Impact of Spatial Layout Design on Energy Consumption of Ice Rinks in Cold Regions

Abstract

The differentiated physical environment requirements within the internal space of ice rinks in cold regions result in a complex heat exchange process, which becomes the primary cause of high energy consumption. Therefore, analyzing the impact mechanisms of spatial layout parameters on the energy consumption of ice rinks is crucial during the early design stages. This study employed the Delphi method to identify the key parameters affecting the total energy consumption of ice rinks. It conducted single-factor experiments using building performance simulations to quantify the relationship between each layout parameter and the energy consumption. Based on the single-factor experiment results, orthogonal experiments were conducted to develop an energy-efficient spatial layout combination. The study indicates that the height-to-width ratio and the mixed area width are the most significant parameters. By adjusting the values of these parameters, the total energy consumption can be reduced by approximately 18% to 31%. The spatial layout strategy for ice rinks in cold regions proposed in this study will help architects make more effective decisions during the early design stages.

1. Introduction

The construction industry significantly contributes to China’s energy consumption and greenhouse gas emissions [1]. Its full-process energy consumption accounts for approximately 45.8% of the country’s total energy. The extremely high energy intensity of public buildings is particularly noteworthy [2], making them important targets for energy conservation.

In recent years, following the successful hosting of the 2022 Beijing Winter Olympics, the nation has vigorously promoted the popularization and development of ice sports across the country, leading to an unprecedented boom in ice rink construction. However, the complex thermal environment and the presence of the ice surface result in particularly severe energy consumption issues [3,4]. According to the U.S. Department of Energy, the average annual electricity consumption of a typical indoor ice rink in Massachusetts is 730 MWh. In comparison, the annual average energy consumption of an ice rink in Sweden is about 1185 MWh [5].

In cold regions, the rink’s ice surface requires continuous refrigeration to maintain its quality and smoothness. Additionally, extremely low temperatures increase heat loss from the building’s interior, raising the heating demand to ensure the comfort of spectators and staff. These environmental conditions make the heat exchange processes inside the skating rink more complex and variable. Therefore, addressing the high energy consumption issues of ice rinks in cold regions is particularly necessary.

1.1. Ice Rink Energy Efficiency Improvement Research

In the energy consumption structure of ice rinks, refrigeration energy consumption accounts for the most significant proportion, ranging from 35% to 75% [5]. One measurement also found that the annual operational electricity consumption of indoor ice-making systems per unit area can reach 300–700 kWh/m², 1.7 to 3.9 times that of office buildings [6]. In response, scholars have proposed corresponding energy-saving measures targeting ice-making energy consumption. Daoud used energy consumption simulation software to model and simulate an ice rink in Montreal, analyzing the impact of supply air temperature and refrigerant temperature on refrigeration energy consumption [7]. Li et al. used CFD to simulate the effects of displacement and jet ventilation systems on humidity and temperature in the competition hall, emphasizing that a well-designed ventilation system can effectively save energy [8]. Erola et al. conducted a comprehensive evaluation of the performance of ice rink refrigeration systems through advanced motion economy analysis [9]. Lin et al. used CFD to study the heat and moisture transfer characteristics in the competition hall of an ice rink, finding that regional mixing generates a large amount of additional energy consumption and highlighting the importance of proper ventilation system design and adequate physical isolation [10]. Mun et al. developed and identified the optimal R-value affecting the ice-making system based on simulations and economics to economically and effectively reduce refrigeration energy consumption economically and effectively [11]. Li et al. experimentally discussed the relationship between the frost formation time and humidity ratio at different stages of the ice surface to reduce dehumidification energy consumption [12]. Additionally, many studies [4,13,14,15,16,17] emphasized that using low-emissivity materials for the ceiling of the competition hall can effectively reduce ice-making energy consumption.

In summary, most existing studies explore energy-saving strategies for ice rinks mainly from the perspectives of equipment systems and ice-making processes.

