centration to the measurement data, in terms of and winter periods. All control strategies were metrics listed in Table 3. Therefore, these values simulated using EP runtime language. were adopted for all later simulation scenarios. Note The four scenarios simulated for the summer that due to a lack of the heating system operational week are described below and summarised in data, the model was only tested for the non-heating Table 5: period. As is shown in Figure 2, the general trends of Baseline. The window is operated as measured using predicted indoor PM concentration and indoor sensors and no HAP is used. 2.5 temperature closely match the measured ones, and the large indoor PM peaks were well captured. HAP mode. The HAP is modelled as being located in 2.5 However, some limitations with the model were also the centre of the living room, close to the occupants’ noted. When the window was open, the estimated seating area. The control logic illustrated in Figure 3 indoor PM concentration could be higher than the operates the window as measured, while the HAP is 2.5 576 Building Services Engineering Research & Technology 43(5) Table 3. Comparison between simulated and measured indoor PM concentration and indoor temperature. 2.5 a b Metrics Indoor PM concentration Indoor temperature 2.5 Mean bias error (MBA) 0.68 (μg/m ) 0.6 (°C) Mean absolute error (MAE) 1.81 (μg/m ) 1.2 (°C) Root mean square error (RMSE) 5.23 (μg/m ) 1.5 (°C) Pearson’s correlation coefficient 0.65 0.70 half-hourly running means of simulated and measured indoor PM concentrations were compared. 2.5 half-hourly running means of simulated indoor air temperatures and measured indoor temperatures were compared. Figure 2. Demonstration of EP model estimates compared with measurements for two weeks. Note that the half-hourly running means for both PM concentration and temperature were used to better illustrate the trend. 2.5 Table 4. Simulation setting for hypothetic summer and winter week. Period Dates Cooking schedules Outdoor PM file 2.5 nd 19 Summer 22 –29 August Emission rate: 1.6 (mg/min) week 2019 Sourced from Greenwich-John Harrison May station of Weekdays: Breakfast (7–7.15 th th am) and dinner (7.30–8 pm) London air quality Network https://www.londonair. Winter 18 –25 org.uk/LondonAir/ week November Weekends: Breakfast (9–9.15 am) and dinner (7.30–8 pm) 2019 Wang et al. 577 Table 5. Controls of windows and HAPs in different scenarios. Period Scenario Window operation HAP operation Summer week Baseline As measured — HAP mode As measured Control logic in Figure 3 Auto-window mode Control logic in Figure 4 — Hybrid mode Control logic in Figure 4 Control logic in Figure 3 Winter week Baseline As measured — HAP mode As measured Control logic in Figure 3 MVHR mode Closed — Hybrid mode Closed Control logic in Figure 3 Figure 3. HAP control logic. activated once the indoor PM concentration rea- set to 303 m /h, corresponding to a medium fan 2.5 ches the ‘HAP-on’ threshold and stops running once speed of the HAP with five different operating modes the concentration falls to the ‘HAP-off’ threshold. used in a previous study. The power of the HAP 3 3 27 The HAP-on threshold was set to be 15 μg/m (the was modelled as 17 W per 100 m /h of CADR. The WHO 24-h limit ) in both HAP and hybrid modes, HAP operation was assumed to be independent of the as daily performance is of interest for this study. The window operations based on the findings of recent 3 12,28 HAP-off threshold was set to 5 μg/m in both the work. HAP and hybrid modes, as preliminary tests found higher HAP-off thresholds could result in cycling on/ Auto-window control mode. Due to security consid- off too often. The clean air delivery rate (CADR) was erations, the window is programmed to be fully 578 Building Services Engineering Research & Technology 43(5) closed at midnight and when people are away, and limit or closed when below the lower limit of EN the flat-level occupancy was determined based on 16798-1 Category II adaptive comfort temperature. both passive infrared (PIR) and CO sensors, with In all other conditions, the default window setting is details available in a previous paper. In brief, this fully open. The control logic for the auto-window method relied upon positive values from the PIR mode is illustrated in Figure 4. sensors and then used the CO concentration to evaluate the negative detection results from the PIR Hybrid control mode. In this mode, the HAP (control sensors. At other times, the window is set to be fully logic shown in Figure 3) and window (control logic open when the indoor temperature is above the upper shown in Figure 4) control functions are running in Figure 4. Auto-window control algorithm. Wang et al. 579 parallel. The window is operated to prioritise thermal reductions in indoor PM concentrations. Formulae 2.5 comfort, as in the auto-window mode. However, if from Miller and Hurley were the basis for the cal- outdoor PM concentration is high, and indoor culation of changes in mortality and life 2.5 32,36 temperature remains within the comfort zone, the expectancy. The life-table model was im- window will be closed and the HAP will be running. plemented with the open-source statistical software This strategy aids in efficient HAP operation. The R. A schematic diagram of the model inputs, HAP was located and operated as described in the structure and flow is presented in Figure 5. The same HAP mode. underlying birth and mortality rates from the starting Another set of four scenarios was also simulated year (2019) were assumed to apply in all future years. for the winter week, as detailed in Table 5. The heating system was set to work with a setpoint temperature of Health model parameterisation. The life-table model 21°C with schedules as found in the UK NCM da- was used to determine the benefit from the reduction tabase. No automatic window mode was modelled of indoor PM in residences such as the case study 2.5 in the wintertime, as opening the window to reduce flat in the UK from the use of building environmental PM concentration in winter would introduce a cold controls that automate the use of HAPs and window 2.5 draught and increase the heating load. Instead, in operations. Reductions in mean daily exposure were MVHR and hybrid modes, the mechanical ventilation from the time spent in the living room where the air was simulated to provide continuous background purifier was located and was estimated to be 7 h per ventilation that met the minimum requirement by the day based on occupancy monitoring and other sur- 2 38 UK government (0.3 l/s/m based on Approved veys. The results from the modelled case study flat Document F Volume 1: Dwellings 2021 edition – for were used for all scenarios: the average of the use in England ) with the window shut to avoid heat modelled concentrations of the summer and winter loss. The hybrid mode for the winter week was a weeks from the baseline, automated window mode in combination of the MVHR system and HAP. The summer with MVHR system mode in winter, HAP power of the MVHR system was modelled as 42 W and hybrid modes in both summer and winter. based on manufacturer information. No filters were Population and age-specific disease and mortality modelled for the MVHR system due to the minimal data for 2019 from the Office for National Statistics filtration of the MVHR system as mentioned above. were used to parameterise the model. Mortality rates and relative risks (RR) for causes the Global Burden of Disease (GBD) found to be associated with PM 2.5 Health impact assessment were included in the model: all-cause, lung cancer, Background and health model description. Quantitative chronic obstructive pulmonary disease (COPD), health impact assessments are used to estimate lower respiratory infection (LRI), stroke and ische- future rates of mortality and morbidity from dif- mic heart disease (IHD). Age-specific all-cause and ferent interventions compared to what is predicted disease-specific mortality rates were taken from the without such changes. These assessments were used 2019 GBD study. The upper and lower limits of the to evaluate the impact of changes to ambient air 95% confidence intervals of the RRs were calculated quality at the urban and regional scales. One and used to test impacts across the range of potential approach to the assessment of changes in population risks (which will be further discussed in the next mortality is life-table models which predict survival section). patterns based on changes in age-specificdeath Previous findings from other research showed that rates. This type of quantification of health impact the use of a lag between the intervention that reduces hasbeen usedtoassess air pollution at national PM concentrations and changes in health out- 2.5 scales, as well as the evaluation of building-level comes (i.e. cessation lag) made relatively little dif- 34,35 33 changes in exposure. ference to the life-table results over the long-term. In the work presented here, life-table models were Therefore, the model used in the work described here used to quantify the impacts on mortality from does not include a cessation lag. 580 Building Services Engineering Research & Technology 43(5) Figure 5. The conceptual framework for life-table calculations of the impact on mortality from automated control of window operations and HAP use. Health model uncertainty analysis. Recognising that the when indoor PM concentration exceeded the 2.5 exposure-response function per change in PM could WHO limit to 4 days (compared to 6 days in the 2.5 introduce uncertainty into the model, the effect of using baseline scenario), while still maintaining thermal the range of values within the 95% confidence intervals comfort. The key action taken that reduced indoor of the RRs derived from the 2019 Global Burden of PM was the automatic opening of the window 2.5 Diseases was tested using the 95% confidence inter- during morning cooking on the first two days. vals. This method was in line with the recommenda- Nevertheless, the result shows that relying solely on tions for sensitivity analysis made by COMEAP. window controls may not be sufficient when both indoor and outdoor pollution are high. Results HAP