Optimizing the Integration of Building Materials, Energy Consumption, and Economic Factors in Rural Houses of Cold Regions: A Pathway

Abstract

Limited material options and economic conditions significantly restrict the potential for energy efficiency improvements in rural houses in China’s cold regions. It is worth exploring how to propose suitable energy-saving renovation plans for rural houses in cold regions under practical constraints. By using Grasshopper within Rhinoceros 8 software, an algorithm integrates material selection, energy consumption calculations, and economic analysis. The method efficiently generates thermal optimization schemes, providing insights into energy use, costs, and payback periods. In a case study of a typical rural house in Daqing City, the optimized scheme achieved over 70% energy savings compared to traditional homes, with renovation costs amounting to less than 40% of residents’ annual income and a 2-year payback period. This significant improvement highlights the potential of the proposed method in enhancing the energy efficiency and economic viability of rural house renovations.

1. Introduction

1.1. Why Choose Rural Houses in Severe Cold Regions for Energy-Saving Renovations

In the severe cold regions of China, outdoor temperatures drop significantly during winter, leading to high demand for building heating. Rural residences, compared to urban houses, often have lower floors, scattered layouts, and poor thermal performance in their building envelopes [1,2]. This results in excessive energy consumption and insufficient indoor thermal comfort. Rural houses have significant potential for energy-saving renovations.

1.2. Existing Studies on Thermal Performance of Building Envelope Structures

From a thermal perspective, the building envelope serves the functions of heat preservation, insulation, and wind resistance, helping to stabilize indoor temperature fluctuations [3,4]. Currently, much research on envelope structures focuses primarily on walls, roofs, and windows [5,6]. Among them, the wall research results account for the largest proportion. In the past few decades, using thermal insulation materials to ameliorate the thermal inertia of building walls is widely considered as the most effective measure [7,8,9]. Considering local climate and customs, a variety of thermal insulation materials, energy-saving measures, and optimization methods are employed. For example, Braulio-Gonzalo and Bovea introduced a method to choose thermal insulation materials based on energy consumption, environmental factors, and economic considerations. They concluded that fiber materials exhibited the best cost performance [10]. Ustaoglu et al. added 3% paper mill sludge (PMS) to polyurethane (PUR) and found that it has good thermal insulation performance in buildings in various climatic areas. In comparison to other insulation materials, these require a smaller thickness, reducing the overall life cycle cost [11]. Khoukhi explored the impact of thermal and humid environments on Polyurethane (PUR) performance, revealing a positive correlation between relative humidity and building cooling and heating loads [12]. Ziapour et al. developed a novel composite prefabricated wall block and assessed its insulation thickness concerning cold and heat loads, establishing a positive correlation between the optimal thickness and air temperature [13]. In addition, Raimundo et al. analyzed thermal insulation solutions for buildings in Portugal considering economic and energy-related issues. They ultimately recommended the adoption of expanded polystyrene board (EPS) sandwich insulation, with insulation thickness positively correlating with the severity of climate change [14].

Heat loss through windows constitutes approximately 20% of the total heat loss [15]. Numerous scholars have analyzed the energy consumption of windows with varying structures to identify the optimal window type [16,17,18]. Hashemi et al. proposed replacing conventional thermal insulation materials with vacuum insulation panels (VIP) in louvers. By eliminating the thermal bridge, heat loss can be reduced by 79% [19]. Lyu et al. set up a vacuum-water flow window, which can save 40% of refrigeration energy consumption after adding normal temperature water [20]. Guo et al. investigated the energy consumption of various photovoltaic (PV) windows in four cities in China, considering factors such as transmittance and orientation angles. The result showed that in Harbin, Beijing, Shanghai, Shenzhen, and other places, the energy-saving performance of PV window with a light transmittance of 10% is better than that of PV window with a light transmittance of 5% [21].

Roofs are particularly susceptible to the impact of solar radiation. Given their significant effect on energy consumption and indoor air temperature, numerous scholars have extensively researched and discussed optimization and improvement methods related to their form and structure [22,23,24,25,26]. Dehwah and Krarti introduced a switchable insulation system (SIS) applied to the roof. The results indicated that SIS reduced the cooling load by 44% and the heating load by 17% [27]. Kivioja and Vinha studied the influence of internal convection of heat insulation materials on the thermal insulation performance in the roof structure, and concluded that the internal convection of high-density heat insulation materials is mainly affected by the temperature difference, while the low-density heat insulation materials are most affected by the surface airflow [28].

In general, current research on the thermal performance of building envelopes predominantly focuses on specific components of the envelope structure. Comprehensive solutions for the overall retrofitting of various parts of building envelopes receive limited attention.

