A Review of Studies on Heat Transfer in Buildings with Radiant Cooling Systems

Abstract

Due to their benefits in interior thermal comfort, energy saving, and noise reduction, radiant cooling systems have received wide attention. Radiant cooling systems can be viewed as a part of buildings’ maintenance structure and a component of cooling systems, depending on their construction. This article reviews studies on heat exchange in rooms utilizing radiant cooling systems, including research on conduction in radiant system structures, system cooling loads, cooling capacity, heat transfer coefficients of cooling surfaces, buildings’ thermal performance, and radiant system control strategy, with the goal of maximizing the benefits of energy conservation. Few studies have examined how radiant cooling systems interact with the indoor environment; instead, earlier research has focused on the thermal performance of radiant cooling systems themselves. Although several investigations have noted variations between the operating dynamics of radiant systems and conventional air conditioning systems, the cause has not yet been identified and quantified. According to heat transfer theory, the authors suggest that additional research on the performance of radiant systems should consider the thermal properties of inactive surfaces and that buildings’ thermal inertia should be used to coordinate radiant system operation.

1. Introduction

Energy demand for buildings has experience a considerable increase, 4% or so since 2020, the highest rate in the preceding ten years. This growth can be ascribed to the rising use of fossil fuel gases in buildings in emerging nations. Therefore, carbon dioxide emissions have also increased noticeably, around 5% since 2020, and are already 2% higher than the peak recorded in 2019 [1]. The energy consumed by buildings over their full life cycles accounted for 45.5% of the total consumption in 2020, according to the “2022 China Buildings Energy Consumption and Carbon Emissions Research Report” [2]. This suggests that a sizeable amount of total building energy consumption is made up of carbon emissions from buildings. Additionally, a large portion of the total carbon emissions from buildings are caused by air conditioning systems. As a result, to achieve energy-saving objectives, it is required to examine workable control mechanisms to lower carbon emissions and study the energy consumption of radiant heating/cooling systems.

Radiant cooling systems can utilize renewable energy sources to provide appropriate water temperatures for high-temperature cooling. They have attracted considerable attention from numerous researchers in recent years. Since the early 20th century, scholars have been dedicated to exploring the usage of radiant cooling systems in indoor comfort and building energy efficiency and have developed different kinds of radiant cooling systems. Generally, radiant systems are roughly categorized into radiant cooling panels (RCP), water-based embedded surface cooling systems (ESCS), and thermally activated building systems (TABS) [3,4].

High-temperature chilling water is used in hydronic circuits of radiant cooling systems to reduce the inside surface temperature of enclosures. Therefore, for various radiant cooling systems, there is a difference between the heat absorbed by the surfaces and the heat taken away by the hydronic loop (Figure 1). The hydronic cooling rate of an RCP system is the same as the cooling output of surfaces if the insulated layer is mounted on the backside of the panel. For ESCS and TABS, the heat extracted from the cooling surface zone is not equivalent to the heat removed by the hydronic loop since hydronic pipes are integrated into the building structure. Particularly, with a suitable management approach, TABS may use the thermal mass between panel surfaces and hydronic pipes to alter the internal heat environment. Therefore, the operation and control strategy of the system affect the rate of heat extracted by the hydronic loop.

Feng et al. [4] compared the cooling rates between a representative radiant ceiling system and an air system, both of which could create a thermal environment with the same operative temperature. They considered that the cooling rates of the radiant system were different from those of the air system and the peak cooling rates of the radiant system were greater than those of the air system (Figure 1). Furthermore, convective heat transfer is the main pattern of operation in traditional air conditioning systems. While the radiant fraction of the heat gain in the room is absorbed by the surrounding buildings, furniture, and other thermal masses, the convective portion immediately becomes the system’s transient cooling load. Once their surface temperature increases, it is released back into the room through convective heat transfer, becoming the cooling load of traditional air conditioning systems.

For a compound radiant cooling system, like the one illustrated in Figure 2, the cooling surface directly absorbs radiant heat gain from the environment, minimizing the influence of the enclosure’s thermal storage ability on the heat received through radiant heat transfer and maximizing transient spatial heat exchange. However, due to the thermal inertia of the structure of the radiant cooling system, the instantaneous heat exchange on the cooling surfaces, meanwhile, might not be equivalent to the heat carried away by the internal water system. So, the load of the compound radiant cooling system is the sum of the two terminal devices and may not be equal to spatial heat exchange rate.

Furthermore, cooling capacity is crucial for radiant system design and performance analysis. In engineering applications, cooling capacity is equal to the surface heat transfer coefficient multiplied by the difference between surface and indoor air temperatures [5]. In Figure 2, the cooling capacity of the combined system is equal to the sum of the heat absorbed by the cooling surface and the load of the fresh air system, while the cooling load of the radiant system is the sum of the heat carried by the water system and the cooling load borne by the fresh air system.

Radiant cooling systems can not only be considered as a part of construction but can also be treated as terminal equipment of radiant cooling systems. Firstly, the cooling surfaces of radiant cooling systems, as a portion of buildings, enhance the radiant heat exchange between surfaces and rooms, thereby altering the thermal inertia and heat storage/release processes of buildings’ structure. Secondly, as a part of a cooling system, this terminal equipment can actively adjust the temperature and heat exchange of the cooling surfaces by modifying water temperature or flow rate. Furthermore, combining radiant cooling systems and passive buildings to regulate indoor air quality and indoor comfort has been studied by some researchers recently. Building energy consumption and carbon emissions can be reduced by utilizing building thermal storage and release characteristics and by adjusting operation strategies of heating, ventilating, and air conditioning (HVAC) systems. However, current control methodologies for these systems are still under theoretical research [6,7,8]. Therefore, it is particularly significant to study the operation dynamics of radiant cooling systems.

Based on the background of the radiant cooling system mentioned above and the assumption of heat transfer characteristics, in this review, the heat exchange between a water system and a cooling surface is categorized as conduction, while the heat exchange between a cooling surface and other inside faces of enclosures is categorized as spatial heat exchange, including convection and radiation. The thermal storage ability of inactive surfaces and the thermal inertia of cooling surfaces are categorized as thermal properties of the building. To promote the development of radiant cooling technology and explore energy-saving potential, this article reviews the latest research on radiant cooling systems from three aspects: building structure heat transfer characteristics, system load and spatial heat transfer processes, building thermal performance and system control strategies. The flowchart in Figure 3 illustrates the logical structure of this review. The aim of this research is to review heat performance analysis and operation control strategies of radiant cooling systems and identify meaningful issues in conduction heat transfer models, the difference between cooling load and cooling capacity, surface heat transfer coefficients, and the interaction between radiant cooling systems and building thermal inertia. Moreover, based on the review, we put forward some suggestions for further study in the conclusion.

