1. Introduction
The International Maritime Organization (IMO) has consistently fortified regulations aimed at diminishing greenhouse gas (GHG) emissions from maritime vessels after unveiling its initial strategy for reduction of the GHG footprint in the maritime sector [
1]. In addition, from the 80th session of the Maritime Environment Protection Committee (MEPC) held in July 2023, a revised strategy was adopted, which includes levels of ambition and indicative checkpoints for the gradual reduction of GHG emissions. According to the revised strategy, the maritime sector must implement a minimum of 5% zero or near-zero GHG emission technologies, fuels, and energy sources by 2030 and strive to achieve at least a 70% reduction in annual GHG emissions by 2040 compared to 2008 levels, with efforts to reach an 80% reduction. Ultimately, the goal is to achieve net-zero GHG emissions by around 2050 [
2].
According to the research by E.A. Bouman et al. (2017), it is suggested that the IMO’s strategy can be achieved through modifications in hull design, changes in power and propulsion systems, the utilization of alternative fuels and energy sources, and operational enhancements [
3]. Among these, the use of low-carbon or zero-carbon alternative fuels is considered an imperative measure for reducing GHG emissions from ships. Alternative fuels with lower carbon content, such as liquefied natural gas (LNG), liquefied petroleum gas (LPG), and methanol, exhibit a high level of technological maturity and can serve as short-term and medium-term measures. However, they do not represent a complete solution for achieving the ultimate goals of the IMO’s strategy since they still include some carbon. On the other hand, the adoption of carbon-free fuels like hydrogen and ammonia with fuel cells for a full-electric propulsion system can effectively eliminate GHG emissions and provide the necessary energy for ship operations [
4]. Among these options, hydrogen fuel can be stored within the vessel either as high-pressure gas (HP-GH
2) or as a liquid (LH
2). The HP-GH
2 offers the advantages of a simpler FGSS configuration and easier operation when compared to the LH
2. However, it has a volumetric energy density of approximately 40 kg/m
3 (for HP-GH
2 of 700 bar), which is roughly half of the LH
2’s density of 70 kg/m
3. Consequently, as the propulsion capacity and bunkering period of a ship increase, the fuel tank’s volume expands excessively. Therefore, it is anticipated that hydrogen fuel for future H
2-fueled ocean-going ships or high-power vessels will likely be stored in the form of LH
2 to manage the challenges associated with volumetric energy density [
5,
6,
7,
8]. On the other hand, hydrogen has a lower boiling point than air, so an insulation system that includes vacuum insulation, which is safe and highly efficient, should be considered. To use LH
2 as a ship fuel, related research about the cryogenic property of LH
2 should also be carried out in advance.
R.T. Madsen et al. conduct a technical, regulatory, and economic feasibility study for a coastal research vessel powered by hydrogen fuel cells, offering viable design, construction, and operational solutions [
9]. L.E. Klebanoff et al. perform a comparison of hazardous areas and annual GHG emissions for diesel-electric, battery-hybrid, and hydrogen-hybrid powertrains applicable to coastal/local research vessels [
10]. Q.H. Nguyen et al. reported the recovery of waste heat from solid oxide fuel cells (SOFC) in a hydrogen-fueled ship powered by SOFCs, which enhances the efficiency of the target system [
11].
Fuel cells exhibit comparatively slower dynamic response characteristics in contrast to the conventional power sources employed in maritime vessels [
12]. Moreover, given that the periodic start/stop operations can accelerate fuel cell degradation, they are typically integrated with other power sources and energy storage systems to form a hybrid propulsion system. Among the various options available, a lithium-ion battery (hereafter battery) is frequently coupled with fuel cells due to their rapid response capabilities, enabling them to effectively manage substantial load fluctuations.