mode. Note that the window opening schedule and temperature profiles for the HAP mode were the Summer week same as in the baseline scenario of the summer week. th Baseline scenario. As seen in Figure 6, the daily mean As shown in Figure 8, there were still two days (25 th of PM concentration exceeds the WHO 24-h limit and 27 August) when, even with the use of HAP, the 2.5 of 15 μg/m on 6 days out of the week, while the daily mean concentration of indoor PM was above 2.5 th th indoor temperature stayed within the comfort range the WHO limit, with another two days (24 and 26 the whole time. August) approaching the limit. The primary factor was that outdoor PM levels were high on those 2.5 Auto-window mode. As shown in Figure 7, the au- days, therefore, leaving the window open for long tomatic window system reduced the number of days periods worsened indoor conditions. Wang et al. 581 Figure 6. Summer week: Baseline. Figure 7. Summer week: Auto-window mode. Hybrid mode. When both automatic HAP and win- indoor temperature was not compromised and stayed dow controls were used, the indoor PM concen- within the comfort range. The main advantage of the 2.5 tration was reduced substantially with no days joint control of HAP and windows was that the exceeding the WHO limit. As shown in Figure 9, the window was shut when outdoor pollution was high, 582 Building Services Engineering Research & Technology 43(5) Figure 8. Summer week: HAP mode. Note that the corresponding window status and temperature profiles were the same as the baseline and are therefore not repeated here. Figure 9. Summer week: Hybrid mode. such that not only the working burden of HAP was time, because the outdoor PM concentration was 2.5 minimised but also the overall indoor PM con- above the defined limit, but the window was still 2.5 centration was lower. On the other hand, the hybrid open for a short period for three times in the after- control algorithm sought opportunities to open the noon and evening. window for ventilation whenever the outdoor con- Metrics from several aspects are provided in th ditions allowed. For example, on 26 August, the Table 6 for each scenario in the summer week. Indoor window was directed to be closed for most of the temperature is consistently maintained within the Wang et al. 583 comfort range in each scenario. As for PM , the as high as the WHO 24-h limit. The indoor tem- 2.5 hybrid mode was the most effective, and noticeably, perature was maintained around the heating point required much less electricity use than the HAP during the scheduled hours due to fixed heating mode. schedules. Mechanical ventilation and heat recovery mode. A Winter week small decrease in indoor PM concentration was 2.5 Baseline. As shown in Figure 10, with the window predicted to be achieved when the MVHR system mainly staying closed and the same cooking was operating to provide the minimum required schedule, the daily mean concentration of indoor ventilation rate without high-grade filters, and the PM was very similar across the week, almost twice window staying closed, as shown in Table 7. 2.5 Table 6. Metrics for evaluation of different control modes in the summer week. Auto-window HAP Hybrid Baseline mode mode mode Mean indoor PM concentration (μg/m ) 26.64 17.45 13.80 7.67 2.5 HAP running time (hours) —— 67.3 19.3 Weekly HAP electricity use (kWh) —— 3.4 1.0 Number of days with the daily PM concentration mean over 64 2 0 2.5 WHO 24-h limit Percentage of time outside comfort temperature range 0% 0% 0% 0% Figure 10. Winter week: Baseline. 584 Building Services Engineering Research & Technology 43(5) HAP mode. Utilising the HAP led to a large re- in terms of mean indoor PM concentration. That is 2.5 duction of the indoor PM concentration. As shown because the outdoor PM concentration was often 2.5 2.5 in Figure 11, the daily mean indoor PM concen- higher than the HAP-on threshold in the studied 2.5 tration was estimated to be below the WHO threshold winter week. on all days. Health assessment Hybrid mode. Same as in the HAP mode scenario, the purification effect was estimated to be substantial, Based on the modelled indoor PM concentrations of 2.5 as reflected in reduced indoor PM levels and all the case study flat, the mean years of life gained 2.5 daily means below the WHO limit, as shown in (YLGs) per 100,000 people in a population across the Table 7. modelled period (97 years) was approximately 19,000, Using an MVHR system without high-grade fil- 43,000, and 51,000 for the automatic window/MVHR, ters was not effective in reducing indoor PM HAP, and hybrid modes respectively. The results for the 2.5 concentration in the simulated winter scenario. The lower and upper confidence intervals of the relative performances of HAP and hybrid modes were similar risks, as well as the means, are shown in Table 8. Table 7. Metrics for evaluation of different control modes in the winter week. Baseline MVHR mode HAP mode Hybrid mode Mean concentration of indoor PM (μg/m ) 33.67 25.75 9.54 9.82 2.5 Weekly HAP running time (hours) —— 23.2 21.1 Weekly HAP electricity use (kWh) —— 1.2 1.1 Weekly MVHR electricity use (kWh) — 4.7 – 4.7 Number of days with the daily mean over WHO 24-h limit 7 7 0 0 Figure 11. Winter week: HAP mode. Wang et al. 585 Table 8. Summary of life-table model estimates of changes in mortality per 100,000 population from different environmental control strategies based on modelled PM concentrations in case study flat. 2.5 Total YLG per 100,000 pop. Total YLG per 100,000 pop. Total YLG per 100,000 pop. Mode (Mean RR) (Lower limit RR) (Upper limit RR) Auto-window 18,723 13,902 23,331 HAP 43,338 30,695 56,505 Hybrid 51,094 34,561 67,965 20,39 cooker extract hoods reported in other studies, Discussion although it could be a useful exploration in future Strengths research. Additionally, the proposed control frame- work accompanied by health impact assessments was The study presented here proposes a novel frame- tested in a case study flat as proof of concept, but it is work that controls both HAP operation and window expected to be more meaningful to extend this work opening to reduce indoor PM concentration 2.5 to large-scale building stock modelling, as the life- without compromising occupant thermal comfort. It table health modelling is a population-based method. should be noted that the presented work focuses on Moreover, this proposed framework that features proposing and testing a building control framework HAPs currently only considers PM as the control 2.5 rather than quantification of the accuracy of the target, but other types of pollutants such as NO simulation results. Considering that the vast majority should be considered in future work. of prior studies focused on thermal comfort and very The work presented here assumes that appropriate few considered indoor PM , this work advances 2.5 safety and protection measures (such as pinch pro- research on smart window control systems. More- tection and finger guards) for automatic windows can over, this framework aims to assess the potential be accommodated in residential applications. Con- health impacts associated with the adoption of venience, safety and security issues, and how they building controls that reduce indoor PM levels in 2.5 affect acceptance and compliance of automated homes. As reduced exposures to PM are expected 2.5 systems, should be considered in future work. The to contribute to improving occupants’ health, an model demonstrated in this work only considered evaluation of intervention measures from the per- fully open or closed window states due to the binary spective of health benefits is meaningful but remains nature of the window sensor data, but future work a missing part of previous work of the same nature. could explore options of incremental openings. It should be acknowledged that the health impact as- Limitations and future work sessment is a population-based average. The avail- ability of data on specific indoor concentrations, The current building model only considered cooking health effects of reductions in indoor PM , and as the indoor PM source alongside a general as- 2.5 2.5 differences in the relative risk due to the primary sumption about the emission rate and cooking source (indoor or outdoor) of PM exposure are schedules. This simplification may not be able to limited. Another limitation is the relative risks used estimate levels and patterns in more complicated situations, e.g. homes with smokers, occupants with in the evaluation were drawn from the GBD which more diverse cooking types (associated with a wide were derived for ambient and household (i.e. solid- range of PM emission rates) and more flexible or fuel combustion indoors) PM exposures. How- 2.5 2.5 unpredictable cooking times. This work also did not ever, previous studies have used the GBD data for the model the range of utilisation rates or efficiencies of estimation of risk, and still other research has 586 Building Services Engineering Research & Technology 43(5) highlighted the importance of indoor PM to total ORCID iDs 2.5 41–43 exposure. Yan Wang  https://orcid.org/0000-0003-4266-1125 Health modelling provides a useful method of Farhang Tahmasebi https://orcid.org/0000-0001-5727- evaluating the impact of interventions on population health. However, the reliability of the results is subject to the accuracy of available sources of information, and Declaration of conflicting interests the ability to add scientific credibility when those sources are uncertain. Building simulations can allow The author(s) declared no potential conflicts of interest for the provision of a rich and readily customisable with respect to the research, authorship, and/or publication dataset to add to the predictive power of health of this article. modelling when empiric data are not available. Greater integration of the building simulation to modelled Funding health outcomes could help inform future iterations of The author(s) disclosed receipt of the following financial the control framework. Additionally, as more infor- support for the research, authorship, and/or publication of mation is gained about user behaviour and the feasi- this article: This work was supported by the EIT-Digital bility of long-term use of HAPs, more robust project, “Quality of Indoor Air on Sites Matched with estimations of actual risk reductions can be incorpo- Outdoor Air Quality Datasets to Improve Wellbeing rated into the health impact assessments. Lastly, the Outcomes” (activity number 19144); UK EPSRC (Project model presented here does not consider morbidities, number 559487). such as asthma, which are associated with PM . 