2. Conduction Heat Transfer in Radiant Cooling Systems

There are numerous works of research on heat conduction models of radiant floor, ceiling, and wall systems. According to the hypothesis and description of the heat exchange process in a radiant cooling system, these models are usually divided into three categories: analytical model, numerical model, and simplified model. The following presents some studies conducted by different researchers on these three kinds of physical models.

2.1. Analytical Models

Wang et al. [9] introduced a single-layer homogeneous heat transfer model specifically designed for rooms with internal heat sources. By applying the separation of variables method, they successfully derived an analytical solution for the temperature field distribution. Larsen [10] also proposed an analytical model specifically designed for a slab with a heating/cooling parallel pipe system. They employed a combination of the separation of variables and superposition methods, as well as the superposition method and orthogonal expansion method, to solve the analytical model. The model was successfully applied in practical engineering cases, demonstrating its effectiveness and practicality in engineering. However, the analytical model posed significant challenges in the solving process.

To address this limitation, some researchers have explored an approach known as semi-analytical models. These models combine analytical methods, numerical methods, and heat transfer function methods to establish more comprehensive heat transfer models. By incorporating the strengths of each approach, semi-analytical models strive to overcome the limitations of purely analytical models. Laouadi [11] proposed a more precise model for radiant cooling systems. The model is divided into two parts: the first calculates the structural heat transfer by using a one-dimensional model, while the second part calculates the flow of heat transfer within pipes by using a two-dimensional model. These two parts are then coupled together based on the temperature outside the pipe wall. Notably, the results obtained from the semi-analytical model show good agreement with the results obtained from the two-dimensional numerical model.

2.2. Numerical Model

Although analytical models can provide accurate predictions of surface temperature field distribution and thermal transfer processes in radiant cooling/heating systems, they usually prove to be complex for practical engineering applications. As a result, researchers have sought alternative methods to overcome this limitation. Many have turned to numerical methods, such as the finite difference method (FDM) and the finite element method (FEM), to study the characteristics of radiant cooling/heating systems. These numerical approaches offer a more practical and computationally efficient means of analyzing such systems.

Liu et al. [12] employed a MATLAB program to solve the theoretical model of a lightweight radiant floor that incorporated an aluminum layer. The researchers discussed the heat action of the radiant floor and specifically focused on the influence of parameters such as water temperature, water flow velocity, and aluminum layer thickness on the performance of the system. Sattari and Farhanieh [13] utilized the finite element method to investigate the impact of various design parameters on the heating capability of a system. Specifically, they analyzed parameters like pipe diameter, pipe material, pipe number, and pipe depth to understand their influence. Notably, the researchers highlighted that the types and thickness of cover layers are crucial factors in the system’s design. Su et al. [14] developed a two-dimensional steady-state heat transfer model by using the finite difference method to analyze the thermal exchange process in a ceiling concrete cooling system. Their findings highlighted the impacts of key parameters, such as supply water temperature, pipe spacing, and water flow rate, on the cooling capability of the system. the results also indicated that due to the significant thermal inertia present in the concrete slab, there were minimal fluctuations observed in both the surface and indoor air temperatures. Additionally, the study revealed that any negative effects resulting from changes in water temperature could be mitigated. Strand and Pederson [15,16] proposed a radiant heating/cooling model for Energy-Plus1.0 software by utilizing the transfer function method. This model accurately calculated the heat transfer between fluid and surrounding materials by using the thermal efficiency-transfer unit number method (e-NTU). However, this calculation model only considered convection and neglected the connection between thermal resistance and the intrinsic thermal resistance of pipe materials. According to the ISO 11855 standard, Jin et al. [17] divided a floor structure into two parts and researched floor surface temperature distribution with composite grids using the finite volume method. In addition, they also established a radiant floor conduction model using a composite grid and the finite volume method to study the effects of water flow velocity and pipe heat resistance on floor cooling performance [18]. The results indicated that the pipe material has impacts on the behavior of radiant floors when the thermal conduction of the pipe is low, but the effect of water velocity can be ignored.

2.3. Simplified Model

The formulae of semi-analytical models are complicated for engineers to apply in practical projects. Numerical models such as the finite element method and finite difference method have difficulties in numerical convergence and integration with existing simulation software and are time-consuming. Simplified thermal transfer models are more convenient to calculate than these models. Therefore, Wu et al. [19] calculated the surface temperature and heat transfer of radiant heating/cooling systems by using the conduction shape factor method. The results demonstrated a strong agreement between the calculation and numerical simulation under the condition that the average water temperature of the modeled floor heating system ranged from 25 °C to 45 °C and the average water temperature of the modeled floor cooling system ranged from 10 °C to 20 °C. Lu et al. [20] introduced a comprehensive thermal resistance and heat capacity network (RC) model for radiant floor heating combined with building envelopes in intermittent operation scenarios. On the other hand, Kilkis et al. [21] proposed a simplified fin model to analyze the heat transfer process within radiant cooling structures. They suggested that heat is conducted in one dimension through the pipe’s surface (rib root) and subsequently transferred laterally through the structure, exchanging heat with the indoor environment.

While fin models offer simplification, their applicability is limited. To overcome these limitations, Liu et al. [22] developed composite fin and equivalent heat resistance models, which account for various radiant floor structures. These models provide a more versatile approach for calculating heat transfer in different radiant floor configurations. Li et al. [23] introduced an equivalent heat resistance method, highlighting the significant influence of the pipe space and the floor-covering material on the heat behavior of radiant systems. Additionally, Li et al. [24] proposed a heat transfer model for a three-layer floor that integrates both two-dimensional and one-dimensional heat transfer processes. They also proposed a simplified method to determine the surface temperature of a radiant floor cooling system with a multiple-layer floor structure. Similarly, Zhang et al. [25] employed the equivalent heat resistance method to simplify the heat transfer process from pipes to an indoor zone, focusing specifically on one-dimensional heat conduction. The study concluded that the rate of heat exchange between floor surfaces and the indoor environment is a critical parameter influencing the cooling/heating capacity of a radiant floor system. Furthermore, the thickness and thermal conductivity of the floor layer were found to impact the system’s cooling/heating performance.