1.3. Existing Studies on Energy-Saving Renovation of Rural Houses in Cold Regions

Shao et al. conducted an energy consumption simulation for rural residential buildings in cold regions using DesignBuilder, taking Zalantun as an example. They analyzed the factors influencing energy consumption and their respective weights. Through orthogonal experimental design, they proposed the optimal parameter combination [29]. Cui et al. found that improving the thermal performance of the building envelope is the most effective measure for enhancing indoor temperature and energy-saving potential. Using DesignBuilder, they compared the effects before and after the retrofitting and proposed suitable renovation methods for different types of rural residential buildings [30]. Li et al. used Lhasa as a case study to validate the energy-saving potential of using passive design in solar-rich regions. Using EnergyPlus 24.1.0 software for optimization, they found that approximately 80% of thermal energy could be saved [31]. Xi et al. aimed to reduce heating carbon emissions and control costs in the Wusu region. They conducted multi-objective optimization using Rhino’s Grasshopper plugin, and most of the proposed solutions met practical requirements [32]. Alev et al. analyzed energy-saving retrofit alternatives for rural residential buildings in the Baltic Sea region (cold climate). They concluded that energy-saving targets, building types, thermal transmittance, and building service systems are the main factors influencing the enhancement of energy-saving potential [33]. Hu et al. used IDA ICE 5.0 and AutoMOO (IDA ICE’s internal integration plugin) software to simulate the renovation costs and carbon emissions of rural residential buildings in cold regions under intermittent and continuous heating. They found that the cost-optimal solution had significant advantages and that intermittent heating was more effective than continuous heating [34]. Jiang et al. proposed achieving net-zero energy consumption for rural residential buildings in severely cold regions by applying a self-developed foam cement insulation material in the building envelope, installing photovoltaic systems, and enhancing the airtightness of doors and windows. They also validated the economic viability of the optimized solutions [35].

In summary, energy-saving retrofits for rural residential buildings in cold regions have garnered widespread attention. Research topics include the analysis of factors influencing energy consumption and their weights, solar energy utilization, improvement of the thermal performance of building envelopes, cost control, and multi-objective optimization from different perspectives. However, no studies have yet proposed a comprehensive set of retrofit schemes that consider material and construction selection, cost investment, payback period, and enhancement of energy-saving potential. This is crucial for selecting the most suitable schemes according to specific needs and applying them in practice.

1.4. Why Consider Energy and Economy Synergistically

According to the National Bureau of Statistics of China’s statistical bulletin on national economic and social development released on 29 February 2024, the disposable income of national residents is divided into five categories: low income, lower middle income, middle income, upper middle income, and high income, with corresponding disposable incomes of 9215 CNY, 20,442 CNY, 32,195 CNY, 50,220 CNY, and 95,055 CNY, respectively (https://www.stats.gov.cn) (accessed on 10 March 2024).

Most rural households correspond to the low-income bracket. It is precisely because the per capita annual disposable income in rural areas is generally low that the willingness of residents to voluntarily renovate depends on the degree of energy saving (encompassing annual total energy savings, energy savings per unit area, and energy saving rate), as well as economic benefits such as cost input and payback period. Therefore, a renovation plan that considers both energy saving degree and economic benefits better aligns with the actual needs of the country, society, and individuals.

In China, the evaluation of building thermal performance is currently mainly based on the specified heat transfer coefficient limits for various parts of the building envelope and the limits of annual total load and energy consumption as stated in the General Norms for Building Energy Efficiency and Renewable Energy Utilization [36]. When using energy-saving reports that comply with national requirements, the types, thickness, and specific thermal parameters of building envelopes are often subjectively adjusted based on experience. However, this method does not clearly identify cost-effective solutions that simultaneously consider both energy and economic.

1.5. Existing Studies on Building Envelope Parameters Considering Energy and Economy

The balance between energy efficiency and economic feasibility is crucial for the feasibility of building energy retrofit project. Wang et al. [37] took Shandong Province in China as an example and proposed an automatic optimization scheme for building envelope structures based on neural networks and multi-objective genetic algorithms, with energy saving and thermal comfort as the guiding principles. The optimal solution improved overall energy consumption and thermal comfort by 25.5% and 21.6%, respectively. Li et al. [38] and Jia et al. [39] discussed the application of Phase Change Materials (PCMs) in buildings under different climatic conditions to enhance thermal performance while also considering economically feasible solutions. Luo et al. [40] aimed to reduce the overall life cycle energy consumption and cost using the HypE algorithm, and the optimized solutions reduced life cycle energy consumption by 15.1–18.8% and cost by 2.2–10.1%. Additionally, if residents have further energy-saving intentions, the life cycle energy consumption can be further reduced by 6.5% and 12.02%. Jiang et al. [41] carried out nearly zero-energy retrofitting of rammed earth rural residential buildings based on energy efficiency and cost–benefit analysis. The retrofitting measures included adding insulation materials, additional sunlight spaces, and PCMs, and the selected optimized solutions achieved an energy saving effect of 92.17%.