1.2. Impact of Ice Rink Spatial Layout on Energy Consumption

It is worth noting that the refrigeration energy consumption of ice rinks is closely related to spatial layout. This may be because to maintain the rink’s thermal environment balance, the diverse temperature control requirements of different functional areas generate a significant amount of additional energy consumption [10]. Additionally, as large space buildings, unnecessary spatial waste in ice rinks may lead to substantial lateral temperature differences [18], further intensifying heat exchange. However, only some have considered the impact of passive design factors such as the building layout on the refrigeration energy consumption of ice rinks, let alone conducted quantitative studies on this aspect.

1.3. Design Decisions Based on BPS

It is important to note that once the spatial layout of a building is determined, it is difficult to modify, highlighting the importance of making correct design decisions in the early conceptual stages [19,20,21,22]. Due to the flexibility of building design, early-stage design decisions offer significant space for adjustment [23]. Building Performance Simulation (BPS) is widely used in the design process, helping designers explore the energy performance potential of buildings and improving design efficiency [24]. Bellache et al. developed a two-dimensional transient model to describe airflow heat and mass transfer within an ice rink. Seghouani et al. further developed a transient three-state model to predict the annual energy consumption of ice rinks [25]. Mon et al. researched a heat conduction model for ice-making systems, integrating it into EnergyPlus 22.1.0 for energy consumption simulation analysis [3]. Teyssedou et al. proposed a mathematical and simulation model for heat transfer in ice rink floors [26]. The development of energy consumption simulation models of ice rinks is becoming increasingly refined. However, only some studies have integrated these models with early-stage architectural design decisions, which hinders architects from making more scientifically informed design decisions at the early stages.

1.4. Aim of This Research

In summary, the following research gaps can be identified: Existing energy-saving designs for ice rinks primarily focus on equipment systems and ice-making processes, needing a comprehensive review of all factors contributing to reducing ice rink energy consumption. During the design proposal stage, current ice rink layout designs rely on the subjective experience of designers, with insufficient quantitative analysis. This imprecise decision-making approach may lead to inadequate building performance in later stages, affecting operational sustainability. Additionally, the layout design of ice rinks in cold regions lacks systematic design guidelines, making it difficult to provide a scientific basis for decision-making.

The innovations of this study include the following:

• A comprehensive summarization and discussion of low-energy consumption design parameters for ice rinks through the Delphi method;
• Combining single-factor experiments and orthogonal experiments based on BPS, the study quantifies the mechanisms of spatial layout regarding energy consumption, ultimately identifying more energy-efficient spatial layout combinations;
• The study develops ice rink layout strategies in cold regions to assist designers in decision-making at the planning stage, enabling passive energy savings in ice venues in cold regions.

2. Methods

Figure 1 illustrates the workflow of this study. The initial stage involved establishing a baseline model of the ice rink through a combination of a literature review, online research, and field investigations. This model was developed using the Rhino platform. In the second stage, the Delphi method was employed to comprehensively evaluate key design parameters influencing the energy consumption of ice rinks. Additionally, through research and references, the spatial layout parameters of ice rinks in cold regions were clarified, and the ranges of these parameters were determined, laying the foundation for subsequent energy consumption simulation experiments. The third stage involved constructing an energy consumption model using the Ladybug 1.5.0 and Honeybee 1.5.0 modules on the Grasshopper platform for Rhino 7.9. The entire energy simulation process was executed in collaboration with plugins, including the OpenStudio 1.4.0 energy simulation engine and Colibri 2.0. Single-factor simulation experiments were conducted on the selected spatial layout design parameters in the fourth stage. Based on these results, parameters significantly impacting energy consumption were chosen for orthogonal experiments. The optimal low-energy consumption parameter combination for ice rinks was determined through multi-factor simulation experiments. Furthermore, the key parameters of the spatial layout affecting ice rink energy consumption were identified by considering the interactions between parameters. Finally, based on these analyses, design strategies for the spatial layout of ice rinks were proposed.

2.1. Climate Information

Thanks to the favorable climatic conditions, ice rink construction in China is mainly concentrated in the three northeastern provinces. Harbin has the highest number of ice rinks in cold regions, making it representative. Additionally, Harbin has hosted international or national ice sports events. Therefore, the meteorological parameters in this study are selected from Harbin, Heilongjiang Province. The meteorological file for Harbin was obtained using LB Download Weather from the Chinese Standard Weather Database (CSWD), as detailed in Figure 2. It can be observed that the climate characteristics of cold regions include extremely low outdoor temperatures in winter and relatively low solar radiation intensities.