The simulation software TRNSYS 18 utilizes a RC (lumped parameter method) model to simulate the heat exchange processes in radiant cooling systems. However, this model has certain limitations regarding the pipe depth [26]. The star-shaped RC-network model, which represents a thermal resistance and thermal capacity network, offers the ability to simulate the thermal dynamic operation of concrete cooling panels and easily integrates with existing building energy simulation software. However, determining the thermal resistance and heat capacity of concrete layers is challenging.

To address this issue, Liu et al. [27] proposed a method that determines these parameters from geometrical perspectives and proposed a corrected RC-network model. Additionally, Ren and Wright [28] also introduced a heat transfer model that involves the interaction between air and the floor by using the lumped parameter method (lumped model). Similarly, Koschenz and Dorer [29] developed a separate RC model to investigate the relationship between heat storage ability and pipes’ geometric shape (1D resistance model). Zhu et al. [30] also developed a theoretical model for structures with embedded pipes and conducted frequency characteristics analysis using the frequency-domain finite difference method (5R2C model). These resistance and capacity models are shown in Figure 4. Similarly, Li et al. [31] divided a concrete radiant floor into the following three parts (Figure 5): cover layer, concrete floor with embedded pipes, and insulation layer. The cover layer and insulation layer were simplified as a one-dimension RC model, which has two resistances and one capacitance. The heat transfer in the slab with embedded pipes was calculated by a coupled RC model, which describes the heat transfer process between the concrete floor and the internal hydronic system. The experiment demonstrated that the errors of the coupled RC model were within 5.5%. The simplified RC network model can predict the dynamic heat performance of a radiant floor with multi-layers.

3. Cooling Load of Radiant Systems and Space Heat Transfer

The cooling load of systems refers to the heat removed from a room to the outside environment and adjacent rooms, which is based on some parameters such as zone climate data, location, usage, and so on [32]. The cooling capacity of a radiant system represents the amount of cooling required at a specific moment to keep a stable room temperature [33]. It is a basic parameter of radiant systems and an important reference for system design and operation.

Cooling load calculation and cooling capacity determination of systems are key steps in sizing radiant systems. When radiant cooling/heating systems stay in a steady state, the cooling load is equal to the cooling capacity. In addition, for a room with an air conditioning system, cooling surfaces alter the heat transfer process in a room, thereby affecting the cooling capacity and the cooling load of the radiant system.

An auxiliary air conditioning system is usually applied for ventilation requirements; these air conditioning systems are also responsible for latent loads [34,35]. Heat gain contains a latent part and a sensible part, and the sensible heat gain can be formed through convection and radiation. The transient cooling load is the cooling rate provided by air conditioning equipment to keep the room temperature steady. Thermal properties of the envelope and types of heat gains determine the relation between heat gain and cooling load, and the cooling load is equal to or lower than the heat gain. In a room with a traditional air conditioning system, latent and convective heat gain can be directly transformed into the transient cooling load, while radiant parts (such as transient solar heat and lighting radiant heat) cannot be immediately turned into the transient cooling load. Moreover, radiant heat is absorbed and conserved by surfaces and then discharged into the air by convection to form a cooling load of air conditioning when the surface temperature rises and is higher than the adjacent air temperature [36]. It is worth noting that in a room with a radiant system, an active surface can directly absorb a part of radiant heat gain through radiation. This could reduce the effect of thermal storage of the envelope on radiant heat gain and strengthen transient space heat exchange. Instantaneous heat exchange on cooling surfaces is not necessarily equal to the heat removed by internal water loops due to structural thermal inertia. The cooling load of a combined system (radiant cooling system + auxiliary air conditioning system) is the sum of heat exchanges in two different terminal circulating water systems, and the value is not necessarily equal to the amount of heat exchange in space.

3.1. Cooling Load of Radiant Systems

Cooling load calculation is an important step for system sizing and equipment selection. Hu and Niu [37] reviewed the applications of radiant heating/cooling systems in China and emphasized that cooling load calculation is the most urgent problem that needs to be solved during the design and application of radiant systems. The current theory and methods for cooling load calculation are primarily based on traditional convective air conditioning systems, which remove heat from rooms through convective heat transfer. However, these methods do not apply to radiant systems, which primarily remove heat through both radiation and convection [38].

Currently, there are several methods for calculating the cooling load of a radiant heating/cooling system, such as the Total Equivalent Temperature Difference Method [39], the radiant time series method [40], and the heat balance method [41]. Mao et al. [42] compared five methods: the Total Equivalent Temperature Difference/Time Averaging (TETD/TA) method, the transfer function method (TFM), the Cooling Load Temperature Difference/Solar Cooling Load/Cooling Load Factor (CLTD/SCL/CLF) method, the heat balance method (HBM), and the radiant time series method (RTSM)) recommended by the ASHRAE handbook for computing cooling load. They concluded that heat balance method gives more accurate results than other methods.

Niu et al. [43] used software called ACCURACY to calculate the cooling load of a radiant slab system and discovered that the direct cooling rate provided by this system increased the peak cooling load. Using the thermal balance method, Ning et al. [44] created a model of cooling load for a radiant ceiling coupled with a fresh air system. Compared to an all-air system, they found that the radiant ceiling system coupled with a fresh air system was able to remove more heat gain when the indoor air temperature was the same, and the peak load of the combined system increased by 16%.

To quantify the thermal transfer of radiant ceiling systems, Ning and Chen [40] introduced the concepts of radiant time series (RTS) and convective time series (CTS) factors and modified them based on the linear superposition principle. They proposed radiation and convection time series methods to calculate the cooling load. However, this method does not apply to systems with embedded pipes. Considering this limitation, Ning et al. [45] proposed another revised radiant time series method (RTSM) for calculating the cooling load. The results demonstrated that this method achieved higher accuracy compared to the traditional heat balance method.

Feng et al. [46] utilized Energy-Plus, an energy simulation software, to investigate the discrepancies of radiant systems and all-air systems in peak cooling load and 24 h cooling energy. The study revealed that the total energy of the modeled radiant cooling system was 2.7–6.5% higher, with a corresponding increase in peak cooling load ranging from 10–40%. Furthermore, Feng et al. [4] observed that the peak load of a modeled radiant system was 7–35% higher when solar heat gain was not taken into consideration. The finding showed that surfaces of radiant systems could reduce the effect of thermal inertia on radiant heat gain; compared to Response Factor Methods, the heat balance method is more precise in calculating dynamic cooling load [36].

