Next Article in Journal
Integrated Evaluation of Bio-Based Phase Change Materials to Reduce Operational and Embodied Carbon in Service Buildings Across Multiple Climate Zones
Previous Article in Journal
Dynamic Simulation of Enterprise-Level Strategic Choices in Intelligent Construction: Integration of Evolutionary Game Theory and System Dynamics
Previous Article in Special Issue
Quantifying the Impact of Energy Storage Capacity on Building Energy Flexibility: A Case Study of the PV-ESS-GSHP System
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Optimization on Ventilation Time in Winter Based on Energy, Thermal Comfortable and Air Quality in Severe Cold Rural Dwellings of Northeast China

by
Xueyan Zhang
1,
Xingkuo Zhang
1,
Yiming Yang
2 and
Jing Li
3,*
1
Laboratory of Building Environment and New Energy Resources, Dalian University of Technology, Dalian 116024, China
2
Qingdao Hisense Hitachi Air-Conditioning System Co., Ltd., Qingdao 266000, China
3
Central Hospital of Dalian University of Technology, Faculty of Medicine, Dalian University of Technology, Dalian 116024, China
*
Author to whom correspondence should be addressed.
Buildings 2025, 15(20), 3718; https://doi.org/10.3390/buildings15203718
Submission received: 1 September 2025 / Revised: 25 September 2025 / Accepted: 29 September 2025 / Published: 15 October 2025
(This article belongs to the Special Issue Sustainable Architecture and Healthy Environment)

Abstract

In Northeast China’s severe cold regions, increasingly airtight rural dwellings face a critical challenge: traditional biomass-fueled heating and cooking generate severe indoor particulate matter (PM) pollution, creating a sharp trade-off between maintaining thermal comfort and ensuring safe indoor air quality through ventilation. While multi-objective optimization is widely applied to urban buildings, its use to develop practical, behavior-based ventilation strategies for resource-constrained rural dwellings in this context represents a significant research gap. This study integrates field measurements of occupant behavior and environmental parameters from 192 households with a coupled thermal-PM2.5 predictive model. The NSGA-II genetic algorithm was employed to perform a multi-objective optimization, balancing PM reduction against thermal comfort. The optimization reveals that short, high-intensity ventilation bursts are allowed. A typical optimized event can reduce post-cooking PM2.5 concentrations to near-guideline levels while maintaining the indoor temperature within the residents’ adaptive comfort zone. This research provides the first evidence-based, region-specific natural ventilation guidelines for these dwellings. The findings offer a practical, no-cost strategy to mitigate health risks from indoor air pollution without significant energy penalties, providing a theoretical basis for future smart ventilation system design.

1. Introduction

In recent years, ensuring indoor environmental health has become a global priority, with household air pollution in rural areas recognized as a significant health risk [1]. Concurrently, as these regions undergo an energy transition, residents’ demand for greater thermal comfort has led to the widespread adoption of measures to improve building airtightness, such as constructing sun-rooms and sealing openings. Simultaneously, traditional cooking and heating methods can generate high concentrations of smoke and pollutants, leading to respiratory diseases among residents. Consequently, rural residents often resort to opening doors or using mechanical smoke exhaust to improve indoor air quality. In severe cold regions, such intuitive ventilation practices are often suboptimal, adversely affecting thermal comfort and increasing energy consumption. Therefore, it is crucial to scientifically determine the optimal duration of natural ventilation to balance these competing factors.
Research has consistently shown that indoor air quality in rural dwellings is a severe health risk. Albalak, R. et al. [2] investigated indoor air pollution from biomass cooking in Bolivian rural dwellings, documenting extreme PM10 concentrations (1830 μg/m3) that indicated significant health risks. Parikh, J. et al. [3] also conducted indoor air quality measurements in 418 rural households, with results showing inhalable particle concentrations ranging from 500 to 2000 μg/m3. Focusing on influencing factors, Zhang et al. [4] conducted a field study in Fuxin, China, and revealed that PM emissions were linked to indoor temperature, humidity, and ventilation practices. Further studies in China’s severe cold regions by Wang et al. [5] confirmed severe wintertime exceedances of a wide range of pollutants, including PM2.5 and CO. Parallel to these air quality concerns, researchers had examined residents’ thermal comfort. Studies by Gong et al. [6] and Ma et al. [7] both concluded that rural residents have different thermal requirements and exhibit greater adaptability to colder conditions. This adaptability is a key factor for improving their quality of life under specific local circumstances. Currently, indoor air pollution control in residential buildings mainly includes measures such as pollution source control [8,9], air purification [10,11], and ventilation dilution. Based on the aforementioned studies, replacing biomass fuel combustion methods in rural areas in the short term is challenging. Furthermore, the excessive pollutants in rural dwellings and the current air purification technologies are insufficient to eliminate all types of indoor pollutants. In contrast, ventilation dilution was found to be more appropriate for rural residential environments. The effectiveness of ventilation was supported by Song et al. [12] and Kong et al. [13], whose studies showed it could control various pollutants. However, the actual practice is complex; Warren et al. [14] found that ventilation behavior is strongly driven by thermal comfort, and Nicol [15] further modeled this behavior in relation to outdoor temperature. This finding underscored that proper natural ventilation control could effectively enhance indoor air quality and safeguard human health.
These aforementioned studies collectively indicate that while natural ventilation is a crucial tool, its effectiveness depends on a complex balance of environmental and behavioral factors, pointing to the need for multi-objective optimization. However, while various optimization methods have been employed in building science, their application has been primarily focused on urban buildings equipped with automated HVAC systems, with limited extension to rural areas. The limitation exists because models for urban contexts are not directly applicable to rural dwellings, which rely on occupant-driven, intermittent ventilation and thus require a tailored approach. These urban-focused studies have demonstrated the power of methodology. For example, Yu Wei et al. [16] used the NSGA-II algorithm with neural networks to optimize energy and comfort. Jin Guohui et al. [17] applied a BPNN model to optimize comfort, energy, and cost. Similarly, Zhang Yong et al. [18] and Li Zhu et al. [19] also applied multi-objective optimization to balance key performance indicators. Therefore, while powerful optimization tools exist and the specific challenges in rural dwellings are known, a critical research gap remains: the lack of an integrated framework capable of applying these methods to simultaneously resolve the conflict between air quality, thermal comfort, and energy use under occupant-driven ventilation strategies.
To address the critical research gap identified in the literature review, this study aims to develop a multi-objective optimization framework for natural ventilation in these specific rural dwellings. The primary objectives are as follows: (1) to characterize the coupled dynamics of the thermal environment and particulate matter concentration under occupant-driven ventilation, and (2) to determine optimal ventilation durations that resolve the trade-off between occupant thermal comfort and indoor air quality. To achieve these objectives, this paper first presents the findings from a comprehensive field survey used to classify dwelling archetypes and establish residents’ thermal demand. Subsequently, a coupled thermal-PM2.5 model, incorporating these behavioral and environmental data, is developed and validated. Finally, the NSGA-II algorithm is employed to conduct a trade-off analysis and propose optimal door-opening durations that provide simple, actionable guidance.
This study conducted field measurements and investigations on the rural house models and indoor heating environments in the severe cold Northeastern region. Firstly, the study classified the baseline rural house models and, in conjunction with the current heating conditions and residents’ behavioral habits, proposed the residents’ thermal demand temperature for indoor heating environments. Secondly, thermal balance and PM2.5 diffusion models for rural houses were established, and functional relationships between natural ventilation duration and area were derived. Thirdly, considering the residents’ thermal demand temperature and the national standard for PM2.5 concentration as constraints, the study utilized the NSGA-II algorithm to conduct a trade-off analysis of the indoor thermal environment and air quality under natural ventilation conditions in rural houses. Finally, the optimal duration for opening doors to ventilate during winter cooking and heating periods in northeastern rural houses was proposed.

