EFFECTS OF FLOW RATE AND INLET TEMPERATURE ON PERFORMANCE OF ANNULUS TYPE LOW-TEMPERATURE LATENT HEAT THERMAL ENERGY STORAGES

Solar energy is one of the largest energy potentials which can be utilized in solar heater and integrated with the latent heat energy storage (LHTES). This research aims to investigate the effects of the operating conditions of flow rate and inlet temperature on the performance of the annulus type Low-Temperature LHTES using Computational Fluid Dynamics method in which the enthalpy-porosity is used as the solidification model. The results indicated that better performance can be obtained by increasing the flow rate and inlet temperature. The increase in flow rate resulted in a higher heat transfer, producing better performance up to 11.91% and 24.91% during the charging and discharging, respectively. Meanwhile, increasing the inlet temperature in the Low-Temperature LHTES system enhanced the performance up to 192.72% during the charging and 13.07% during the discharging. This is an open access article under the CC BY-NC license


Introduction
The need of global energy has been increasing exponentially since the 1900s as depicted in Fig. 1 [1]. Considering the Paris Agreement in 2015, the world has started to see the potential of renewable energy to reduce fossil fuel emissions [2]. The largest renewable energy potential in Indonesia is solar with 207.8 GWp [3]. However, for harvesting solar energy, one of the biggest challenges is the intermittent characteristics of solar energy or only can be harvested in a limited period and in a fluctuated nature, as shown in Fig. 2 [4]. The use of the renewable energy worldwide is still low due to the imbalance of energy supply and demand period. Currently, the technology of energy storages such as sensible heat energy storages, latent heat energy storages, and chemical energy storages have been developed to solve the problem in the most effective, efficient, and safest way [5,6]. Scapino, et. al. [7] reviewed that the latent heat thermal energy storages (LHTES) is one of the technologies that could achieve the three indicators explained before. LHTES is more effective in cost, has a high efficiency, and has a minimum response time [7]. Almost all LHTES applications have the same phase change material (PCM) geometry configuration of doughnut-shape, where the PCM will cover the tube in which heat transfer fluid is flowing inside the tube, as shown in Fig. 3 a) [5,[8][9][10]. One of methods to improve the thermal performance can be done by extending the heat transfer area and by making the fluid flows more turbulent [11,12], by changing the geometry configuration of the PCM, where the heat transfer fluid covers the annulus tube that contains the PCM. Based on the applications, LHTES relies on the PCM or material for absorption of the thermal energy (heat storage) and classified as Low-Temperature (below 150°C, like for solar heater), Medium-Temperature (between 150°C-400°C), and High-Temperature (above 400°C, for Concentrated Solar Power, Steam Power Plant, etc.) applications [6,13]. In this study, low temperature LHTES is selected for solar applications by using paraffin wax like n-octadecane for the PCM and evaluated in charging and discharging region because the properties of paraffin wax have a high density, stability of termal properties, unreactive, and the most important is the cycle resistance up to 2000 cycles [14,15].

Model Formulation
Here, the PCM is placed in the inner tube and the heat transfer fluid (HTF) flows through a pipe in the outer side of tube and cover the PCM tube (shell-side) with adiabatic wall as depicted in Fig. 3 b). Various configuration of inlet temperature and flow rate of HTF are considered in this research. The evaluation of performance of charging and discharging is evaluated by enthalpy of the process.

A. Governing Equations
The governing equations are employed for two components, water and PCM. During charging, the water, which flows in the outer side tube, transferred the heat to melt the PCM and stored the thermal energy in a liquid phase. While during discharging, the heat that stored in PCM is then taken by the cold water and thus solidified the PCM. In this research, a three-dimensional model is simplified into two-dimensional to reduce the cost of computation.
In the HTF, the convection of heat and fluid flow will be considered by using the Navier-Stokes equations. These equations focused on conservation of mass, momentum, and energy [5,9].
During the process of melting and solidification, the molten PCM will form a mushy zone where it will decrease the value of porosity. To evaluate solidification process, the enthalpy-porosity formulation is represented as a source term, , and it is added to the conservation momentum Equation (2) with the value defined as: where, ζ is mushy zone constant that has value of 1×10 5` for n-octadecane. Then, the enthalpy is added on energy conservation Equation (3) and defined as where, h is sensible heat of PCM and defined as: where, ℎ , is reference enthalpy on reference temperature and ΔHpcm is latent heat of PCM which is a function of liquid fraction, β of the PCM.