Woolley et al. [47] conducted a comparative experiment to investigate the differences in cooling load between radiant cooling systems and all-air systems while maintaining parallel room temperature (operative temperature). The findings revealed that compared with the all-air system, radiant systems are more effective in removing heat. Additionally, the gap in peak cooling load between the two systems becomes deeper as the radiant heat gain in a room increases [48]. In another study, Hu et al. [49] examined the influence of building thermal mass on the cooling loads of radiant systems in different cities. The findings demonstrated that the peak cooling load and accumulative load of a radiant cooling system coupled with a fresh air system were 9% to 11% and 3% to 4% higher, respectively, compared to an all-air system.

3.2. Cooling Capacity of Radiant Systems

Radiant system cooling capacity refers to the capacity determined during the system design phase. Generally, the cooling capacity of a metal cooling ceiling can achieve a cooling capacity of up to 100 W/m², whereas the cooling capacity of a cooling floor only reaches 40 W/m². Olesen [50] concluded that the cooling capacity of a radiant floor system typically ranges from 35 W/m² to 50 W/m². When direct solar radiation falls on the floor, the cooling capacity increases dramatically, ranging from 100 W/m² to 150 W/m².

Zhang et al. [51] increased the cooling capacity by 19% by installing inclined aluminum sheets in a metal cooling ceiling system. Jeong and Mumma [52] concluded that the overall cooling capacity of a radiant ceiling system increased by 5% to 35% under mixed convection conditions. And they created an innovative simplified method to assess the cooling capacity of metal ceiling panels. This validated analytical model was then utilized to evaluate the cooling efficiency for various designs [53]. Additionally, Jeong and Mumma [54] presented a more advanced model to compute the cooling capacity of metal ceiling panels under both natural and mixed convection conditions. When computing the cooling capacity of radiant systems, Andrés-Chicote et al. [55] highlighted the importance of considering both radiation and convection separately, rather than relying solely on the operative temperature. Furthermore, Tian er al. [56] analyzed the properties of a radiant ceiling system without mechanical ventilation by conducting experiments in an office subjected to various conditions. They developed a thermal transfer model and found that the internal surface temperature and surrounding wind speed significantly influence the cooling performance, with wind speed accounting for approximately 75% of the impact. Some researchers have established formulas to compute the cooling capacity of systems, but the variables involved are different. Under a steady-state situation, the heat exchange between active surfaces and spaces approaches the heat removed by circulating water systems, and to calculate the cooling capacity of systems, some equations proposed by different researchers are summarized in

Table 1. Cooling capacity calculation equations.

Author/ Standard	Heat Flux Calculation (Steady-State Situation)	Measurement Conditions
		Dimension W × L × H (m × m × m)	Surface Area (m2)	Length (m)
Acikgoz and Kincay [57]	$q_t=\frac{m_c{c}_p\left(t_{in}-t_{out}\right)}{A_s}$ $\begin{array}{cc}{q}_t& =q_r+q_c\\ & =4.3\left(AUST-t_s\right)+2.7\left(t_a-t_s\right)\end{array}$	1.8 × 1.8 × 2.85	5.1	2.2
Zhang et al. [51]	$q_t=\frac{m_c{c}_p\left(t_{in}-t_{out}\right)}{A_s}$ $\begin{array}{cc}{q}_t& =q_r+q_c\\ & =5.3\left(\mathrm{AUST}-t_s\right)+\frac{A_c}{A_p}\times 3.2\left(t_a-t_s\right)\end{array}$	12 × 6 × 3, 10.5 × 6 × 5	26/72.8/19.5/54.6	-
Andrés Chicote et al. [55]	$q_t=\frac{m_c{c}_p\left(t_{in}-t_{out}\right)}{A_s}$ $\begin{array}{cc}{q}_t& =q_r+q_c\\ & =5.4\left(AUST-t_s\right)+4.2\left(t_a-t_s\right)\end{array}$	3.6 × 3.6 × 3	4.8	2.2
Causone et al. [58]	$q_t=\frac{m_c{c}_p\left(t_{in}-t_{out}\right)}{A_s}-\frac{q_{out}}{A_s}$$\begin{array}{cc}{q}_t& =q_r+q_c\\ & =5.6\left(AUST-t_s\right)+4.4\left(t_a-t_s\right)\end{array}$	4.3 × 2.7 × 2.56	11.6	3.3
Cholewa et al. [59,60]	$\begin{array}{cc}{q}_t& =q_r+q_c\\ & =5.2\left(AUST-t_s\right)+h_c\left(t_a-t_s\right)\end{array}$	1.56 × 1.56 × 2.21	2.4	1.6
	$\begin{array}{cc}{q}_t& =q_r+q_c\\ & =5.2\left(AUST-t_s\right)+0.1\left(t_a-t_s\right)\end{array}$
Olesen et al. [61]	$q_t=\frac{m_c{c}_p\left(t_{in}-t_{out}\right)}{A_s}-\frac{U\left(t_w-t_{down}\right)}{A_s}$ $\begin{array}{cc}{q}_t& =q_r+q_c\\ & =h_r\left(t_{ref}-t_s\right)+h_c\left(t_{ref}-t_s\right)\end{array}$	6 × 4 × 2.8	24	4.8
Acikgoz et al. [62]	$q_t=\frac{m_c{c}_p\left(t_{in}-t_{out}\right)}{A_s}$ $\begin{array}{cc}{q}_t& =q_r+q_c\\ & =h_r\left(AUST-t_s\right)+h_c\left(t_a-t_s\right)\end{array}$	1.8 × 1.8 × 2.25	3.24	-
Zhang et al. [25]	$\begin{array}{cc}{q}_t& =q_r+q_c\\ & =h_r\left(AUST-t_s\right)+h_c\left(t_a-t_s\right)\\ & =h_{tot}\left(t_{op}-t_s\right)\end{array}$ $T_{op}=\frac{h_c{t}_a+h_{r}AUST}{h_c+h_r}$ and $q_t=\frac{\left(t_s-t_m\right)}{R}$ where $\begin{array}{cc}R=& \frac{l}{\pi \delta }\left(\frac{1}{h_c}+\frac{t}{K_p}\right)+\frac{l}{2\pi K_0}\left[\ln \left(\frac{l}{\pi \delta }+{\displaystyle {\displaystyle \sum }_{s=1}^{\infty }}\frac{G\left(s\right)}{s}\right)+\frac{2\pi d_0}{l}\right]\\ & +{{\displaystyle \sum }}_{i=1}^2\frac{d_i}{K_i}\end{array}$	-	-	-
REHVA [3]	$q_t=6\left(t_{op}-t_s\right)$	-	-	-
	$q_t=7\left(t_{op}-t_s\right)$	-	-	-
	$q_t=8\left(t_{op}-t_s\right)$	-	-	-

.

