Research Progress on the Performance Enhancement Technology of Ice-on-Coil Energy Storage

Abstract

Ice-on-coil energy storage technology has been widely used in air conditioning systems and industrial refrigeration as an efficient energy storage technology. This paper reviews the research progress of ice-on-coil energy storage technology, including its working principle, system design, key parameter optimization, and practical application challenges and solutions. Three kinds of ice melting systems are introduced. The internal ice melting system has the largest cold storage density and the slowest rate of ice melting. The external ice melting system has the lowest cold storage density and the fastest rate of ice melting. The combined ice melting system can have the highest density of cold storage density and a high rate of ice melting. By comparing the results of different studies, the influence of fin and thin ring application on the heat transfer enhancements of the ice-on-coil storage system is summarized. It is found that the ice storage time can be reduced by 21% and 34% when the annular fin and thin ring are set. Regarding system control, adopting the ice-melting priority strategy increases operating energy consumption, but the economy improves; using the unit priority strategy lowers operating energy consumption, but the economy suffers slightly. When the cooling demand exceeds the cooling capacity of the chiller, an ice melting priority control strategy is more economical. Some suggestions for future research are presented, such as optimizing the shape and arrangement of coil fins and ice storage systems integrated with renewable energy. It provides guidance for the further development of ice storage air conditioning technology.

1. Introduction

Energy storage technology is the key to solving the imbalance between energy supply and demand. It not only helps power grid peak regulation but also enhances the proportion of renewable energy consumption. As a crucial support for achieving the “dual carbon” goals, it has great significance for building a new power system [1,2,3]. Nowadays, the energy industry remains heavily reliant on coal and other fossil fuels as primary sources of energy. The demand for energy in the contemporary world is increasing, and the cost of fossil fuels is rising. Renewable energy sources offer the benefits of minimal environmental impact and sustainable utilization. This contributes to reducing global carbon emissions [4,5,6]. However, the volatility and uncertainty of renewable energy generation, influenced by weather conditions, pose significant challenges to grid integration and lead to frequent curtailment of wind and solar energy phenomena [7,8]. Given these challenges, relying solely on generation and grid-side regulation is insufficient to achieve high penetration of new energy into the grid [9]. There is an urgent need to develop new energy storage technologies on the user side. This will not only break through the current dilemma but also pave the way for a comprehensive energy system that integrates generation, transmission, load, and storage, and leverages the complementary use of multiple energy sources. In this context, user-side energy storage technology emerges as a key driver for the development of integrated energy systems [10,11].

In recent years, the global energy demand has surged dramatically, with an average annual growth rate of 2.4% [12]. Notably, the energy consumption of residential and office buildings accounts for about 40.0% [13,14], with the majority of this energy being consumed by air conditioning systems for indoor temperature regulation [15]. Given this significant energy footprint, the application of energy storage technology in air conditioning systems is of paramount importance for optimizing building energy supply [16,17]. As illustrated in Figure 1, integrating cold storage technology with air conditioning systems offers a strategic solution. During the valley period of the power grid or when there is an excess of wind and solar resources during the day, electricity can be used for cooling, and the generated cooling energy can be stored in the form of low-temperature water or ice [18]. When users need cooling, the stored cold energy can be directly supplied. This approach can effectively regulate the power load, increase the consumption of renewable energy, and reduce the dependence on traditional energy sources.

Figure 1. Basic schematic diagram of ice storage air conditioning system.

Cold storage technology can be divided into two types based on different storage methods, namely sensible heat storage and latent heat storage [19,20]. Sensible heat storage is the storage and release of heat without chemical changes. It relies on the thermophysical properties of the heat storage material to store and release heat. In this process, only the material’s temperature changes. Latent heat energy storage, also known as phase-change energy storage, mainly utilizes heat absorption or exothermic behavior during the phase change of materials to store or release thermal energy. It typically has a relatively high heat storage density and a small temperature change. Its storage capacity is 5–14 times higher than that of sensible heat storage [21], which is a widely considered energy storage technology at present.

Table 1 [22,23,24] lists the current typical cold storage methods and their technical features. Compared with water, eutectic salt [25], and gas hydrate storage [26], ice storage emerges as a superior option. It boasts several significant advantages such as higher energy storage density, better heat transfer performance, and a lower temperature of cold release. Moreover, ice storage facilitates the storage and transportation of cold and heat energy at an almost constant temperature [27,28]. While water storage offers better heat exchange performance and lower operational costs, its energy density is lower than that of ice storage. On the other hand, eutectic salt storage has higher operational costs and lower energy storage density, resulting in lower cost-effectiveness compared to ice storage. Additionally, although gas hydrate refrigeration boasts higher density and system efficiency, its operational and maintenance costs exceed those of ice refrigeration. These factors have contributed to the increasing attention that ice storage technology has garnered in recent years.