B. Constitutive Relations
The water liquid defines as HTF that has temperature dependent polynomial functions for the thermophysical properties. The density, viscosity, and thermal conductivity of water are defined in (8-10) respectively. The other thermophysical properties are presented in Table 1 The PCM (n-octadecane) also has temperature dependent polynomial functions for the thermophysical properties that are defined in (11)(12)(13) and the other properties are presented in Table 1. = 774 9 × 10 −4 ( − 300.65) + 1 To evaluate the heat transfer rate, ̇ on charging and discharging condition can be evaluated by Journal of Emerging Supply Chain, Clean Energy, and Process Engineering Vol.

C. Boundary Conditions
In this research, these following boundary conditions are employed on the model.
• Inlet: at the inlet, a various inlet mass flow rate and inlet temperature is specified as shown in Table 2.
• Outlet: at the outlet, stream-wise gradient of temperature and gauge pressure are set to be zero.
• PCM wall is set to be no-slip condition, zero roughness height, and coupled • HTF wall is set to be insulated wall (no heat flux), no-slip condition, and zero roughness height.

Numerical Methodology
In this research, the LHTES model and grid generation are conducted in ANSYS Workbench 2019 R2 in which simplified symmetrical 2D model is used as depicted in Fig. 4. Then, the mathematical model formulation (governing equations, constitutive relations, and boundary conditions) is implemented and solved. For several thermo-physical properties of HTF and PCM, the C-Language is coded and compiled for user-defined functions in the model. To ensure the grid generation is independence, the grid independent study was conducted by dividing the domain into areas and set the number of divisions in c-side to increase the number of fluid elements (cells), the summary of the test shown in Table 3 and the results is presented in Fig. 5. It is seen that, there is no significant differences on the average liquid fraction of PCM due to the time which means the number of cells is not affecting the value significantly. Hence, the grid size of 71,568 cells was chosen for all cases of this study. The results of this grid is shows in Fig. 6 with the orthogonal quality is 0.9998 which means the setting of grid is good.
Semi-Implicit Pressure Linked Equation (SIMPLE) algorithm is chosen for solving the numerical model of this research. The momentum and energy equations are solved by the second order upwind discretization and pressure correction equation is adopted in the PRESTO! scheme. A time step of 0.1 second is taken with the number of maximum iterations of 20 times for each time step. Convergence criterion of 1 × 10 -3 is selected for the momentum and turbulent equations and 1 × 10 -5 is set for the energy.

A. Model Validation
To ensure correct selection and settings of the mathematical model, the validation is made by comparing the results of the simulation with the results of previous experiment [8]. The comparison is focusing on two temperature locations (T1 and T2) and the results are shown in Fig. 7. As can be seen from Fig. 7, a relatively good behavior is shown with small relative error of 0.357% for T1 and 0.152% for T2. This explain that the settings and parameters of this research are valid and thus the simulation results sufficiently describe the behavior of the LHTES.