In conclusion, much further research has been conducted on the collaborative optimization of building energy and economics, yielding relatively ideal results. However, for building designers, most of the optimized paths constructed by existing research cannot be widely popularized for the following reasons: First, the optimized paths constructed by existing research are relatively complex, while simple, intuitive, and easy-to-use software programs are usually required. Second, multiple optimization solutions that comply with industry requirements need to be obtained in a short period of time, and the optimized paths constructed by existing research do not pay sufficient attention to simulation time and compliance with standard regulations.

1.6. Climate Condition

The research took place in Xingshugang Village, Daqing City, Heilongjiang Province, China. According to China’s Civil Building Thermal Design Code [42], the climate in this region is classified as severe cold 2B (equivalent to a temperate continental monsoon climate according to the Köppen climate classification). The design principle requires meeting winter insulation requirements, with summer heat prevention generally being less of a concern.

The climate conditions in Daqing exhibit distinct seasonal variations. Specifically, winter begins in November and lingers until the end of March the following year, with temperatures dropping between 0 and −30 °C. Summer, on the other hand, extends from early June to the end of August, with temperatures rising between 20 and 30 °C. Spring and autumn have relatively mild temperatures, making them suitable for outdoor activities and agricultural production. This information was verified by comparing the EPW format meteorological data from the EnergyPlus official website (https://energyplus.net/) (accessed on 10 March 2024) with Anda City as a reference (Daqing does not have a dedicated weather file and often uses the weather data of Anda City as a reference). The monthly average air temperatures throughout the year can be seen in Figure 1, and the hourly air temperatures can be seen in Figure 2. It can be observed that the air temperature data align well with the actual conditions.

Moreover, the area receives ample precipitation, with significant annual rainfall predominantly during the summer months, notably from June to August. In winter, precipitation mainly takes the form of snow, which might cover the ground for a certain period, impacting local life and economy. The region is relatively dry in winter and spring with low humidity; summer has the highest humidity, often exceeding 70%; autumn has moderate humidity and is more comfortable. Monsoons play a pivotal role in shaping the region’s climate. In winter, cold and dry northwest winds prevail, while in summer, warm and humid monsoons saturate the area. Overall, Daqing’s climate is characterized by distinct seasons, making it suitable for agriculture and energy development. Nevertheless, challenges arise due to the harsh cold and heavy snowfall during winter, necessitating appropriate countermeasures.

1.7. Basic Information of Rural Houses

(1)

Geographical overview

The location and overview of the village relative to the city are shown in Figure 3. There are over 200 rural houses in the village, with a permanent population of approximately 600, of which 2 (couples) and 3 (couples and children) make up the majority.

Most rural residential buildings in this village are nearly 30 years old. These homes are primarily self-built, with villagers often assisting each other to save on labor costs. Renovations are also typically carried out by residents themselves, focusing mainly on material costs. The proposed performance optimization solutions for these buildings can also provide guidance for the construction of new rural homes.

(2)

Heating load and equipment

In the Daqing area, most urban residences are heated through centralized heating systems, with the heating period lasting from 20 October to 20 April of the following year. According to the regulation [36], the average annual heating consumption per unit area for newly constructed residential buildings in the severe cold zone B is 178 MJ/(m²·a) (equivalent to 49.44 kW·h/(m²·a)). This underscores a substantial annual demand for heating energy. However, field surveys reveal that the heating demands of most existing rural houses in Xingshugang Village significantly exceed this limit, the heating demands of most existing rural houses far exceed the specified limit in the regulation [36].

It was observed that centralized heating facilities are not established in this area. Instead, rural houses rely on domestic cooking stoves located in the kitchens, where straw, wood, and coal combustion are used to generate heat. This heat is transferred through water pipes to radiators in various rooms to raise the overall room temperature. Additionally, the heat from the combustion is used to warm the heated Kang (a traditional Chinese bed with built-in heating). Nevertheless, in comparison to centralized heating systems, these methods generally exhibit lower thermal efficiency. An illustrative real-life indoor example is presented in Figure 4.

(3)

Materials and construction

Basic information from 81 local rural households was collected, and the types of materials used in these houses, along with their thermal parameters and the construction of various parts of the building envelope, are documented in

Table 1. Parameters of materials used in existing rural houses.

Material	Conductivity (W/(m·K))	Density (kg/m3)	Specific Heat (J/(kg·K))
Cement mortar	0.93	1800	1050
Plastering	0.81	1600	1050
Clay brick	0.81	1800	1050
Polymer-modified asphalt waterproof coiled material	0.17	600	1470
Polyurethane (PUR) waterproof coating	0.17	600	1005
Reinforced concrete roof slab	1.74	2500	920
Color steel plate	58.2	7850	480
Floor tile	1.59	1920	1260
Concrete	1.51	2300	920
Glass	0.76	2500	840
Air layer	0.017	1	1010
Expanded polystyrene board (EPS)	0.039	20	1380
Extruded polystyrene board (XPS)	0.03	35	1380
PUR	0.024	50	1380
Rock wool (RW)	0.041	120	1220
Expanded perlite (EP)	0.058	100	1170

and

Table 2. Existing rural house construction.