2. Methods

2.1. Field Survey and Case Selection

A comprehensive field survey was conducted from 2020 to 2023, encompassing 192 rural households in Northeast China’s severe cold region. Through questionnaires and on-site measurements, extensive data were collected on building characteristics, heating methods, and occupant demographics. This dataset was then analyzed using the K-means clustering method [20] to identify representative dwelling archetypes based on factors including heating costs and household demographics. Three representative types of rural residences in Northeast region were identified, as shown in Table 1. In winter, residents usually take measures such as sealing the building envelope with plastic films to enhance the airtightness of the building and reduce the intrusion of cold air through the gaps around doors and windows. Therefore, the ventilation method for buildings in winter is unilateral ventilation. The second category of residential buildings accounts for 55% of the surveyed households. It can be representative to a certain extent and has been identified as the research object of this study for the in-depth investigation. Focusing on this predominant dwelling type ensures that the findings and the proposed optimization strategies are directly relevant and applicable to the majority of the region’s rural dwellings.

2.2. Data Collection and Measurement

This study primarily focuses on Type II rural dwellings, characterized by brick-concrete structures with uninsulated 370 mm walls, primarily constructed from fired clay bricks. The building design features an improved three-bay layout, with rooms lacking buffer spaces or vestibules, and the kitchen is not equipped with auxiliary devices such as range hoods or exhaust fans. The household consists of two elderly or middle-aged individuals and one younger adult, with an annual heating expenditure ranging between 4000 and 6000 RMB. The windows are single-pane, 3 mm glass, and are covered with plastic film for insulation during winter. A rural dwelling in Fushun City was selected as the prototype, with its three-dimensional space and floor plans shown in Figure 1.
The selected rural residence was oriented southward, with a total floor area of 79 m2. It was constructed in 2000 and was inhabited by two residents during the monitoring period. The household used straw and firewood as the primary fuels for both heating and cooking. The heating system consisted of a traditional Kang bed-stove combined with a rudimentary radiator system. The measurements covered indoor and outdoor air temperature and relative humidity, concentrations of particulate matter (PM10, PM2.5), door opening frequency, CO2 concentration, and air velocity. The layout of the measurement points was shown in Figure 2, and all measurement setups were implemented in accordance with relevant national standards [21,22]. Testing instruments and their measurement accuracy are listed in Table 2.

2.3. Coupled Thermal-PM2.5 Dynamic Model

2.3.1. Assumption Conditions

Assumption conditions for indoor thermal environment under natural ventilation were as follows:
(1)
Assuming the initial condition for the study of ventilation behavior was during farmers’ cooking and heating activities, at this point, the variation in indoor air temperature was considered to be primarily caused by the entry of cold outdoor air into the room.
(2)
The extent of indoor temperature variation depended on the door opening area A and the duration of door opening τ.
Assumption conditions for the mathematical model of the impact of natural ventilation on the rate of indoor PM2.5 variation were as follows:
(1)
The impact of outdoor PM2.5 penetration was considered negligible (Qf ≈ 0), as indoor PM2.5 concentrations during the studied combustion events are typically an order of magnitude higher than ambient outdoor levels, making the indoor source dominant. This assumption is acknowledged as a model limitation during periods of severe outdoor air pollution.
(2)
Neglecting the effects of gas absorption, chemical reactions, condensation, and re-suspension on indoor particle concentration, i.e., RLfAf + S + H + F + C ≈ 0.
(3)
Only considering the PM2.5 generated during cooking and heating processes, neglecting the effects of cleaning activities and smoking.
(4)
Under the condition of sealed doors and windows, the airflow velocity inside the rural residence was assumed to be low. At this point, the inter-zone airflow rates (Qik) between the experimental room and various functional rooms were considered constant at 5 m3/h [23].

2.3.2. Model Establishment

(1)
Mathematical Model for the Impact of Ventilation Behavior on Indoor Thermal Environment
To quantify the impact of a door-opening event on the indoor thermal environment, a mathematical model was established based on the principles of physics. This model predicts the final indoor air temperature by establishing a dynamic heat balance for the room. The key variables include the initial indoor (ta) and outdoor (t0) temperatures, room volume (V), door opening area (A), and ventilation duration (τ). The total heat change (Qtotal) is calculated as the sum of heat exchanged through ventilation (Qven) and other pathways (Qother) to determine the final indoor temperature (tᵢ). The core heat balance equation is expressed as follows:
Q v e n + Q o t h e r = Q t o t a l
In this equation,
Q t o t a l = ρ i C ρ V ( t i t a )
Q v e n = C ρ m i ( t i + t a 2 t 0 )
Q o t h e r = 1.3 3600 τ + 0.108 3600 τ K F t i + t a 2 t 0
m = v ρ i A τ
solved for
t i = C P ρ i v + 0.00039 τ + K F 1000 t i + t a 2 t 0 τ C P ρ i v t a 2 t 0 K F t a τ 2000 C p ρ i v + K F 2000 + C p ρ i υ A τ 2
(2)
Mathematical Model for the Influence of Ventilation Behavior on Indoor PM2.5 Variation Rate
To model the temporal variation in indoor PM2.5 concentration, this study adapted the established particle mass balance model proposed by Tian [24]. The general model was simplified to reflect the conditions of the studied dwellings, where particle decay is primarily governed by ventilation and deposition, with key parameters set according to previous research [25]. The full derivation of the resulting PM2.5 model, C4(τ), is provided in Appendix A. This PM2.5 model was then integrated with the thermal model (t) to establish the complete system of equations for optimization. This system defines the relationship between the key ventilation variables (area A, time t) and the two objective functions (temperature (T) and PM2.5 (f2)), as represented in Equation (7).
f 1 A , τ = t i = C P ρ i v + 0.00039 τ + K F 1000 t i + t a 2 t 0 τ C P ρ i v t a 2 t 0 K F t a τ 2000 C p ρ i v + K F 2000 + C p ρ i v A τ 2 f 2 A , τ = C 4 τ = υ C 3 υ A + 0.00657 exp υ A + 0.00657 48.6 + C 0 υ C 3 υ A + 0.00657 * exp υ A + 0.00657 48.6 τ

2.3.3. Model Verification

The initial optimization model was validated by calculating the initial temperature and PM2.5 concentration in the bedroom before opening the door as 23 °C and 321 μg/m3, respectively. After one hour, the bedroom temperature and PM2.5 concentration were calculated to be 7.81 °C and 38.09 μg/m3. These results were close to the measured data of 11.81 °C and 44 μg/m3 before and one hour after opening the door, demonstrating the credibility of the calculations. Additionally, two evaluation indicators recommended by ASHRAE 14-2014, the Normalized Mean Bias Error (NMBE) and the Coefficient of Variation of the Root Mean Square Error (CVRMSE), were considered for validation. ASHRAE 14-2014 suggests that NMBE should be within 10%, and CVRMSE should be within 30% for the results in unit hours, indicating good agreement between the established model and the actual data.
N M B E = 100 × i = 1 n E s , j E m , j C i τ C i 0 n p × E m
C V R E M S E = 100 × i = 1 n E s , j E m , j 2 ÷ n p 2 1 2 E m
The variables in the equation consist of Es,i for simulated data, Em,i for measured data, Em as the average of measured data, n representing the calculated number of hours, and p set to 1. The calculated NMBE and CVRMSE for the bedroom based on measured PM2.5 and temperature are 6.4% and 27.6%, respectively. Both values fall within the specified range of ASHRAE14-2014 regulations, indicating that the multi-objective optimization model established using MATLAB 2022b was accurate and meet the acceptable criteria.