2.2. Building Energy Model

This study employed literature reviews, online data collection, and field surveys to extensively gather essential information on 65 ice rinks in cold regions of China, excluding speed skating rinks, air-supported structures, and commercial ice rinks. Twenty-five typical ice rinks in the three northeastern provinces, including Harbin, Shenyang, Jilin, and Changchun, were selected for field surveys. As shown in Figure 3, the survey statistics revealed that medium-sized ice rinks with 3000 seats and double-sided seating are the most representative of constructing ice rinks in cold regions. This type of ice rink had the largest number and the most diverse spatial layout patterns. According to the requirements of ice sports events, venues with over 2500 seats need at least one training ice rink in addition to the competition rink [27,28]. The survey of ice rinks in cold regions found that 57.3% of the medium-sized ice rinks have two ice surfaces, making dual-ice rinks the primary focus of this study.

The statistical survey results show that the ice rink floor plans are primarily right-angled rectangles. The layout of the two ice surfaces mainly follows the pattern where the long side of the competition rink is perpendicular to the long side of the training rink. Considering the need to accommodate various ice sports such as short-track speed skating, curling, figure skating, and ice hockey, the selected ice area dimensions are 70 m × 40 m [27,28]. The total building area is 9600 m², with the height of the competition hall set at 15 m. Other detailed dimensions are provided in Figure 4. The seating arrangement in the competition hall is designed with double-sided seating around the ice surface, accommodating 3000 spectators.

Since there is no specialized thermal code for ice rinks to guide the design of the envelope, the thermal performance of the envelope in this study refers to the standard [29] for public buildings in the cold B zone. Furthermore, to enhance modeling and calculation efficiency, a scientific simplification method is used in this study, abstracting the enclosure interfaces of the building as zero-thickness planes while retaining their thermal resistance characteristics [30].

This model integrates an air-cooled ice-making system into the baseline model. In terms of physical environmental parameter settings, due to the complex thermal environment inside the ice rink, the operational settings required for different functional areas are different. According to the standards and references, the parameters for temperature, humidity, lighting, and occupancy are provided in

Table 1. Physical Environmental Parameter Settings for Different Zones [10,27,31].

Functional Zone	Temperature (°C)	Humidity (%)	Occupancy (Person/m2)	Lighting (w/m2)
Ice field	−9–−3	35	0.036	—
Competition hall/ Training hall	8–10	40	—	48
Spectator stand	18–24	65	0.8	15
Auxiliary room	21–25	65	0.1	20

.

2.3. Selection of Parameters Affecting Energy Consumption

2.3.1. The Definition of Key Parameters Used in the Delphi Method

The Delphi method was used for parameter selection, as existing studies do not provide an in-depth analysis or summary of low-energy design factors for large spaces like ice rinks. This study first classified the factors influencing ice rink energy consumption through literature reviews and interviews. Twenty experts and architects in the field of ice rink design were invited to rate each parameter to clarify the design variables that were most relevant and modifiable by architects during the early design phase.

2.3.2. Definition of Spatial Layout Parameters

This study organized and labeled the spatial layout parameters that require decision-making (Figure 5). Based on the field survey measurements of 65 typical ice rinks and references [29,32,33], the range of input parameters was determined (

Table 2. Classification of ice rink spatial layout parameters.