2.5 Future work would include a fuller range of health References outcomes beyond mortality. 1. Chenari B, Dias Carrilho J and Gameiro da Silva M. Towards sustainable, energy-efficient and healthy ventilation strategies in buildings: a review. Renew Conclusion Sustain Energy Rev 2016; 59: 1426–1447. 2. Pope CA III, Coleman N, Pond ZA, et al. Fine par- This study develops a novel control framework that ticulate air pollution and human mortality: 25+ years integrates portable home air purifiers and window of cohort studies. Environ Res 2020; 183: 108924. control systems with the aim to reduce indoor PM 2.5 3. Fan J, Li S, Fan C, et al. The impact of PM2. 5 on concentrations whilst maintaining thermal comfort. asthma emergency department visits: a systematic The proposed framework was demonstrated through a review and meta-analysis. Environ Sci Pollut Res series of building simulations for an apartment as the 2016; 23(1): 843–850. virtual testbed. The results show that the joint control of 4. Schraufnagel DE, Balmes JR, De Matteis S, et al. HAP and window operation based on indoor and Health benefits of air pollution reduction. Ann Am outdoor environmental conditions is one control Thoracic Soc 2019; 16(12): 1478–1487. mechanism that has the potential to not only maintain 5. Psomas T, Heiselberg P, Lyme T, et al. Automated thermal comfort but also achieve effective PM re- 2.5 roof window control system to address overheating moval which, consequently, can lead to considerable on renovated houses: summertime assessment and health benefits at a relatively low extra energy cost. The intercomparison. Energy Build 2017; 138: 35–46. impact on population health via the implementation of 6. Psomas T, Fiorentini M, Kokogiannakis G, et al. the type of control logic demonstrated in this work is Ventilative cooling through automated window predicted to be substantial. The work presented here is opening control systems to address thermal discom- the first known of its kind to integrate the assessment of fort risk during the summer period: framework, potential changes to mortality from the implementation simulation and parametric analysis. Energy Build of advanced building control systems that measure and 2017; 153: 18–30. predict PM concentrations indoors. 2.5 Wang et al. 587 7. Fiorentini M, Serale G, Kokogiannakis G, et al. 18. He C. Contribution from indoor sources to particle Development and evaluation of a comfort-oriented number and mass concentrations in residential houses. control strategy for thermal management of mixed- Atmos Environ 2004; 38(21): 3405–3415. mode ventilated buildings. Energy Build 2019; 202: 19. Shrubsole C, Ridley I, Biddulph P, et al. Indoor PM2.5 109347. exposure in London’s domestic stock: modelling 8. Stazi F, Naspi F, Ulpiani G, et al. Indoor air quality current and future exposures following energy effi- and thermal comfort optimization in classrooms de- cient refurbishment. Atmos Environ 2012; 62: veloping an automatic system for windows opening 336–343. and closing. Energy Build 2017; 139: 732–746. 20. Lunden MM, Delp WW and Singer BC. Capture 9. An Y, Xia T, You R, et al. A reinforcement learning efficiency of cooking-related fine and ultrafine par- approach for control of window behavior to reduce ticles by residential exhaust hoods. Indoor Air 2015; indoor PM2.5 concentrations in naturally ventilated 25(1): 45–58. buildings. Build Environ 2021; 200: 107978. 21. Chen C and Zhao B. Review of relationship between 10. Cooper E, Wang Y, Stamp S, et al. Use of portable air indoor and outdoor particles: I/O ratio, infiltration purifiers in homes: operating behaviour, effect on factor and penetration factor. Atmos Environ 2011; indoor PM2.5 and perceived indoor air quality. Build 45(2): 275–288. Environ 2021; 191: 107621. 22. Taylor J, Shrubsole C, Davies M, et al. The modifying 11. Arcipowska A, Mariottini F, Anagnostopoulos F, effect of the building envelope on population expo- et al. Energy performance certificates across the sure to PM2.5 from outdoor sources. Indoor Air 2014; EU–a mapping of national approaches.BPIE, 24(6): 639–651. 2014. 23. Ji W and Zhao B. Contribution of outdoor-originating 12. Wang Y, Tahmasebi F, Cooper E, et al. Exploring the particles, indoor-emitted particles and indoor sec- relationship between window operation behavior ondary organic aerosol (SOA) to residential indoor and thermal and air quality factors: a case study of PM2.5 concentration: a model-based estimation. UK residential buildings. JBuild Eng 2022; 48: Build Environ 2015; 90: 196–205. 103997. 24. Xie W, Fan Y, Zhang X, et al. A mathematical model 13. Taylor J, Shrubsole C, Biddulph P, et al. Simulation of for predicting indoor PM concentration under dif- 2.5 pollution transport in buildings: the importance of ferent ventilation methods in residential buildings. taking into account dynamic thermal effects. Build Build Serv Eng Res Technol 2020; 41(6): 694–708. Serv Eng Res Technol 2014; 35(6): 682–690. 25. World Health Organization. Air quality guidelines: 14. Jain N, Burman E, Stamp S, et al. Cross-sectoral global update 2005: particulate matter, ozone, ni- assessment of the performance gap using calibrated trogen dioxide, and sulfur dioxide. Genève: World building energy performance simulation. Energy Health Organization, 2006. Build 2020; 224: 110271. 26. World Health Organization. WHO global air quality 15. Tahmasebi F, Wang Y, Cooper E, et al. Window guidelines. particulate matter (PM and PM ), 2.5 10 operation behaviour indoor air quality during lock- ozone, nitrogen dioxide, sulfur dioxide and carbon down: a monitoring-based simulation-assisted study monoxide. Geneva: World Health Organization, 2021. in London. Build Serv Eng Res Technol 2021; 43: 27. Ji W, Chen C and Zhao B. A comparative study of the 01436244211017786. effects of ventilation-purification strategies on air 16. Building Research Establishment. National Calcu- quality and energy consumption in Beijing, China. lation Method (NCM) database (Activity, Construc- Build Simul 2020; 14(3): 813–825. tion, and Glazing), https://www.uk-ncm.org.uk 28. Wang Y, Tahmasebi F, Cooper E, et al. An investigation (2018, accessed 08 August 2021). of the influencing factors for occupants’ operation of 17. Jones N, Thornton C, Mark D, et al. Indoor/outdoor windows in apartments equipped with portable air pu- relationships of particulate matter in domestic homes rifiers. Build Environ 2021; 205: 108260. with roadside, urban rural locations. Atmos Environ 29. Standardization Ecf EN 16798–1:2019. Energy perfor- 2000; 34(16): 2603–2612. mance of buildings, in: ventilation for buildings. Indoor 588 Building Services Engineering Research & Technology 43(5) Environ input parameters for des and assess of energ COMEAP report, https://www.gov.uk/government/ perform of buildings addressing indoor air qual, thermal publications/comeap-mortality-effects-of-long-term- environment, lighting and acoustics.CEN,2019. exposure-to-particulate-air-pollution-in-the-uk (2010, 30. HM Government. Approved document F: volume 1 accessed 17 May 2021). applies dwellings (2021 Edition – Use in England), 37. R Core Team. R: A language and environment for https://assets.publishing.service.gov.uk/government/ statistical computing. Vienna, Austria: R Found uploads/system/uploads/attachment_data/file/1045918/ Statistical Computing, 2018. ADF1.pdf (2021, accessed 08 April 2022). 38. Klepeis NE, Nelson WC, Ott WR, et al. The National 31. Manisalidis I, Stavropoulou E, Stavropoulos A, et al. Human Activity Pattern Survey (NHAPS): a resource for Environmental and health impacts of air pollution: a assessing exposure to environmental pollutants. JExpo review. Front Public Health 2020; 8: 14. Anal Environ Epidemiol 2001; 11(3): 231–252. 32. Miller B and Hurley J. Life table methods for 39. Zhao H, Chan WR, Delp WW, et al. Factors impacting quantitative impact assessments in chronic mortality. range hood use in california houses and low-income J Epidemiol Commun Health 2003; 57(3): 200–206. apartments. Int J Environ Res Public Health 2020; 33. COMEAP. The mortality effects of long-term expo- 17(23): 8870. sure to particulate air pollution in the United King- 40. WHO. Global burden of disease, https://www.who. dom. COMEAP, 2010. Retrieved from London, UK: int/data/gho/data/themes/mortality-and-global-health- http://comeap.org.uk/documents/reports/128-the-mortality- estimates (2019, accessed 07 June 2021). effects-of-long-term-exposure-to-particulate-air-pollution- 41. Chi R, Chen C, Li H, et al. Different health effects of in-the-uk.html indoor-and outdoor-originated PM on cardiopul- 2.5 34. Milner J, Chalabi Z, Vardoulakis S, et al. Housing monary function in COPD patients and healthy elderly interventions and health: Quantifying the impact of adults. Indoor Air 2019; 29(2): 192–201. indoor particles on mortality and morbidity with 42. Martins NR and Carrilho da Graça G. Impact of PM 2.5 disease recovery. Environ Int 2015; 81: 73–79. in indoor urban environments: a review. Sustain Cities 35. Hamilton I, Milner J, Chalabi Z, et al. Health effects of Soc 2018; 42: 259–275. home energy efficiency interventions in England: a 43. Li Z, Wen Q and Zhang R. Sources, health effects and modelling study. BMJ Open 2015; 5(4): e007298. control strategies of indoor fine particulate matter 36. Miller BG and Hurley JF. Technical aspects of life (PM ): a review. Sci Total Environ 2017; 586: 2.5 table analyses. supporting paper for the 2010 610–622. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Building Services Engineering Research and Technology: An International Journal SAGE

Improving indoor air quality and occupant health through smart control of windows and portable air purifiers in residential buildings

Download PDF
18 pages

Loading next page...