Batteries, as they charge and discharge electric power through electrochemical reactions, are sensitive to temperature variations, affecting their performance and lifespan. When the operating temperature deviates from the normal operating range (roughly
T < 10 °C or
T > 40 °C), it results in reduced energy capacity, diminished round-trip efficiency, and an increased aging rate, consequently reducing the battery’s lifetime [
13]. According to previous research, it has been reported that, when the C-rate of batteries is not high (under 1C), an increase in the depth of discharge leads to an increase in the aging rate, resulting in a reduction in the lifetime of the batteries [
14]. Moreover, operating at higher temperatures not only impairs battery performance and shortens its lifespan, but also poses a serious safety risk due to thermal runaway. To mitigate these issues, a thermal management system is installed to maintain the battery’s temperature within the normal operating range during vessel operation. Various heat transfer media can be considered for battery thermal management, including air, liquid, or phase change materials (PCM) [
13,
15,
16,
17]. Among these, forced air cooling and liquid cooling methods have undergone significant technological development. Liquid cooling, in particular, efficiently maintains battery cell temperatures, reducing temperature differences between cells. Also, the choice of thermal management method is influenced by space limitations on ships. When using the forced air cooling, an increase in ship capacity necessitates a larger space allocation.
On the other hand, when storing hydrogen fuel in a liquid state within a ship, a heating source is required to vaporize and heat the stored LH
2 to meet the temperature conditions demanded by fuel cells. Similar to LH
2-fueled ships, LNG-fueled ships, which use fuel stored at low temperatures, commonly utilize an ethylene glycol/water mixture (GW) as a heating medium to supply heat energy [
18]. It is expected that this same approach of using GW as a heating medium can be applied to future LH
2-fueled ships, which will similarly require the vaporization and heating of their stored fuel.
In the case of LNG-fueled ships, there has been significant research by many scholars on the recovery of cold energy (i.e., the heat energy that can be obtained when a fluid stored in an extremely low-temperature state is heated and vaporized) generated during the vaporization and heating processes of LNG. F. Han et al. (2019) propose a triple organic Rankine cycle (ORC) system to recover waste heat and cold energy during the operation of LNG-fueled vessels, followed by an energy analysis and multi-objective optimization [
19]. T. W. Lim and Y.S. Choi (2020) conduct research on the thermal design and performance evaluation of shell and tube heat exchangers installed in ORC systems for recovering cold energy in LNG-fueled ships [
20]. Additionally, J. Koo et al. (2019) propose six different ORC systems for LNG-fueled ships powered by high-pressure and medium-pressure dual-fuel engines and performed optimization based on exergy analysis and economic analysis for nine different working fluids [
21]. While research on the recovery of cold energy from LNG-fueled ships is abundant, there is a lack of similar studies for LH
2-fueled ships. Lee et al. (2023) introduce a system that combines a polymer electrolyte membrane fuel cell (PEMFC) system with an organic Rankine cycle-direct expansion cycle (ORC-DEC) and conducted thermodynamic and economic analyses of the system [
22].
The LH2-fueled hybrid electric propulsion system (LH2-HSPS) utilizes LH2 as fuel, which inherently possesses cold energy. The process of heating and vaporizing LH2 and managing the thermal conditions of battery cells consumes electric power. By integrating the use of the same heat transfer medium, such as GW, for the heating source in the LH2 fuel gas supply system (FGSS) and the thermal management system for the battery, the total electric power consumption can be reduced. This integration offers the advantage of not requiring additional equipment, thus eliminating the need for additional capital expenditure. Despite these benefits, research on the design and efficiency improvement of LH2-HSPS has been insufficient, leading to a lack of studies in this area to date.
Therefore, this study aims to propose a method to enhance the efficiency of the LH
2 FGSS and battery system within the LH
2-HSPS by integrating the GW systems installed in each of them. The objective is to analyze the effects of this heat integration. Additionally, the study investigates whether the operating temperature can be maintained within the acceptable range. The remaining sections of this paper are structured as follows. In
Section 2, the system and model configuration of the LH
2 FGSS and battery system are introduced. Based on this,
Section 3 presents the methodology for integrating the GW system and the efficiency analysis method.
Section 4 covers the research results and discussions, while
Section 5 provides the conclusion of this study.