2.4. Multi-Objective Optimization Framework

2.4.1. Multi-Objective Genetic Algorithm

Optimizing both temperature and pollutant concentration is a typical multi-objective optimization problem. The complexity arises from the competitive relationship between indoor temperature and pollutant levels. Achieving better optimization results for indoor temperature may lead to suboptimal indoor thermal comfort. Therefore, the final outcome is often not a unique optimal solution but rather a set of Pareto solutions. For a multi-objective optimization problem min f(x), Pareto solutions were defined by Equation (10):
P = X = X Ω X Ω , f j X f j X , j = 1 , 2 , r
The multi-objective genetic algorithm is an established method for identifying Pareto-optimal solution sets, with increasing application in architectural research [26]. Accordingly, this study selected the highly efficient Non-dominated Sorting Genetic Algorithm II (NSGA-II) proposed by Deb et al. [27]. The algorithm’s effectiveness stems from key mechanisms, including fast non-dominated sorting to rank solutions, crowding distance sorting to maintain diversity, and an elitist strategy to preserve the best-performing solutions across generations. As conceptually illustrated in Figure 3, the algorithm iteratively evolves a population of potential solutions (rectangles) toward the Pareto-optimal front (circles). In this study, this front represents the best possible trade-offs between the two competing objectives: maximizing indoor temperature f1(x) and minimizing pollutant concentration f2(x)).
To ensure the reproducibility of the optimization, the algorithm was implemented in MATLAB. The key parameters were set as follows: a population size of 100, a maximum of 200 generations, a crossover probability of 0.9, and a mutation probability of 0.1. These parameters were determined through preliminary testing to ensure robust convergence to a stable Pareto front.

2.4.2. Demand Temperature During the Heating Season

The two objective functions were the demand temperature during the heating season and the indoor PM2.5 mass concentration. The indoor air temperature represented the air temperature values in various functional rooms during the winter heating season. The higher the indoor air temperature within the acceptable comfort temperature range for farmers, the better the thermal environment of the rural residence. The indicator for indoor air PM2.5 mass concentration represented the PM2.5 concentration values in various functional rooms under different door opening and ventilation time conditions. Considering that the direction of temperature parameter demand was contradictory to the genetic algorithm direction, the negative value of the temperature-solving objective function was taken. Therefore, the objective functions were defined as follows:
M i n f 1 = t a , m i n = M i n f 1 A , τ M i n f 2 = C i , m i n = M i n f 2 A , τ
where ta is the indoor air temperature in degrees Celsius, Ci is the PM2.5 mass concentration in room i in micrograms per cubic meter (μg/m3), A is the door opening area, and is the ventilation time.

2.4.3. Constraint Conditions

(1)
The temperature range acceptable to 90% of the residents was between 12.0 °C and 21.2 °C, with a neutral operating temperature of 16.84 °C.
(2)
According to the standard “Indoor Air Quality Standards” GB/T18883-2022 [28], the recommended value for indoor PM2.5 is set at 50 μg/m3. Therefore, in this study, the indoor pollutant concentration standard for rural residences was defined as CPM2.5 ≤ 50 μg/m3.

3. Results

3.1. Field Measurement Findings

3.1.1. Dwelling Characteristics and Thermal Performance

The surface temperature and heat flux density of the enclosure structure were measured by positioning the measurement points at the center of the wall surface, avoiding direct heat sources. A JTNT-A multi-channel temperature and heat flux testing system (Jantytech, Beijing, China, temperature accuracy: ±0.1 °C, heat flux sensitivity: ±0.1 W/m2) was used for one week of continuous monitoring, with representative results from a selected day shown in Figure 4.
The recorded data in Figure 4 illustrate the variations in surface temperature and heat flux of the rural residence walls. In Figure 4a, the temperature variations on the inner surfaces of the enclosure structure were relatively consistent, with the largest amplitude observed on the south-facing wall, primarily due to solar radiation. The wall experienced two temperature-rise stages throughout the day, with the north-facing wall reaching its peak first. Subsequently, the east-facing wall, shed roof, west-facing wall, south-facing wall, and floor achieved their peaks approximately 20 min later. This sequential pattern was closely linked to the cooking activities of the residents.
Additionally, the heat flux on the east bedroom wall, as shown in Figure 4b, exhibited a similar trend to the wall temperature. In contrast, the west-facing wall displayed more significant fluctuations, primarily due to its proximity to the heated brick bed (Kang) and the adjacent living room, where ventilation from the entrance door during cooking contributed to the observed variations.
To assess the thermal environment of rural residences, a WSZY-1 temperature and humidity data logger was used for one week of continuous monitoring. The outdoor sensor was installed 1.5 m above the ground in a well-ventilated, shaded area, away from heat sources, with aluminum foil shielding to reduce radiation interference. The indoor sensor was placed at the center of the room, at a height of 700–1800 mm, avoiding direct sunlight.
The recorded temperature variations were presented in Figure 5, the indoor and outdoor air temperature variations in the second type of rural residence were generally similar. In Figure 4a, the outdoor temperature ranges between −22 °C and 13 °C, while the bedroom temperature fluctuated between 8 °C and 23 °C., indicating a significant diurnal temperature difference. The average temperature of the east bedroom was 18.1 °C, followed by the west bedroom (13.6 °C), the living room (12.8 °C), and the kitchen, which had the lowest average temperature (7.8 °C). Upon comparison, only the indoor temperature of the east bedroom met the design requirements of the rural residential building energy efficiency design standard (GB/T50824) [21]. Additionally, the relative humidity in kitchen during cooking could exceed 90%, while the fluctuation range in other functional rooms was between 30% and 50%, falling within comfort zone.

3.1.2. Indoor Environmental Quality

To evaluate indoor air quality, PM10 and PM2.5 concentrations were continuously measured using a QZKQ-4G-2 multifunctional air quality transmitter (Renke Control Technology, Jinan China), with a measurement accuracy of ±1 μg/m3. The sensor was positioned at the center of the room, at a height of 1.2 m above the ground, following standard indoor air monitoring protocols. Data were recorded at 2 min intervals throughout the monitoring period to capture temporal variations in particulate matter concentrations.
The variation in indoor PM10 and PM2.5 mass concentrations in different functional rooms was illustrated in Figure 6. For most of the day, the indoor concentrations of PM10 and PM2.5 had exceeded the national standard limits [28]. Peak concentrations were observed during heating and cooking periods, with each room reaching its peak at different times, indicating a time lag in particle diffusion from the kitchen to other rooms. The lowest concentrations were recorded between midnight (0:00) and 6:00 AM.
During the household interviews, several households near cooking times reported a noticeable increase in smoke density. Table 3 provided a summary of residents’ subjective assessments of indoor air quality. The largest proportion of residents, 80.49%, felt no discomfort regarding air quality during sleep, while during heating and cooking periods, the majority reported felt the air quality but found it acceptable. This suggested the need for increased awareness and promotion of air quality improvement among rural residents to enhance health consciousness.

