*3.2. System Temperature*

Three system temperatures were examined and processed to obtain the average system temperature data. Figure 5 shows the diurnal variations of the system temperatures on 14 February and 2 June 2021 based on a minute interval. All temperatures increase in the middle of the day and decrease towards the end of the day; however, the trend of MI is distinctly different from those of the two roof temperatures, MR and RR, with the latter trends being quite similar. The trend similarity was also reported by Saleh et al. [33] for the PV module temperature and Chung and Park [34] for the roof temperature. Similar to the case in Figure 3, the difference in temperature between the two days was due to seasonal differences, in South Korea, February is winter and June is summer. The analyzed data from 2020 until the middle of 2022 are represented in Figure 6 and the temperature data from 2020 to 2022 are summarized in Table 3. The trend shown in Figure 6 displays an increase from the start of the year until the middle of the year and a subsequent decline until the end of the year. The variation shown throughout the year is caused by seasonal changes, as South Korea experiences four seasons throughout the year. The increases in system temperatures were correlated with—and caused by—the seasonal changes from a cold winter to a warm spring to an even warmer summer. Subsequently, the system temperature dropped in autumn, and the lowest temperature occurred in winter, thereby completing the annual cycle.

**Figure 5.** Diurnal variations of the system temperatures on (**a**) 14 February and (**b**) 2 June 2021.

**Figure 6.** Average system temperature in (**a**) 2021, (**b**) 2021, and (**c**) 2022.


**Table 3.** Average system temperature from January 2020 to June 2022.

The average value of MI was higher than MR and RR throughout the year, except in the middle of the year, from June to August, when those of RR matched its values. The higher MI compared to the other system temperatures was caused by the activity of the PV modules. The absorption of solar energy in the form of light and its conversion to electricity was conducted at the PV modules' capacity, with the other form of solar energy, i.e., heat energy, being absorbed. Heat absorption was subsequently aggravated by the dark color of the PV modules, thereby increasing the heat absorption capacity of the PV modules [11,12]. A high MI value is characteristic of most BIPV systems, as roofing systems encapsulate them without enough air ventilation compared to the cases of conventionally installed PV modules and more traditional BAPV systems [11,12].

The MR values were always the lowest among the values of the three measured temperatures and were always approximately 1–2 ◦C lower than the RR values and 1–5 ◦C lower than the MI values. The annual differences between RR and MR are shown in Figure 7, where MR is almost always lower than RR throughout the 2.5-year observation period, except in July 2021. The small differences between MR and RR were likely due to the different roof tile configurations used in this system when the PV modules were installed on top of the tiles. The installation of PV modules on the roof tiles left a little gap which facilitated the heat transfer caused by more wind. In contrast, the original roof tiles retained more heat because the gap was nonexistent [35–37]. The finding implies that the PV system could be installed as the entire roofing system with this configuration without sacrificing the roof temperature.

**Figure 7.** Monthly differences between the roof rear temperature (RR) and module rear temperature (MR) throughout the observation period.