B. Effect of Mass Flow Rate
The heat transfer in LHTES is affected by the flow condition, especially flow rate. Hence, this research is carried out to prove and find the behavior of heat transfer of PCM in three variation flow rates of 0.01575 kg/s; 0.03150 kg/s; and 0.15750 kg/s. The simulation is evaluated by flowing the hot water with 310.65 K (10 K above the melting temperature of PCM, n-octadecane) in the inlet. Fig. 8 presents the average PCM liquid fraction during charging-discharging process for the flow rate variation. This figure shown that from the initial until charging process ended, the increase of flow rate will increase the heat transfer rate. This is also implied that at higher flow rate the liquid fraction of PCM is faster to reach value of 1.0 which means the PCM has melted completely into the liquid phase. Higher flow rate creates more turbulence eddies. These eddies will mix the fluid elements and speed up the convection rate in the heat transfer process, thus give faster melting process in the PCM [17].  Fig. 8 also shows that the rate of liquid fraction is slower towards the end of charging cycle because the heat transfer in PCM is started to be dominated by convection process. This phenomenon also depicted in Figs. 9 and 10 where the melting process in the center region of the PCM is reduced due to the change of conduction process into a convection process near the wall. Figs. 9 and 10 also shows that towards the end of the charging cycle, the temperature of PCM will approach the hot water temperature and in the higher flow rate and the melted area in the PCM tends to be larger.
The behavior of flow rate effect on the discharging process is similar to the charging process. When the flow rate is increased, the solidification of PCM is increased because turbulence effect on heat transfer. Fig. 11 and 12 show that at higher flow rate, the area of solidification near the wall in PCM is larger and accompanied with lower temperature distribution.    Table 4 is the summary for the evaluation of the overall heat transfer performance (Thermal Enhancement Ratio, TER) for each case. The value shows that by increasing the flow rate the heat transfer process will be faster. Increasing 2 times the flow rate will give 6.17% and 10.90% higher efficiency of heat transfer on charging and discharging, respectively. Meanwhile, increasing 10 times of the flow rate will improve the efficiency by 11.91% in charging and 24.91% in discharging.

C. Effect of Inlet Temperature
The inlet temperature gives more significant effect towards the average PCM liquid fraction as depicted in Fig.  13, especially during the charging process. As can be seen from Fig. 13, the heat transfer rate increases as the inlet temperature of the HTF increases and promotes faster melting processes of the PCM. Figs. 14 and 15 shows supporting evidence for this tendency that in the fifth minute of charging, the PCM is melted in the area near wall and the area of liquid PCM is larger at the higher temperature water. The similar behavior also depicted at the 20 th minute of charging. Thus, with increasing of the inlet temperature of HTF, the temperature difference will also increase and will drives the heat transfer process both by conduction (in the solid state) and convection (in the liquid state) faster [8,18] and gives higher trend of the curve gradient.
During discharging process, it shows slightly different behavior from the charging one. Fig. 13 shows the rate of the average PCM liquid fraction during discharging is almost similar or looks like there is no effect by the temperature variation. The temperature and liquid fraction distribution that shown in Figs. 16 and 17 respectively also shows almost no differences due to the inlet temperature. However, if the graphs in Fig. 13 are enlarged in Fig. 13, the curve with 320.65 K was reaching the zero liquid fraction which slightly faster than the other two which indicate that the heat transfer process for 320.65 K slightly faster than the others.    Looking at the heat transfer performance for different inlet temperatures, as summarized in Table 5, it is found that the inlet temperature will gives relatively a higher effect on the heat transfer performance (thermal enhancement ratio, TER) especially during charging cycle. The highest temperature (adding 20 K above the melting point of PCM) will give relatively a higher heat transfer performance up to 192.72% during the charging and 13.07% during the discharging.

Conclusions
Based on this numerical simulation, the behavior of flow rate and inlet temperature on annulus type LHTES with n-octadecane as PCM could give significant effect in the heat transfer performance, both during charging and discharging process.
• Maximizing the flow rate could give more turbulence effect on LHTES system so it could speed up the heat transfer process. In this research, increasing the flow rate 10 times will give 11.91% and 24.91% efficiency during charging and discharging respectively.
• Rising the inlet temperature will give more significant temperature difference and drives the heat transfer process. By setting the temperature 10 K above melting point of PCM could give the efficiency of heat transfer up to 192.72% during charging and 13.07% during discharging.
• Based on the characteristics above, preferred thermal performance could be designed by carefully at first evaluate the operating condition of the inlet temperature and then followed by the flow rate.