3.1.3. Occupant Door-Opening Behavior

In the surveyed dwellings, wintertime indoor airflow changes were primarily driven by occupant door-opening behavior, as exterior windows were typically kept sealed for insulation. To quantify the frequency of door opening, this study employed magnetic switch recorders for continuous monitoring of door-opening activities. As shown in Figure 7, the vertical axis represented the state of the door, where “2” represented a fully open door, “0” represented a closed door, and “1” represented a partially open door. Residents habitually opened doors to ventilate, with peak frequencies consistently coinciding with high-pollution events such as cooking and heating periods. However, field observations indicated that the duration and intensity of these ventilation events were arbitrary, guided by subjective feelings of stuffiness or cold rather than by any scientific principle for effective pollutant removal. This gap between the residents’ need to ventilate and their lack of an optimized strategy constitutes the central problem that the modeling and multi-objective optimization in this study aim to solve. Therefore, scientifically optimizing this occupant-driven behavior is critical, as it directly governs the inter-zonal airflow and subsequent pollutant diffusion that dictates indoor air quality.

3.2. Residents’ Thermal Demand

The main temperature indicators affecting farmers’ thermal comfort vary depending on the characteristics of the heating system. Currently, over 90% of farmers still use heated brick beds (Kang) as the heating system, which have a significantly higher radiant effect than convective heat. The selection of indoor air temperature operating temperature (t0) provided a comprehensive evaluation of both indoor radiant temperature and air temperature, which could be considered as the perceived temperature. The calculation method for the operating temperature (t0) was given by Equation (12):
t 0 = A t a + ( 1 A ) t m ¯
where t0 is the operating temperature, °C, A is the correction coefficient. When the wind speed is less than 0.2 m/s, it is set to 0.5, ta is the air temperature, °C, and tm is the mean radiant temperature, °C.
The pattern of operating temperature variation was obtained through calculations, as depicted in Figure 8. The graph showed that the daytime operating temperature was higher than the air temperature whereas at night, the air temperature consistently exceeded the operating temperature, exhibiting an entirely consistent trend with an average difference of only 0.87 °C. As the fuel in the heated brick bed (Kang) system was depleted, both the air temperature and operating temperature decreased uniformly from 17 °C to 7 °C.
The regression analysis of MTS (Mean Thermal Sensation) and operating temperature (t0) was conducted using the temperature frequency method (Bin method). The operating temperature corresponding to MTS (Mean Thermal Sensation) equals 0 was considered as farmers’ subjective thermal demand temperature, also known as the thermal neutral temperature. The regression analysis of MTS and operating temperature t0 was shown in Figure 9, with the regression equation:
M T S = 0.226 t 0 3.804 R 2 = 0.7136
As shown in Figure 9, a linear regression analysis was conducted between the occupants’ Mean Thermal Sensation (MTS) and the operating temperature (t0), yielding a strong correlation (R2 = 0.714). From this regression, the thermal neutral temperature (MTS = 0) was determined to be 16.84 °C. Furthermore, the analysis established the residents’ adaptive comfort zone, with an acceptable temperature range of 14.6 °C to 19.0 °C for 80% of occupants (MTS = [−0.5, 0.5]), and a wider range of 12.0 °C to 21.2 °C for 90% of occupants (MTS = [−1.0, 1.0]). Notably, this thermal neutral temperature of 16.84 °C is significantly higher than the average measured operating temperature of 14.6 °C, highlighting a clear gap between the residents’ thermal needs and their actual indoor environmental conditions [29].

3.3. Optimization Results

To facilitate daily operations for rural residents, this study selected parameter values corresponding to the maximum and minimum outdoor temperatures during the daytime from three regions, as shown in Table 4. The values for the two time periods, noon and evening, were substituted into the formula for CVRMSE (Coefficient of Variation Root Mean Square Error). Finally, through multiple iterations of multi-objective optimization, the Pareto front solutions that satisfy the acceptable indoor temperature and PM2.5 concentration were obtained, as shown in Figure 9 and Figure 10.
In Figure 10a, the ranges of air temperature and PM2.5 concentration changes were (7.61 °C, 16.86 °C) and (11.4 μg/m3, 146.41 μg/m3), respectively, when considering the maximum measured outdoor air temperature. In Figure 10b, the ranges were (7.34 °C, 18.90 °C) and (12.81 μg/m3, 168.52 μg/m3), respectively, when considering the average measured outdoor air temperature. Figure 10c showed the ranges as (8.08 °C, 19.29 °C) and (14.07 μg/m3, 173.12 μg/m3), respectively, when considering the maximum measured outdoor air temperature.
Figure 10 presented the Pareto front optimization results for natural ventilation control with farmers opening doors after evening cooking. In Figure 11a, the ranges of air temperature and PM2.5 concentration variations were (7.82 °C, 18.83 °C) and (18.68 μg/m3, 163.76 μg/m3), respectively, considering the maximum measured outdoor air temperature. Figure 10b had shown the ranges as (7.86 °C, 19.35 °C) and (19.09 μg/m3, 173.09 μg/m3), respectively, considering the average measured outdoor air temperature. Figure 10c had illustrated the ranges as (9.80 °C, 19.58 °C) and (17.97 μg/m3, 186.31 μg/m3), respectively, considering the highest measured outdoor air temperature.
The Pareto fronts generated in this study (Figure 10 and Figure 11) quantitatively visualize the fundamental trade-off between thermal comfort and indoor air quality. Analysis of these fronts reveals an optimal operational zone—often near the “knee point” of the curve—where a minimal investment in ventilation time yields a substantial improvement in air quality without a severe thermal penalty. To derive practical control strategies from these optimal zones, the results corresponding to the maximum and minimum outdoor temperatures for each region were analyzed to establish the upper and lower limits for ventilation duration. The resulting summarized optimization strategies for each region are presented in Table 5 and Table 6.

4. Discussion

The Pareto fronts generated in this study quantitatively visualize the fundamental trade-off between thermal comfort and indoor air quality. The shape of these fronts reveals a critical insight: the “knee point” of the curve represents the most efficient operational zone, where a minimal investment in ventilation time yields a substantial improvement in air quality without a severe thermal penalty. This finding is the basis for the core practical strategy derived from our results, which centers on brief, high-intensity ventilation events post-combustion. This approach effectively balances the trade-off by strategically mitigating pollutant exposure within the residents’ adaptive comfort limits.
To quantify the benefits of this approach, it was compared to a baseline scenario representing typical, unoptimized occupant behavior. This comparative analysis reveals a substantial advantage: an optimized 30 s ventilation can achieve a similar PM2.5 reduction as a 3 min unguided ventilation, but with 2–3 °C less temperature drop. This stark contrast demonstrates that residents can achieve a healthier indoor environment with a much lower energy penalty by adopting an efficiency-driven strategy over a reactive, sensation-driven one. Beyond this immediate guidance for residents, the quantitative relationships established herein provide foundational data for designers and engineers to develop low-cost smart ventilation systems, such as automated alerts or actuators, tailored specifically for rural dwellings.

5. Conclusions

In conclusion, this study identified the predominant rural dwelling archetypes in Northeast China and confirmed the critical challenge of substandard thermal conditions coexisting with hazardous PM2.5 concentrations from domestic fuel use. A validated multi-objective optimization model, based on residents’ adaptive comfort needs was developed. The findings reveal that brief, high-intensity ventilation is the optimal strategy. This provides a practical, low-cost solution to mitigate public health risks within the existing energy constraints of rural households and serves as a theoretical basis for future smart ventilation systems.
However, this study has several limitations that should be acknowledged. The findings are based on the most prevalent dwelling type (Type II), and their applicability to other archetypes requires further investigation. The analysis was focused solely on PM2.5, while other gaseous pollutants from combustion were not included. Furthermore, the model relies on certain simplifying assumptions that may affect its accuracy under severe haze conditions.
Future research should expand on this study by addressing its limitations and broadening its scope. Immediate next steps should include incorporating other key pollutants from combustion into the model and extending the analysis to other dwelling archetypes. Furthermore, the integrated methodology developed herein is broadly applicable to other rural regions facing similar energy and climate challenges, provided the framework is locally calibrated for different building forms and occupant habits.