2. System Description
In this study, to analyze the reduction in power consumption resulting from the integration of the GW system in the balance of plant of the LH
2 FGSS and the thermal management system of the battery system, the LH
2 FGSS and the battery system were modeled as shown in
Figure 1. The LH
2 FGSS comprises a fuel tank for LH
2 storage, a pump for LH
2 transfer, a vaporizer for heat exchange between LH
2 and GW, and control valves to regulate the flow rate, temperature, and pressure of H
2 and GW in response to changes in the output of the PEMFC system. The battery system charges and discharges electric power over time to meet the power demand of the target vessel with variation of the PEMFC system, and the thermal management system absorbs the generated heat through the GW to maintain the temperature of battery cells within an acceptable range.
The GW system, which supplies the heat energy required for the vaporization and heating of LH2 or is used in the thermal management of the battery system, is a closed system consisting of a GW tank, GW pump, GW heater (or radiator), and control valves. For the LH2 FGSS, GW passing through the LH2 vaporizer experiences a decrease in temperature and pressure. The temperature and pressure of the circulating GW are then increased by the GW heater and GW pump before being supplied back to the LH2 vaporizer. The GW passing through the battery system’s cold plates experiences an increase in temperature and a decrease in pressure. It circulates back to the cold plates again after passing through the GW radiator and GW pump. Various fluids, including air and seawater, can be considered as heating media for supplying or removing thermal energy. However, the GW was assumed to be the heating medium in this study due to its relatively better heat transfer characteristics and wide operating temperature range (approximately −35 °C to 100 °C for a mole fraction of 0.5), making it suitable for heat exchange with LH2, which has a lower temperature.
To analyze the effects of integrating GW systems installed in the LH
2 FGSS and battery system, two GW systems were modeled, as shown in
Figure 2.
Figure 2a represents the independent GW system, where each GW system comprises a GW pump and a GW heater (or radiator). These components supply the heat energy required for heating LH
2 and managing the thermal conditions of battery cells.
Figure 2b illustrates a schematic diagram of the integrated GW system, which combines the two GW systems introduced in
Figure 2a. In this integrated system, the GW, which has been used for LH
2 vaporization and heating, provides some cooling energy for the battery cells. The remaining thermal energy is supplied from the GW heater, similar to the independent GW system. Furthermore, by integrating the two GW systems, it is possible to reduce redundant components, which can also lead to a partial reduction in equipment costs for the installation of the target system.
A 2 MW class platform supply vessel (PSV) was considered as the target ship to quantitatively assess the efficiency improvement achieved by the application of the integrated GW system and the temperature variations within the battery system. PSVs are vessels that transport materials and crew to offshore platforms, and their power demand varies significantly when operating in the marine environment due to the use of dynamic positioning. PSVs are well suited for effectively addressing the rapid changes in power demand that occur during operation in dynamic positioning mode, thanks to the battery system’s assistance. Research on the application of hybrid power systems for PSVs, including alternative fuels such as H
2, has been actively conducted [
23,
24,
25].
Figure 3 represents the power demand profile of the PSV used in this study for the performance analysis of the integrated GW system [
5,
23,
24,
25].
The LH
2 FGSS used for vaporizing and heating LH
2 to supply it to the PEMFC system was simulated using Aspen HYSYS based on the previous research in [
5,
6,
26]. Aspen HYSYS is commercial software that simulates systems and processes composed of mechanical equipment, such as FGSS, based on thermodynamic, fluid mechanics, and heat transfer equations, and various FGSS for different fuels have been simulated through this software in previous studies.
Table 1 presents the main specifications of the equipment comprising the LH
2 FGSS, along with the parameters for simulation. The volume of the LH
2 fuel tank was determined based on the total energy required for vessel operation, as indicated by the previously mentioned PSV’s power demand. This determination was made based on the efficiency of a typical PEMFC system (50~60%) and the design parameters of the fuel tank. Furthermore, to model the pressure and temperature variations of the heated GW stream after thermal management of the battery cells, a heater component was added to equivalently simulate the variation. The heat flow rate and pressure drop, calculated in the battery system model described later, were used as input values for the heater component to perform the calculations of the FGSS. All systems, including the LH
2 FGSS, were confirmed to be specifiable during the specification determination stage based on the general arrangement of a 2.5 MW class PSV for feasible placement within the vessel [
27].