Table 1. Comparison of the characteristics of different cold storage technologies.

Type	Technical Characteristics	Energy Storage Density (kJ/kg)	Evaporating Temperature (°C)	System Efficiency (%)	Cooling Temperature (°C)	Cold Storage Medium	Heat Transfer Performance	Operation and Maintenance Costs
Water storage	Water storage	20.9	0.0	90.0	5.0–13.0	Water	relatively good	Low
Eutectic salt	Eutectic salt	96.0	0.0	90.0	9.0–10.0	Eutectic salt	general	High
Ice storage	Coil	384.0	−10.0	70.0	1.0–3.0	ice	good	High
	ice ball	350.0	−10.0	70.0	1.0–3.0	ice	general	High
	slushy	208.0	−3.0	85.0	1.0–3.0	slushy	good	High
Gas hydrate	Gas hydrate	302.0–464.0	0.0–3.0	90.0–95.0	6.0–3.0	hydrate	good	High

There are two main types of ice storage technologies, depending on the method of ice production. These are, namely, coil-type ice storage technology for static ice production [29] and ice slurry technology for dynamic ice production [30]. Although ice slurry cold storage technology is more advanced than coil ice storage technology, it requires higher control precision and involves more complex control mechanisms. The stability and reliability of the ice slurry ice storage system mostly depend on the self-control of the system’s capabilities. Table 2 lists several typical ice storage cases where ice-on-coil storage technology has been commercially applied on a large scale.

Table 2. Some ice storage cases.

Ice Storage Cases	RT	Type	Refrigeration Machine	Cold Storage Technology
Marina Bay Cooling Project, Singapore	26,000	Office building	16 refrigeration units	Ice-on-coil
Qatar Pearl Tower	30,000	Shopping centers	52 refrigeration units	Ice-on-coil
Cooling for Business Bay Administration Building, UAE	32,500	High-rise office Building	16 refrigeration units	Ice-on-coil
Texas Medical Center Cooling System	120,000	Buildings	Absorption and electric chillers	Ice-on-coil

The review paper is structured as follows. First, we construct and analyze schematic diagrams to explore the operational modes and technical characteristics of internal ice melting, external ice melting, and combined internal and external ice melting. Then, we delve into the enhanced heat transfer mechanisms of ice-on-coil energy storage and summarize the principles of enhanced heat transfer under typical structural parameters, such as fins and thin rings. Adding fins and thin rings to the outside of the ice-on-coil energy storage not only enlarges the heat transfer area and improves the speed of ice formation but also ensures uniform ice thickness exterior of the tube during the ice storage process. Finally, we summarize the load distribution characteristics of three distinct operational modes, namely full capacity ice storage, unit priority cooling, and ice melting priority cooling in different periods. Additionally, in terms of operating strategies for ice storage systems, we compare and summarize the application scope and optimization conclusions of various strategies through diverse technical approaches, including mathematical modeling, software simulation, and model optimization.

At present, there are few studies on the integration of ice-on-coil energy storage with air conditioning technology, and the system analysis is not deep and specific enough. This paper provides a comprehensive overview of the current status and future directions of key aspects related to ice-on-coil energy storage technology. It analyzes and highlights the primary main problems and bottlenecks encountered when applying ice storage technology to air conditioning systems and provides ideas and solutions for optimizing ice storage air conditioning technology. The future research direction, including the development of new high-efficiency heat transfer materials and intelligent control systems, is proposed, which provides guidance for further research in this field. The purpose of this paper is to provide theoretical reference for the further research and development of ice coil cold storage technology, help realize the goal of “double carbon”, and serve the power grid peak regulation.

2. Research on Ice-on-Coil Type Ice Melting Methods

2.1. Ice-on-Coil Type Ice Melting Method

Ice-on-coil storage systems are classified into two types based on their ice-melting methods, namely internal ice melting and external ice melting [31]. The internal ice melting system is completely frozen, while the external ice melting system is incompletely frozen. The completely frozen system has a considerable amount of cold storage capacity, but the cooling temperature fluctuates significantly. Conversely, the incompletely frozen system has a smaller storage capacity, yet it maintains a more stable cooling temperature and achieves a faster cooling rate.

2.1.1. Internal Ice Melting Method

When the internal ice melting method is used for cooling, the high-temperature heat transfer fluid, which has been heated by the air conditioning load, circulates through the coil. The ice on the outside of the tube melts from the inside out to provide the cooling load [32]. Figure 2 illustrates the operational mechanism of the internal ice melting system. The system comprises a base load refrigerating machine, a dual-working-condition refrigerating machine, a heat exchanger, a coil, an ice storage tank, a cooling tower, and other pumps and valves [33,34].

Figure 2. Schematic diagram of a typical internal ice melting system.