Author Contributions

X.Z. (Xueyan Zhang) and X.Z. (Xingkuo Zhang) designed the study; X.Z. (Xueyan Zhang) performed the theoretical calculation and wrote the manuscript; X.Z. (Xingkuo Zhang) wrote the first draft of the manuscript and conducted the results calculation and discussion. Y.Y. participated in the entire experimental and calculation process. J.L. provided revision suggestions for this manuscript. All authors have read and agreed to the published version of the manuscript.

Funding

This work was funded by the 14th Five-Year Plan for National Key Research and Development Project (No. 2022YFC3803205-02), the National Nature Science Foundation of China (No. 52078098 and No. 51608092), and National Science Foundation of Liaoning Province (No. 2019-ZD-0022).

Institutional Review Board Statement

In accordance with the “Measures for the Ethical Review of Life Sciences and Medical Research Involving Humans,” jointly issued in 2023 by the National Health Commission, the Ministry of Education, the Ministry of Science and Technology, and the National Administration of Traditional Chinese Medicine of the People’s Republic of China. This study does not fall under the jurisdiction of the aforementioned “Measures,” it is not subject to the requirement of an ethical review.

Informed Consent Statement

Informed consent was obtained from all subjects involved in the study.

Data Availability Statement

The original contributions presented in this study are included in the article. Further inquiries can be directed to the corresponding author.

Acknowledgments

The authors would like to thank the survey participants who allowed us to carry out this research.

Conflicts of Interest

Author Yiming Yang was employed by the company Qingdao Hisense Hitachi Air-conditioning System Co., Ltd. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Appendix A. Basic Information Table of Building Envelope Structure

Year and Area of Construction
Exterior WallMaterial□ Solid Brick Wall
□ Concrete
□ Adobe Wall
□ Stone Wall and Mud Grass
□ Others, Thickness 240 mm □ 370 mm □
Decoration□ Face Brick
□ Coating
□ Cement Mortar
□ Others
External Wall InsulationTypes□ External Insulation
□ Internal Insulation
□ Intermediate Insulation
□ None, Thickness____(mm)
Material□ Polystyrene Board
□ XPS
□ Polyurethane
□ Phenolic Board
□ Rock Wool Board
□ Others
External windowMaterial□ Aluminum Alloy
□ Plastic Steel Window
□ Wooden Frame Window
□ Others
Layers□ One
□ Double
Glass Type□ Single Glazed
□ Double Glazed
□ Others
Seal□ Winter
□ Annual
□ None
External doorMaterial□ Aluminum Alloy
□ Plastic Steel
□ Wood
□ Others
Door Dou□ Yes
□ No
Layers□ One
□ Double
Window to Wall RatioSouth:
□ ≥45%
□ <45%
□ none
North:
□ ≥25%
□ <25%
□ none
East:
□ ≥30%
□ <30%
□ none
West:
□ ≥30%
□ <30%
□ none
RoofMaterial□ Wooden House Tile Roof
□ Cast-in-Place Concrete
□ Precast Concrete
□ Others
Heat Preservation□ Perlite Sawdust
□ Polystyrene Board
□ XPS
□ Polyurethane
□ Phenolic Board
□ Rock Wool Board
□ Foam Insulation Board
□ Thick Forage and Straw
□ Others, Thickness____(mm)
Suspended Ceiling□ Yes
□ No
GroundMaterial□ Land Surface
□ Brick Covered Ground
□ Brick Covered Ground
□ Ceramic Tile
□ Wooden Floor
Heat Preservation□ Polystyrene Foam Plastic Board
□ Polystyrene Particle Insulation Slurry
□ Composite Silicate Board
□ Inorganic Insulation Mortar
□ Cast in Situ Foam Concrete
□ Ground Insulation Coating
□ Others, Thickness____(mm)

Appendix B. Survey Questionnaire on Subjective Information Needs of Residents

  • Basic information of personnel:
Gender:
□ Man
□ Woman
Age:
□ <20
□ 20–30
□ 30–40
□ 40–50
□ 50–60
□ >60
Recent health status:
□ fine
□ cold
□ pernio
□ inflame
□ Skin and lips are prone to dryness others_______
Educational level:
□ Not receiving education
□ primary school
□ middle school
□ high school
□ junior college
□ Bachelor degree or above
Family composition:
□ live alone
□ 2 people
□ 3 people
□ 3~5 people
□ More than 5 people
Children’s Home Situation:
□ Yes
□ No
2.
Personnel behavior habits:
TimeActivity ContentNotes
Wake Up and Rest TimeIgnition and
Cooking Time
Home Activities (1–8)Egress Times
Morning
Noon
Afternoon
Night
Category of activity content: 1 Read Book; 2 Watch TV; 3 Clean; 4 Meeting and chatting with guests; 5 Eating; 6 Go Out; 7 Sleeping; 8 Others.
3.
Heating method?
□ Fire Kang and Earth Heating System
□ Fire Kang
□ Earth Heating System
□ Electric Heating
□ Floor Heating
Basis for burning firewood:
□ Fixed time
□ After the Kang surface cools down
□ The surface of the Kang is slightly not hot
□ After entering the house
4.
Cooking methods in winter? (multiple choice)
□ kitchen range
□ gas cooker
□ induction cooker
□ Others
5.
Preparation before Winter? (multiple choice)
□ Plastic cloth sealed window
□ Southbound plastic shed
□ Northbound plastic shed
□ Branch
□ Leaf
□ Straw
□ corncob
□ firewood
□ coal
6.
Winter ventilation habits
Ventilation frequency:
□ Ventilate more than three times a week
□ Ventilate 1–2 times a week
□ Ventilate once every two weeks
□ Ventilate once a month
□ Not ventilated
Ventilation method:
□ Only open the bedroom window
□ Only open the door
□ Only open the kitchen window
□ Open doors and windows
□ Other ways
Ventilation time:
□ Less than 5 min
□ 5–10 min
□ 10–30 min
□ More than 30 min
Subjective needs survey:
7.
The feeling of warmth and coldness in the body indoors at this moment:
□ Very cold
□ cold
□ Slightly cool
□ tepid
□ Slightly warm
□ warm
□ Very hot
8.
When do you usually feel the temperature is too low at home (multiple choice)
□ 5:00–7:00
□ 7:00–11:00
□ 11:00–13:00
□ 13:00–17:00
□ 17:00–24:00
9.
Do you think the indoor temperature is comfortable at this moment:
□ comfort
□ Slightly uncomfortable
□ uncomfortable
□ uncomfortable
□ unbearable
Do you want the indoor temperature:
□ rise
□ unchanged
□ reduce
10.
How do you feel about the humidity inside at this moment:
□ Very Humid
□ Little Humid
□ moderate
□ Little Dry
□ Very Dry
record: Indoor temperature:__________ Indoor humidity:___________
11.
How do you feel about indoor air quality at different times at home?
During ignition and cooking:
□ Indoor smoke is very heavy
□ Perceptible yet acceptable
□ Imperceptible
□ Favorable
During the heating period:
□ Indoor smoke is heavy
□ Perceptible yet acceptable
□ Imperceptible
□ Favorable
During sleep:
□ Indoor smoke is heavy
□ Perceptible yet acceptable
□ Imperceptible
□ Favorable
Air quality: CO2: _________
Adaptive behavior:
12.
What measures do you take when you feel cold in the past week?
□ Close doors and windows
□ Add fuel/coal
□ Wear more clothes
□ Drink hot beverages
□ Engage in physical activity
□ Sit on a heated bed
□ Other measures ___________
□ Did not feel cold
13.
What measures do you take when you feel that the air quality is poor (e.g., stuffy, with irritating odors, heavy smoke) in the past week?
□ Slightly open windows/doors
□ Open windows/doors fully
□ Turn on exhaust fans
□ No measures taken
□ Other measures_________
(Slightly open means less than half open, fully open means more than half or completely open)
14.
What measures do you take when there is a lot of smoke or steam during cooking?
□ Turn on the extractor fan
□ Open windows
□ Open doors
□ Other smoke removal methods
□ No measures taken
15.
Clothing Record:
Upper Garments:
□ Down jacket
□ Thick cotton coat
□ Jacket
□ Down vest
□ T-shirt
□ Knitted sweater
□ Thick sweater
□ Long-sleeve flannel shirt
□ Long-sleeve shirt
□ Thin sweater
Lower Garments:
□ Thin trousers
□ Thick trousers
□ Thermal pants/cotton pants
□ Thin thermal leggings
□ Thick thermal leggings
Skirts:
□ Skirt (thin)
□ Skirt (thick)
□ Thick long-sleeve dress
□ Thin long-sleeve dress
Underwear:
□ Women’s bra
□ Women’s panties
□ Men’s underwear
□ Thermal shirt
□ Thermal pants
Footwear:
□ Plastic slippers
□ Cotton slippers
□ Thick-soled sports shoes
□ Boots/leather shoes
Socks:
□ Thin socks
□ Thick socks
□ Thick long socks