In this study, we did not directly perform modeling of a PEMFC system that generates power using hydrogen fuel and operates alongside the balance of plant by connecting multiple stacks. Instead, we determined the hydrogen flow that can produce varying PEMFC system outputs over time based on the model of a 2 MW class PEMFC system developed in previous research and the reported efficiency based on lower heating value of the entire system [
5]. This determined hydrogen flow was then used as the setpoint for the LH
2 FGSS model.
Electrochemical models, such as the single particle model, porous electrode model, and pseudo two-dimensional model, are highly accurate for simulating the charging and discharging processes of battery cells. However, they come with the drawback of high model complexity and significant computational costs [
13]. Consequently, these electrochemical models are challenging to apply in power system studies. Therefore, simplified models are commonly employed in such research.
The equivalent circuit model (ECM) allows for the simulation of battery cells’ behavior by implementing changes in cell voltage, electric current, and state of charge (SOC) based on the electrochemical reactions occurring within the battery cell. This is achieved using electric components, such as voltage source, resistor, and capacitor [
28]. In this study, the battery cell was simulated based on a Thevenin model that utilizes four components, as shown in
Figure 4 (two resistors, one voltage source, and one capacitor). The parameters for these four components were determined based on a two-dimensional look-up table for SOC and cell temperature, referencing the research results from [
5,
6,
29]. This approach was employed to simulate the changes in cell performance with varying SOC and cell temperature. Equation (1) represents the voltage of a battery cell calculated based on the Thevenin model.
Meanwhile, the charging and discharging operations of the battery system were simulated through the serial and parallel connection of cell models. The generated heat by battery cells during operation and the GW-based thermal management system designed to manage it were modeled based on Equations (2)–(5) [
5,
6,
17]. In Equation (2), the first term on the right-hand side represents the ohmic heat of the battery cell, while the second term signifies the heat generated by the change in entropy due to electrochemical reactions. Additionally, the pressure drop of the GW flowing through the channels inserted into the cold plates installed in the battery system is calculated through Equation (5), and the pressure drop is compensated by the GW pump.
Table 2 presents the specifications of the battery system, determined with reference to [
5,
6].
: Heat generation of the battery cell.
: Electric current through the battery cells.
: Open circuit voltage.
: Cell voltage.
: Mass of the cold plate.
: Specific heat of the cold plate.
: Thermal conductivity.
: Convective heat transfer coefficient of GW.
: Heat transfer area of cooling channels.
: Friction factor with averaged condition between the inlet and outlet.
: Reynolds number with averaged condition between the inlet and outlet.
: Prandtl number with averaged condition between the inlet and outlet.
: Roughness of tube.
: Diameter of tube.
Table 2.
Specifications of the battery system.
Table 2.
Specifications of the battery system.
Item | Unit | Value |
---|
Nominal capacity | kWh | 2007.04 |
Available DoD | % | 80.00 |
Available energy | kWh | 1605.63 |
Nominal voltage | V | 980.00 |
Cells configuration | - | 272S 432P |
Initial SOC | - | 0.50 |
Range of temperature variation (for independent GW system) | °C | 20.50 ± 0.50 |
3. Heat Integration Methodology
To integrate the GW system, a heater component is added to the LH
2 FGSS model, which can reflect the temperature increase and pressure decrease of GW after cooling the battery system. The Aspen HYSYS library provided the heater component, which calculates the outlet temperature based on the input power. As the GW stream flows through the channel inserted into the cold plates of the battery packs and cools down the battery system, the simulated heat flow rate from the battery system to GW is used as the input value for the heater component. In order to simulate the integrated GW system, a GW control scheme is presented in
Figure 5. The control valve regulates the mass flow rate of GW supplied to the battery system, while the bypass stream is not controlled since controlling both streams can cause control failure during transient states. The heat duty of the GW heater is controlled for the same GW inlet temperature as the LH
2 vaporizer, which is the same as the control method used for the independent GW system.
The performance of the integrated GW system is analyzed according to the degree of load change of the LH
2-HSPS with three different simulation cases, as shown in
Table 3. The three simulation cases represent different operating conditions of the PEMFC system, including full-load operation, low-load operation, and constant output operation. As the output range of the PEMFC system varies, the power required for charging and discharging of the battery system also changes. Through this, we analyzed the heat integration effects based on various operating strategies for the battery system.