Envelope	Construction (from Outside to Inside)
Outer wall	Cement mortar (15 mm), Clay brick (490 mm), Plastering (20 mm)
Interior wall	Plastering (20 mm), Clay brick (240 mm), Plastering (20 mm)
Flat roof	1:3 Cement mortar (20 mm), Low-strength cement mortar (10 mm), Polymer-modified asphalt waterproof coiled material (4 mm), 1:3 Cement mortar (20 mm), XPS (140 mm), PUR waterproof coating (1.2 mm), 1:3 Cement mortar (20 mm), Reinforced concrete roof slab (120 mm)
Sloping roof	Color steel plate (wood beam support) (0.6 mm)
Ground	Floor tile (8 mm), 1:3 Cement mortar (30 mm), Concrete (60 mm)
Outer window	Glass (5 mm)
Inner window	Glass (3 mm)

. Among them, the thermal parameters of each material are determined based on the values given in the corresponding standard specifications according to the actual situation of the material. During the research process, it was also observed that most rural houses have been built for over 30 years, leading to issues such as structural aging, limited variety in materials, wall cracks, and moisture problems. These factors are also one of the reasons for high energy consumption and poor indoor thermal comfort during the heating season. It should be noted that the commonly used materials in the region mentioned in the study are the result of long-term considerations by the local population based on factors such as the ease of material acquisition, thermal performance, economic factors, and environmental and health aspects. The differences in the environmental impact of these materials over their entire life cycle are minimal.

1.8. Assessment of the Thermal Performance of Existing Rural House Envelope Structures

Table 3. Comparison of standards and actual K-values in rural houses.

	Minimum K-Value Satisfying the Limit Specified in [36]	Actual Situation	K Difference
	K (W/(m2·K))	K (W/(m2·K))	K (W/(m2·K))
Wall	0.25	1.25	1
Roof	0.2	0.2	0
Window	1.4	Subject to specific circumstances

compares the minimum K-values required by the regulation [36] with the actual K-values of rural houses, highlighting the discrepancies between the two. Variations in materials, techniques, and sealing properties of window frames lead to differing heat transfer coefficients, making it challenging to establish a uniform K-value difference for windows. It is evident that simply increasing the thickness of the walls does not significantly enhance their insulation performance. In contrast, the roof’s insulation performance significantly improves due to the presence of 140 mm of XPS, surpassing that of the walls and aligning relatively closely with the values specified in regulation [36]. Hence, the primary focus of rural house renovations should center on strategically selecting insulation materials concerning their types and thickness.

2. Methodology

2.1. Software Selection

Following the research requirements, the simulation software must possess functionalities such as parametric modeling, energy consumption calculation, data processing, and multi-objective optimization. Taking these factors into account, Rhino was selected as the simulation software. The simulation program was predominantly developed using Grasshopper and its plugins—Ladybug, Honeybee, and Octopus—which together provide a comprehensive array of tools for research purposes.

2.2. Simulation Path

To provide cost-effective envelope retrofit solutions for rural residences that take into account both energy and economic, a simulation pathway, as illustrated in Figure 5, has been established. This process was divided into four phases: modeling, energy consumption calculation, data processing, and multi-objective optimization. During the modeling phase, an energy consumption model was developed using Grasshopper and its plugins Ladybug and Honeybee, incorporating construction and thermal disturbance parameters. During the energy consumption calculation phase, Honeybee utilized the EnergyPlus engine to obtain data on the building’s heat load. The data processing phase involved the use of Grasshopper to create computational procedures for energy saving degree and economic benefits. During the multi-objective optimization phase, the Octopus plugin iteratively executed the aforementioned steps, cost-effective retrofit solutions are obtained through genetic iterations.

3. Result

3.1. Preparation before Simulation

Before conducting the formal simulation study, preparations were made in three areas: simulation validation, selection of research subjects, and parameter setting. For material thermal parameters, construction settings, software modeling, and cost and payback period calculations, information was not only obtained from field surveys but also compared with content from the literature referenced in the introduction to ensure the accuracy of the information.

(1)

Simulation verification

The study involves energy consumption simulation of individual buildings, and it is necessary to verify the differences between the simulated values and the measured values. Due to constraints such as time, equipment, and personnel, it is often not easy to obtain actual measured annual energy consumption values for buildings under ideal conditions. However, relevant standards and regulations [36] clearly specify the limit of annual heating energy consumption for new residential buildings in severe cold region 1B as 178 MJ/m²·a, which is equivalent to 49.44 kW·h/m²·a. Although this value is also obtained through simulation, its reliability has been widely confirmed. Comparing the simulated values with the standard limit values, it can be verified whether the simulated energy consumption values have practical reference significance.