Appendix C. Resident Behavior and Activity Tracking Record

Household
Location
Housing Type Floor
Height
Orientation
Record
Subject
Age Gender Date
TimeActivity Duration (h)Window
Ventilation Behavior
Start and Duration of Pollution Sources (h)
CleaningSmokingHeatingCooking
0:00–6:00
6:00–7:00
7:00–8:00
8:00–9:00
9:00–10:00
10:00–11:00
11:00–12:00
12:00–13:00
13:00–14:00
14:00–15:00
15:00–16:00
16:00–17:00
17:00–18:00
18:00–19:00
19:00–20:00
20:00–21:00
21:00–22:00
22:00–23:00
23:00–24:00

Appendix D. Derivation of the PM2.5 Concentration Model

To more accurately describe how farmers’ door-opening ventilation behavior influenced the concentration of indoor PM2.5, the particle model proposed by Tian Liwei [24] was introduced for quantitative analysis. The control model for door-opening ventilation behavior in rural households in this study was built upon this model. The model parameters were optimized, considering the movement characteristics of PM2.5 particles. The natural ventilation process was simplified into permeable natural ventilation and ventilation caused by residents’ door-opening and closing. The PM2.5 concentration control model for fine particle pollution, based on the law of mass conservation, was presented in Equation (A1):
V d C i d τ = Q f C 0 + ρ Q s C 0 + E Q 0 + K V C i i = 1 n Q i k C i C K + R L f A f + S + H + F + C
where V is the indoor volume in cubic meters (m3); Qf is the fresh air volume in cubic meters per hour (m3/h); Qs is the infiltration air volume in cubic meters per hour (m3/h); p is the PM2.5 penetration rate per hour (1/h); E is the emission rate of indoor PM2.5 pollution sources in micrograms per hour (μg/h); Q0 is the airflow to the outside in cubic meters per hour (m3/h); K is the PM2.5 settling rate per hour (1/h); Qik is the inter-room airflow transmitted from room i to room k in cubic meters per hour (m3/h); Ci is the PM2.5 mass concentration in room i in micrograms per cubic meter (μg/m3); C0 is the mass concentration of outdoor PM2.5 in micrograms per cubic meter (μg/m3); Ck is the PM2.5 mass concentration in room k in micrograms per cubic meter (μg/m3); R is the secondary suspension rate of PM2.5 per hour (1/h); Lf is the PM2.5 mass per unit area of the floor in micrograms per square meter (μg/m2); Af is the floor area in square meters (m2); S is the mass of gas converted to PM2.5 per unit time in micrograms per hour (μg/h); H is the mass of PM2.5 generated by moisture absorption per unit time in micrograms per hour (μg/h); F is the mass of PM2.5 generated by chemical reactions per unit time in micrograms per hour (μg/h); C is the mass of PM2.5 generated by condensation per unit time in micrograms per hour (μg/h); t is time in hours; h = 1, 2, …, k = 1, 2, …, n.
In this mathematical model, due to the use of plastic film sealing for windows in the studied households, it was known from the fitting results and the research of Yang Yiming [25] that as the inter-room airflow at the opening of the indoor door decreases, the impact of ventilation caused by infiltration from the building envelope on indoor PM2.5 concentration gradually increases. For the state where the interior door was closed, the impact of infiltration ventilation on indoor PM2.5 concentration was significant and needs to be taken into consideration. Therefore, the attenuation of indoor PM2.5 was solely influenced by the residents’ door-opening ventilation behavior and infiltration ventilation.
After substituting the values of various parameters, Equation (9) was revised to Equation (A2):
V d C i d τ = ρ Q s C i + E Q 0 + K V C i i = 1 n Q i k C i C K
where p is the PM2.5 penetration rate, and in this study, p was set to 1 [30].
Selecting the bedroom as the target function Ci, the variation of PM2.5 concentration in the bedroom could be expressed by Equation (10), obtained by substituting Equation (A3) into Equation (A2) and rearranging:
V 4 d C 4 τ d τ = Q 34 C 3 τ C 4 τ Q 40 C 4 τ K V 4 C 4 τ
For Equation (10), assuming the initial conditions τ = 0 and C = C4, after rearranging through separation of variables and integrating both sides, the PM2.5 concentration in the indoor environment at time τ was given by Equation (A4):
C 4 τ = υ C 3 υ A + 0.00657 exp υ A + 0.00657 48.6 + C 0 υ C 3 υ A + 0.00657 * exp υ A + 0.00657 48.6 τ
where C4(τ) represents the PM2.5 concentration in the bedroom at time τ, and Equation (13) represents the rate of change of PM2.5 concentration in the east bedroom with time.
Establishing a system of equations for temperature (T) and PM2.5 (f2) in relation to ventilation area (A) and ventilation time (t), as represented by Equation (A5).
f 1 A , τ = t i = C P ρ i v + 0.00039 τ + K F 1000 t i + t a 2 t 0 τ C P ρ i v t a 2 t 0 K F t a τ 2000 C p ρ i v + K F 2000 + C p ρ i v A τ 2 f 2 A , τ = C 4 τ = υ C 3 υ A + 0.00657 exp υ A + 0.00657 48.6 + C 0 υ C 3 υ A + 0.00657 * exp υ A + 0.00657 48.6 τ