 
/lp/sage/improving-indoor-air-quality-and-occupant-health-through-smart-control-yELlWA27kZ

References (39)

Publisher
SAGE
Copyright
© The Author(s) 2022
ISSN
0143-6244
eISSN
1477-0849
DOI
10.1177/01436244221099482
Publisher site
See Article on Publisher Site

Abstract

Indoor exposure to PM (particulate matter with aerodynamic diameter less than 2.5 μm) has a substantial 2.5 negative impact on people’s health. However, indoor PM can be controlled through effective ventilation and 2.5 filtration. This study aimed to develop a smart control framework that (1) combines a portable home air purifier (HAP) and window control system to reduce indoor PM concentrations whilst maintaining thermal comfort; (2) 2.5 evaluates the associated health impacts and additional energy use. The proposed framework was demonstrated through a simulation-based case study of a low-energy apartment. The simulation results showed that joint control of HAP and window openings has great potential to not only maintain thermal comfort but also achieve effective PM removal which, consequently, can lead to considerable health benefits at a low additional energy cost. 2.5 Compared to similar previous studies, the strength of the proposed control framework lies in combining window operations and HAPs in the same system and including both thermal comfort and indoor PM as the control 2.5 targets. This work also introduces a novel concept of linking a building control system with a health impact assessment, an important and innovative step in the creation of holistic and responsive building controls. Practical application: This study proposes a novel control framework that jointly controls portable home air purifiers (HAPs) and windows to maintain thermal comfort and achieve effective PM removal. The 2.5 simulation results suggest that such a hybrid control strategy can result in considerable health benefits at low additional energy costs. Keywords Window operation, indoor air quality, thermal comfort, health impact assessment, air purifier, smart building control Received 12 January 2022; Revised 22 April 2022; Accepted 22 April 2022 UCL Institute for Environmental Design and Engineering, London, UK Department of Civil Engineering, Tampere University, Tampere, Finland Corresponding author: Yan Wang, The Bartlett School for Environment, Energy and Resources, UCL Institute for Environmental Design and Engineering, 14 Upper Woburn Place, London WC1H 0NN, UK. Email: y.wang.18@ucl.ac.uk 572 Building Services Engineering Research & Technology 43(5) modern 1-bedroom apartment in London, UK. Com- Introduction pared to similar previous studies, the strength of the Considerable research efforts have been made in re- control framework proposed lies in combining window cent decades to improve indoor air quality (IAQ) to operations and HAPs in the same system and including provide healthy indoor environments for occupants. both thermal comfort and indoor PM as the control 2.5 Currently, carbon dioxide (CO ) sensors are com- targets. This work also introduces a novel concept of monly used in building control systems, but CO is linking a building control system with a health impact not representative of all indoor air pollutants including assessment, an important and innovative step in the particulate matter (PM). PM refers to a mixture of creation of holistic and responsive building controls. airborne liquid droplets and solid particles and is categorised as PM (≤1 μm), PM (≤2.5 μm) and 1 2.5 Material and methods PM (≤10 μm) based on aerodynamic diameter. PM is of particular concern because it can infiltrate 2.5 Description of case study deeply into the respiratory system, causing severe health problems including cardiovascular diseases and The case study residence is a 1-bedroom flat, ap- 2,3 2 asthma. A link has been established between ex- proximately51m , located on the ninth floor of a 13- posure to PM and an increase in all-cause mor- storey residential building built in 2015. The building is 2.5 tality. Thus, reductions in PM are estimated to have sited in a busy urban area in London, UK, adjacent to 2.5 major health benefits. two heavily trafficked roads. The Energy Performance Due to the important role window opening plays Certificate (EPC) for the flat is band B, with band A in shaping the indoor environment, implementing being the highest and band D being the average rating automatic window control systems has been deemed for dwellings in England and Wales. The monitored flat a promising building control strategy. Several papers was located within a building equipped with decen- reported the findings of deploying automated win- tralised mechanical ventilation and heat recovery 5–7 dow systems to facilitate ventilative cooling, or (MVHR) without mechanical cooling in each dwelling. The operation of the MVHR system was, therefore, minimise the amount of time with high indoor CO individually controlled by the occupants of each flat. concentration. In comparison, very limited studies The filtration of the MVHR system in the case study of window control systems considered indoor PM . 2.5 building was found to be minimal (ISO Coarse 45%) in As one rare example, An et al. recently used the a previous study. There was a cooking extract hood reinforcement learning approach to develop an au- available in the open plan kitchen-living room. During tomatic window control system to mitigate indoor the semi-structured interviews, residents from the case PM . However, when outdoor air quality is poor, 2.5 study flat reported that they turned on the MVHR this approach cannot reduce indoor PM concen- 2.5 system only occasionally, although the design intent trations. In this regard, it is worthwhile to consider was to provide continuous background ventilation. In alternative strategies such as portable home air pu- regards to cooking, they reported preparing simple and rifiers (HAPs) that use high-efficiency particulate air (HEPA) filters. Notably, the new generation of HAPs quick breakfasts without using the oven or cooktop, (such as those used in a recent study ) has built-in and cooked dinner about twice a week using the front PM sensors and can be connected to the internet to burner of the cooktop with the extract hood turned on. 2.5 realise instant remote control, showing great po- Temperature, relative humidity, CO and PM 2 2.5 tential to be part of an advanced building automation were measured by air quality sensors (Eltek AQ 110) in system. the living room of the flat and outside the building. As This paper presents a novel control framework that shown in Figure 1, the outdoor sensor was placed on integrates HAPs and automatic window systems to the ground floor at the left façade of building A directly reduce indoor PM concentrations and maintain facing a road. The monitored flat is situated in Building 2.5 thermal comfort. The proposed framework was dem- B, with only the balcony side having external walls and onstrated through a simulation-based case study of a the other boundary walls adjacent to neighbouring flats Wang et al. 573 Figure 1. Indoor and outdoor monitoring locations (left) and the floor plan of the case study flat (right). Note that to protect the residents’ privacy, schematic drawings were used for illustration. or the inner corridor. The indoor sensor was placed on simulation software because it has been previously an internal wall of the living room (about 1.6 m above validated in simulations of indoor pollutants and the floor), while the status (open or closed) of the thermal environment. The key input parameters double-glazed balcony door in the living room was and assumptions for the EP model are detailed in monitored by magnetic reed switches sensors (Eltek Table 2. The U-values for the building envelope GS34). This balcony door is referred to as ‘window’ in and windows were determined in a previous case the following text. The sampling frequency for all study of another flat from the same building. The sensors was every 5 min. The equipment specifications key parameters (e.g. discharge coefficient) related are detailed in Table 1. More details about the envi- to the airflow model were estimated based on a ronmental monitoring and participant interviews and recent calibration effort focusing on the indoor surveys (including sleep and wellbeing surveys, IAQ CO concentration of this flat. Note that because opinions, and occupant behaviours) can be found in the MVHR system was rarely used, it was not 10,12 previous publications. considered during the EP model development phase. Building model development Indoor PM modelling. Cooking schedules, deposi- 2.5 Model inputs and assumptions. A building physics- tion rate and penetration factor were determined to based model was developed in EnergyPlus 9.4 (EP) model indoor PM concentration. The following 2.5 to simulate indoor temperature and PM in the provides a description of the process of determina- 2.5 living room of the flat. EP was chosen as the tion for these factors. 