To compare the performance of integrated GW system quantitatively, an energy-saving ratio is defined for heating and cooling of the GW with Equations (6)–(9).
PFGSS,id,heater and
PFGSS,id,pump refer to the heating duty of the GW heater and power consumption for the GW pumps, respectively, obtained through calculations of LH
2 FGSS with the independent GW system. Variables with ‘it’ in subscript all represent the same physical quantity, which refers to the calculated results of the LH
2 FGSS using the integrated GW system.
Pbattery,id,radiator denotes the power used for cooling the GW after the battery system is cooled, which is necessary for the circulation of the GW when using an independent GW system. The cooling system assumes the use of a GW radiator, and the power was calculated using a COP of 2.5, which is commonly used for GW radiators in battery or power systems.
Pbattery,id,pump is the required power for the GW pump in the battery system. When the integrated GW system is applied, it is assumed that no power is required for the GW pump in the battery system.
The above equations can be used to determine the amount of energy saved by using the integrated GW system instead of the independent GW system in the LH2-HSPS. Furthermore, an analysis of the temperature of the battery system is conducted to determine whether temperature variations are within acceptable limits.
4. Results and Discussion
Figure 6 shows the power demand and output of the battery system. In Case 1, where the PEMFC system mostly produces propulsion power that varies over time, it is observed that output power can be higher or lower than the propulsion power to maintain the battery system’s SOC within a certain range. Even when the minimum output of the PEMFC system is set to 200 kW, it satisfactorily meets this constraint during transient states. On the other hand, in Cases 2 and 3, where the PEMFC system cannot produce all the propulsion power, it is confirmed that there is a more significant variation in the output of the battery systems. In particular, in Case 3, where the PEMFC system produces almost constant output, it can be observed in
Figure 6c that the battery system exhibits the most significant output variation. When the Ah-throughput of the battery system or the C-rate is the highest, it results in increased heat generation within the battery cells. Therefore, it can be expected that the effects of integrating the GW system, as analyzed in the following paragraphs, will be most effective in Case 3.
The power consumption of the GW heater for both systems is compared under different cases, as illustrated in
Figure 7. The comparison results clearly indicate that the reduction of power consumption is closely related to the output power of the battery system, since the heat generated in battery cells is proportional to the electric current. Moreover, the heating duty of the GW for LH
2 FGSS is proportional to the output power of the PEMFC system. The average power consumption of the GW heater with the independent GW system was found to be 27.60, 26.22, and 24.60 kW in Cases 1, 2, and 3, respectively. However, the average power consumption was reduced to 27.22 (1.38% savings), 21.95 (16.29% savings), and 17.83 kW (27.52% savings), respectively, when the integrated GW system was applied. Consistent with the aforementioned results, the greatest reduction in power consumption occurred in Case 3, which has the largest amount of charged and discharged energy of the battery system.
Figure 8 shows the temperature variation of the battery cells when using the independent GW system and integrated GW system. As explained in the previous section, the temperature of the battery system is maintained within a tolerance range of ±0.5 K when the independent GW system is applied. However, with the application of the integrated GW system, temperature variation is observed within a tolerance range of ±3 K compared to the set point. Although the temperature of the battery system deviates from the set point when using the integrated GW system, it is generally acceptable for normal operation of the battery system. Furthermore, if precise temperature control is required, additional GW cooling can be utilized to sufficiently cool the battery system with partial reduction of efficiency of the GW system.
Figure 9 provides a summary of the performance of the integrated GW system based on the energy-saving ratio defined in
Section 3. Although the average power consumption of the GW heater, as presented in
Figure 7, decreased by approximately 1.38% (Case 1) to 27.52% (Case 3), the total energy required for the GW system decreased by 1.86% (Case 1) to 33.80% (Case 3), when the required GW cooling duty in the independent GW system is considered. In other words, when the GW system is integrated, it can reduce the electric power consumption of the GW heater used in the LH
2 FGSS and the electric power of the GW radiator used in the battery system. Therefore, a higher energy-saving ratio was calculated compared to the reduction in GW heater energy consumption as introduced in
Figure 7.