References

  1. Chowdhury, S.; Pillarisetti, A.; Oberholzer, A.; Jetter, J.; Mitchell, J.; Cappuccilli, E.; Aamaas, B.; Aunan, K.; Pozzer, A.; Alexander, D. A global review of the state of the evidence of household air pollution’s contribution to ambient fine particulate matter and their related health impacts. Environ. Int. 2023, 173, 107835. [Google Scholar] [CrossRef] [PubMed]
  2. Albalak, R.; Keeler, G.J.; Frisancho, A.R.; Haber, M. Assessment of PM10 Concentrations from Domestic Biomass Fuel Combustion in Two Rural Bolivian Highland Villages. Environ. Sci. Technol. 1999, 33, 2505–2509. [Google Scholar] [CrossRef]
  3. Parikh, J.; Balakrishnan, K.; Laxmi, V.; Biswas, H. Exposure from cooking with biofuels: Pollution monitoring and analysis for rural Tamil Nadu, India. Energy 2001, 26, 949–962. [Google Scholar] [CrossRef]
  4. Zhang, X.Y.; Chen, B. Analysis on the PM2.5 Concentration During Heating and Cooking in Cold Rural Areas of Northern China. In Proceedings of the International Conference on Sustainable Energy and Environmental Engineering (ICSEEE), Shenzhen, China, 20–21 December 2015; pp. 1083–1086. [Google Scholar]
  5. Wang, Z.J.; Xie, D.D.; Tang, R. Indoor air pollutants and their correlation at rural houses in severe cold region in winter. J. Harbin Inst. Technol. 2014, 46, 60–64. [Google Scholar]
  6. Gong, X.; Yu, Y.; Lai, S.; Wang, X.; Duan, X.; Liu, J.; Wang, W. A Field Investigation on Summer Thermal Comfort of Occupants of Rural Houses in the North Area of Hot Summer and Warm Winter Zone, China. Buildings 2025, 15, 715. [Google Scholar] [CrossRef]
  7. Ma, L.; Shao, N.; Zhang, J.; Zhang, T.; Xu, M. A Study on Indoor Thermal Environment of Rural Residence in South Liaoning Province. Procedia Eng. 2016, 146, 366–374. [Google Scholar] [CrossRef]
  8. Li, B.; Sun, Y.; Zheng, W.; Zhang, H.; Jurasz, J.; Du, T.; Wang, Y. Evaluating the Role of Clean Heating Technologies in Rural Areas in Improving the Air Quality. Appl. Energy 2021, 289, 116693. [Google Scholar] [CrossRef]
  9. Wu, J.; Cheng, L.; Lu, Y.; Song, L.-L. Life cycle inventory analysis of clean alternatives to scattered coal heating. China Environ. Sci. 2018, 38, 1570–1578. [Google Scholar]
  10. Han, S.W. Research on the problems and solutions existing in the development and utilization of clean energy in rural areas. J. Agric. Catastropholgy 2020, 10, 120–121. [Google Scholar]
  11. Zhang, Y.; Mo, J.; Li, Y.; Sundell, J.; Wargocki, P.; Zhang, J.; Little, J.C.; Corsi, R.; Deng, Q.; Leung, M.H.K.; et al. Can commonly-used fan-driven air cleaning technologies improve indoor air quality? A literature review. Atmos. Environ. 2011, 45, 4329–4343. [Google Scholar] [CrossRef] [PubMed]
  12. Song, J.S.; Huang, K.L.; Feng, G.H. Analysis and on-site measurement of indoor formaldehyde in residential buildings during transition season in Liaoning province. Constr. Sci. Technol. 2019, 22, 22–24. [Google Scholar]
  13. Kong, W.C. Monitoring and Controlling of Indoor Air Pollutants in Cold Rural Areas in Winter; Harbin Institute of Technology: Harbin, China, 2015. [Google Scholar]
  14. Warren, P.R.; Parkins, L.M. Window-opening behavior in office buildings. Build. Serv. Eng. Res. Technol. 1984, 23, 89–101. [Google Scholar] [CrossRef]
  15. Nicol, J.F.; Humphreys, M.A. Thermal Comfort as Part of a Self-Regulating System. Build. Res. Pract. 1973, 1, 174–179. [Google Scholar] [CrossRef]
  16. Yu, W. The Low Energy-Consumption Strategy for Improving Indoor Thermal Environment Quality in Residential Building. Ph.D. Thesis, Chongqing University, Chongqing, China, 2011. [Google Scholar]
  17. Jin, G.H.; Ma, J.; Yang, P. MABC-BPNN based multi -objective optimization prediction model for ultra-low energy consumption residential houses in western Inner Mongolia grassland. J. Arid Land Resour. Environ. 2021, 35, 73–79. [Google Scholar]
  18. Zhang, Y.; Liang, X.K.; Yuan, L.J. Multi-objective optimization algorithm for building energy-saving design. Huazhong Univ. Sci. Technol. (Nat. Sci. Ed.) 2021, 49, 107–112. [Google Scholar]
  19. Zhu, L.; Wang, B.; Sun, Y. Multi-objective optimization for energy consumption, daylighting and thermal comfort performance of rural tourism buildings in north China. Build. Environ. 2020, 176, 106841. [Google Scholar] [CrossRef]
  20. Gao, X. Improvement and Empirical Study of K-Means Clustering Algorithm on Panel Data Analysis. Master’s Thesis, Taiyuan University of Technology, Taiyuan, China, 2015. [Google Scholar]
  21. GB/T 50824-2013; Energy Efficiency Design Standards for Rural Residential Buildings. China Construction Industry Press: Beijing, China, 2013.
  22. GB3095-2012; Ambient Air Quality Standards. China Environmental Press: Beijing, China, 2022.
  23. Zhao, W.X.; An, B.G.; Liu, J.P. Analysis of Indoor Thermal Environment Test of New Civil Housings in Guanzhong Area, Shanxi Province in Winter. Build. Sci. 2013, 29, 72–76. [Google Scholar]
  24. Tian, L.W. The Research on Modeling of Indoor Particulate Matter of Outdoor Origin and Control Strategies. Ph.D. Thesis, Hunan University, Hunan, China, 2009. [Google Scholar]
  25. Yang, Y.M. Diffusion Characteristics of Indoor PM2.5 and Health Risk Assessment of Rural Dwellings in Liaoning Province in Winter. Master’s Thesis, Dalian University of Technology, Dalian, China, 2022. [Google Scholar]
  26. Attia, S.; Hamdy, M.; O’Brien, W.; Carlucci, S. Assessing Gaps and Needs for Integrating Building Performance Optimization Tools in Net Zero Energy Buildings Design. Energy Build. 2013, 60, 110–124. [Google Scholar] [CrossRef]
  27. Deb, K.; Pratap, A.; Agarwal, S.; Meyarivan, T. A fast and elitist multi-objective genetic algorithm: NSGA-II. IEEE Trans. Evol. Comput. 2002, 6, 182–197. [Google Scholar] [CrossRef]
  28. GB/T18883-2022; Indoor Air Quality Standard. China Construction Industry Press: Beijing, China, 2022.
  29. Li, J.K. Study on Thermal Environment Control and Energy-Saving Strategies Based on the Operative Temperature. Master’s Thesis, Qingdao University of Technology, Qingdao, China, 2023. [Google Scholar]
  30. Toftum, J.; Jørgensen, A.S.; Fanger, P.O. Upper limits for indoor air humidity to avoid uncomfortably humid skin. Energy Build. 1998, 28, 1–13. [Google Scholar] [CrossRef]
Figure 1. Description of the physical model of the farmhouse.
Figure 1. Description of the physical model of the farmhouse.
Buildings 15 03718 g001
Figure 2. Measurement point layout diagram.
Figure 2. Measurement point layout diagram.
Buildings 15 03718 g002
Figure 3. Calculation results of NSGA-II.
Figure 3. Calculation results of NSGA-II.
Buildings 15 03718 g003
Figure 4. Thermal parameters of the East bedroom envelope; (a) Temperature of the East bedroom, (b) Heat flow of the East bedroom.
