Comparative Analysis of System Performance and Thermal Comfort for an Integrated System with PVT and GSHP Considering Two Load Systems: Convective Heating and Radiant Floor Heating

The zero-energy building (ZEB) concept has a high potential for securing energy savings in the building sector. To achieve ZEB, various active systems, including renewable systems such as photovoltaic, solar heating, and geothermal systems, have been developed. However, the existing systems are costly or not optimized. To overcome these issues, the authors previously developed an integrated tri-generation system. In this research, the previously developed system was comprehensively analyzed considering the indoor thermal comfort and energy efficiency to develop a design and operation method for the integrated system. Two different heating systems (convective heating and radiant floor heating) were employed in the tri-generation system, and their system performance, predicted mean vote (PMV), and predicted percentage of dissatisfied (PPD) were compared using simulations. The results showed that the heating coefficient of power of the radiant floor heating system was 18.8% higher than that of the convective heating system. Moreover, the radiant floor heating system (Case 4) met the PMV and PPD standards during all the heating periods. Overall, radiant floor heating was found to be more efficient than convective heating. The results confirm that radiant floor heating is more suitable than convective heating considering the indoor thermal comfort of occupants.


Introduction
In recent years, issues in energy saving in the building sector have become significant from the aspects of not only global warming and energy security but also economic stability. Energy cost accounts for 50% of the total heating, ventilation, and air conditioning (HVAC) system cost of the building life cycle [1]. Therefore, energy costs must be considered during the building design stage to ensure energy saving in the building sector. The zero-energy building (ZEB) concept is one of the most suitable approaches aimed at achieving this purpose. Many leading countries have actively participated in the dissemination of ZEB through policy support or financial incentives. In the U.S., Executive Order (EO) 13,834 requires all new federal buildings entering the planning process in 2020 to be designed to achieve annual reductions in building energy use and implement energy efficiency measures that reduce costs [2]. In South Korea, ZEB became mandatory for new public buildings from 2020, and this will be mandated for new private buildings by 2025 [3]. The expected market size of ZEB in Korea will also reach up to 17 billion dollars by 2024 [4]. Under this background, various active and energy efficiency. To this end, an integrated simulation was performed with building, individual equipment, and weather models. In the integrated renewable energy system, the PVT and GSHP systems were used as energy sources, and two different heating systems were compared in terms of the system performance and PMV based on annual dynamic energy simulation. Figure 1 shows a summary of the integrated simulation model used in this research. The system has PVT and GSHP as heat source systems, a fan coil unit (FCU) and floor heating as space heating systems, and an FCU as a cooling system. The integrated model has a thermal storage tank for storing solar and ground heat. The building is based on ISO 13,790 and the calculation condition presented in [18]. The energy efficiency was calculated based on the total energy production and consumption by all equipment in the system. Furthermore, the PMV was calculated by the indoor air temperature, air velocity, humidity, mean radiation temperature (MRT), metabolic rate (MET), and clothing factor (Clo.), which are based on ISO 7730 [22].

Overview of the Integrated Simulation Model
Energies 2020, 13 , x  3 of 18 building, individual equipment, and weather models. In the integrated renewable energy system, the PVT and GSHP systems were used as energy sources, and two different heating systems were compared in terms of the system performance and PMV based on annual dynamic energy simulation. Figure 1 shows a summary of the integrated simulation model used in this research. The system has PVT and GSHP as heat source systems, a fan coil unit (FCU) and floor heating as space heating systems, and an FCU as a cooling system. The integrated model has a thermal storage tank for storing solar and ground heat. The building is based on ISO 13,790 and the calculation condition presented in [18]. The energy efficiency was calculated based on the total energy production and consumption by all equipment in the system. Furthermore, the PMV was calculated by the indoor air temperature, air velocity, humidity, mean radiation temperature (MRT), metabolic rate (MET), and clothing factor (Clo.), which are based on ISO 7730 [22]. The PVT model, thermal storage model, heat pump model, and ground heat exchanger (GHX) model used in the integrated simulation are based on the tri-generation model built by the author in a previous study [7]. In this study, an convective heating system model using an FCU and a radiant floor heating system was constructed to analyze the energy efficiency and thermal comfort according to the heating type and operating conditions. Figure 2 shows a schematic of the integrated model for the convective heating system. Figure 3 shows the control strategy of the integrated system. The integrated system is operated based on the load of the building and the indoor setpoint temperature. The operation of each unit (PVT module and heat pump) is determined by the setpoint tempearture of the heat storage tank (HST) and outlet temperature of the PVT module. Figure 3 shows the control strategy of the integrated system. The control strategy includes the heating operation and heat storage operation. A heating operation is performed when the average temperature of the heat storage tank is 45 °C or higher.