During nighttime ice storage, the dual-working-condition refrigerating machine operates in ice-making mode. The valves V3 and V7 on the ethylene glycol circuit side are closed, while V1 and V4 are open. Cooled by the dual-working-condition refrigerating machine, the ethylene glycol solution flows into the coil. The temperature of the supply water is maintained at 10.5 °C, whereas the return water temperature is kept at 3.5 °C. It indirectly exchanges heat with the cold water in the ice storage tank (10.0–15.0 °C) to freeze the water completely or partially.

During daytime ice melting, the water in the ice storage tank remains static. On the glycol circuit side, valve V3 is closed, while valves V1, V4, and V7 are open. The dual-working-condition refrigerating machine operates in ice-melting mode. The ethylene glycol pump drives the solution through the coil. The ice in the ice storage tank gradually melts with the continuous flow of glycol solution. In this process, the latent heat of water fusion is released and transmitted to the user side through the heat exchanger, which provides heat energy to the chilled water supply system. This maintains the supply water temperature at 5.0 °C and the return water temperature at 15.0 °C.

The ice storage capacity of the internal melting system is large, but its cooling rate is relatively slow. This limitation stems from its reliance on the forced flow of glycol solution within the coil to indirectly exchange heat with the ice and water outside the tube. The heat transfer on the outside of the tube depends on the thermal conductivity of the ice and the natural convection of the water. Both of these mechanisms exhibit poor heat transfer performance and are challenging to enhance further [35,36].

2.1.2. External Ice Melting Method

Unlike internal ice melting, external ice melting occurs in the opposite direction: the ice on the coil surface gradually melts from the outside inward, which is referred to as external ice melting [37]. Figure 3 shows a typical external ice melting system. Although the equipment components of this system are similar to those of the internal ice melting system, the piping connections are different.

Figure 3. Schematic diagram of a typical external ice melting system.

When ice is stored at night, the dual-working-condition refrigerating machine operates in ice-making mode, the valve V3 on the glycol circuit side is closed, while V1 and V4 are open. The dual-working-condition refrigerating machine cools the ethylene glycol solution, which flows through the coil. The supply and return temperatures are −5.6 °C and −2.5 °C, respectively. It indirectly exchanges heat with the water in the ice storage tank at a temperature between 10.0 °C and 15.0 °C. Over time, this process gradually forms an ice layer with a thickness of 20.0 mm to 30.0 mm on the outer surface of the tube. Since the cold water in the ice storage tank is not completely frozen, it is necessary to prevent the formation of ice bridges between the coils.

During the day when the ice melts, the dual-working-condition refrigerating machine stops running, while the valves of V5 and V6 are opened. The system circulation cold water pump draws unfrozen cold water from the ice storage tank and directly delivers it to the user to provide cooling. The supply water temperature is 2.0 °C, while the return water temperature is 12.0 °C. During the operation of the external ice melting system, the cold water in the ice storage tank directly contacts the ice on the coil. This contact mode facilitates the forced convection heat transfer process, which significantly increases the cooling capacity of the system. However, the system is not completely frozen, and its cold storage capacity is lower than that of the internal ice melting system.

2.1.3. Combined Internal and External Ice Melting Methods

Figure 4 illustrates a combined internal and external ice melting system, which shares the same equipment composition as the individual internal and external ice melting systems. The ice storage process in this combined system is identical to that of the external ice melting method. In the ice-melting process, the glycol solution circulates inside the coil and gradually melts the ice on the outside of the coil. Concurrently, the return water from the air conditioning system also circulates the exterior of the coil, which facilitates the melting of the external ice layer from the outer surface inward. This dual-action approach significantly enhances the system’s cooling rate. The melting rate achieved by the combined method surpasses that of the external or internal ice melting method alone. This improvement addresses the issue of ice bridging in the external ice melting system and improves the overall ice storage capacity.

Figure 4. Schematic diagram of the combining internal and external ice melting system.

Summarizing the three ice storage methods, the following conclusions can be drawn:

① Internal ice melting systems have a higher storage capacity than external ice-melting systems;

② External ice melting systems provide lower water temperatures and lower pump consumption;

③ The combined internal and external ice-melting system can optimize ice storage capacity by eliminating the constraints posed by ice bridge formation. However, this system demands more sophisticated switching and control mechanisms, and the pump requires a higher energy input during operation.