A rectangular model measuring 8 m in length, 4 m in width, and 3 m in height was created in Rhinoceros 8 software, as depicted in Figure 6. Its construction parameters, indoor thermal disturbance parameters, and schedule were set according to the specifications [36], as detailed in

Table 4. Model parameter.

Construction Parameters			Unit
Window	South-facing window-to-wall ratio (WWR)	0.3	-
	Solar heat gain coefficient (SHGC)	0.35	-
External wall heat transfer coefficient		0.25	W/(m2·K)
Roof heat transfer coefficient		0.2	W/(m2·K)
Overhanging floor slab heat transfer coefficient		0.25	W/(m2·K)
Indoor thermal disturbance parameters			Unit
People		0.04	number/m2
Lighting		5	W/m2
Equipment		3.8	W/m2
Heating control (setpoint)		18	°C

and Figure 7. The model consisted of a single room, which served as the heating zone. An ideal air conditioning system was used to determine the heat load, and the simulation value for the building’s energy consumption during the heating season was calculated.

Following the simulation, the model’s annual heating consumption per unit area was calculated to be 54.10 kW·h/(m²·a). When compared to the specifications [36], it was determined that the discrepancy between the simulated and specified values was 8.61%. The possible reasons for this deviation are as follows: On the one hand, there could be slight differences in the weather files used for simulation. On the other hand, errors might have been introduced due to simplifications in the model and parameter settings. Additionally, differences between the averages of Daqing and the severe cold zone B could also contribute to the disparity. Despite these factors, the error falls within the acceptable 10% margin, allowing the research to proceed with this energy consumption calculation method.

(2)

Selection of research subjects

The study collected information from 81 residential units in the village, conducting a comprehensive assessment of factors such as the area, form, building materials and construction, and household population composition. From this, the most representative typical rural residential buildings that reflect the majority of the village’s characteristics were selected. This specific household exhibits typical characteristics of rural houses in the village. Real-life photographs of the house are depicted in Figure 8, and the architectural floor plan is presented in Figure 9. The constructed model is illustrated in Figure 10, with construction parameters matching those specified in Table 2.

(3)

Parameter setting and program construction

Five types of insulation materials, namely EPS, XPS, PUR, RW, and EP, were chosen to enhance the thermal performance of the walls and roof. To establish the simulation range for insulation material thickness, the minimum thickness of each insulation material is calculated, ensuring adherence to the heat transfer coefficient limits specified in regulation [36] (rounded to the nearest 10 mm increment). The minimum thicknesses of the five insulation materials are listed in

Table 5. Minimum thickness of insulation material required for renovation.

	EPS	XPS	PUR	RW	EP
Wall	120	100	80	130	180
Roof	180	140	110	190	270

. For comprehensive and consistent simulation, the simulation range was expanded to 50–300 mm, with 10 mm increments.

Considering the original rural house featured a single-layer window with a thickness of 5 mm, its insulation performance was relatively weak. Therefore, discussions were conducted regarding the number of window layers and thickness. Window thickness and the thickness of the air gap layer were chosen based on commonly observed local values. The final simulated range for window thickness was set at 3–12 mm, and the air gap layer thickness was set at 6–20 mm, both varying in increments of 1 mm. In the multi-objective optimization process, the independent and dependent variables are specified in

Table 6. Setting of independent and dependent variables.

Independent	Types of Insulation Materials	Insulation Material Thickness	Number of Window Layers	Glass Thickness	Air Gap Thickness
	EPS, XPS, PUR, RW and EP	50–300 mm	Single-Layer/Double-Layer	3–12 mm	6 mm–20 mm
Dependent	Energy saving rate	Cost input	Payback period

. It is worth noting that annual total energy savings, energy savings per unit area, and energy saving rate all reflect the extent of energy conservation. For multi-objective optimization, only the energy saving rate was chosen as the criterion for energy conservation, while cost input and payback period were used to evaluate economic benefits.

Among the 81 rural residential units surveyed, all used household stoves for heating, with Kang and radiators serving as heat transfer media to distribute warmth across various rooms. The thermal efficiency of household stoves across these units showed minimal variation. The primary heating fuel was coal, supplemented by straw and firewood. This choice was based on a comprehensive consideration of factors such as the availability of coal, its heat output, burning duration, and compatibility with heating equipment. For ease of cost estimation, coal was used as the heat source to determine the energy demand for the heating season.

The heating costs for rural residential buildings are determined based on the materials and construction of the buildings, the area of the heating zone, the building form, and the thermal conversion efficiency of the energy and equipment. These details need to be input as known information before the simulation. If a different fuel source is desired, one only needs to substitute the standard heat value and thermal conversion efficiency of the new fuel, along with its unit mass price, while keeping all other information in the program unchanged. In the constructed program, the heat load demand of the rural residential buildings is calculated using an ideal air conditioning system. Then, based on the energy conversion efficiency of the heating equipment, the heating energy consumption of the building is calculated. Finally, the fuel consumption is determined based on the building’s heating energy consumption and the fuel’s heat value. Therefore, the results are precisely calculated based on the provided information.