Overview of the Integrated Simulation Model
The heat storage operation using the PVT module is conducted when the outlet temperature of the PVT module is 2 °C higher than the average temperature of the HST. If the heat storage operation cannot be conducted using the PVT module, the heat storage operation is performed with the heat pump. At this time, the heat pump is used the solar-thermal storage tank (SST) and GHX; the heat pump is utilized with a higher outlet temperature between SST and GHX. The PVT module stores heat to the SST when the HST average temperature is above 45 °C. The PVT model, thermal storage model, heat pump model, and ground heat exchanger (GHX) model used in the integrated simulation are based on the tri-generation model built by the author in a previous study [7]. In this study, an convective heating system model using an FCU and a radiant floor heating system was constructed to analyze the energy efficiency and thermal comfort according to the heating type and operating conditions. Figure 2 shows a schematic of the integrated model for the convective heating system. Figure 3 shows the control strategy of the integrated system. The integrated system is operated based on the load of the building and the indoor setpoint temperature. The operation of each unit (PVT module and heat pump) is determined by the setpoint tempearture of the heat storage tank (HST) and outlet temperature of the PVT module. Figure 3 shows the control strategy of the integrated system. The control strategy includes the heating operation and heat storage operation. A heating operation is performed when the average temperature of the heat storage tank is 45 • C or higher.   Figure 4 shows the structure of the system of thermo-active construction elements. Thermoactive elements, such as the radiant floor model, are used to condition buildings by integrating a fluid system into the building.   Figure 4 shows the structure of the system of thermo-active construction elements. Thermoactive elements, such as the radiant floor model, are used to condition buildings by integrating a fluid system into the building. The heat storage operation using the PVT module is conducted when the outlet temperature of the PVT module is 2 • C higher than the average temperature of the HST. If the heat storage operation cannot be conducted using the PVT module, the heat storage operation is performed with the heat pump. At this time, the heat pump is used the solar-thermal storage tank (SST) and GHX; the heat pump is utilized with a higher outlet temperature between SST and GHX. The PVT module stores heat to the SST when the HST average temperature is above 45 • C. Figure 4 shows the structure of the system of thermo-active construction elements. Thermo-active elements, such as the radiant floor model, are used to condition buildings by integrating a fluid system into the building. A two-dimensional temperature field occurs in the plane of the cross-section of the thermally active component due to the finite distance between the pipes. Thermal input or output through the pipe loop changes the water temperature in the pipe [23]. A stationary solution for the temperature distribution in the x-y plane is represented by Equation (1) for heat flow from the surface towards the indoor space:

Analysis Method (Equation) of Integrated Simulation Model
The calculation of MRT was defined by Fanger [24]. MRT is the uniform temperature of an imaginary enclosure in which the radiant heat transfer in the actual nonuniform enclosure. MRT varies according to the person's posture and the color of clothing for short-wave radiation, and calculates based on a person in relation to the surrounding wall and the surface temperature of the enclosure [25].
where is the surface temperature of the area and is the angular factor between a person and the area.
The calculation model of PMV is based on ASHRAE(American Society of Heating, Refrigerating and Air-Conditioning Engineers) Standard 55-2013 [26]. This PMV model can be calculated for a wide range of six parameters (air temperature, MRT, relative humidity, air velocity, Clo., and MET) based on the thermal load of a body in the environment: = 0.202 .
where PMV is the predicted mean vote, is the thermal load, is the metabolic rate, is the rate rate of mechanical work accomplished, is the total rate of evaporative heat loss from the skin, is the rate of evaporative heat loss from respiration, is the rate of convective heat loss from respiration, and are the heat losses, is the respiratory quotient, is the volumetric rate A two-dimensional temperature field occurs in the plane of the cross-section of the thermally active component due to the finite distance between the pipes. Thermal input or output through the pipe loop changes the water temperature in the pipe [23]. A stationary solution for the temperature distribution in the x-y plane is represented by Equation (1) for heat flow from the surface towards the indoor space: The calculation of MRT was defined by Fanger [24]. MRT is the uniform temperature of an imaginary enclosure in which the radiant heat transfer in the actual nonuniform enclosure. MRT varies according to the person's posture and the color of clothing for short-wave radiation, and calculates based on a person in relation to the surrounding wall and the surface temperature of the enclosure [25].