Classification	Parameter	Abbreviation	Range	Unit	Step Size
Spatial orientation	Orientation	O	[0, 360]	°	1
Spatial dimension	Length-to-width ratio	LW	[1.12, 1.40]	-	0.01
	Height-to-width ratio	HWH	[10, 23]	m	1
		HWW	[70, 80]	m	1
Spatial organization	Ice surface spacing	D	[−5.0, 5.0]	m	0.01
	Mixed area width	d1	[3.0, 9.0]	m	0.5
		d2	[3.0, 9.0]	m	0.5
		d3	[3.0, 9.0]	m	0.5
	Enclosure mode of auxiliary space	S1	[−3, 4]	m	0.5
		S2	[−5, 6]	m	0.5

). The specific definitions and explanations of the parameters’ ranges are as follows:

• Orientation: The parameter value obtained by clockwise rotating the ice rink along its long side. The statistical analysis showed no clear pattern in the distribution of building orientations among the venues, so in the simulation, the orientation was varied from 0° to 360°, with a step size of 1°.
• Length-to-width ratio: The parameter is defined as the ratio of length to width. The survey found that the venues’ length-to-width ratio mostly ranges between 1.0 and 1.7, with a step size of 0.01.
• Height-to-width ratio: The parameter is defined as the height-to-width ratio. According to relevant regulations, the space above the ice surface should have a minimum height of 7 m. Combining the survey data of the competition hall height, the parameter was simulated with a step size of 1 m, varying from 10 m to 23 m. Similarly, based on the survey results, the width variation range is set to 70 m to 80 m, with a step size of 1.
• Ice surface spacing: The parameter is defined as the distance between the adjacent edges of the competition and training ice surface. Based on survey statistics, a baseline distance of 12.4 m was set, with positive values indicating an increase and negative values indicating a decrease in spacing. The simulation was conducted with a step size of 0.1 m, ranging from −5 m to 5 m.
• Mixed area width: The parameter is defined as the distance from the outer edge of the artificial ice surface to the first row of seats. d₁, d₂, and d₃ represent the widths of the mixed areas in different orientations (Figure 5). According to ice sports event requirements, the minimum width of the ice surface mixed area should be at least 3 m. Referring to survey data, the parameter was simulated with a step size of 0.5 m, ranging from 3 m to 9 m.
• Enclosure mode of auxiliary space: The research found that the most typical enclosure pattern places the center of the ice rink at the geometric center of the venue’s floor plan, with auxiliary spaces symmetrically distributed around it. This parameter takes this layout as the initial state for the auxiliary space enclosure pattern. An offset of the ice rink center towards the north or east is defined as a positive value, while an offset towards the south or west is defined as a negative value. Here, S1 represents the horizontal offset distance of the ice rink center, ranging from −3 m to 4 m; S2 represents the vertical offset distance of the ice rink center, ranging from −5 m to 6 m. In the simulation, the step size was set to 0.5 m.

2.4. Eenergy Consumption Simulation

2.4.1. Simulation Tool

This study focuses on spatial layout parameters most relevant to architects, but the significant variability of these parameters leads to extensive modeling requirements. Compared to energy simulation software such as Energy Plus, Ecotect, and Design Builder, the Ladybug and Honeybee modules, embedded within the Grasshopper platform, can meet the need for creating batch parametric models. This study used the Open Studio energy simulation engine built into Honeybee, working in collaboration with plugins like TT Tool Box, to efficiently and effectively integrate the entire energy simulation process, making it the chosen tool for energy simulation experiments in this research.

2.4.2. Energy Performance Objective

To evaluate the impact of the spatial layout on the overall energy performance of the ice rink, this study primarily selected the annual Energy Use Intensity (EUI) as the energy performance objective. Additionally, refrigeration energy consumption (Er₁) (including refrigeration energy consumption for the training rink (Er₂) and competition rink (Er₃)), heating energy consumption (Eh), and cooling energy consumption for auxiliary rooms (Ec) were used as auxiliary objectives to more precisely examine the impact of the spatial layout on each type of energy consumption.

2.5. Energy Model Calibration

The baseline model in this study has a floor plan and scale highly similar to the Baqu Ice Hockey Arena in Harbin, Heilongjiang Province. According to technical drawings (Figure 6), a parametric model of the Baqu Ice Hockey Arena in Harbin was established, and energy simulation experiments were conducted following the above processes and parameter settings.