Figure 11 illustrates the automated simulation program designed according to the specified requirements, with different sections distinguished by background colors for clarity. Among these, Part 1 includes the battery group for setting enclosure structure materials and construction parameters. Part 2 encompasses the battery group for setting indoor thermal disturbance parameters. Part 3 signifies the battery group for parametric modeling. Part 4 contains the battery group for energy consumption calculation and result output. Part 5 comprises the battery group for energy saving and economic benefits calculations. Part 6 includes the battery group for multi-objective optimization. The detailed procedures for each part can be found in Appendix A.

During the energy saving and economic benefits calculations, the following formulas were used:

$${\mathrm{Q}}_{\mathrm{E}\mathrm{S}\mathrm{A}}={\mathrm{Q}}_1-{\mathrm{Q}}_2$$

(1)

$${\mathrm{Q}}_{\mathrm{SR}}{=\mathrm{Q}}^{\mathrm{ESA}}{/\mathrm{Q}}_1$$

(2)

$${\mathrm{m}}_{\mathrm{CC}}{=\mathrm{Q}}_{\mathrm{ESA}}/(\mathrm{q}\times \mathsf{\eta })$$

(3)

$${\mathrm{C}}_{\mathrm{ACS}}{=\mathrm{m}}_{\mathrm{CC}}\times {\mathrm{P}}_{\mathrm{C}}$$

(4)

$${\mathrm{C}}_{\mathrm{I}}{=\mathrm{C}}_{\mathrm{W}}{+\mathrm{C}}_{\mathrm{R}}{+\mathrm{C}}_{\mathrm{Win}}$$

(5)

$${\mathrm{Y}}_{\mathrm{I}}={\mathrm{C}}_{\mathrm{I}}{/\mathrm{C}}_{\mathrm{ACS}}$$

(6)

where:

${\mathrm{Q}}_{\mathrm{E}\mathrm{S}\mathrm{A}}$—Energy saving amount; ${\mathrm{Q}}_1$—Original rural house’s annual total load; ${\mathrm{Q}}_2$—Modified rural house’s annual total load; ${\mathrm{Q}}_{\mathrm{SR}}$—Energy saving rate; ${\mathrm{m}}_{\mathrm{CC}}$—Coal consumption; $\mathrm{q}$—Calorific value of standard coal, taken as 29,288,000 J, approximately 8.14 kW·h/kg after conversion; $\mathsf{\eta }$—Thermal utilization efficiency, assumed as 0.7 (for domestic cooking stoves, actual value varies based on the specific circumstances); ${\mathrm{C}}_{\mathrm{ACS}}$—Annual cost saving; ${\mathrm{P}}_{\mathrm{C}}$—Price of coal, assumed as 1200 CNY per ton (actual value varies based on the specific circumstances); ${\mathrm{C}}_{\mathrm{I}}$—Cost input; C_W—Cost of wall renovation; ${\mathrm{C}}_{\mathrm{R}}$—Cost of roof renovation; ${\mathrm{C}}_{\mathrm{Win}}$—Cost of window renovation; ${\mathrm{Y}}_{\mathrm{I}}$—Investment payback period.

Because finished windows exhibit significant price variations due to different materials and manufacturing processes, cost calculations become challenging. Therefore, the pricing calculation for windows excludes the frame section and concentrates solely on the valuation based on glass area and thickness. Prices for insulation materials and glass are determined based on rates provided by local manufacturers on the Construction Cost Network website (https://ah.zjtcn.com/) (accessed on 10 March 2024). Please refer to

Table 7. Material prices.

	EPS	XPS	PUR	RW	EP	Glass
Price	302.8 CNY/m3	607.06 CNY/m3	1268.69 CNY/m3	346.65 CNY/m3	462.84 CNY/m3	7 CNY/(mm·m2)

for specific prices.

The multi-objective optimization is achieved using a genetic algorithm modeled after an octopus. The set objectives include maximizing energy efficiency while minimizing the payback period and cost investment. Since the octopus can only automatically optimize the minimum values of the objectives, the energy efficiency is converted into a negative number using a Negative cell before being input. Within the octopus, the History parameter is set to 20, the Population Size is 50, and other parameters are kept at their default values.

Although the program constructed is quite extensive, it only requires a few inputs, which are highlighted in red. After clicking Start in the Octopus plugin, the iteration process begins, and once the results fully converge, the iteration can be stopped. A point from the Pareto Front can then be selected, and the Reinstate solution function can be used to observe the corresponding optimized solution and results in the constructed program. With video tutorials or personal guidance, proficiency in using this program can be achieved in just one day.