where T n is the surface temperature of the area and F p is the angular factor between a person and the area. The calculation model of PMV is based on ASHRAE(American Society of Heating, Refrigerating and Air-Conditioning Engineers) Standard 55-2013 [26]. This PMV model can be calculated for a wide range of six parameters (air temperature, MRT, relative humidity, air velocity, Clo., and MET) based on the thermal load of a body in the environment: A D = 0.202m 0.425 l 0.725 , where PMV is the predicted mean vote, L is the thermal load, M is the metabolic rate, W is the rate of mechanical work accomplished, E s is the total rate of evaporative heat loss from the skin, E re is the rate of evaporative heat loss from respiration, C re is the rate of convective heat loss from respiration, R and C are the heat losses, RQ is the respiratory quotient, Qo 2 is the volumetric rate of oxygen consumption at 0 • C, A D is the DuBois surface area, m is the person's weight, and l is the person's height. After calculating PMV, the predicted percentage dissatisfied (PPD) can also be estimated as: In ASHRAE Standard 55-2013, the satisfaction range of PMV is ±0.5 (−0.5 < PMV < +0.5), and the range of PPD is below 10% (PPD < 10%) [26].
The heat pump model is based on a user-supplied file [27]. This file contains catalog data for the source-side and load-side flow rate, entering load temperature, entering source temperature, capacity, and power. The heating and cooling coefficient of performance (COP) of the heat pump can be calculated by Equations (11) and (12) [27].
where COP HP,heating is the heat pump COP in heating mode, Q absorbed is the energy absorbed by the heat pump in heating mode, P heating is the power drawn by the heat pump in heating mode, COP HP,cooling is the heat pump COP in cooling mode, Q rejected is the energy rejected by the heat pump in cooling mode, and P cooling is the power drawn by the heat pump in cooling mode. The heat pump is controlled by fluid flow through absorbing energy in the heating mode or rejecting energy in the cooling mode. COP was calculated considering the total heat production and total energy consumption. The total heat production is the sum of the heat production of the photovoltaic-thermal module and the heat pump. The total energy consumption is the sum of the power consumption of the FCU, the circulating pumps, and the heat pump.
where COP SYS is the coefficient of the performance of the system, Q PVT is the heat production by the photovoltaic-thermal module, Q HP is the heat production by the heat pump, P FCU is the power consumption of the fan coil unit, P pump is the sum of the power consumption of all circulating pumps, and P HP is the power consumption of the heat pump. The thermal and electrical performance of the PVT module model was calculated based on the solar radiation, module area, electrical efficiency, and slope. The calculation method of the PVT module model is based on the book "Solar Engineering of Thermal Process," described by Duffie and Beckman [28]. The GHX model is based on the duct ground heat storage (DST) developed by Hellstrom [29]. In addition, the heat storage tank can be modeled assuming that the tanks are made up of equal volume segments [30]. The degree of stratification and tank volume are determined by the user. More details of the system unit (PVT module, heat pump, GHX, and HST) can be found in the author's previous study [7]. the ceiling was 2.4 m. The window to wall ratio (WWR) according to the cardinal direction was 1.2% in the east, 9.9% in the north, and 14.1% in the south.