In the future, the external ice-melting process can be optimized by adjusting the flow direction and flow rate of the water distributor. This will lead to a more uniform temperature distribution within the ice storage tank and reduce the energy consumption of the pump during operation

3. Research on Enhanced Heat Transfer of Ice-on-Coil

In the ice-on-coil energy storage system, the heat transfer inside the coil includes the forced convection between the ethylene glycol solution and the coil and the heat conduction through the tube wall. Outside the coil, the heat transfer mechanisms include heat conduction between the coil and the ice and natural convection between the ice and the surrounding water. Figure 5 illustrates the heat transfer process of the ice storage unit and its thermal resistance model. With the thermal resistance model, the total heat transfer coefficient of the ice-on-coil energy storage process can be obtained with the equation [38].

$${\displaystyle \frac{1}{\mathrm{kA}}}={\displaystyle \frac{1}{{\mathrm{A}}_{\mathrm{tube},\mathrm{i}}{\mathrm{h}}_{\mathrm{H}\mathrm{T}\mathrm{F}}}}+{\displaystyle \frac{\ln \left(\frac{{\mathrm{D}}_{tube,o}}{{\mathrm{D}}_{tube,i}}\right)}{2\pi L_{tube}{\lambda }_{tube}}}+{\displaystyle \frac{\ln \left(\frac{{\mathrm{D}}_{ice}}{{\mathrm{D}}_{tube,o}}\right)}{2\pi L_{tube}{\lambda }_{ice}}}+{\displaystyle \frac{1}{{\mathrm{A}}_{\mathrm{ice}}{\mathrm{h}}_{\mathrm{water}}}},$$

(1)

where $\mathrm{k}$ is the total heat transfer coefficient, W/(m²·K); $\mathrm{A}$ is the total heat transfer area, m²; ${\mathrm{A}}_{\mathrm{t}\mathrm{u}\mathrm{b}\mathrm{e},\mathrm{i}}$ is the inner surface area of the tube wall, m²; ${\mathrm{h}}_{\mathrm{H}\mathrm{T}\mathrm{F}}$ is the forced convection heat transfer coefficient of heat transfer fluid in the tube, W/(m²·K); ${\mathrm{D}}_{tube,o}$ and ${\mathrm{D}}_{tube,i}$ are the inner and outer diameter of the tube, respectively, m; ${\mathrm{L}}_{tube}$ is the length of the tube, m; ${\lambda }_{tube}$ is the thermal conductivity of the tube wall, W/(m·K); ${\mathrm{D}}_{ice}$ is the diameter of the ice layer formed outside the tube, m; ${\lambda }_{ice}$ is the thermal conductivity of the ice layer, W/(m·K); ${\mathrm{A}}_{\mathrm{i}\mathrm{c}\mathrm{e}}$ is the area of the ice layer formed outside the tube, m²; ${\mathrm{h}}_{\mathrm{w}\mathrm{a}\mathrm{t}\mathrm{e}\mathrm{r}}$ is the convection heat transfer coefficient of water, W/(m²·K).

Figure 5. (a) Schematic diagram of the heat transfer process in the ice storage unit; (b) Thermal resistance diagram.

Ice has relatively low thermal conductivity compared to materials commonly used for coils; for example the thermal conductivity of carbon steel is generally 45~50 W/(m·K), while the thermal conductivity of ice is 2.2 W/(m·K).

$$\begin{array}{l}{R}_{tube}={\displaystyle \frac{\ln \frac{{\mathrm{D}}_{tube,o}}{{\mathrm{D}}_{\mathrm{tube},\mathrm{i}}}}{2\pi {\lambda }_{tube}}}\end{array},$$

(2)

$$R_{ice}={\displaystyle \frac{\ln \frac{{\mathrm{D}}_{\mathrm{Ice}}}{{\mathrm{D}}_{\mathrm{tube},\mathrm{o}}}}{2\pi {\lambda }_{ice}}},$$

(3)

where $R_{tube}$ is the thermal conductivity resistance of the ice coil, K/W; $R_{ice}$ is the heat conduction resistance of ice, K/W.

Taking an ice coil with a tube diameter of 20 mm and a tube wall thickness of 1 mm as an example, before the formation of ice layer, $R_{tube}$ is 0.00037 K/W, when the ice thickness is 1 mm, $R_{ice}$ is 0.0069 K/W, and when the ice thickness is 10 mm, $R_{ice}$ is 0.0501 K/W. It is obvious that in the process of ice storage, the thermal conductivity resistance of ice is significantly greater than that of ice coil, which becomes the bottleneck affecting the ice storage process. Therefore, it is necessary to adopt appropriate enhanced heat transfer methods.

Enhancing the heat transfer performance of ice storage can be achieved through various methods, such as utilizing extended heat transfer surfaces or increasing the thermal conductivity of phase change materials [39,40]. Xu et al. [41] discovered that once the thermal conductivity of the coil material exceeds that of ice, further increasing the coil material’s thermal conductivity has minimal impact on the ice formation time. Consequently, this paper primarily focuses on summarizing the influence of fin addition on the icing and melting times of the ice storage system.