574 Building Services Engineering Research & Technology 43(5) Table 1. The specifications of sensors. Sensor Parameter Range Resolution Accuracy Eltek AQ 110 Temperature 30.0 to 65.0°C 0.1°C ±0.2°C (at 20°C) ±0.4°C (5 to 40°C) ±1.0°C (20 to 65°C) RH 0.0–100.0% 0.1% ±2% RH (0–90% RH) ±4% RH (0–100% RH) CO 0–5000 ppm 1 ppm ±50 ppm 3 3 PM (≤2.5 μm) 0.00–500.00 μg/m 0.01 μg/m 2.5 PM (≤10.0 μm) Eltek GS34 Window status 0 (closed) or 1 (open) Table 2. Key inputs and assumptions for the living room in the EP model. Category Values/settings 2 1 External wall U-value: 0.18 (Wm k ) Air mass flow coefficient: 0.0011 (kgs ) Window Height: 2.0 m, width: 0.9 m 2 1 U-value: 0.92 (Wm k ) Width factor for the open state: 1 Discharge coefficient when open: 0.65 1 1 Air mass flow coefficient when opening is closed: 0.0001 (kgs m ) 1 1 Air mass flow coefficient when opening is closed: 0.0001 (kgs m ) Opening schedule: Measurement data Wall, floor, ceiling adjacent to Assumed to be adiabatic neighbouring flats Door to corridor Always fully open based on occupant survey 2 16 Internal gain Lighting: Power density (7 W/m ) and schedule as per UK NCM Equipment: 30% of the area as a kitchen and 70% of the area as a lounge; the power 2 2 density (kitchen:30.28 W/m , lounge: 3.9 W/m ) and schedules as per UK NCM Hourly external weather Including air temperature, air relative humidity, global, diffuse and direct irradiance, wind speed and direction, obtained from London City Airport station, about 4 km away from the case study building Emission rates (E) and cooking schedules. Smoking 5 μg/m ) in the morning and occasional large peaks and cooking were previously found to be the primary (usually over 50 μg/m ) in the evening. Thus, two sources of high indoor PM concentrations. Since rules were used to generate breakfast and dinner 2.5 the occupants of the case study flat were not smokers, schedules for the EP model: only cooking was modelled here. Consistent with occupant survey results, preliminary observations of 1) when there was a morning peak of measured the measured indoor PM concentrations found that PM concentrations of 5–10 μg/m at 6–9am, 2.5 2.5 there were frequent small peaks (typically around 5min’ use of microwave (E: 0.03 mg/min ) Wang et al. 575 and toasting (E: 0.11 mg/min ) were as- measured indoor PM concentration, for example, 2.5 th sumed to occur during the corresponding 6–9am on 4 August. This difference was likely due period; to the location of the outdoor sensor which was at the 2) when there was an evening peak of PM ground level, directly adjacent to a busy road. The 2.5 concentrations over 50 μg/m at 6–9 pm, measured flat, meanwhile, was located on the other cooking (E: 1.60 mg/min ) was assumed to side of the building on the ninth floor (as illustrated in happen during the corresponding rising pe- Figure 1). Another drawback was the disparity be- riod, and a 20% capture efficiency (CE) of the tween the large peaks in the simulation and measured extract hood was assumed (a midrange of the values, likely a consequence of using general as- CE for the front burner that was estimated to sumptions about cooking emission rates and cooking be 4%–39% ). schedules. Generally, the PM model underesti- 2.5 mates the emission rate, which leads to large errors Penetration factor (P) and deposition rate (K). From (especially RMSE). the literature, the values of both P and K are de- pendent on particle size. For PM , the range for the 2.5 Control strategies and simulation scenarios penetration factor (P) is 0.7–1.0 ; P is less than one when the window is closed, while it should be ap- The mean of the measured indoor PM concen- 2.5 proximately equal to one when the window is open trations during the monitoring period in this flat was 21,22 3 3 for naturally ventilated buildings. The measured 4.90 μg/m , below the WHO annual limit of 5 μg/m , outdoor PM data was used as the outdoor con- and no days exceeded the WHO 24-h limit of 15 μg/ 2.5 3 25 taminant source in the EP model. The deposition rate m . Thus, to create a case that more closely re- was reported to be more varied, e.g. 0.06–0.39 h , sembles PM concentrations modelled in other 2.5 123 1 24 0.21–0.63 h , 0.30–0.69 h . After comparing intervention studies, hypothetical scenarios were the simulated and measured indoor PM concen- developed for one summer week and one winter 2.5 trations, the best-fit values of P (ranging between 0.7 week. The modelled parameters are illustrated in and 1.0) and K (ranging between 0.06 and 0.69) were Table 4, and represent scenarios with both high in- found using the assumed cooking emission rate door and outdoor PM concentrations. A 15-min 2.5 (from above), the inferred cooking schedule, and the breakfast and a 30-min dinner were set to repeat measured outdoor PM data. every day, based upon the cooking schedule adopted 2.5 in a previous study. The outdoor PM data was 2.5 Model tests. The outcomes of test simulations sourced from an outdoor air quality station about showed that the combination of K = 0.69 and p = 0.7 2.2 km away from the case study building. The (when the window is closed), 1 (when the window is chosen weeks saw higher-than-average levels of open) gave the best fit of the simulated PM con- outdoor PM concentrations in both the summer 2.5 2.5 centration to the measurement data, in terms of and winter periods. All control strategies were metrics listed in Table 3. Therefore, these values simulated using EP runtime language. were adopted for all later simulation scenarios. Note The four scenarios simulated for the summer that due to a lack of the heating system operational week are described below and summarised in data, the model was only tested for the non-heating Table 5: period. As is shown in Figure 2, the general trends of Baseline. The window is operated as measured using predicted indoor PM concentration and indoor sensors and no HAP is used. 2.5 temperature closely match the measured ones, and the large indoor PM peaks were well captured. HAP mode. The HAP is modelled as being located in 2.5 However, some limitations with the model were also the centre of the living room, close to the occupants’ noted. When the window was open, the estimated seating area. The control logic illustrated in Figure 3 indoor PM concentration could be higher than the operates the window as measured, while the HAP is 2.5 576 Building Services Engineering Research & Technology 43(5) Table 3. Comparison between simulated and measured indoor PM concentration and indoor temperature. 2.5 a b Metrics Indoor PM concentration Indoor temperature 2.5 Mean bias error (MBA) 0.68 (μg/m ) 0.6 (°C) Mean absolute error (MAE) 1.81 (μg/m ) 1.2 (°C) Root mean square error (RMSE) 5.23 (μg/m ) 1.5 (°C) Pearson’s correlation coefficient 0.65 0.70 half-hourly running means of simulated and measured indoor PM concentrations were compared. 2.5 half-hourly running means of simulated indoor air temperatures and measured indoor temperatures were compared. Figure 2. Demonstration of EP model estimates compared with measurements for two weeks. Note that the half-hourly running means for both PM concentration and temperature were used to better illustrate the trend. 2.5 Table 4. Simulation setting for hypothetic summer and winter week. Period Dates Cooking schedules Outdoor PM file 2.5 nd 19 Summer 22 –29 August Emission rate: 1.6 (mg/min) week 2019 Sourced from Greenwich-John Harrison May station of Weekdays: Breakfast (7–7.15 th th am) and dinner (7.30–8 pm) London air quality Network https://www.londonair. Winter 18 –25 org.uk/LondonAir/ week November Weekends: Breakfast (9–9.15 am) and dinner (7.30–8 pm) 2019 Wang et al. 577 Table 5. Controls of windows and HAPs in different scenarios. Period Scenario Window operation HAP operation Summer week Baseline As measured — HAP mode As measured Control logic in Figure 3 Auto-window mode Control logic in Figure 4 — Hybrid mode Control logic in Figure 4 Control logic in Figure 3 Winter week Baseline As measured — HAP mode As measured Control logic in Figure 3 MVHR mode Closed — Hybrid mode Closed Control logic in Figure 3 Figure 3. HAP control logic. activated once the indoor PM concentration rea- set to 303 m /h, corresponding to a medium fan 2.5 ches the ‘HAP-on’ threshold and stops running once speed of the HAP with five different operating modes the concentration falls to the ‘HAP-off’ threshold. used in a previous study. The power of the HAP 3 3 27 The HAP-on threshold was set to be 15 μg/m (the was modelled as 17 W per 100 m /h of CADR. The WHO 24-h limit ) in both HAP and hybrid modes, HAP operation was assumed to be independent of the as daily performance is of interest for this study. The window operations based on the findings of recent 3 12,28 HAP-off threshold was set to 5 μg/m in both the work. HAP and hybrid modes, as preliminary tests found higher HAP-off thresholds could result in cycling on/ Auto-window control mode. Due to security consid- off too