Building Model Conditions
Energies 2020, 13, x 7 of 18 ceiling was 2.4 m. The window to wall ratio (WWR) according to the cardinal direction was 1.2% in the east, 9.9% in the north, and 14.1% in the south.  Table 1 shows the U-values of the building model, which are based on the "Green Buildings Construction Support Act" proposed by the Korean Ministry of Land, Infrastructure, and Transport [3]. The U-values were input assuming the building model to be located in Seoul.  Figure 6 shows the shematic diagram of the floor structure. The floor structure is composed of a finishing material, mortar, a pipe, insulation, and a concrete slab. Table 2 shows the thermal properties of the construction material.   Table 1 shows the U-values of the building model, which are based on the "Green Buildings Construction Support Act" proposed by the Korean Ministry of Land, Infrastructure, and Transport [3]. The U-values were input assuming the building model to be located in Seoul.  Figure 6 shows the shematic diagram of the floor structure. The floor structure is composed of a finishing material, mortar, a pipe, insulation, and a concrete slab. Table 2 shows the thermal properties of the construction material.   Table 1 shows the U-values of the building model, which are based on the "Green Buildings Construction Support Act" proposed by the Korean Ministry of Land, Infrastructure, and Transport [3]. The U-values were input assuming the building model to be located in Seoul.  Figure 6 shows the shematic diagram of the floor structure. The floor structure is composed of a finishing material, mortar, a pipe, insulation, and a concrete slab. Table 2 shows the thermal properties of the construction material.     Figure 7 shows the schedules of occupants, lights, equipment, and system for a day. The schedule values of internal heat gain were based on ASHRAE 90.1-2004 [32]. The system operated for 24 h.   Figure 7 shows the schedules of occupants, lights, equipment, and system for a day. The schedule values of internal heat gain were based on ASHRAE 90.1-2004 [32]. The system operated for 24 h.  Table 3 shows the indoor conditions of the building model. The occupants were assumed to be seated and relaxed. The number of occupants was set to four, considering the floor area. In addition, the lighting power density (LPD), equipment power density (EPD), and infiltration were set based on ASHRAE 90.   Table 4 shows the thermal comfort conditions. The occupants, who were wearing simple clothes, were assumed to be seated and relaxed.  Table 3 shows the indoor conditions of the building model. The occupants were assumed to be seated and relaxed. The number of occupants was set to four, considering the floor area. In addition, the lighting power density (LPD), equipment power density (EPD), and infiltration were set based on ASHRAE 90.   Table 4 shows the thermal comfort conditions. The occupants, who were wearing simple clothes, were assumed to be seated and relaxed.

Simulation Cases
In this study, a case study was conducted on the heating type and operating conditions. Table 5 shows the case study conditions. The heating type was classified as (1) the convective heating system using FCU and (2) the radiant floor heating system. Thermal comfort, according to the heating type, was analyzed through Case 1 and Case 2. The indoor setpoint temperatures for Case 1 and Case 2 were based on the ASHRAE 90.2-2004 single zone [33]. Case 3 and Case 4 focused on the thermal comfort of the occupant rather than the energy efficiency of the system. For the indoor setpoint temperatures for Case 3 and Case 4, the range of comfortable heating and cooling temperatures suggested by ASHRAE standard 55-2013 was considered [26].  Figure 8 shows the ambient temperature of Seoul and the monthly energy demand of the building model. The annual average temperature in Seoul is 11.8 • C; the minimum is −14.5 • C and the maximum is 33.8 • C. The annual heating and cooling loads were calculated to be 11,039 kWh and 1335 kWh, respectively. Furthermore, the heating and cooling peaks were 9.35 and 2.72 kW, respectively. Therefore, the building model of the case study was assumed to be a heating-dominant building. The system capacity of each component was set according to the results of load calculation, and Table 6 presents the specific information of the system.

Simulation Cases
In this study, a case study was conducted on the heating type and operating conditions. Table 5 shows the case study conditions. The heating type was classified as (1) the convective heating system using FCU and (2) the radiant floor heating system. Thermal comfort, according to the heating type, was analyzed through Case 1 and Case 2. The indoor setpoint temperatures for Case 1 and Case 2 were based on the ASHRAE 90.2-2004 single zone [33]. Case 3 and Case 4 focused on the thermal comfort of the occupant rather than the energy efficiency of the system. For the indoor setpoint temperatures for Case 3 and Case 4, the range of comfortable heating and cooling temperatures suggested by ASHRAE standard 55-2013 was considered [26].  