However, adding fins also reduces the volume of the cold storage working medium within the same volume of the ice storage tank, which decreases the ice storage capacity [42,43]. There is an optimal number of fins and width. Beyond these two parameters, the heat transfer performance will not be further enhanced [44,45]. Implementing thin ring connections between the coils can improve temperature distribution and enhance heat transfer efficiency. Increasing the thermal conductivity and area of thin rings can significantly improve the ice storage rate [46]. Specifically, compared to smooth tubes, using annular fins can reduce ice storage time by 21%, while thin rings can reduce ice storage time by 34% [47]. Furthermore, the staggered arrangement of thin rings outperforms the parallel arrangement in terms of ice storage efficiency [48]. The primary findings of various enhanced heat transfer studies are summarized in Table 3.

Table 3. Study on heat transfer enhancement of ice-on-coil.

Type	Num.	Parameters and Dimensions	Main Contents	Main Conclusions	Figures	Ref.
Fins	1	(a) Four fins of 4 × 7 mm; (b) Four fins of 2 × 8 mm; (c) Eight fins of 2 × 3.5 mm; (d) Eight fins of 1 × 4 mm.	Comparison of the enhancement effect of four different parameters of fins on the ice storage capacity of coils.	As the number of fins increases and their height decreases, the solidification rate of the fins diminishes.		[42]
	2	Pitch: 20, 30, 40, 50, 60, 70 mm; height: 30 mm ring fins.	Study of optimal spacing between ring fins.	The 50 mm spacing is optimized and the ring fins have a high ice storage rate of 21%.		[47]
	3	The number of fins is 6, 8, and 10; the height of the fins is 20, 30, and 40 mm; and the thickness of the fins is 1, 3, and 5 mm.	The effects of fin height, fin thickness, and number of fins on the ice storage process are analyzed. New evaluation criteria are proposed to optimize the fin parameters.	The optimal fin thickness and fin number are recommended to be 3 mm and 8 mm to achieve the largest performance enhancement with the least mass penalty.		[49]
	4	Six fins; fins and tube thicknesses are the same; Ro = 4 Ri; Hf = 2.5 Ri;	Study the effect of fins of different lengths and PCM materials on heat transfer time.	Increasing fin length reduces melting and solidification time by over 14%, while composite PCM reduces it by over 20%.		[50]
	5	Four fins; the height is 45 mm.	Study of optimum height and angle of variable thickness fins.	Fastest heat transfer at half-fin angle (β = 8°).		[51]
	6	(a) Pinned tube: 16 pins, spacing 3~9.5 mm, length 20~40 mm; (b) Ring finned tube: five fins, spacing 52~92 mm, thickness 0.3~1.0 mm.	Analyze and compare the effects of different parameters of pinned and finned tubes on the heat transfer rate.	Finned tubes have a larger heat transfer area than pinned tubes and store ice faster.		[52]
	7	Y-fins; L1 = 0~11 mm; L2 = 0~28.5 mm; L3 = 4.5 mm; L4 = 28.5 mm; β = 13.4°~35.4°.	Single and double bifurcated Y-fins are proposed and geometrically optimized.	Double bifurcation increases discharge efficiency by about 24%; a smaller beta angle results in shorter run times.		[53]
	8	Five fins; Height is 0.2, 0.4, 0.6, 0.8 times pipe diameter; $\mathrm{\theta }=3$0°, 45°, 60°, 72°.	Study of the effect of angle between fins on heat transfer performance.	Using fins concentrated at the bottom improves efficiency and reduces heat transfer time by up to 50%.		[54]
	9	New symmetrical fins.	Fills a gap in previous studies when buoyancy effects are present in the solidification process.	Efficiency increased by 14.3% with the addition of nano thermal conductive particles combined with fins.		[55]
	10	Eight V-shaped fins; The angle of the fin is 45°.	Study of the effect of V-fin and nanoparticles on heat transfer time in triplex-tube heat transfer.	The use of fins has a greater impact on heat transfer than the addition of nanoparticles.		[56]
	11	New V-fins.	The ice storage performance of V-shaped fins was compared with that of straight fins; the effects of V-shaped fin angle, length, and sub-width on ice storage were investigated.	The smaller length or width ratio of the V-fin has a shorter ice storage time. Compared to straight fins, V-shaped fins reduce the total ice storage time by 64.2%.		[57]
	12	Bionic fins.	Proposed new annular and axial fins, design and optimization of tree branch bionic fins.	The radial fin design in Figure (a) has a tube spacing of 100 mm. The axial fin design with a tube spacing of 160 mm, as shown in Figure (b), was designed for longer charging and discharging times. The new axial fin design with a tube spacing of 70 mm, as shown in Figure (c). The Axial-70 fin design has a high power rate during the heat discharging of the storage unit.		[58]
Thin rings	1	Thin ring thickness is 0.25, 0.5, 1, 2, and 3 mm.	Study of the effect of thin rings of different thicknesses on the rate of ice storage.	Thin ring tubes had a 34% higher ice storage rate than smooth tubes.		[47]
	2	Single parallel; Double parallel; Staggered; The thickness is 0.25, 1, 2, 3 mm.	Study of the effect of the parameters of thin rings on the heat transfer time.	Staggering the thin rings and setting the thickness to 1 mm optimizes the overall heat transfer performance.		[48]