The simulation data were compared with and calibrated against the monthly electricity consumption data for the ice rink in 2023, obtained from field surveys. Due to the complexity and variability of factors such as human activity, equipment operation management, and heat exchange between different areas during the actual operation of the venue, the simulation software cannot achieve the same level of precision as real-world conditions, leading to some discrepancies in the simulation data. As shown in Figure 7, the fit between the simulation data and the measured data is good, with a monthly absolute error range of 1.6% to 9.8%, verifying the feasibility of the entire simulation experiment process and the rationality of the parameter settings.

2.6. Orthogonal Experiment

An Orthogonal Experiment (OE) is a design method used to study multiple factors at different levels. This method is based on orthogonality, selecting representative partial experimental points from a comprehensive set of experiments, thereby significantly reducing the number of experiments, improving experimental efficiency, and lowering experimental costs.

In this study, considering the interaction between parameters, an orthogonal experiment was conducted after completing single-factor simulations of six spatial layout parameters of the ice rink. According to the number and levels of parameters, an L18 (3⁶) orthogonal table was generated using SPSSAU [34], an online statistical analysis tool platform based on cloud computing technology. Then, energy consumption simulations were performed for each parameter combination in the orthogonal table to determine the combination with the lowest energy consumption.

3. Results

3.1. Delphi Method Results

The scoring results were then imported into Yaahp software (https://www.metadecsn.com/yaahp/ (accessed on 2 June 2024)) for weight calculation. The statistical results showed that the consistency was less than 0.10, passing the consistency test.

Figure 8 shows the final results of the Delphi method. Parameters related to the building dimension have the most significant impact on energy consumption. Equipment processes and envelopes also significantly influence the energy consumption of ice rinks. Additionally, essential parameters include spatial organization and orientation. This result further confirms that, in the eyes of designers, studying the spatial layout of ice rinks has substantial energy-saving potential.

3.2. Single-Factor Simulation Results

This section discusses the impact of changes in each spatial layout parameter on the venue’s energy consumption when considered as a single variable. To better understand the relationship between spatial layout parameters and various types of energy consumption, we used parallel plots of decision variables and objectives to present the results. In the parallel plots, red lines indicate schemes with relatively low total energy consumption, while bold red lines represent the most energy-efficient schemes. The plots provide specific values for refrigeration energy consumption for the training and competition rink, heating energy consumption, and cooling energy consumption for auxiliary rooms under the total energy consumption scenarios. Additionally, the parallel plots visualize both more energy-efficient and less energy-efficient spatial schemes, which quantify the relevant values and make it easier for readers to understand the results intuitively.

3.2.1. Spatial Orientation

As shown in Figure 9a, the impact of orientation on EUI is non-linear. As the ice rink rotates clockwise, the building’s total energy consumption first increases and then decreases. The optimal energy-saving orientation range is 244° to 248°, while the least favorable orientation is 104° to 108°. Overall, the impact of orientation on ice rink energy consumption is relatively small.

Figure 9b further illustrates the influence of orientation on refrigeration and heating energy consumption. It can be observed that the optimal orientation can achieve lower refrigeration energy of the training hall (Er₃) and heating energy (Eh). In contrast, the orientation has a relatively smaller impact on the refrigeration energy consumption of the competition hall (Er₂). Additionally, Figure 9 also presents the most energy-efficient scheme for the orientation of the ice rink.

3.2.2. Spatial Dimension

As shown in Figure 10a, there is a linear relationship between the length-to-width ratio and the total energy consumption. The optimal value for the length-to-width ratio is around 1.12 to 1.16. This indicates that the ice rink is most energy-efficient when its layout is approximately a compact square. Similarly, the height-to-width ratio exhibits a clear linear relationship (Figure 10b), with the optimal range for the height-to-width ratio being between 0.140 and 0.142.

In comparing the length-to-width ratio and the height-to-width ratio, the length-to-width ratio can cause a change in the total energy consumption by 2.64 kWh/m². In contrast, the height-to-width ratio can cause a change in the total energy consumption by 10.50 kWh/m². It is evident that, for ice rinks, the height significantly impacts the total energy consumption. The lower the arena height, the lower the total energy consumption of the ice rink.