often. The clean air delivery rate (CADR) was erations, the window is programmed to be fully 578 Building Services Engineering Research & Technology 43(5) closed at midnight and when people are away, and limit or closed when below the lower limit of EN the flat-level occupancy was determined based on 16798-1 Category II adaptive comfort temperature. both passive infrared (PIR) and CO sensors, with In all other conditions, the default window setting is details available in a previous paper. In brief, this fully open. The control logic for the auto-window method relied upon positive values from the PIR mode is illustrated in Figure 4. sensors and then used the CO concentration to evaluate the negative detection results from the PIR Hybrid control mode. In this mode, the HAP (control sensors. At other times, the window is set to be fully logic shown in Figure 3) and window (control logic open when the indoor temperature is above the upper shown in Figure 4) control functions are running in Figure 4. Auto-window control algorithm. Wang et al. 579 parallel. The window is operated to prioritise thermal reductions in indoor PM concentrations. Formulae 2.5 comfort, as in the auto-window mode. However, if from Miller and Hurley were the basis for the cal- outdoor PM concentration is high, and indoor culation of changes in mortality and life 2.5 32,36 temperature remains within the comfort zone, the expectancy. The life-table model was im- window will be closed and the HAP will be running. plemented with the open-source statistical software This strategy aids in efficient HAP operation. The R. A schematic diagram of the model inputs, HAP was located and operated as described in the structure and flow is presented in Figure 5. The same HAP mode. underlying birth and mortality rates from the starting Another set of four scenarios was also simulated year (2019) were assumed to apply in all future years. for the winter week, as detailed in Table 5. The heating system was set to work with a setpoint temperature of Health model parameterisation. The life-table model 21°C with schedules as found in the UK NCM da- was used to determine the benefit from the reduction tabase. No automatic window mode was modelled of indoor PM in residences such as the case study 2.5 in the wintertime, as opening the window to reduce flat in the UK from the use of building environmental PM concentration in winter would introduce a cold controls that automate the use of HAPs and window 2.5 draught and increase the heating load. Instead, in operations. Reductions in mean daily exposure were MVHR and hybrid modes, the mechanical ventilation from the time spent in the living room where the air was simulated to provide continuous background purifier was located and was estimated to be 7 h per ventilation that met the minimum requirement by the day based on occupancy monitoring and other sur- 2 38 UK government (0.3 l/s/m based on Approved veys. The results from the modelled case study flat Document F Volume 1: Dwellings 2021 edition – for were used for all scenarios: the average of the use in England ) with the window shut to avoid heat modelled concentrations of the summer and winter loss. The hybrid mode for the winter week was a weeks from the baseline, automated window mode in combination of the MVHR system and HAP. The summer with MVHR system mode in winter, HAP power of the MVHR system was modelled as 42 W and hybrid modes in both summer and winter. based on manufacturer information. No filters were Population and age-specific disease and mortality modelled for the MVHR system due to the minimal data for 2019 from the Office for National Statistics filtration of the MVHR system as mentioned above. were used to parameterise the model. Mortality rates and relative risks (RR) for causes the Global Burden of Disease (GBD) found to be associated with PM 2.5 Health impact assessment were included in the model: all-cause, lung cancer, Background and health model description. Quantitative chronic obstructive pulmonary disease (COPD), health impact assessments are used to estimate lower respiratory infection (LRI), stroke and ische- future rates of mortality and morbidity from dif- mic heart disease (IHD). Age-specific all-cause and ferent interventions compared to what is predicted disease-specific mortality rates were taken from the without such changes. These assessments were used 2019 GBD study. The upper and lower limits of the to evaluate the impact of changes to ambient air 95% confidence intervals of the RRs were calculated quality at the urban and regional scales. One and used to test impacts across the range of potential approach to the assessment of changes in population risks (which will be further discussed in the next mortality is life-table models which predict survival section). patterns based on changes in age-specificdeath Previous findings from other research showed that rates. This type of quantification of health impact the use of a lag between the intervention that reduces hasbeen usedtoassess air pollution at national PM concentrations and changes in health out- 2.5 scales, as well as the evaluation of building-level comes (i.e. cessation lag) made relatively little dif- 34,35 33 changes in exposure. ference to the life-table results over the long-term. In the work presented here, life-table models were Therefore, the model used in the work described here used to quantify the impacts on mortality from does not include a cessation lag. 580 Building Services Engineering Research & Technology 43(5) Figure 5. The conceptual framework for life-table calculations of the impact on mortality from automated control of window operations and HAP use. Health model uncertainty analysis. Recognising that the when indoor PM concentration exceeded the 2.5 exposure-response function per change in PM could WHO limit to 4 days (compared to 6 days in the 2.5 introduce uncertainty into the model, the effect of using baseline scenario), while still maintaining thermal the range of values within the 95% confidence intervals comfort. The key action taken that reduced indoor of the RRs derived from the 2019 Global Burden of PM was the automatic opening of the window 2.5 Diseases was tested using the 95% confidence inter- during morning cooking on the first two days. vals. This method was in line with the recommenda- Nevertheless, the result shows that relying solely on tions for sensitivity analysis made by COMEAP. window controls may not be sufficient when both indoor and outdoor pollution are high. Results HAP mode. Note that the window opening schedule and temperature profiles for the HAP mode were the Summer week same as in the baseline scenario of the summer week. th Baseline scenario. As seen in Figure 6, the daily mean As shown in Figure 8, there were still two days (25 th of PM concentration exceeds the WHO 24-h limit and 27 August) when, even with the use of HAP, the 2.5 of 15 μg/m on 6 days out of the week, while the daily mean concentration of indoor PM was above 2.5 th th indoor temperature stayed within the comfort range the WHO limit, with another two days (24 and 26 the whole time. August) approaching the limit. The primary factor was that outdoor PM levels were high on those 2.5 Auto-window mode. As shown in Figure 7, the au- days, therefore, leaving the window open for long tomatic window system reduced the number of days periods worsened indoor conditions. Wang et al. 581 Figure 6. Summer week: Baseline. Figure 7. Summer week: Auto-window mode. Hybrid mode. When both automatic HAP and win- indoor temperature was not compromised and stayed dow controls were used, the indoor PM concen- within the comfort range. The main advantage of the 2.5 tration was reduced substantially with no days joint control of HAP and windows was that the exceeding the WHO limit. As shown in Figure 9, the window was shut when outdoor pollution was high, 582 Building Services Engineering Research & Technology 43(5) Figure 8. Summer week: HAP mode. Note that the corresponding window status and temperature profiles were the same as the baseline and are therefore not repeated here. Figure 9. Summer week: Hybrid mode. such that not only the working burden of HAP was time, because the outdoor PM concentration was 2.5 minimised but also the overall indoor PM con- above the defined limit, but the window was still 2.5 centration was lower. On the other hand, the hybrid open for a short period for three times in the after- control algorithm sought opportunities to open the noon and evening. window for ventilation whenever the outdoor con- Metrics from several aspects are provided in th ditions allowed. For example, on 26 August, the Table 6 for each scenario in the summer week. Indoor window was directed to be closed for most of the temperature is consistently maintained within the Wang et al. 583 comfort range in each scenario. As for PM , the as high as the WHO 24-h limit. The indoor tem- 2.5 hybrid mode was the most effective, and noticeably, perature was maintained around the heating point required much less electricity use than the HAP during the scheduled hours due to fixed heating mode. schedules. Mechanical ventilation and heat recovery mode. A Winter week small decrease in indoor PM concentration was 2.5 Baseline. As shown in Figure 10, with the window predicted to be achieved when the MVHR system mainly staying closed and the same cooking was operating to provide the minimum required schedule, the daily mean concentration of indoor ventilation rate without high-grade filters, and the PM was very similar across the week, almost twice window staying closed, as shown in Table 7. 