It is crucial to highlight that ice storage systems must consider both the ice storage and melting processes. Although the enhanced heat transfer by increasing the heat transfer area can shorten the ice storage time, it reduces the volume of ice produced in the ice storage tank and makes it easier to form ice bridges between the coils, which affects the melting efficiency of the external ice melting system. Therefore, considering the characteristics of the solidification and melting process of the coil, future research will focus on further optimizing the shape and arrangement of the coil fin to enhance its heat transfer performance. Additionally, future studies should also consider the economic and environmental impacts of reducing the melting time of ice storage systems.

4. Research on Operation Strategy of the Ice Storage System

The ice storage air conditioning system is more complicated than the conventional one. To cut operating costs, it is vital to reasonably control the amount of ice stored while the system is running. The amount of ice stored is closely related to the amount of cooling released the next day. Hence, the corresponding strategy must be formulated based on the predicted cooling load to allocate the proportion of the ice storage tank and the main chiller’s cooling output. Implementing these optimization strategies can reduce the system’s annual operating costs by over 30% [59].

Ice storage systems can be classified into two types based on their cooling modes, namely full and partial ice storage [60]. Full ice storage utilizes off-peak electric power to produce ice at night. It completely relies on ice melting during the daytime to meet the user’s cooling demands. Partial ice storage utilizes chiller units and ice-melting cooling loads simultaneously during the daytime.

When using the full ice storage mode, the air conditioning’s cooling load is entirely fulfilled by the latent heat released from melting ice. The mode can maximize the realization of the system’s “Load-Shifting” effect, as shown in Figure 6. However, this mode necessitates substantial storage capacity, extensive space, and high investment cost. It is particularly suitable for buildings with relatively concentrated cooling time, such as data centers, industrial plants, shopping malls, etc.

Figure 6. Load time distribution of full ice storage mode air conditioning.

Partial ice storage is divided into two modes, namely unit priority and ice melting priority. As shown in Figure 7 and Figure 8, the unit priority requires a smaller ice storage tank capacity, but the operation cost is high. The ice melting priority mode has lower operating costs but a larger ice storage tank capacity. In practical engineering applications, according to different cooling demands, the amount of ice storage and the cooling loads of units can be reasonably controlled.

Figure 7. Air conditioning load distribution in the operation mode of unit priority cooling.

Figure 8. Air conditioning load distribution in the operation mode of ice melting priority cooling.

Table 4 summarizes the technical approaches and main conclusions of different control strategies. The ice melting priority strategy, despite its higher operating energy consumption, demonstrates better economic performance. Conversely, using the unit priority strategy, the operation’s energy consumption is low, but the economy is slightly worse. When the required cooling capacity exceeds the capacity of the unit, the ice melting priority control strategy is more cost-effective. The ice storage cooling system can also be integrated with renewable energy generation systems, such as wind and solar photovoltaic, to reduce the dependence on grid power [61]. This integrated system can mitigate the volatility and intermittency of renewable energy and promote the use of solar and wind power. This achieves a green, energy-saving, and low-cost cooling supply solution.

Table 4. Research on the operation strategy of ice storage system.