Figure 8 shows the ambient temperature of Seoul and the monthly energy demand of the building model. The annual average temperature in Seoul is 11.8 °C; the minimum is −14.5 °C and the maximum is 33.8 °C. The annual heating and cooling loads were calculated to be 11,039 kWh and 1335 kWh, respectively. Furthermore, the heating and cooling peaks were 9.35 and 2.72 kW, respectively. Therefore, the building model of the case study was assumed to be a heating-dominant building. The system capacity of each component was set according to the results of load calculation, and Table 6 presents the specific information of the system.    Figure 9 shows the thermal and electrical performance of the system on a representative day (January 28) in winter. The heat exchange rate (HER) of the heat pump was calculated as 105 and 110 kW for convective heating (Case 1) and radiant floor heating (Case 2), respectively. There was no significant difference in the amount of heat and electricity produced by the PVT collector according to the heating type.   Figure 9 shows the thermal and electrical performance of the system on a representative day (January 28) in winter. The heat exchange rate (HER) of the heat pump was calculated as 105 and 110 kW for convective heating (Case 1) and radiant floor heating (Case 2), respectively. There was no significant difference in the amount of heat and electricity produced by the PVT collector according to the heating type. Under the same heating load condition, the convective heating (Case 1) system operated continuously, and the radiant floor heating (Case 2) system operated intermittently. The electricity consumption of the heat pump was calculated as 24.3 kW for convective heating (Case 1) and 20.9 kW for radiant floor heating (Case 2). The operation time of the heat pump was 50% longer for convective heating than for radiant floor heating. More heat was produced via radiant floorheating Under the same heating load condition, the convective heating (Case 1) system operated continuously, and the radiant floor heating (Case 2) system operated intermittently. The electricity consumption of the heat pump was calculated as 24.3 kW for convective heating (Case 1) and 20.9 kW for radiant floor heating (Case 2). The operation time of the heat pump was 50% longer for convective heating than for radiant floor heating. More heat was produced via radiant floorheating than via convective heating, although the operating time of the heat pump was shorter. Figure 10 shows the COP of the heat pump on a representative day for heating. The average COP of the heat pump on the representative day of heating was calculated as 4.3 and 5.2 for convective heating (Case 1) and radiant floor heating (Case 2), respectively. The heating COP of radiant floor heating was 18.8% higher than that of convective heating, indicating that the heating performance of the heat pump of the radiant floor heating was more efficient. For the cooling COP, there was no difference between the two systems. Under the same heating load condition, the convective heating (Case 1) system operated continuously, and the radiant floor heating (Case 2) system operated intermittently. The electricity consumption of the heat pump was calculated as 24.3 kW for convective heating (Case 1) and 20.9 kW for radiant floor heating (Case 2). The operation time of the heat pump was 50% longer for convective heating than for radiant floor heating. More heat was produced via radiant floorheating than via convective heating, although the operating time of the heat pump was shorter. Figure 10 shows the COP of the heat pump on a representative day for heating. The average COP of the heat pump on the representative day of heating was calculated as 4.3 and 5.2 for convective heating (Case 1) and radiant floor heating (Case 2), respectively. The heating COP of radiant floor heating was 18.8% higher than that of convective heating, indicating that the heating performance of the heat pump of the radiant floor heating was more efficient. For the cooling COP, there was no difference between the two systems.  Figure 11 shows the HER, electric consumption, heat pump COP, and system COP from January 12 to January 18. The average COP of the heat pumps for convective heating (Case 1 and Case 3) and radiant floor heating (Case 2 and Case 4) was calculated as 4.3, 5.3, 4.3, and 5.3, respectively. The average COP of the system for convective heating (Case 1 and Case 3) and radiant floor heating (Case 2 and Case 4) was calculated as 3.5, 4.5, 3.5, and 4.5, respectively. The heating type had a more substantial influence on the COP of the heat pump and the system than the indoor set temperature. Table 7 shows a summary of the average COP in Figure 11c,d.  Figure 11 shows the HER, electric consumption, heat pump COP, and system COP from January 12 to January 18. The average COP of the heat pumps for convective heating (Case 1 and Case 3) and radiant floor heating (Case 2 and Case 4) was calculated as 4.3, 5.3, 4.3, and 5.3, respectively. The average COP of the system for convective heating (Case 1 and Case 3) and radiant floor heating (Case 2 and Case 4) was calculated as 3.5, 4.5, 3.5, and 4.5, respectively. The heating type had a more substantial influence on the COP of the heat pump and the system than the indoor set temperature. Table 7 shows a summary of the average COP in Figure 11c,d. Figure 12 shows the monthly performance of the heat pump and the system. The average COP of the heat pump in the winter season was calculated as 4.6, 5.4, 4.6, and 5.4 from Case 1 to Case 4, respectively. In addition, the average system COP was calculated as 4.5, 5.3, 4.3, and 5.2. The heating performance varied with the heating type. The heat pump COP and system COP for radiant floor heating were 18.7% and 21.8% higher than those for convective heating. These results clearly show the advantage of radiant floor heating over convective heating in terms of the HER and electricity consumption (operation time), as shown in Figure 9.   