2.5 Table 6. Metrics for evaluation of different control modes in the summer week. Auto-window HAP Hybrid Baseline mode mode mode Mean indoor PM concentration (μg/m ) 26.64 17.45 13.80 7.67 2.5 HAP running time (hours) —— 67.3 19.3 Weekly HAP electricity use (kWh) —— 3.4 1.0 Number of days with the daily PM concentration mean over 64 2 0 2.5 WHO 24-h limit Percentage of time outside comfort temperature range 0% 0% 0% 0% Figure 10. Winter week: Baseline. 584 Building Services Engineering Research & Technology 43(5) HAP mode. Utilising the HAP led to a large re- in terms of mean indoor PM concentration. That is 2.5 duction of the indoor PM concentration. As shown because the outdoor PM concentration was often 2.5 2.5 in Figure 11, the daily mean indoor PM concen- higher than the HAP-on threshold in the studied 2.5 tration was estimated to be below the WHO threshold winter week. on all days. Health assessment Hybrid mode. Same as in the HAP mode scenario, the purification effect was estimated to be substantial, Based on the modelled indoor PM concentrations of 2.5 as reflected in reduced indoor PM levels and all the case study flat, the mean years of life gained 2.5 daily means below the WHO limit, as shown in (YLGs) per 100,000 people in a population across the Table 7. modelled period (97 years) was approximately 19,000, Using an MVHR system without high-grade fil- 43,000, and 51,000 for the automatic window/MVHR, ters was not effective in reducing indoor PM HAP, and hybrid modes respectively. The results for the 2.5 concentration in the simulated winter scenario. The lower and upper confidence intervals of the relative performances of HAP and hybrid modes were similar risks, as well as the means, are shown in Table 8. Table 7. Metrics for evaluation of different control modes in the winter week. Baseline MVHR mode HAP mode Hybrid mode Mean concentration of indoor PM (μg/m ) 33.67 25.75 9.54 9.82 2.5 Weekly HAP running time (hours) —— 23.2 21.1 Weekly HAP electricity use (kWh) —— 1.2 1.1 Weekly MVHR electricity use (kWh) — 4.7 – 4.7 Number of days with the daily mean over WHO 24-h limit 7 7 0 0 Figure 11. Winter week: HAP mode. Wang et al. 585 Table 8. Summary of life-table model estimates of changes in mortality per 100,000 population from different environmental control strategies based on modelled PM concentrations in case study flat. 2.5 Total YLG per 100,000 pop. Total YLG per 100,000 pop. Total YLG per 100,000 pop. Mode (Mean RR) (Lower limit RR) (Upper limit RR) Auto-window 18,723 13,902 23,331 HAP 43,338 30,695 56,505 Hybrid 51,094 34,561 67,965 20,39 cooker extract hoods reported in other studies, Discussion although it could be a useful exploration in future Strengths research. Additionally, the proposed control frame- work accompanied by health impact assessments was The study presented here proposes a novel frame- tested in a case study flat as proof of concept, but it is work that controls both HAP operation and window expected to be more meaningful to extend this work opening to reduce indoor PM concentration 2.5 to large-scale building stock modelling, as the life- without compromising occupant thermal comfort. It table health modelling is a population-based method. should be noted that the presented work focuses on Moreover, this proposed framework that features proposing and testing a building control framework HAPs currently only considers PM as the control 2.5 rather than quantification of the accuracy of the target, but other types of pollutants such as NO simulation results. Considering that the vast majority should be considered in future work. of prior studies focused on thermal comfort and very The work presented here assumes that appropriate few considered indoor PM , this work advances 2.5 safety and protection measures (such as pinch pro- research on smart window control systems. More- tection and finger guards) for automatic windows can over, this framework aims to assess the potential be accommodated in residential applications. Con- health impacts associated with the adoption of venience, safety and security issues, and how they building controls that reduce indoor PM levels in 2.5 affect acceptance and compliance of automated homes. As reduced exposures to PM are expected 2.5 systems, should be considered in future work. The to contribute to improving occupants’ health, an model demonstrated in this work only considered evaluation of intervention measures from the per- fully open or closed window states due to the binary spective of health benefits is meaningful but remains nature of the window sensor data, but future work a missing part of previous work of the same nature. could explore options of incremental openings. It should be acknowledged that the health impact as- Limitations and future work sessment is a population-based average. The avail- ability of data on specific indoor concentrations, The current building model only considered cooking health effects of reductions in indoor PM , and as the indoor PM source alongside a general as- 2.5 2.5 differences in the relative risk due to the primary sumption about the emission rate and cooking source (indoor or outdoor) of PM exposure are schedules. This simplification may not be able to limited. Another limitation is the relative risks used estimate levels and patterns in more complicated situations, e.g. homes with smokers, occupants with in the evaluation were drawn from the GBD which more diverse cooking types (associated with a wide were derived for ambient and household (i.e. solid- range of PM emission rates) and more flexible or fuel combustion indoors) PM exposures. How- 2.5 2.5 unpredictable cooking times. This work also did not ever, previous studies have used the GBD data for the model the range of utilisation rates or efficiencies of estimation of risk, and still other research has 586 Building Services Engineering Research & Technology 43(5) highlighted the importance of indoor PM to total ORCID iDs 2.5 41–43 exposure. Yan Wang  https://orcid.org/0000-0003-4266-1125 Health modelling provides a useful method of Farhang Tahmasebi https://orcid.org/0000-0001-5727- evaluating the impact of interventions on population health. However, the reliability of the results is subject to the accuracy of available sources of information, and Declaration of conflicting interests the ability to add scientific credibility when those sources are uncertain. Building simulations can allow The author(s) declared no potential conflicts of interest for the provision of a rich and readily customisable with respect to the research, authorship, and/or publication dataset to add to the predictive power of health of this article. modelling when empiric data are not available. Greater integration of the building simulation to modelled Funding health outcomes could help inform future iterations of The author(s) disclosed receipt of the following financial the control framework. Additionally, as more infor- support for the research, authorship, and/or publication of mation is gained about user behaviour and the feasi- this article: This work was supported by the EIT-Digital bility of long-term use of HAPs, more robust project, “Quality of Indoor Air on Sites Matched with estimations of actual risk reductions can be incorpo- Outdoor Air Quality Datasets to Improve Wellbeing rated into the health impact assessments. Lastly, the Outcomes” (activity number 19144); UK EPSRC (Project model presented here does not consider morbidities, number 559487). such as asthma, which are associated with PM . 2.5 Future work would include a fuller range of health References outcomes beyond mortality. 1. Chenari B, Dias Carrilho J and Gameiro da Silva M. Towards sustainable, energy-efficient and healthy ventilation strategies in buildings: a review. Renew Conclusion Sustain Energy Rev 2016; 59: 1426–1447. 2. Pope CA III, Coleman N, Pond ZA, et al. Fine par- This study develops a novel control framework that ticulate air pollution and human mortality: 25+ years integrates portable home air purifiers and window of cohort studies. Environ Res 2020; 183: 108924. control systems with the aim to reduce indoor PM 2.5 3. Fan J, Li S, Fan C, et al. The impact of PM2. 