Main Factors	Numbers	Technical Approach	Main Contents	Main Conclusions	Ref.
Cold storage capacity	1	Cooling model optimization.	Taking four typical buildings as the research object, the influence of different control strategies and electricity price structures on the optimal ice storage rate was studied.	When the ice storage rate is less than 0.267, the cooler priority control strategy will be more appropriate to the optimized control strategy; when the ice storage rate is >0.35, the ice storage priority control strategy is more appropriate to the optimized control strategy.	[62]
	2	Particle swarm optimization.	Outlines a study of load-shifting control strategies using different hot and cold energy storage facilities.	Load-shifting controls using building thermal mass can achieve more than a 30% reduction in daily peak loads and significant total cost savings from 8.5% to 29%.	[63]
	3	Model optimization.	A model-based optimal design method using genetic algorithms is developed to actively optimize the storage capacity.	Significant net annual cost savings of up to over USD 80,000 can be achieved by utilizing relatively small-scale active thermal and cooling energy storage systems, which is equivalent to 6.7% of a typical daily cooling load.	[64]
			A simulation-based approach to optimize the design of small-scale active energy storage systems in buildings to limit their power demand is proposed.	Demand-limiting controls using small storage tanks can save about 7% of a building’s total annual power consumption costs each year. The optimal storage capacity is less than 5% of the daily cooling load.	[65]
running cost	1	Forward dynamic programming algorithm.	An efficient model prediction controller for the charging and discharging of the ice storage part was developed (energy costs, equipment costs).	The model-predictive controller (MPC) receives tariff updates and re-optimizes the cooling system strategy according to the new prices. The open-loop optimal control does not have this information and operates the chiller at a very high price, resulting in operating costs that are approximately 40% higher.	[66]
	2	Dynamic programming algorithm.	Discussing the optimization of the ice storage air conditioning system under the condition of considering the minimum life cycle cost and efficiency of ice storage tank.	In the ice priority mode, life cycle costs are lower than for a conventional air conditioning system from the fourth year of operation, and under the 10-year life cycle, nine ice-storage units with an ice charging rate of 36.2% incur minimum cost.	[67]
			The cost optimization analysis is carried out on the selection of cold storage equipment and chiller, and the influence of charge and discharge strategy and electricity price strategy on system operation is determined.	The optimal operating protocol for storage charging and discharging is determined by a dynamic programming method that minimizes the operating cost over an entire year.	[68]
	3	Nonlinear programming methods.	A strategy based on mixed integer nonlinear programming is proposed to optimize the running schedule of building energy systems.	This strategy can significantly reduce operating energy costs (about 25%), reducing or even increasing to about 47% when using thermal energy storage systems.	[69]
	4	Model optimization.	Model-based real-time predictive optimal control of active and passive building thermal storage inventories in a test facility using time-of-use differentiated tariffs without demand charges is demonstrated.	When the optimal controller is given an imperfect weather forecast, the utility cost of the energy resource station can save 17% over the base case and 27% over the reference case.	[70]
	5	Mathematical modeling.	Optimized for life cycle economics.	Under the full load storage scenario and the base tariff structure, the daily savings would be USD 549.4/day, with the energy storage capital costs being paid off over 10 years, afterwards the daily saving in operational cost will be USD 4011.76/day.	[71,72]
	6	Neural network model.	Combination of neural network-based model prediction and genetic algorithm.	When time-of-use pricing or real-time pricing are adopted, the operating costs of district cooling networks are reduced by approximately 16% and 13%, respectively.	[73,74,75]
	7	Software modeling.	A model-based predictive control algorithm for cooling systems of small commercial buildings is proposed.	The proposed MPC algorithm can save about 5–20% per year, and the chiller priority strategy can save 20–30% per year.	[76]
Operational Energy Consumption	1	Particle swarm optimization.	Study of system energy consumption and CO2 emissions.	The optimum capacity of the chiller is estimated to be 250 RT when the ice storage capacity is set to 1800 RT-h.	[77]
	2	Model optimization.	A methodology is presented for determining the optimal chilled water storage (CWS) capacity and corresponding operating strategy for air conditioning loads at different electricity rates.	The optimal CWS strategy reduces peak demand by 38% under time-of-use (TOU) tariffs. Accordingly, customers can save 5.9% on operating costs.	[78]
	3	Strategy optimization.	The energy performance of a distributed energy system with a district cooling system (DES&DCS) was evaluated under four different control strategies. Comparisons were made with DCS and stand-alone cooling systems that are fully dependent on the grid.	Compared with the system that also adopts DCSs but only depends on the grid, the DES&DCS can save more than 10% of primary energy. Compared with the system that adopts an individual cooling system and only depends on the grid, the energy saving can be more than 16% and up to 19.1%.	[79]
	4	Software modeling.	Using EnergyPlus software modeling to validate a prediction methodology specifically designed to control multi-zone heating, ventilation, and air conditioning systems.	Regardless of the mode of operation (heating or cooling) and the time of year, energy consumption is significantly reduced by about 5 to 15%.	[80]
			Development of a dynamic computer model for ice thermal storage systems to compare energy use in conventional air-cooled systems and ice thermal storage systems.	Full ice storage can save up to 55% of the monthly cost of electricity needed for cooling compared to traditional air-cooling systems. Using the full storage option can reduce the total energy consumption of a selected building by up to 5%.	[81]
	5	Mathematical modeling.	A detailed mathematical model is proposed for the combination of heating and cooling chemistries in district energy to achieve the best performance of the whole system.	Effective savings in total annual costs and CO2 emissions. More than 67% of the CO2 reductions are achieved by mixing heat and district cooling.	[82]
External factors	1	Weather incorporated into the control system.	Consider the temperature parameters required for the indoor environment.	Significant increase in system operational capacity.	[83]
			A novel HVAC control method is proposed to minimize energy consumption while maintaining comfortable indoor temperatures based on short-term future predictions from weather forecasting models.	Energy savings of up to 58.79% have been demonstrated in EnergyPlus simulations.	[84]

The operating parameters of the ice storage air conditioning system have significant influence on the system performance. Reducing the carrier coolant’s inlet temperature can improve the heat transfer efficiency of the ice storage coil as well as the ice formation rate [85]. However, a lower inlet temperature of the carrier coolant may reduce the chiller unit’s coefficient of performance (COP). Increasing the carrier coolant flow rate can shorten the ice storage and melting time [86,87]. Furthermore, the longer the coil tube, the more uneven the ice thickness [88]. Several parameters, including the initial inlet temperature, mass flow rate, and the structure and arrangement of the coil tubes, will significantly influence the ice storage process of the system [89,90,91,92,93].