3.2.3. Spatial Organization

Figure 11a shows the relationship between the ice surface spacing and the energy consumption of the ice rink. It can be seen that the ice surface spacing can cause a variation in total energy consumption within a range of 2.69 kWh/m². Reducing the distance between the two ice surfaces can achieve better energy-saving effects, with the most energy-efficient distance being between 10.90 m and 11.30 m (Figure 11b).

As shown in Figure 12a, the mixed area width has a linear relationship with energy consumption. The narrower the mixed area width, the more energy-efficient the ice rink is. Figure 12b further illustrates more energy-efficient schemes for the mixed area width. The more energy-efficient combination is the mixed area width of 3 m along the long axis, 3 m along the short axis for the competition hall, and 1.5 m for the training hall. It is worth noting that while reducing the width of the mixed area decreases refrigeration and HVAC energy consumption, it can lead to higher heating energy consumption.

As shown in Figure 13a, the enclosure mode of auxiliary space has a non-linear relationship with energy consumption, and the variation in energy consumption caused by the mode is relatively small. Figure 13b further shows that the optimal parameter combination is a 0.5 m offset along the long axis and a 5 m offset along the short axis.

By thoroughly analyzing all the experimental results, the energy consumption range was determined by first calculating the difference between the maximum and minimum energy values (V_d) simulated from the six spatial layout parameters and then dividing this difference by the maximum energy consumption value (V_max). The specific calculation process can be found in

Table 3. Calculation of the influence of spatial layout parameters.

Parameter	Vd (kWh/m2)	Vmax (kWh/m2)	Impact (%)
Orientation	0.26	403.03	0.07%
Length-to-width ratio	26.50	399.66	6.63%
Height-to-width ratio	148.18	472.53	31.36%
Ice surface spacing	2.69	412.27	0.65%
Mixed area width	107.80	473.99	22.74%
Enclosure mode of auxiliary space	0.55	403.25	0.14%

.

The parameters were ranked according to their impact on energy consumption: height-to-width ratio (31.358%) > mixed area width (22.743%) > length-to-width ratio (6.630%) > ice surface spacing (0.653%) > enclosure mode of auxiliary space (0.136%) > orientation (0.066%).

3.3. Orthogonal Experiment Results

We obtained 1633 sets of single-parameter simulation results from the above simulation experiments for the six spatial layout parameters. Considering the potential interactions among these six factors, further multi-parameter simulation experiments were conducted. The orthogonal experiment method was used to reduce the large number of experimental groups resulting from the free combination of multiple factors. An L18 (3⁶) orthogonal table was utilized based on the number and levels of the parameters. Table S1 in the Supplementary Materials lists these 18 parameter combinations. Finally, the parameter combination with the lowest energy consumption level is shown in

Table 4. The spatial layout parameter combination with the lowest energy consumption.

Model	Orientation	Length-to-Width Ratio	Height-to-Width Ratio	Ice Surface Spacing	Mixed Area Width	Enclosure Mode of Auxiliary Space
	247°	1.12	0.14	[3 m, 3 m, 3 m]	−1.2m	[−0.5 m, 6 m]

.

In the orthogonal experiment, the coefficient of determination (R²) was 0.986, which validates the effectiveness of the parameter selection. This study further uses the K_avg value to analyze the impact of spatial layout parameter levels on total energy consumption. K_avg represents the average response value of each parameter at different levels, which can identify parameters that significantly affect the response variable and provide a basis for selecting the optimal level combination. Figure 14 shows the parameter effect plot of K_avg, where the X-axis represents the different levels of each parameter, and the Y-axis represents the average value of the response variable, i.e., K_avg. Each data point represents the K_avg at a specific level of a certain factor, and the lines show the trend changes. The steeper the slope of the line, the more significant the impact of that parameter on the response variable at different levels. In other words, if a line in the figure shows significant changes, it indicates that the factor significantly impacts the experimental results.

Figure 14 shows that in the orthogonal experiment, orientation has the most significant impact on the total energy consumption of the skating rink, which was not reflected in the single-parameter experiment results. Additionally, consistent with the single-parameter experiment results, the aspect ratio also significantly impacts total energy consumption.

4. Discussion

4.1. The Spatial Layout Design Strategies

In practical application scenarios, the ranking of energy consumption impact levels from the multi-parameter experiments can be referenced, and the optimal parameter value ranges obtained from the single-parameter experiments can be considered for selection. Finally, after comparing with existing studies, the spatial layout design strategies proposed in this paper are as follows:

4.1.1. Spatial Orientation Results

For ice rinks in cold regions, the outdoor wind environment is an important factor influencing building energy consumption. Different orientations affect the surface area exposed to the dominant wind directions (hot and cold winds) and the effectiveness of natural indoor ventilation. Based on the simulation results mentioned earlier, the recommended optimal energy-efficient orientation for such venues is approximately 244° to 248°, while the less favorable orientation is around 104° to 108°.

The impact of orientation on the refrigeration energy consumption of the training hall is greater than that of the competition hall. This may be due to the lack of seating around the ice surface in the training hall, making it more susceptible to heat exchange with the external environment. Therefore, the focus should be on reducing heat exchange between the training ice rink and the outside environment to enhance energy efficiency. When selecting the orientation, the training hall should avoid the dominant hot wind direction (Figure 15). This ensures sufficient surface area is exposed to the dominant cold wind direction, thereby reducing the energy consumption required for ice-making in the training hall.

4.1.2. Spatial Dimension Results

According to existing research, reducing the shape coefficient is an effective way to lower energy consumption [35,36]. This study further found that when the length-to-width ratio of the ice rink is closer to a square, its energy-saving effect is more significant. Therefore, the length and width difference in the plan layout should be minimized to optimize energy efficiency. Given a fixed total building area for auxiliary spaces, designers can adjust the overall length-to-width ratio by reasonably arranging these spaces. Optimizing the plan layout by dividing and utilizing corner spaces in the competition hall can lead to more efficient energy-saving designs.

The comparative experiments between the height-to-width ratio and the length-to-width ratio of the ice rink show that height significantly impacts energy consumption. Undoubtedly, the lower the height of the ice rink, the more energy-efficient it is in cold regions [36]. At the same time, the lower the position of the air conditioning return vent, the more influential the dehumidification of fresh air [8]. Therefore, reasonable height settings can significantly reduce energy consumption and minimize the risk of fogging on the ice rink. However, lower competition hall spaces may affect the viewing experience. Thus, this study suggests that the design of ice rinks should balance energy efficiency with the audience experience. For example, controlling the average height of the competition hall by reducing the height of non-essential spaces (such as the space above the corners) is advisable. Moreover, the air conditioning return vents can be reasonably placed on both sides of the ceiling to avoid direct airflow on the ice surface while ensuring the audience’s view is unobstructed.

4.1.3. Spatial Organization Results

The experimental results indicate a specific linear relationship between the ice surface spacing and the total energy consumption of the ice rink. The closer the distance between the two ice rinks, the lower the energy consumption, suggesting that the ice surface layout should be as compact as possible. To minimize the distance between the two ice surfaces, a space equivalent to the width of a column grid can be reserved between the competition and training hall of ice rinks to accommodate a set of shared auxiliary facilities, such as an ice resurfacing machine garage, equipment rooms, and a heat exchange station. Shared facilities help reduce the distance between the two ice surfaces and shorten pipeline lengths, thereby reducing energy losses caused by excessively long pipelines. However, existing engineering projects often overlook this energy-saving strategy [32].

Regarding the impact of mixed area width, the results show that the narrower the mixing zone, the lower the energy consumption of the ice rink, especially when the shorter side of the mixing zone is narrower, leading to more significant energy savings. Field studies indicate that ice rinks in cold regions often have more radiators installed on the longer side of the mixing zone. Increasing the width of the long side of the mixed area may cause the heat emitted by the radiators to interfere with the temperature field of the ice surface, thereby increasing the energy consumption for ice-making. Therefore, adding auxiliary rooms in the mixed area can effectively reduce its width and address issues of space waste and temperature imbalance. Meanwhile, it can alleviate the shortage of rooms for athletes.

The results of the enclosure mode of auxiliary space are shown in Figure 16. The experiments suggest that avoiding arranging auxiliary spaces in an ‘L’ shape around the competition hall is advisable. In other words, architects should avoid placing the competition hall in the corner of the ice rink’s layout. Thus, optimizing the relative position of auxiliary rooms and the competition hall is crucial. Architects can flexibly adjust the layout of various functional rooms, such as entrance lobbies and equipment rooms. Related studies have also emphasized the significant impact of spatial organization on sports facilities’ performance [37]. This study further proposes specific spatial organization strategies for ice rinks.

Consistent with the ‘temperature onion’ theory in the field of ecological architecture, the results of this study confirm that auxiliary rooms should be arranged around the competition hall to prevent the ice surface from directly contacting the external environment. It can establish a natural temperature gradient between the cold outdoor air and the comfortable indoor environment, achieving passive energy savings. Existing engineering cases show that when auxiliary spaces are arranged in this manner, the surrounding area can serve as a pre-warm-up activity area for athletes, and it is also common practice to create a circular running track by opening up the space.

Furthermore, the entrance lobby should minimize its contact area with the outside to reduce energy loss and cold air intrusion. A deep-plan design is recommended, and the entrance lobby’s opening should not directly face the dominant wind direction in winter to reduce direct cold air invasion. It is advisable to centrally locate various rooms for equipment rooms to minimize refrigeration capacity waste. If the distance between equipment rooms is too great, it may increase the difficulty of arranging refrigeration pipelines and reduce refrigeration efficiency.

4.2. Limitations and Future Research

This study has certain limitations. The typical ice rink model and spatial layout parameter ranges extracted in the study were based on surveys and statistical results from ice rinks in cold regions. So, the energy-saving parameter values derived only apply to ice rinks with double ice surfaces in cold regions. The conclusions may be inaccurate if applied to other regions or types of ice rinks.

Future research should broaden its scope to enhance the applicability of the findings. On one hand, the study can be extended to different climate zones, such as temperate regions and areas with hot summers and cold winters. On the other hand, the diversity of ice rink energy structures and layouts can be considered, for example, by studying ice rinks with other layout forms (such as single ice rinks). Additionally, convenient simulation tools can be developed in the future to help practitioners quickly assess the energy consumption levels of design schemes during the design process.

5. Conclusions

This study distilled a typical spatial layout model based on extensive survey data. The Delphi method identified key parameters affecting the total energy consumption of ice rinks. By combining single-factor and orthogonal experiments, the study quantified the impact mechanisms between the spatial layout design parameters and total energy consumption, ultimately deriving a spatial layout parameter combination with lower energy consumption. Based on the experimental results, spatial layout design strategies for ice rinks in cold regions were proposed, providing scientific and feasible guidance for designers and promoting the green and sustainable development of ice rinks.

The main contributions of this study are as follows:

• This study breaks away from the traditional qualitative decision-making in the early stages of ice rink spatial layout design by combining qualitative and quantitative experiments. It fills the research gap on the impact of spatial layout design on energy consumption in ice rinks in cold regions. For ice rinks with similar spatial layouts in cold regions, the energy-saving parameter values and their combination patterns obtained from the experiments can directly assist designers in making early-stage decisions.
• The results of the single-factor study show that the spatial layout design parameters that significantly affect the energy consumption of ice rinks in cold regions include the height-to-width ratio, mixed area width, and length-to-width ratio. In contrast, the enclosure mode of auxiliary space and orientation impact ice rink energy consumption less. Adjusting the height-to-width ratio and mixed area width parameters can reduce the annual energy consumption by approximately 18% to 31%. In the orthogonal experiment, considering the interactions between parameters, the orientation has the greatest impact on the ice rink’s total energy consumption.
• The spatial layout design strategies for ice rinks in cold regions include (1) selecting a reasonable orientation, paying particular attention to the impact of hot winds on the training hall, (2) appropriately reducing the height of the ice rink hall, (3) minimizing the length-to-width ratio, (4) evenly distributing auxiliary space around the competition hall, (5) adding auxiliary rooms in the mixed area, and (6) utilizing shared auxiliary facilities to reduce the distance between the two ice surfaces.