*2.2. System Design*

The battery pack design in this study is shown in Figure 1. The interconnection of the battery cells developed within the battery pack created an electrical and mechanical connection, and the casing covered the mechanical requirements for the battery pack. The battery pack also consisted of stiffeners and acrylic to support the battery pack under static and dynamic loads transferred from an electric frame or chassis. The conductor plate was used to flow electricity from the battery to other electrical components and vice versa. The BMS, which could be an active or passive balancer, was placed inside the BMS casing. The inlet and outlet of the airflow were created on the lateral side of the battery pack casing to investigate the effect of forced convection by the cooling fan on the temperature distribution within the battery pack. The battery cells used in this battery pack were lithium nickel manganese cobalt oxide (NMC) 18,650 cylindrical batteries, and the specification is presented in Table 1. The 18,650 cylindrical battery type was employed due to its popularity in its application for electric vehicle battery. This module was suitable for electric trike

and city cars. The electric trike, in particular, is currently being developed in our research facilities and has been presented in several studies, such as by Reksowardojo et al. [34].

**Figure 1.** Battery pack design and its simplification model.

The original geometry of the battery pack was complex, so it was difficult to configure the airflow within the battery pack iteratively with the Ansys Discovery Live software. This situation could also lead to continuous errors and long simulation time. To achieve the optimum results using the software, the battery pack was modeled as a simplified form without removing the components sensitive to the heat generation within the battery pack. The battery cells themselves consisted of layers of cathode, anode, separator, and current collector. Although these different layers had their thermal properties, the detailed structure of the cylindrical cell presented in Figure 1 had an insignificant impact on the thermal performance of the battery, according to many references [35,36]. Therefore, the battery parameters listed in Table 1 were utilized as the equivalent values representing a whole battery cell.

The simplified model of the battery pack was created in the form of 240 cells enclosed by a box, as shown in Figure 1. Each battery cell is represented as a cylinder with a diameter of 18.3 mm and a height of 64.5 mm. The offset between the 240 cylinders and the box was 1 mm on each surface. This close proximity was applied to ensure the designed battery pack had a high density and compact geometry as space and weight in EV are the main constraints to increase energy density [37]. The distance between the center of the two adjacent cylinders was 20 mm, and the gap between them was 1.7 mm. The pack would act as the fluid, and the inlet came from one lateral side and the outlet on the other.

Forced convection utilized in this system was carried out using a cooling fan. The available area for the cooling fan was 240 mm × 40 mm; therefore, the maximum dimension of the cooling fan was 40 mm × 40 mm. The selected cooling fan was a chip cooler AP0405MX-J70 from Adda Corp. with a rated power of 0.7 W and a maximum airflow of 4.7 CFM or 7.99 m3/h.

The effect of fluid flow within the battery pack on the temperature distribution was studied by conducting LBM-based transient CFD simulations. For the simulation, the airflow inlet and outlet were positioned at the lateral opposing end of the battery pack, while the other sides were set as the wall. Each inlet was accompanied by one cooling fan. For simplicity, from this point forth, the number of cooling fans and inlets will be called the number of inlets only, without mentioning the cooling fans. The number of inlets was varied into four configurations of 1-, 2-, 3-, and 4-inlets to investigate the effect of airflow configuration on temperature distribution inside the battery pack. Afterward, the inlet air temperature was varied by 20, 25, and 30 ◦C for each airflow configuration to investigate the effect of inlet air temperature on temperature distribution and to calculate the power required for each variation. Finally, the configuration with minimum temperature distribution and power consumption was chosen as the optimum cooling strategy.