In the future development of the ice storage system, mathematical modeling and software analysis methods will be employed to predict the user side of the required cold load while calculating the ice storage system in different periods of ice storage and ice melting rate. This approach can facilitate the establishment of a feedback loop between cold storage systems and user demand. The system determines whether it is necessary to adjust the cooling capacity based on the cooling demand. Additionally, wind, solar, and other renewable energy generation systems are integrated into the ice storage energy supply system to establish a balanced energy supply and demand relationship in the process of ice storage and ice melting. By incorporating the power output, node voltage, and cooling demand of the chiller into the control system, the volatility and intermittency of renewable energy sources can be effectively mitigated. Furthermore, the control system can adjust the output power in time, ensuring the stable and efficient operation of the ice storage system. This method enables the efficient use of renewable energy and provides users with low-cost cooling capacity.

In summary, the control strategy of ice storage air conditioning systems can be developed and optimized from four aspects, namely storage capacity, operating cost, operating energy consumption and external factors. Studies have identified and compared various operating modes, such as full-capacity ice storage, unit priority cooling, and ice melt priority cooling. This helps us understand which models are most effective under different conditions. By optimizing the operational strategy of the ice storage system, it is possible to significantly improve its coefficient of performance (COP) and reduce energy consumption, thereby reducing operating costs. The study of operation strategy can realize the accurate adjustment of the building cooling load and rationally arrange the ice preparation and melting process according to the cooling load demand in different time periods. Furthermore, it can improve the response speed and flexibility of the system. With the development of the Internet of Things and intelligent control technology, it can be applied to the research of the operation strategy of the ice storage system, which can realize more intelligent and automated management.

5. Discussion and Conclusions

In this paper, the operating principle of the ice storage system is demonstrated, and the problems of the slow rate of generating ice layer and low density of cold storage in the current system are deeply analyzed. However, the current research also has some limitations. For example, the research on heat transfer enhancement in ice storage systems is extensive, but much theoretical work has focused on the solidification process of ice storage systems. Therefore, more investigations are required on ice melting in the presence of various simple and hybrid heat transfer enhancement techniques. This paper analyzes and summarizes the key technology research achievements and existing problems of coil-type ice storage air conditioning.

(1) During the freezing process of water, the total thermal resistance increases as the ice layer thickens. When the ice thickness reaches a critical point, its thermal resistance becomes the dominant factor influencing the ice formation time. Meanwhile, the proportion of the coil tube material’s thermal resistance in the total thermal resistance gradually decreases. Consequently, it is unnecessary to excessively enhance the thermal conductivity of the coil tube material. It has been observed that when the thermal conductivity of the polymer coil material reaches 2.2 W/(m·K), the ice storage and melting rates approach those of metal coils.

(2) When thin rings are installed outside the coil tube, the ice storage rate is better than adding fins or smooth tubes outside the coil, which significantly improves the heat transfer performance, and the enhancement of ice melting and ice storage rate can reach 34%. However, it is necessary to avoid the formation of ice bridges, as they affect the speed of ice melting and cooling.

(3) External ice melting releases cold energy with greater flexibility and faster cooling rates. Internal ice melting has a greater storage capacity. Future research should focus on the system design and operation control of combining internal and external ice melting to realize the two-way improvement of the cold storage capacity and cooling rate.

(4) The control strategy for the ice storage air-conditioning system should be formulated and optimized by considering the following four aspects: cold storage capacity, operating costs, energy consumption, and external factors. (i) In the optimization of the amount of cold storage, by simulating and comparing different models of cold storage and selecting the control scheme that can achieve the maximum amount of ice storage within a certain cost range, the load in the peak period can be reduced by 50%. (ii) By adapting the control system and building predictive models, it is possible to minimize operating expenses and save 5% to 30% of operating costs. (iii) Different algorithms and mathematical models are used to predict and analyze the system energy consumption, which helps to reduce the peak load and carbon dioxide emission. Future research could consider combining weather forecasting, load forecasting, and operation strategies to optimize operational strategies dynamically, thereby achieving more substantial reductions in operation costs. (iv) Combining the ice storage system with renewable energy generation systems, such as wind power and photovoltaic power, can effectively mitigate the volatility and intermittency of renewable energy sources and promote their consumption.

By expanding the heat transfer surface of the coil tube, adopting a combination of internal and external ice melting methods, and optimizing the operation strategy, it can simultaneously achieve the improvement of ice storage and melting rate, energy storage density, and economic performance.