3.2. Comparison of Selected Options

The multi-objective optimization process was stopped after the curve converged, resulting in a total of 20 iterations. Under the set boundary conditions, approximately 30 points, corresponding to 30 optimized solutions, were generated after the 20th iteration. To accommodate the preference for reducing the time required to obtain results and select the optimal solution from a limited set, 8 points with significantly different spatial distributions were selected from the 30 points generated in the 20th iteration. The corresponding optimized solutions were then presented using the Reinstate solution function.

Based on the obtained set of eight optimal solutions, the independent and dependent variables were organized as shown in

Table 8. Renovation plans for rural house.

Solution	Wall		Roof		Window	Energy Saving Rate (%)	Annual Energy Consumption per Unit Area (kW·h/(m2·a))	Cost Input (CNY)	Payback Period (Year)
	Insulation Material	Thickness (mm)	Insulation Material	Thickness (mm)	Thickness (mm)
1	XPS	290	XPS	300	11 + 16A + 11	83.29	31.28	18,502	11.68
2	XPS	290	EPS	270	10 + 16A + 10	81.35	34.92	13,522	8.74
3	RW	300	RW	100	11 + 16A + 11	80.19	37.09	6240	4.09
4	RW	50	RW	260	10 + 20A + 10	78.23	40.77	2567	1.73
5	RW	130	RW	80	9 + 16A + 9	65.57	64.47	1401	1.12
6	RW	50	RW	90	10 + 16A + 10	55.79	82.78	−273	−0.26
7	RW	50	EPS	90	8 + 16A + 8	46.81	99.60	−751	−0.84
8	RW	50	RW	60	9 + 12A + 9	41.48	109.58	−918	−1.16

. It is evident that most solutions utilized RW as the insulation material for walls and roofs, highlighting its favorable cost-effectiveness. Double-layered windows were selected in all optimized solutions, showcasing the significant insulation capacity of the air gap layer. Solutions 6 to 8 had negative retrofit costs. This occurred because the original rural house in the simulation had a 140 mm XPS layer on the roof. The roof retrofit cost was calculated by subtracting the cost of the materials used for the retrofit from the cost of the original insulation material. This indicates that a judicious choice of insulation materials can substantially enhance energy efficiency and reduce costs.

However, considering the regulation stipulating that the energy consumption per unit area should be less than 49.44 kW·h/(m²·a), Solutions 5 to 8 did not meet the requirement. Among Solutions 1 to 4, taking into account the variations in energy saving rate, cost input, and payback period, Solution 4 exhibited a higher energy saving rate and was more favorable in terms of cost input and payback period.

Upon comparing the minimum required RW thickness in walls, as indicated in Table 5, it was discovered that the 50 mm RW in Solution 4 from Table 8 did not meet the thermal conductivity limit requirements for walls. To address this issue, the simulation range was narrowed down to 130 mm–300 mm, considering RW as the insulation material. The program was then restarted for further iterations. After conducting the updated multi-objective optimization, the obtained solutions are presented in

Table 9. Recommended plans for rural house renovation.

Solution	Insulation Material	Wall	Roof	Window	Energy Saving Rate (%)	Annual Energy Consumption per Unit Area (kW·h/(m2·a))	Cost Input (CNY)	Payback Period (Year)
		Thickness (mm)
1	RW	140	280	11 + 16A + 11	79.36	38.65	5278	3.5
2	RW	130	280	10 + 16A + 10	79.26	38.83	4886	3.24
3	RW	130	280	5 + 6A + 5	75.68	45.53	4166	2.9
4	RW	150	200	5 + 16A + 5	73.81	49.03	3326	2.37
5	RW	130	200	3 + 20A + 3	73.54	49.54	2542	1.82
6	RW	150	150	4 + 16A + 4	68.24	59.47	2347	1.81
7	RW	130	170	3 + 20A + 3	64.87	65.78	2041	1.66
8	RW	130	130	3 + 9A + 3	63.34	68.67	1373	1.14

. Analysis revealed that Solutions 1 to 4 in Table 9 fully met the regulatory requirements. The earlier the scheme number, the better the energy-saving effect of the building, but it may lead to increased costs and longer payback periods. Therefore, the selection of the scheme should also take into account the energy-saving willingness of the owner. If there are no specific requirements, scheme 4 should be chosen, as it offers the optimized solution with the least cost investment and the shortest payback period, while still meeting the standard regulatory requirements.

3.3. Performance Comparison of Programs

This section aims to compare the automatic optimization path for building envelope thermal performance improvement with the commonly used energy saving software in China’s construction industry (Green Building SWARE 2024 (GBSWARE) Software) in terms of functionality, response time, and output results. See

Table 10. Performance comparison of the programs.