3.3. Surface Heat Transfer Coefficient

The surface heat transfer coefficient is a crucial parameter utilized in cooling/heating load calculation and radiant system design. It is vital in characterizing the thermal behavior of radiant systems [63]. The radiant heat transfer coefficient of an active surface refers to the radiant heat transfer between a cooling surface and other surfaces in the room. The factors affecting this coefficient are heat flux density, reference temperature, view factor, and so on. The average unheated surface temperature (AUST), which is an average temperature weighted by surface area, is used as the reference temperature for calculating the radiant heat transfer coefficient.

Moreover, the convective heat transfer coefficient signifies the heat exchange between a particular surface and the air in a room. It primarily relies on air velocity and air temperature. The air temperature is used as reference temperature for calculating the convective heat transfer coefficient.

The total heat transfer coefficient represents a combined radiation and convection phenomenon. It is not the sum of the radiant and convective heat transfer coefficients since convection and radiation are different physical phenomena and their reference temperatures used in calculations are different. Currently, a reference temperature for calculating this coefficient has not been definitively determined. However, the operative temperature serves as the most appropriate and widely accepted reference temperature [57].

Convective and radiant heat transfer coefficients of active surfaces are vital for investigating the heating/cooling capacity and thermal properties of radiant heating/cooling systems [64]. To compare the heat transfer coefficients of different cooling surface types at different reference temperatures, the heat transfer coefficients of different cooling surface types are summarized in

Table 2. Heat transfer coefficient of different active surfaces.

Author/ Standard	Radiant Surface Type	Heat Transfer Coefficient (W/m2·K)			Reference Temperature
		Total	Convective	Radiant	Total	Convective	Radiant
Acikgoz and Kincay [57]	Wall cooling	8.1	2.7	4.3	$T_{op}$	$T_a$	AUST
Zhang et al. [51]	Ceiling cooling	8.4	3.2	5.3
Andrés Chicote et al. [55]	Ceiling cooling	8.5	4.2	5.4
Causone et al. [58]	Ceiling cooling	13.2	4.4	5.6
Cholewa et al. [59,60]	Ceiling cooling	$9.36 {(T_{op}-T_s)}^{0.1}$	2.04–3.65	5.2
	Floor cooling	5.7	0.1	5.0
Olesen et al. [61]	Floor cooling	7.5	1.0	5.5
Acikgoz et al. [62]	Floor heating	8.5–10.5	2.3–4.1	5.2–5.9	$T_{op1.4}$	$T_{a1.1}$ $T_{a1.4}$
Zhang et al. [25]	Floor cooling	-	-	-	$T_{op}$	$T_a$
REHVA [3]	Ceiling cooling	11	-	5.5	$T_{op}$	-	-
	Floor cooling	7	-	5.5		-	-
	Wall cooling	8	-	5.5		-	-

Note: Assume the active surface radiant heat transfer coefficient at 15–35 °C is 5.5 W/m²·K.

. The radiant heat transfer coefficient of radiant ceilings is always higher than that of radiant floors. Furthermore, radiant heat transfer coefficients of active surfaces, be it on the ceiling or floor, generally average around 5.5 W/m²·K.

The total transfer coefficient serves as a significant parameter that quantifies both radiant and convective phenomena simultaneously. Presently, some standards and guidelines adopt this comprehensive heat transfer coefficient as a reference for selecting radiant systems. However, due to variations in reference temperatures and heat flow calculations across different research studies, the surface heat transfer coefficient does not accurately reveal the heat exchange process between active surfaces and their surrounding atmosphere. Typically, the operative temperature is used as the reference temperature to calculate the total heat transfer coefficient. This value is obtained by referencing the Newton cooling formula, while the surface heat flow is measured directly. In terms of calculating the radiant heat transfer coefficient for active surfaces, the reference temperatures usually involve the mean radiant temperature. This temperature represents the average temperature of non-active surfaces, weighted by their respective areas. The radiant heat flux density can be determined using the Stefan–Boltzmann law. The convective heat transfer coefficient of active surfaces can be calculated based on the air temperature in the zone and can be calculated using various methods such as those employed by Khalifa [65,66] and Awbi and Hatton [67,68]. The convective heat flow of surfaces is obtained by subtracting the radiative heat flux from the total heat flow.

Koca et al. [69] investigated radiant cooling walls in residential buildings by experimental study and advised that the total and radiant transfer coefficients of active surfaces should be 7.83 W/m²·K and 5.14 W/m²·K, respectively. Cholewa et al. [59] considered that the height and position of the reference temperature should be involved when computing surface heat transfer coefficients in the future. Mustakallio et al. [70] contrasted the effects of internal heat sources and heated surfaces on the heat transfer coefficient and concluded that the ratio of radiant and convective heat transfer from heat sources significantly impacts the cooling capacity of systems. Karadag [71] observed the influences of different room dimensions and heat conditions on the heat transfer coefficient and proposed an equation of the total heat transfer coefficient that includes surface emissivity. Additionally, Karadag [72] also studied the relationship between the convective and radiant heat transfer coefficients of cooling ceilings for different room sizes and temperatures. Zhang et al. [73] examined the changes in indoor parameters during the start-up process of ceiling radiant cooling panels in experiments. The results demonstrated that the surface temperature rapidly decreased when the active surface was activated, while the temperature of other inner surfaces and the indoor air maintained the same trend. The proportion of heat radiation in the total heat transfer of the ceiling decreases with the stability of the indoor thermal environment. Hu et al. [74] surveyed actual systems and discovered that the total heat exchange increased by 48% when the radiant cooling ratio changed from 60% to 74%. This indicates that surface heat transfer coefficients also vary with changes in the radiant cooling ratio.

4. Building Thermal Performance and System Regulation

Implementation of active and passive energy strategies and technologies helps to improve building energy efficiency. Active control strategies include ameliorations made to HVAC or electrical lighting systems, whereas passive control strategies focus on enhancing the building envelope structure [75].

Recently, passive control strategies have garnered significant attention in research and development. The internal environment can be improved by using thermal inertia of building materials, achieving the aim of shifting peak loads and tackling energy crises and environmental pollution. With passive strategies, buildings can reach higher energy efficiency and create a more sustainable built environment.