5 on concentrations whilst maintaining thermal comfort. asthma emergency department visits: a systematic The proposed framework was demonstrated through a review and meta-analysis. Environ Sci Pollut Res series of building simulations for an apartment as the 2016; 23(1): 843–850. virtual testbed. The results show that the joint control of 4. Schraufnagel DE, Balmes JR, De Matteis S, et al. HAP and window operation based on indoor and Health benefits of air pollution reduction. Ann Am outdoor environmental conditions is one control Thoracic Soc 2019; 16(12): 1478–1487. mechanism that has the potential to not only maintain 5. Psomas T, Heiselberg P, Lyme T, et al. Automated thermal comfort but also achieve effective PM re- 2.5 roof window control system to address overheating moval which, consequently, can lead to considerable on renovated houses: summertime assessment and health benefits at a relatively low extra energy cost. The intercomparison. Energy Build 2017; 138: 35–46. impact on population health via the implementation of 6. Psomas T, Fiorentini M, Kokogiannakis G, et al. the type of control logic demonstrated in this work is Ventilative cooling through automated window predicted to be substantial. The work presented here is opening control systems to address thermal discom- the first known of its kind to integrate the assessment of fort risk during the summer period: framework, potential changes to mortality from the implementation simulation and parametric analysis. Energy Build of advanced building control systems that measure and 2017; 153: 18–30. predict PM concentrations indoors. 2.5 Wang et al. 587 7. Fiorentini M, Serale G, Kokogiannakis G, et al. 18. He C. Contribution from indoor sources to particle Development and evaluation of a comfort-oriented number and mass concentrations in residential houses. control strategy for thermal management of mixed- Atmos Environ 2004; 38(21): 3405–3415. mode ventilated buildings. Energy Build 2019; 202: 19. Shrubsole C, Ridley I, Biddulph P, et al. Indoor PM2.5 109347. exposure in London’s domestic stock: modelling 8. Stazi F, Naspi F, Ulpiani G, et al. Indoor air quality current and future exposures following energy effi- and thermal comfort optimization in classrooms de- cient refurbishment. Atmos Environ 2012; 62: veloping an automatic system for windows opening 336–343. and closing. Energy Build 2017; 139: 732–746. 20. Lunden MM, Delp WW and Singer BC. Capture 9. An Y, Xia T, You R, et al. A reinforcement learning efficiency of cooking-related fine and ultrafine par- approach for control of window behavior to reduce ticles by residential exhaust hoods. Indoor Air 2015; indoor PM2.5 concentrations in naturally ventilated 25(1): 45–58. buildings. Build Environ 2021; 200: 107978. 21. Chen C and Zhao B. Review of relationship between 10. Cooper E, Wang Y, Stamp S, et al. Use of portable air indoor and outdoor particles: I/O ratio, infiltration purifiers in homes: operating behaviour, effect on factor and penetration factor. Atmos Environ 2011; indoor PM2.5 and perceived indoor air quality. Build 45(2): 275–288. Environ 2021; 191: 107621. 22. Taylor J, Shrubsole C, Davies M, et al. The modifying 11. Arcipowska A, Mariottini F, Anagnostopoulos F, effect of the building envelope on population expo- et al. Energy performance certificates across the sure to PM2.5 from outdoor sources. Indoor Air 2014; EU–a mapping of national approaches.BPIE, 24(6): 639–651. 2014. 23. Ji W and Zhao B. Contribution of outdoor-originating 12. Wang Y, Tahmasebi F, Cooper E, et al. Exploring the particles, indoor-emitted particles and indoor sec- relationship between window operation behavior ondary organic aerosol (SOA) to residential indoor and thermal and air quality factors: a case study of PM2.5 concentration: a model-based estimation. UK residential buildings. JBuild Eng 2022; 48: Build Environ 2015; 90: 196–205. 103997. 24. Xie W, Fan Y, Zhang X, et al. A mathematical model 13. Taylor J, Shrubsole C, Biddulph P, et al. Simulation of for predicting indoor PM concentration under dif- 2.5 pollution transport in buildings: the importance of ferent ventilation methods in residential buildings. taking into account dynamic thermal effects. Build Build Serv Eng Res Technol 2020; 41(6): 694–708. Serv Eng Res Technol 2014; 35(6): 682–690. 25. World Health Organization. Air quality guidelines: 14. Jain N, Burman E, Stamp S, et al. Cross-sectoral global update 2005: particulate matter, ozone, ni- assessment of the performance gap using calibrated trogen dioxide, and sulfur dioxide. Genève: World building energy performance simulation. Energy Health Organization, 2006. Build 2020; 224: 110271. 26. World Health Organization. WHO global air quality 15. Tahmasebi F, Wang Y, Cooper E, et al. Window guidelines. particulate matter (PM and PM ), 2.5 10 operation behaviour indoor air quality during lock- ozone, nitrogen dioxide, sulfur dioxide and carbon down: a monitoring-based simulation-assisted study monoxide. Geneva: World Health Organization, 2021. in London. Build Serv Eng Res Technol 2021; 43: 27. Ji W, Chen C and Zhao B. A comparative study of the 01436244211017786. effects of ventilation-purification strategies on air 16. Building Research Establishment. National Calcu- quality and energy consumption in Beijing, China. lation Method (NCM) database (Activity, Construc- Build Simul 2020; 14(3): 813–825. tion, and Glazing), https://www.uk-ncm.org.uk 28. Wang Y, Tahmasebi F, Cooper E, et al. An investigation (2018, accessed 08 August 2021). of the influencing factors for occupants’ operation of 17. Jones N, Thornton C, Mark D, et al. Indoor/outdoor windows in apartments equipped with portable air pu- relationships of particulate matter in domestic homes rifiers. Build Environ 2021; 205: 108260. with roadside, urban rural locations. Atmos Environ 29. Standardization Ecf EN 16798–1:2019. Energy perfor- 2000; 34(16): 2603–2612. mance of buildings, in: ventilation for buildings. Indoor 588 Building Services Engineering Research & Technology 43(5) Environ input parameters for des and assess of energ COMEAP report, https://www.gov.uk/government/ perform of buildings addressing indoor air qual, thermal publications/comeap-mortality-effects-of-long-term- environment, lighting and acoustics.CEN,2019. exposure-to-particulate-air-pollution-in-the-uk (2010, 30. HM Government. Approved document F: volume 1 accessed 17 May 2021). applies dwellings (2021 Edition – Use in England), 37. R Core Team. R: A language and environment for https://assets.publishing.service.gov.uk/government/ statistical computing. Vienna, Austria: R Found uploads/system/uploads/attachment_data/file/1045918/ Statistical Computing, 2018. ADF1.pdf (2021, accessed 08 April 2022). 38. Klepeis NE, Nelson WC, Ott WR, et al. The National 31. Manisalidis I, Stavropoulou E, Stavropoulos A, et al. Human Activity Pattern Survey (NHAPS): a resource for Environmental and health impacts of air pollution: a assessing exposure to environmental pollutants. JExpo review. Front Public Health 2020; 8: 14. Anal Environ Epidemiol 2001; 11(3): 231–252. 32. Miller B and Hurley J. Life table methods for 39. Zhao H, Chan WR, Delp WW, et al. Factors impacting quantitative impact assessments in chronic mortality. range hood use in california houses and low-income J Epidemiol Commun Health 2003; 57(3): 200–206. apartments. Int J Environ Res Public Health 2020; 33. COMEAP. The mortality effects of long-term expo- 17(23): 8870. sure to particulate air pollution in the United King- 40. WHO. Global burden of disease, https://www.who. dom. COMEAP, 2010. Retrieved from London, UK: int/data/gho/data/themes/mortality-and-global-health- http://comeap.org.uk/documents/reports/128-the-mortality- estimates (2019, accessed 07 June 2021). effects-of-long-term-exposure-to-particulate-air-pollution- 41. Chi R, Chen C, Li H, et al. Different health effects of in-the-uk.html indoor-and outdoor-originated PM on cardiopul- 2.5 34. Milner J, Chalabi Z, Vardoulakis S, et al. Housing monary function in COPD patients and healthy elderly interventions and health: Quantifying the impact of adults. Indoor Air 2019; 29(2): 192–201. indoor particles on mortality and morbidity with 42. Martins NR and Carrilho da Graça G. Impact of PM 2.5 disease recovery. Environ Int 2015; 81: 73–79. in indoor urban environments: a review. Sustain Cities 35. Hamilton I, Milner J, Chalabi Z, et al. Health effects of Soc 2018; 42: 259–275. home energy efficiency interventions in England: a 43. Li Z, Wen Q and Zhang R. Sources, health effects and modelling study. BMJ Open 2015; 5(4): e007298. control strategies of indoor fine particulate matter 36. Miller BG and Hurley JF. Technical aspects of life (PM ): a review. Sci Total Environ 2017; 586: 2.5 table analyses. supporting paper for the 2010 610–622.

Journal

Building Services Engineering Research and Technology: An International Journal – SAGE

Published: Sep 1, 2022

Keywords: Window operation; indoor air quality; thermal comfort; health impact assessment; air purifier; smart building control

There are no references for this article.