Figure 4. Thermal parameters of the East bedroom envelope; (a) Temperature of the East bedroom, (b) Heat flow of the East bedroom.
Buildings 15 03718 g004
Figure 5. Indoor parameters of a dwelling; (a) Indoor temperature of farmhouse, (b) Indoor and outdoor relative humidity of farmhouse.
Figure 5. Indoor parameters of a dwelling; (a) Indoor temperature of farmhouse, (b) Indoor and outdoor relative humidity of farmhouse.
Buildings 15 03718 g005
Figure 6. Distribution of RSP in different rooms; (a) Distribution of PM10 in Different Rooms, (b) Distribution of PM2.5 in Different Rooms.
Figure 6. Distribution of RSP in different rooms; (a) Distribution of PM10 in Different Rooms, (b) Distribution of PM2.5 in Different Rooms.
Buildings 15 03718 g006
Figure 7. Statistics of the inner door opening of House 1; (a) West Inner Door, (b) East Inner Door.
Figure 7. Statistics of the inner door opening of House 1; (a) West Inner Door, (b) East Inner Door.
Buildings 15 03718 g007
Figure 8. Operating temperature and air temperature variations.
Figure 8. Operating temperature and air temperature variations.
Buildings 15 03718 g008
Figure 9. Operating temperature and MTS fitting results (farmer thermoneutral temperature).
Figure 9. Operating temperature and MTS fitting results (farmer thermoneutral temperature).
Buildings 15 03718 g009
Figure 10. Optimization results under daytime outdoor temperature at noon; (a) Minimum Value, (b) Average Value, (c) Maximum Value [28].
Figure 10. Optimization results under daytime outdoor temperature at noon; (a) Minimum Value, (b) Average Value, (c) Maximum Value [28].
Buildings 15 03718 g010
Figure 11. Optimization results under daytime outdoor temperature in the evening; (a) Minimum Value, (b) Average Value, (c) Maximum Value [28].
Figure 11. Optimization results under daytime outdoor temperature in the evening; (a) Minimum Value, (b) Average Value, (c) Maximum Value [28].
Buildings 15 03718 g011
Table 1. Cluster analysis results.
Table 1. Cluster analysis results.
TypesCategory
Type 1Type 2Type 3
Population characteristicsElderly couple living aloneMiddle-aged and elderly couple living with a childMultigenerational cohabitation
Characteristics of heating expensesLower levelModerate levelHigh level
Architectural featuresFloor areaSmallNormalLarge
Building formThree-room layout dominates.Improved three-room layout dominates.Improved three-room and four-room layouts dominate.
Building structureOnly master bedroom and secondary bedroom used as storage or living room.Master bedroom, secondary bedroom, living room, and kitchen are all availableMultiple bedrooms shared collectively.
Heating SystemKangEarth Stove/KangEarth Stove, Floor Heating
Door Pocket, Buffer SpaceNoneNoneExist
Number of Cases33%55%12%
Table 2. Testing instruments and their measurement accuracy.
Table 2. Testing instruments and their measurement accuracy.
Test ParametersTest InstrumentsInstrument PrecisionsManufacturers
PM10 and PM2.5 concentrationPM2.5 recorder developed based on Plan tower a003 sensor ZF-R30–2999 μg/m3
±1 μg/m3
Beijing Co-Cloud (Beijing, China)
Air temperature Relative humidityAir temperature and relative humidity recorder WEZY-2−40–100 °C (±0.1 °C)
0–100%RH (±0.1%RH)
TIAN JIAN HUA YI Technology Co., Ltd. (Tianjin, China)
CO2 concentrationCO2 recorder WEZY-10–5000 ppm
±75 ppm
Door switching frequencyMagnetic switch recorder CKJM-1Maximum sensing distance 30 mm
Wind speed and wind temperatureOmnidirectional Wind Speed and Temperature Data Logger WFWZY-1−20–80 °C (±0.5 °C)
0.05~30 m/s (5% ± 0.05 m/s)
Table 3. Indoor personnel evaluation of air quality at different times.
Table 3. Indoor personnel evaluation of air quality at different times.
PeriodHeavy Indoor
Smoke
Noticeable but
Acceptable
Not NoticeableFeel the Air
Was Good
Ignition, Cooking period26.83%36.59%26.83%9.76%
Heating period15.85%46.34%26.83%9.76%
Sleeping period0%3.66%80.49%15.85%
Table 4. Experimental area parameter selection.
Table 4. Experimental area parameter selection.
Parameters/LocationFushunChangchunHarbin
Outdoor Average Temperature (°C)−6.7−14.2−19.3
Outdoor Maximum Temperature (°C)1.5−8.1−10.9
Outdoor Minimum Temperature (°C)−23.7−22.3−27.9
Outdoor Average Wind Speed (m/s)2.02.73.2
Table 5. Optimized ventilation time during fuels combustion in northeast rural dwellings at noon.
Table 5. Optimized ventilation time during fuels combustion in northeast rural dwellings at noon.
RegionConditionVentilation
Scheme
Ventilation
Time (s)
Air
Temperature
(°C)
Pollutant
Concentration
(μg/m3)
FushunMaximum ValueFully Open2812.567.1
Half Open3214.269.2
Minimum ValueFully Open1713.370.3
Half Open2713.565.4
ChangchunMaximum ValueFully Open2112.367.9
Half Open2912.774.1
Minimum ValueFully Open1312.382.9
Half Open2211.879.1
HarbinMaximum ValueFully Open1612.867.4
Half Open2913.068.2
Minimum ValueFully Open1512.365.8
Half Open1913.169.1
Table 6. Optimized ventilation time during fuels combustion in northeast rural dwellings in the evening.
Table 6. Optimized ventilation time during fuels combustion in northeast rural dwellings in the evening.
RegionConditionVentilation
Scheme
Ventilation
Time (s)
Air
Temperature
(°C)
Pollutant
Concentration
(μg/m3)
FushunMaximum ValueFully Open2612.558.6
Half Open3513.867.4
Minimum ValueFully Open1712.064.4
Half Open2312.373.6
ChangchunMaximum ValueFully Open1613.169.2
Half Open2812.065.7
Minimum ValueFully Open1412.973.1
Half Open2212.771.4
HarbinMaximum ValueFully Open1612.668.3
Half Open2612.768.8
Minimum ValueFully Open1311.767.9
Half Open1912.370.1
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Zhang, X.; Zhang, X.; Yang, Y.; Li, J. Optimization on Ventilation Time in Winter Based on Energy, Thermal Comfortable and Air Quality in Severe Cold Rural Dwellings of Northeast China. Buildings 2025, 15, 3718. https://doi.org/10.3390/buildings15203718

AMA Style

Zhang X, Zhang X, Yang Y, Li J. Optimization on Ventilation Time in Winter Based on Energy, Thermal Comfortable and Air Quality in Severe Cold Rural Dwellings of Northeast China. Buildings. 2025; 15(20):3718. https://doi.org/10.3390/buildings15203718

Chicago/Turabian Style

Zhang, Xueyan, Xingkuo Zhang, Yiming Yang, and Jing Li. 2025. "Optimization on Ventilation Time in Winter Based on Energy, Thermal Comfortable and Air Quality in Severe Cold Rural Dwellings of Northeast China" Buildings 15, no. 20: 3718. https://doi.org/10.3390/buildings15203718

APA Style

Zhang, X., Zhang, X., Yang, Y., & Li, J. (2025). Optimization on Ventilation Time in Winter Based on Energy, Thermal Comfortable and Air Quality in Severe Cold Rural Dwellings of Northeast China. Buildings, 15(20), 3718. https://doi.org/10.3390/buildings15203718

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop