Indoor Environment Improvement and Energy-Saving Effects of Light Shelf System with Integrated Radiant Heating and Cooling Panel

Abstract

Achieving good daylighting while maintaining thermal comfort and reducing perimeter energy use is a key challenge in low-energy office buildings. This study developed a thermally activated light shelf (TALS) system that integrates multiple functions into a conventional light shelf. The top surface blocks excessive perimeter light and reflects daylight deeper into the room, while the bottom surface operates as a radiant heating and cooling panel using circulating warm or cool water. To evaluate the system, full-scale empirical experiments were conducted in a mock-up test bed with two identical office-like cells under the same boundary conditions; one cell was equipped with TALS and the other served as a reference. Indoor thermal environment indices and heating and cooling energy use were monitored during winter and summer. The TALS room achieved ISO 7730 Category A comfort more frequently, with Category A cumulative duration approximately 3.4 times longer in winter and 7.8 times longer in summer compared with the non-TALS room. In addition, heating and cooling energy were reduced by about 39.2% and 7.7%, respectively. These promising results are based on a single prototype and climate, and further studies are needed to optimize TALS capacity and window-related heat loss.

1. Introduction

Perimeter zones in modern office buildings with large glazed façades are exposed to significant solar and conductive heat gains and losses. Consequently, designers must simultaneously achieve sufficient daylight, visual and thermal comfort, and reduced heating and cooling energy use. A light shelf is a widely used passive façade component that is installed on a building envelope to block direct sunlight and mitigate glare. It improves the indoor light environment with natural light and reduces the lighting load by allowing daylight to enter the building via the reflective upper surface [1,2,3,4,5]. According to “Green Building Design Practices Manual” the Ministry of Land, Infrastructure, and Transport in Korea, light shelves improve the indoor light environment by dispersing the light that enters the building, and they are included among the component technologies for implementing low-energy buildings [6]. Light shelves are used in buildings with various purposes to improve the indoor light environment and provide lighting energy savings [7,8,9].

Extensive research and development on light shelves have therefore been pursued, mainly focusing on improving daylighting performance and providing lighting energy savings. In most previous studies, the lighting performance was analyzed by season and by hour of day according to the reflectance of the light-shelf surface and the color of the indoor finishing materials, and efficient lighting devices were proposed. Additionally, researchers have proposed improvements to light shelves and ceilings based on the sun’s altitude to improve indoor illumination and uniformity for enhancing the daytime sunlight environment. Thus far, these studies have concentrated on the development of single-function light shelves that are optimized through performance evaluations of various combinations of light-shelf variables such as size, shape, and placement, based on indoor illuminance, uniformity, and glare. Recently, numerous hybrid light shelves have been proposed that incorporate building engineering component technologies to ensure energy-saving performance. For example, some systems integrate PV panels into the light shelf’s reflective surface and evaluate power generation performance according to the panel angles, whereas other systems combine light shelves with shading devices and verify improvements in daylight uniformity, solar radiation blocking, and energy savings through test-bed experiments [10,11,12]. However, these hybrid concepts have still mainly targeted the visual environment and lighting or electrical energy, and have rarely addressed the thermal environment in perimeter zones.

In contrast, the perimeter zones of office buildings, which are generally composed of windows and curtain walls, experience substantial thermal loads [13,14]. In summer, the operative temperature increases owing to elevated window surface temperatures, and in winter, radiative imbalances occur because of temperature differences between the perimeter and interior zones, leading to phenomena such as cold drafts caused by window cooling and local discomfort for occupants. A single daylighting component such as a conventional light shelf cannot resolve these indoor environment control and building energy-saving difficulties, and separate heating and cooling terminals are usually required to treat the perimeter load. This indicates that a gap remains in existing systems: there is a lack of façade-integrated components that can simultaneously regulate daylight and mitigate perimeter thermal loads within a unified device.

From existing studies, it has been suggested that the purpose of light shelves can be expanded to that of radiant heating and cooling panels that are useful for reducing the heating and cooling load along the building perimeter zone, in addition to their original purpose of regulating the indoor light environment [15]. Combining a light shelf with a radiant panel is particularly attractive because (i) the shelf is located near the window where thermal loads are concentrated, (ii) its unobstructed surface can function as an effective radiant emitter or absorber without occupying floor area or interfering with occupancy, and (iii) integrating both functions into a single façade element can reduce system complexity while jointly improving visual and thermal comfort. Nevertheless, there has been little empirical work that quantitatively evaluates the indoor thermal environment and heating and cooling energy impacts of such a hybrid light shelf–radiant panel system under realistic operating conditions.