Passive energy-saving strategies leverage the concept of thermal inertia, with buildings with immense thermal mass in its envelope exhibiting a reduced and delayed response to sudden temperature fluctuations. Thermal inertia is an intricate phenomenon, which is affected by building use, climate, and other factors [1,76]. By embedding thermal mass materials into buildings’ maintenance structure, the building can effectively store and release thermal energy, helping to keep room temperatures stable and decreasing the demand for active heating or cooling. This approach is particularly beneficial in climates with significant temperature variations, as it helps to smooth out temperature fluctuations and improve occupant comfort while reducing energy consumption.

4.1. Thermal Inertia Parameters

The thermal storage ability of a building pertains to the capacity of building materials to absorb heat when there are fluctuations in temperature or heat flux on surfaces. Currently, research on the thermal inertia of envelope structures relies on two main methods: the harmonic analysis method [77] and the transfer function method [78]. Zhao and Dang [79] conducted a comprehensive comparison between the harmonic analysis and transfer function methods. The results are presented in

Table 3. Comparative study of some parameters for the heat storage performance of building envelope structures [79].

Harmonic Analysis Method	Transfer Function Method	Relationship between Parameters
Heat storage coefficient $\mathrm{S}=\frac{A_q}{A_t}=\sqrt{\frac{2\pi }{T}\lambda c\rho }$	Heat permeability coefficient $b=\sqrt{\lambda c\rho }$	$\mathrm{S}=b\sqrt{\frac{2\pi }{T}}$
Thermal inertia index $\mathrm{D}=\mathrm{RS}=d\sqrt{\frac{2\pi }{T}\frac{\rho c}{\lambda }}$	Heat–thickness ratio $\xi =\frac{d}{{\delta }_e}=d\sqrt{\frac{\pi }{T}\frac{\rho c}{\lambda }}$	$\mathrm{D}=\sqrt{2}\in$
Intense fluctuation of layer thickness ${\delta }_t=\frac{\lambda }{S}=\sqrt{\frac{T}{2\pi }\frac{\lambda }{\rho c}}$	Periodic thermal penetration depth ${\delta }_e=\sqrt{\frac{T}{\pi }\frac{\lambda }{\rho c}}$	${\delta }_e=\sqrt{2}\delta$
Heat absorption coefficient $\mathrm{B}=\frac{A_q}{A_{ti}}=\frac{1}{\frac{1}{Y_i}+\frac{1}{{\alpha }_i}}$	Thermal admittance $Y_{11}=\frac{Aq}{A_{ti}}=-\frac{Z_{11}}{Z_{12}}$	$\mathrm{B}=Y_{11}$
Attenuation multiplier $v=0.92e^{\frac{\sum \mathrm{D}}{\sqrt{2}}}\cdot \frac{S_1+{\alpha }_i}{S_1+Y_1}\cdot \frac{S_2+Y_1}{S_2+Y_2}\cdot \frac{S_n+Y_{n-1}}{S_n+Y_n}\cdot \frac{Y_n+{\alpha }_e}{{\alpha }_e}$	Periodic heat transfer coefficient $Y_{12}=-\frac{1}{Z_{12}}$	$v=\frac{{\alpha }_i}{Y_{12}}$
Delay time ${\in }_0 = \frac{1}{15}\left(40.5D-arctan\frac{{\alpha }_i}{{\alpha }_i+\sqrt{2}{Y}_i}+arctan\frac{Y_e}{Y_e+\sqrt{2{\alpha }_e}}\right)$	Delay time $\Delta t=\frac{T}{2\pi }arctan{Z}_{12}$	${\in }_0 = \Delta t$

.

These comparative findings provide valuable insights into the strengths and limitations of each method, aiding researchers in selecting the most appropriate approach for studying thermal inertia in building envelope structures. In addition to the harmonic analysis method and transfer function method, various parameters have been proposed by researchers to quantify the thermal inertia of building envelopes. Table 4 provides a summary of four such parameters. For instance, in the book Energy Simulation in Building Design [80], JA Clarke emphasized the need to accurately quantify the influence of thermal inertia on buildings by solving a temporal differential heat transfer algebraic equation. Asan [81] demonstrated that thickness and material category have an impact on time delay and the decrement coefficient.

Roucoult et al. [82] introduced a method for describing thermal inertia performance based on a three-criteria calculation. Jaunzems and Veidenbergs [83] investigated the impact of thermal inertia and thermal mass on building envelopes from two perspectives: reduction factors and time lags. They employed the concept of equivalent thermal mass area as a parameter to quantify thermal inertia. Al-Sanea et al. [84] used wall dynamic thermal resistance to quantify thermal inertia and determined optimal insulation layer thicknesses of buildings under stable conditions based on the climate of Riyadh, Saudi Arabia. Further, Al-Sanea et al. [85] studied the impact of thermal mass on some parameters such as the wall transient load and reduction factors by studying walls with similar resistance values R. They proposed concepts of thermal mass saving potential and crucial thermal mass to determine the thickness required for the ideal energy conservation of thermal mass layers.

Tsilingiris [86] studied the impacts of thermal capacity combined with spatial distribution and thermal resistance on wall transient thermal behavior. They compared this physical system with an ideal wall using a lumped parameter model and found that the thermal time constant of a physical system is related to the effective heat capacity of the wall. The effective heat capacity ratio is defined as the ratio of the effective heat capacity involved in the transient heat process to the concentrated heat capacity. An effective heat capacity ratio is used to quantify the thermal inertia of envelope structures.

Table 4. Other parameters to quality thermal inertia.

Author	Parameter Name	Equations
Dougal Drysdale [87]	Thermal inertia	$\sqrt{k\rho c} \left(\frac{J}{m^{2}K\sqrt{s}}\right)$ or $k\rho c\left(\frac{W^{2}s}{m^4{K}^2}\right)$
Ng et al. [88]	Time lag	-
Kolokotroni [89]	Decrement factor	$f=\frac{A_i}{A_e}=\frac{T_{i\left(max\right)}-T_{i\left(ave\right)}}{T_{e\left(max\right)}-T_{e\left(ave\right)}}$
Butcher et al. [90]	Thermal response factor	$f_r=\frac{\sum AY+C_v}{\sum AU+C_v}\left(C_v=\frac{1}{3}NV\right)$

Table 4. Other parameters to quality thermal inertia.

Author	Parameter Name	Equations
Dougal Drysdale [87]	Thermal inertia	$\sqrt{k\rho c} \left(\frac{J}{m^{2}K\sqrt{s}}\right)$ or $k\rho c\left(\frac{W^{2}s}{m^4{K}^2}\right)$
Ng et al. [88]	Time lag	-
Kolokotroni [89]	Decrement factor	$f=\frac{A_i}{A_e}=\frac{T_{i\left(max\right)}-T_{i\left(ave\right)}}{T_{e\left(max\right)}-T_{e\left(ave\right)}}$
Butcher et al. [90]	Thermal response factor	$f_r=\frac{\sum AY+C_v}{\sum AU+C_v}\left(C_v=\frac{1}{3}NV\right)$

4.2. Thermal Inertia of Structure with Phase Change Materials

Gil et al. [91] divided heat storage media into three categories and each of them has its own advantages and disadvantages. Sensible heat energy storage depends on the temperature gradient within the medium (solid or liquid) to store or release energy. Latent heat energy storage utilizes PCMs to store heat energy based on thermal characteristics. Thermochemical energy storage refers to using chemically reversible reactions to store or release energy. Sensible thermal energy storage and latent thermal energy storage can be applied in passive building systems, while thermochemical storage is often used in active building systems [92,93].

Passive buildings incorporate phase change materials (PCMs) into building units such as floors, walls, and ceilings. By employing the latent heat storage principle, PCMs enhance the thermal storage capacity of buildings, effectively reducing peak cooling loads when the thermal storage capacity is utilized optimally [94]. Sun et al. [95] conducted a study on the impact of six parameters (outdoor and indoor air temperature, insulation level, thickness of PCM, location of PCM within the wall, and transient temperature of the phase change material) on the thermal response of walls. They concluded that the heat transmission can be reduced by over 50% when the PCM is appropriately positioned. Pandey et al. [96] reviewed the current applications of PCM in solar energy systems. The results showed that PCMs have been implemented in various solar energy systems, but their application remains limited on account of some irresistible factors, so they are not yet widely available commercially.

The thermal mass of building structures typically exhibits low thermal storage capacity, which limits demand-side management. To address this issue, Koschenz and Lehmann [97] developed a thermally activated ceiling panel suitable for lightweight and retrofitted buildings. This panel can utilize renewable energy to provide heating and cooling for office and industrial buildings. In addition, Lin et al. [98] investigated a novel electrical floor heating system that incorporates a shape-stabilized PCM. This system is capable of providing space heating throughout the day and can be controlled conventionally. However, it may not be suitable for office buildings where nighttime space heating is not necessary. To overcome the limitations of passive under-floor electric heating systems, Lin et al. [99] conducted further research and proposed an alternative electrical floor heating system that incorporates a ductless air supply and shape-stabilized PCM for thermal storage.

Although TABS have a significant amount of thermal mass, traditional active systems are difficult to install. Based on this consideration, Jobli et al. [100] established an experimental model and simplified simulation model to study a new system in which capillary tubes are integrated with a phase change material (CT-PCM). The results demonstrated that the simplified model agreed well with experimental results when the flow rate of thermal fluids in the CT-PCM system was greater than 800 mL/min.

Weinlader et al. [101] monitored two different PCM ceiling cooling systems in two office rooms in Germany. One system had the PCM placed upon a water pipe, while the other had the PCM placed below the water pipe. The findings indicated that the latter showed better heat conduction between the PCM and the room. Moreover, Weinlader et al. [102] analyzed and monitored the reuse behavior of two diverse cooling ceilings with PCM, utilizing two different methods. One method involved analyzing the measured PCM temperature, while the other focused on analyzing the measured heat flux passing through the PCM module. The investigation found that, for the cooling ceiling system in which the PCM was placed above the water pipe, only 5% of the study period recorded completely liquid PCM. In comparison, for the cooling ceiling system wherein the PCM was placed below the water pipe, over 60% of the investigation days experienced complete liquefaction of the PCM. Bogatu et al. [103] studied operational characterization of a new PCM plate, integrated with pipes, used as an active cooling ceiling. The results demonstrated that this system could serve as a thermally active building component that can achieve the purposes of building refurbishment.

Based on the literature mentioned above, although research has focused on examining the performance of radiant ceiling panels integrating PCM, few have delved into the energy flexibility benefits offered by such solutions. To address this gap, Gallardo and Berardi [104] evaluated the volume and resulting internal heat atmosphere. The results showed that the RCP-PCM system can transfer the cooling loads to non-peak hours while maintaining a comfortable atmosphere during working hours from 8:00 to 18:00; the mean daytime cooling capacity was 17.3 W/m² and the absorbed heat was from 180 Wh/m² to 230 Wh/m².

4.3. Optimization of HVAC Control Strategy by Using Thermal Inertia

In a humid and hot climate, to ensure indoor hygiene and comfortable conditions, it is necessary to prevent condensation on radiant surfaces. Generally, this is achieved by controlling the temperature of chilled water within radiant cooling systems to ensure that its surface temperature is higher than the dew point temperature of the adjacent air. Furthermore, due to the excessive thermal mass of radiant systems and long response times, some excellent control strategies, for example, predictive control strategies, are crucial in controlling these systems [105].

For commercial buildings, it is a valuable control strategy to shift the cooling load during the demand response (DR) by using the thermal inertia of the buildings themselves. However, previous research on DR in commercial buildings has predominantly emphasized the thermal inertia of the buildings, leading to an underestimation of the true DR ability. Recognizing this limitation, Li et al. [106] established a coupled model for building thermal inertia and air conditioning systems and validated the model through experiments. They found that the cooling demand for air conditioning systems has a considerable reduction.

The usage of thermal inertia can decrease the investment in radiant heating systems and transfer peak heating or cooling loads to off-peak hours through appropriate control strategies. Edna Shavi et al. [107] simulated summer night ventilation in four regions in a thermal and moist climate in Israel. They studied the thermal environment under different conditions, i.e., different structures with various thermal masses (light, medium–light, medium–heavy, and heavy), and different ventilation strategies (no night ventilation, night natural ventilation, night mechanical ventilation, and mechanical night ventilation with different number of air changes). The findings demonstrated that the building’s thermal storage properties were linear with maximum and minimum temperature differences between the inside and outside rooms. Thus, they believed that the application of the thermal inertia of different materials and ventilation could reduce zone temperature by 3–6 °C.