Figure 12 shows the monthly performance of the heat pump and the system. The average COP of the heat pump in the winter season was calculated as 4.6, 5.4, 4.6, and 5.4 from Case 1 to Case 4, respectively. In addition, the average system COP was calculated as 4.5, 5.3, 4.3, and 5.2. The heating performance varied with the heating type. The heat pump COP and system COP for radiant floor heating were 18.7% and 21.8% higher than those for convective heating. These results clearly show the advantage of radiant floor heating over convective heating in terms of the HER and electricity consumption (operation time), as shown in Figure 9.    11. Weekly performance of the heat pump and system.  Figure 12 shows the monthly performance of the heat pump and the system. The average COP of the heat pump in the winter season was calculated as 4.6, 5.4, 4.6, and 5.4 from Case 1 to Case 4, respectively. In addition, the average system COP was calculated as 4.5, 5.3, 4.3, and 5.2. The heating performance varied with the heating type. The heat pump COP and system COP for radiant floor heating were 18.7% and 21.8% higher than those for convective heating. These results clearly show the advantage of radiant floor heating over convective heating in terms of the HER and electricity consumption (operation time), as shown in Figure 9.   Figure 13a shows the indoor air temperatures (T a ) of convective heating and radiant floor heating on a representative day in winter (28 January). The indoor air temperature of convective heating was calculated as 21.9 • C and 22.2 • C for Case 1 and Case 3, respectively. In addition, the indoor air temperature of radiant floor heating was calculated as 19.9 • C and 21.9 • C for Case 2 and Case 4, respectively. For radiant floor heating, the indoor air temperature was confirmed to be similar to Energies 2020, 13, 5524 13 of 19 the indoor setpoint temperature. For convective heating, the indoor air temperature was similar to or higher than the indoor setpoint temperature.

Thermal Environmental Parameters
heating, the indoor air temperature was higher than the indoor set-point temperature by approximately 1 °C. This is because the convective heating system can supply high-temperature heat to the zone in a short time compared to the radiant floor heating system. In contrast, for radiant floor heating, the indoor air temperatures were 20.6 and 22.3 °C in Case 2 and Case 4, respectively. Figure 15b shows the monthly average MRT for one year. The changes in MRT, according to the heating type, were more pronounced at higher heating loads (January, February, and December). This is because the radiant floor heating system takes more time to heat the floor structure and supply radiant heat to the space at a higher heating load. On the other hand, the MRT was similar for radiant floor heating and convective heating at low heating loads (April, May, and September).   Figure 13b shows the indoor air temperatures (T a ) of the convective heating and radiant floor heating for a week period (January 25-31). Convective heating exhibited a larger variation in indoor air temperature than radiant floor heating. On the other hand, the indoor air temperature of the radiant floor heating is almost constant. This is because the heat output from the heated floor is controlled according to the indoor air temperature due to the self-control effect of radiant floor heating [34]. Figure 14a shows the MRT on a representative day. Changes in the MRT of convective heating and radiant floor heating were similar to those of indoor air temperature ( Figure 12). For radiant floor heating, MRT was higher than the indoor air temperature by approximately 1.3 • C. In contrast, for convective heating MRT was lower than the indoor air temperature by approximately 1.4 • C.   Figure 16a shows the PMV on a representative day. The red line represents the one of the ranges of the PMV standard (−0.5 < PMV < 0.5) proposed by ASHRAE Standard 55-2013 [22]. The cases of convective heating show that the PMV was below −0.5 throughout the day and did not reach the thermal comfort zone. PMV was higher with radiant floor heating than with convective heating. In particular, the average PMV of Case 4 for radiant floor heating met the thermal comfort zone.  The radiant floor heating system is a higher MRT than the convective heating system because of the increased radiant heat exchange. Figure 15a shows the monthly average indoor air temperature (T a ) for one year. For convective heating, the indoor air temperature was higher than the indoor set-point temperature by approximately 1 • C. This is because the convective heating system can supply high-temperature heat to the zone in a short time compared to the radiant floor heating system. In contrast, for radiant floor heating, the indoor air temperatures were 20.6 and 22.3 • C in Case 2 and Case 4, respectively.   