Name	GBSWARE	Automatic Optimization Path
Ease of operation	Both of them require the model to be constructed first, followed by inputting material parameters and structures, and then the simulation can be carried out.
Function	The focus is on checking whether the heat transfer coefficients of various parts of the building envelope and the annual total energy consumption are within the specified limits. Only one result can be generated at a time, and if the thermal design does not meet the requirements of Regulation [36], adjustments need to be made by the practitioners themselves.	Support is provided for multi-variable, multi-objective automatic parameter optimization. Multiple renovation plans considering both energy saving degree and economic benefits can be generated in one go. Practitioners only need to select the plan they deem most reasonable.
Response time	Simulations with energy consumption as the single objective can be completed within a few minutes.
Simulation engine	DOE-2	EnergyPlus
Output results	The accuracy of energy consumption simulations has been widely validated by the industry.	The simulation verification section in Section 3.1 (1) demonstrated the accuracy of energy consumption simulations.

for details. It should be noted that the premise of the response time comparison is that both are simulated based on the main hardware parameters shown in

Table 11. The primary hardware parameters used for the simulation.

Central Processing Unit	Memory	Graphics Card	Mainboard
Intel Core i7-8750H @2.20GHz Hexa	32 GB (Hynix DDR4 2666 MHz 16 GB/Micron DDR4 2666 MHz 16 GB)	NVIDIA GeForce GTX 1070 (8 GB/Asus)	Asus GL504 GS (HM370 Chipset)

.

The number of independent variables, the range of their variations, and the number of dependent variables are the main factors causing changes in response time, among which the number of dependent variables has the most significant impact on response time. When the dependent variable is only the energy consumption as a single target, there is no significant difference in simulation time (usually within a few minutes) and results between the GBSWARE software and the automatically optimized pathway constructed in the study. However, when the dependent variables are the synergy control of multiple objectives such as energy saving rate, cost input, and payback period, only the self-constructed automatic optimization pathway can meet the requirements. The simulation time varies with the settings of Population Size and the control of iteration times. In Section 3.1, with a Population Size set to 100 and 20 iterations, a cost-effective rural residential building envelope renovation scheme was obtained within approximately 4 h.

Before conducting the formal simulation study, the research utilized simulation validation methods to demonstrate the accuracy of the constructed simulation program’s results. Parameters such as material properties, construction details, energy types, and thermal conversion efficiencies were input as known information prior to the simulation, significantly enhancing the consistency with actual data. The various optimization schemes obtained through the multi-objective automatic optimization method each emphasize different aspects, such as investment cost, payback period, and energy-saving potential. Therefore, each result can serve as an effective scheme to be selected based on specific needs.

The purpose of the study is to provide an automatic optimization path for building envelope structure retrofitting that considers energy and economics in a collaborative manner. The innovative aspect of the research is that this path is constructed from the perspective of building designers, enabling simple operations to quickly obtain multiple cost-effective retrofitting solutions that comply with standard regulations, and revealing the energy-saving and cost increment associated with each solution. The study is mainly divided into three parts: the construction of an automatic optimization path for building envelope structure retrofitting that considers energy and economics in a collaborative manner, methods for recording and reducing simulation time, and the evaluation and selection of the optimized solutions.

However, this approach has limitations. Material prices can vary due to geographical, temporal, quality, and performance parameters. The price of coal and the thermal utilization efficiency of domestic cooking stoves also need to be determined based on actual conditions. Therefore, it is necessary to determine the numerical values of various parameters before performing automatic optimization. Additionally, the calculation cost of windows is relatively underestimated since it is only evaluated based on the number of glass layers and thickness, leading to a potential discrepancy with actual costs.

4. Conclusions

This study has pioneered an automated optimization approach for building envelope renovation, focusing on both energy conservation and economic efficiency. Taking Chinese cold region rural houses as an example, the focus of the renovation was placed on the types and thicknesses of insulation materials for walls and roofs, as well as the selection of window layers and thickness. Multiple renovation solutions were generated through multi-objective optimization. The optimal renovation solution for rural houses in the Daqing region, chosen in compliance with the regulations [36], involves utilizing 150 mm RW for wall insulation, 200 mm RW for roof insulation, and a double-layer window with a 5 mm + 16 mm air gap + 5 mm glass. Compared to the original rural house, this solution achieves an energy saving rate of 73.81%, annual unit area energy consumption of 49.03 kW·h/(m²·a), a cost input of 3326 CNY, and a payback period of 2.37 years. The research demonstrates that through this pathway, several renovation schemes for rural residential building envelopes, guided by energy saving degree and economic benefits, can be swiftly obtained. The program meets practitioners’ requirements in terms of operational ease, functionality, response time, and output results. It has the potential for application in different climates and regions.

Subsequent research could delve deeper into the applicability of this approach for envelope structure renovations in diverse environmental conditions and specific optimization solutions, offering valuable insights for building renovations centered on energy conservation and economy.