Cui et al. [108] utilized TRNSYS to analyze the utilization potential of different radiant floor cooling systems and different control strategies in different climate zones in China, and the results revealed that the energy saving reached 25–66% by directly using underground cooling and continuously operating radiant systems. Hu et al. [109] studied interim running peculiarities of a radiant cooling system and proposed a standby cooling operation strategy: the radiant system turns on when the indoor dry global temperature is close to or greater than 30 °C and internal dew point temperature is below 18 °C. The simulation results showed that this operation strategy can meet indoor heat and moisture conditions.

Chen et al. [110] investigate the demand response potential and characteristics of internal thermal masses by experimentation. They found that short-term (0.5 h) and medium-term (2 h) demand response (DR) can ensure indoor thermal comfort levels within an appropriate range. Additionally, they emphasized that the combination of passive heat masses and initiative energy storage technologies can achieve better demand management outcomes.

These studies highlight the potential energy savings and controllable indoor environments that can be achieved through the implementation of different radiant cooling systems and optimal control strategies. By leveraging the advantages of radiant systems, such as direct cooling from underground sources and smart intermittent operation, significant energy efficiency improvements and enhanced thermal comfort can be realized in various climate zones.

5. Discussion

Based on the thermal transmit process of radiant cooling systems, heat exchange between the cooling water and radiant panel surfaces of radiant cooling systems is considered as conduction. Furthermore, the definition of cooling load and cooling capacity are also determined in this review. The cooling load represents the amount of heat removed by the hydronic system per unit time, while the cooling capacity refers to the heat absorbed by cooling surfaces per unit time. Although both values contribute to maintaining indoor comfort levels, they are not necessarily the same due to the heat mass surrounding the water loop.

This review summarized different works of research on conduction heat transfer models, heat exchange between radiant cooling surfaces and zones in rooms, and the coupling between radiative cooling systems and buildings’ thermal inertia. Moreover, some potential future research directions are also highlighted in this review: heat transfer coefficients of non-active surfaces, the impact of thermal mass on the cooling capacity and cooling load of systems, defining the optimal parameters of building thermal inertia, and methods of using heat masses to coordinate system operation.

Unlike traditional air conditioning systems, radiant cooling systems are capable of directly absorbing the radiant heat in a room. This leads to enhanced heat transfer and reduces the impact of the thermal mass of inactive surfaces on radiant heat transfer. Therefore, thermal mass or thermal inertia is crucial in computing cooling loads and selecting proper radiant cooling systems [49]. Additionally, building insulation levels have been improving by recent regulations for building energy efficiency. However, further increasing the thickness of the insulation layer does not necessarily result in additional economic or environmental benefits. Consequently, the influence of thermal mass in indoor environments and its potential for energy conservation have garnered wide attention from researchers and engineers [85,111]. Many believe that using the effects of thermal mass can decrease energy consumption for cooling while maintaining indoor comfort. Therefore, it is necessary to review studies on the heat transfer process of radiant systems within rooms and consider how to effectively utilize thermal mass and develop system operation strategies to achieve energy efficiency.

Currently, heat transfer coefficients of cooling surfaces in rooms with radiant systems have been determined by many researchers through simulations or experiments. However, the heat transfer coefficients of inactive surfaces in these rooms have not attracted as much attention. In rooms with radiant systems, a portion of the radiant heat gain is directly absorbed by the cooling surfaces, and radiation occurs as long as a temperature difference exists between an inactive surface and the cooling surface. As a result, compared to environments dominated by convective air conditioning systems, cooling surfaces can increase heat transfer to inactive surfaces and even enhance heat transmission through external walls. Therefore, the surface heat transfer coefficient of inactive surfaces should increase, but the specific values have not yet been determined.

At present, the heat storage and release characteristics of building envelope structures have not been clearly defined. Specifically, the relationship between the amount of heat storage or release and the thermal mass for various operation strategies has not been defined. Understanding this relationship is important for system design and building construction, as it can help reduce cooling consumption by effectively utilizing thermal mass and improving air conditioning system strategies.

Moreover, Hu et al. [112] emphasized the necessity of combining artificial environment technologies with building passive technologies to reduce energy consumption and carbon emissions. Passive technologies should be employed to extend the duration without cooling or heating, while active systems can be utilized in extreme weather conditions. Many researchers believe that radiant cooling systems offer advantages in low-grade energy utilization and environmental sustainability, and further comprehensive studies on their operational dynamics are necessary.

Overall, the integration of passive and active technologies, along with a deeper understanding of the heat transfer processes and characteristics of radiant cooling systems, can contribute to energy efficiency and environmental protection.

6. Conclusions

Considering heat transfer theory, this article reviews heat transfer in buildings with radiant cooling systems, including conduction in the structure of radiant systems, system cooling load and cooling capacity, surface heat transfer coefficients, building thermal performance, and system performance. The aim of this review is to provide scientific support for optimizing building thermal performance and radiant cooling system design and for promoting passive cooling buildings.

Numerous researchers have established physical models to obtain the temperature field distribution on the cooling surfaces of radiant cooling systems. These models can be broadly categorized into three groups based on the assumptions and descriptions of the heat transfer mechanism in the associated radiant cooling system: analytical models, numerical models, and simplified models.

Analytical models can accurately calculate the analytical solution of the temperature field distribution. However, they are difficult for designers to understand due to their complex derivation process and are not suitable for practical engineering applications. Numerical models, such as the finite element method, finite difference method, and finite volume method, have their drawbacks including difficulty in convergence, difficulty in integrating with existing simulation software, and their being time-consuming. To overcome these shortcomings, some researchers have proposed simplified models.

Radiant cooling load and cooling capacity are vital factors for sizing a system. For radiant cooling systems, the radiant heat gain can be directly absorbed by cooling surfaces. Due to the thermal mass surrounding a hydronic system, the instantaneous heat transfer on cooling surfaces is not equal to the heat removed by internal water systems. The presence of a cooling surface also affects the dynamic heat gain in the room, and researchers generally use the surface heat transfer coefficient to quantify a room’s heat transfer.

Furthermore, the system peak cooling load can be shifted by making use of the thermal inertia of building structures and optimizing system operation to reduce carbon emissions and energy savings. However, there is currently no unified characterization parameter for the structural heat storage performance of buildings. Some parameters are based on indoor and outdoor air temperature changes, without considering the impact of radiant heat exchange on the indoor environment.

Based on the literature review, some issues should be addressed in future studies, including the heat transfer coefficient of inactive surfaces, the effect of thermal mass in system cooling load or cooling capacity, and the definition of thermal inertia for buildings with radiant cooling systems. Moreover, the method of using thermal mass to coordinate system operation should be studied for energy conservation and carbon emission reduction.