Figure 16a shows the PMV on a representative day. The red line represents the one of the ranges of the PMV standard (−0.5 < PMV < 0.5) proposed by ASHRAE Standard 55-2013 [22]. The cases of convective heating show that the PMV was below −0.5 throughout the day and did not reach the thermal comfort zone. PMV was higher with radiant floor heating than with convective heating. In particular, the average PMV of Case 4 for radiant floor heating met the thermal comfort zone. Figure 16b shows PMV for a week. The average PMVs in Case 1 to Case 4 were estimated at −0.8, −0.9, −0.7, and −0.4, respectively. The MRT levels for radiant floor heating (Case 2) and convective heating (Case 3) were similar; however, the PMV was higher for convective heating (Case 3) than for radiant floor heating (Case 2). Despite the higher MRT, the PMV decreased with a decreasing indoor air temperature. On the other hand, convective heating (Case 3) reached a period of satisfying PMV on January 27. However, most other periods did not meet the PMV standard. Radiant floor heating (Case 4) satisfied the PMV for a week. These results confirm that the radiant floor heating system is superior to the convective heating system in terms of indoor thermal comfort. Figure 17a shows the predicted percentage of dissatisfied (PPD) on a representative day. The red line represents the PPD standard (PPD < 10%) suggested by ASHRAE Standard 55-2013 [22]. From Case 1 to Case 4, the PPDs were 18.6%, 24.6%, 16.5%, and 9.5%, respectively. With radiant floor heating (Case 4), though the PPD standard (PPD < 10%) was exceeded once, the standard was sustained throughout the most of the time. Figure 17 shows the PPD for a week. For convective heating (Case 3), the standard was met once but was not satisfied during the remaining period. The radiant floor heating system satisfied the PPD standard only when the indoor set-point temperature was 22 °C. Even when the indoor air temperature reached a generally acceptable level, it hardly met the PPD standard at a low MRT. Figure 18a illustrates the monthly average PMV. Neither convective heating (Case 1) nor radiant floor heating (Case 2) met the PMV standard during the heating periods, except in May and October, when the heating load was low. Meanwhile, convective heating (Case 3) did not meet the PMV standard, even though the average indoor air temperature exceeded 22 °C in January and February.  Figure 15b shows the monthly average MRT for one year. The changes in MRT, according to the heating type, were more pronounced at higher heating loads (January, February, and December). This is because the radiant floor heating system takes more time to heat the floor structure and supply radiant heat to the space at a higher heating load. On the other hand, the MRT was similar for radiant floor heating and convective heating at low heating loads (April, May, and September). Figure 16a shows the PMV on a representative day. The red line represents the one of the ranges of the PMV standard (−0.5 < PMV < 0.5) proposed by ASHRAE Standard 55-2013 [22]. The cases of convective heating show that the PMV was below −0.5 throughout the day and did not reach the thermal comfort zone. PMV was higher with radiant floor heating than with convective heating. In particular, the average PMV of Case 4 for radiant floor heating met the thermal comfort zone. In contrast, radiant floor heating (Case 4) met the PMV standard during all the heating periods. The convective heating system did not meet the PMV standard due to the low MRT. Figure 18b summarizes the monthly average PPD. Both convective heating (Case 1) and radiant floor heating (Case 2) failed to meet the required PPD standard during all the heating periods. Convective heating (Case 3) did not meet the required PPD either in January, February, or December, whereas radiant floor heating (Case 4) met the PPD criteria in all the heating periods. This result confirms that radiant floor heating is more suitable for the indoor thermal comfort of occupants than convective heating.    and convective heating (Case 3) were similar; however, the PMV was higher for convective heating (Case 3) than for radiant floor heating (Case 2). Despite the higher MRT, the PMV decreased with a decreasing indoor air temperature. On the other hand, convective heating (Case 3) reached a period of satisfying PMV on January 27. However, most other periods did not meet the PMV standard. Radiant floor heating (Case 4) satisfied the PMV for a week. These results confirm that the radiant floor heating system is superior to the convective heating system in terms of indoor thermal comfort. Figure 17a shows the predicted percentage of dissatisfied (PPD) on a representative day. The red line represents the PPD standard (PPD < 10%) suggested by ASHRAE Standard 55-2013 [22]. From Case 1 to Case 4, the PPDs were 18.6%, 24.6%, 16.5%, and 9.5%, respectively. With radiant floor heating (Case 4), though the PPD standard (PPD < 10%) was exceeded once, the standard was sustained throughout the most of the time.

Conclusions
An integrated simulation was performed to develop the design and operation method for a renewable energy system considering the indoor thermal comfort and energy efficiency. The energy efficiency and indoor thermal comfort of two different heating systems were compared based on dynamic energy simulations. The conclusions of this study can be summarized as follows.
(1) The average COPs of the heat pump on a representative day in the winter season were 4.3 and 5.2 for convective heating (Case 1) and radiant floor heating (Case 2), respectively. The heating  Figure 17 shows the PPD for a week. For convective heating (Case 3), the standard was met once but was not satisfied during the remaining period. The radiant floor heating system satisfied the PPD standard only when the indoor set-point temperature was 22 • C. Even when the indoor air temperature reached a generally acceptable level, it hardly met the PPD standard at a low MRT. Figure 18a illustrates the monthly average PMV. Neither convective heating (Case 1) nor radiant floor heating (Case 2) met the PMV standard during the heating periods, except in May and October, when the heating load was low. Meanwhile, convective heating (Case 3) did not meet the PMV standard, even though the average indoor air temperature exceeded 22 • C in January and February. In contrast, radiant floor heating (Case 4) met the PMV standard during all the heating periods. The convective heating system did not meet the PMV standard due to the low MRT.
In contrast, radiant floor heating (Case 4) met the PMV standard during all the heating periods. The convective heating system did not meet the PMV standard due to the low MRT. Figure 18b summarizes the monthly average PPD. Both convective heating (Case 1) and radiant floor heating (Case 2) failed to meet the required PPD standard during all the heating periods. Convective heating (Case 3) did not meet the required PPD either in January, February, or December, whereas radiant floor heating (Case 4) met the PPD criteria in all the heating periods. This result confirms that radiant floor heating is more suitable for the indoor thermal comfort of occupants than convective heating.

Conclusions
An integrated simulation was performed to develop the design and operation method for a renewable energy system considering the indoor thermal comfort and energy efficiency. The energy efficiency and indoor thermal comfort of two different heating systems were compared based on dynamic energy simulations. The conclusions of this study can be summarized as follows.
(1) The average COPs of the heat pump on a representative day in the winter season were 4.3 and 5.2 for convective heating (Case 1) and radiant floor heating (Case 2), respectively. The heating  Figure 18b summarizes the monthly average PPD. Both convective heating (Case 1) and radiant floor heating (Case 2) failed to meet the required PPD standard during all the heating periods. Convective heating (Case 3) did not meet the required PPD either in January, February, or December, whereas radiant floor heating (Case 4) met the PPD criteria in all the heating periods. This result confirms that radiant floor heating is more suitable for the indoor thermal comfort of occupants than convective heating.

Conclusions
An integrated simulation was performed to develop the design and operation method for a renewable energy system considering the indoor thermal comfort and energy efficiency. The energy efficiency and indoor thermal comfort of two different heating systems were compared based on dynamic energy simulations. The conclusions of this study can be summarized as follows.
(1) The average COPs of the heat pump on a representative day in the winter season were 4.3 and 5.2 for convective heating (Case 1) and radiant floor heating (Case 2), respectively. The heating COP for radiant floor heating was 18.8% higher than that for convective heating. Overall, the radiant floor heating system was more efficient than the convective heating system. (2) The radiant floor heating system was more effective for MRT than the convective heating system.
The MRT showed a difference of 2.3 • C according to the heating type. The MRT difference was larger in January, February, and December, when the heating load was high. This is because the available time for radiant floor heating to heat the floor structure increased with the increasing heating load. However, the MRT was similar for both radiant floor heating and convective heating at low heating loads. (3) Although the MRT was higher for radiant floor heating (Case 1) than for convective heating (Case 3), the PMV standard (−0.5 < PMV < 0.5) was not met when the indoor air temperature was low. This result confirms that MRT has a strong influence on PMV only if an appropriate level of indoor air temperature (22 • C) is secured. For convective heating (Case 3), the standards of PMV and PPD were not met in January and February when the heating load was high. For radiant floor heating (Case 4), the PMV and PPD standards were met in all heating periods. This result confirms that the radiant floor heating system is more suitable than the convective heating system for the indoor thermal comfort of occupants.
Author Contributions: The authors S.B. and Y.N. wrote the full manuscript and comparatively analyzed the energy efficiency between convective heating and radiant floor heating. The author J.-H.C. analyzed the thermal comfort according to the heating type of the tri-generation system and reviewed this manuscript. All authors have read and agreed to the published version of the manuscript.