Therefore, in this study, the top surface of the light shelf is assigned its original function of blocking excessive light in the perimeter zone and allowing light reflected from a highly reflective surface to penetrate deep into the interior of the room. For the bottom surface, we develop a thermally activated light shelf (TALS) system that integrates a hydronic radiant heating and cooling panel into the light shelf based on the circulation of hot and cold water. Full-scale mock-up experiments are performed in a twin test bed with two identical office-like cells—one equipped with the TALS and the other serving as a reference without the TALS—to evaluate the system’s indoor thermal environment improvement effects and building heating and cooling energy-saving effects. Specifically, this study addresses the following research questions: (1) To what extent does the proposed TALS system improve vertical temperature distribution, local discomfort indices, and predicted mean vote (PMV) in the perimeter zone compared with a non-TALS reference? (2) How much heating and cooling energy can be saved by applying the TALS to the building envelope under typical winter and summer operating conditions? We hypothesize that, relative to the reference case, the TALS will (i) increase the cumulative time during which the indoor environment satisfies the ISO 7730 comfort Categories A and B and (ii) reduce the heating and cooling energy demand in the perimeter zone by providing additional radiant heat exchange near the window.

The purpose of this study is to develop and experimentally evaluate a thermally activated light shelf (TALS) system that integrates a hydronic radiant heating and cooling panel into the bottom surface of a conventional light shelf while preserving its daylight-redirecting function on the top surface. Using full-scale mock-up experiments in twin office-like test cells with and without TALS, the study aims to quantify the effects of the proposed system on the indoor thermal environment in perimeter zones—including vertical temperature distribution, local discomfort indices, and predicted mean vote (PMV)—and to assess the associated reductions in heating and cooling energy use when the TALS is applied as a façade-integrated component.

2. Methodology

2.1. TALS System Concept

To design the TALS, a prototype light shelf with an integrated radiant cooling and heating panel was fabricated. The basic configuration of the system was designed, and the structural members for each component were selected. Using this prototype light shelf with a radiant cooling and heating panel, the relationship between solar altitude and louver angle was examined, and preliminary tests of the radiant panel’s cooling and heating performance were carried out (Figure 1).

The TALS incorporates a radiant heating and cooling system, and it consists of a top surface, middle part, and bottom surface. Its size is 1.2 m (L) × 0.6 m (W) × 0.05 m (H), and its materials and system configuration are shown in Figure 2. The top surface is covered with a highly reflective sheet with 95% reflectivity and has the function of allowing the daylight that is reflected off the light shelf to flow into the inner part of the room. In the middle part, the piping that supplies the heating and cooling source was installed over the bottom surface. The piping performs the function of heating or cooling the bottom surface as hot and cold water circulate, and the bottom surface was covered with a copper thermal conduction layer to increase the heating and cooling capacity. In addition, insulation was installed to block the solar heat transmitted from the top surface, and the shelf was designed so that radiant heating and cooling heat could be transmitted from the bottom surface only. The bottom surface was prepared with an aluminum radiative surface that had high thermal conductivity and emissivity; thus, it was capable of promoting radiant heat transfer. The TALS modules were designed to be compatible with the ceiling height and grid of the full-scale test bed. Each module (1.2 m × 0.6 m × 0.05 m) was installed at 2.2 m above the floor under the perimeter skylights in a room with a ceiling height of 2.7 m, thereby maintaining typical room clearance while allowing daylight redirection and radiant exchange in the perimeter zone. The hydronic connection to the air-source heat pump and circulation pumps follows common practice for radiant ceiling panels and fan-coil systems, so that the proposed TALS concept can be implemented using commercially available equipment. Nevertheless, the present prototype was evaluated in a simple open-plan configuration without interior partitions; thus, future work is required to examine installation details, maintenance access, and interaction with lighting and ceiling systems in more complex real office environments.

The dimensions and installation height of the TALS were selected to be compatible with typical office ceiling and façade configurations. A module size of 1200 mm in width, 600 mm in depth, and 50 mm in thickness corresponds to a standard ceiling grid module and a commonly used light-shelf depth, allowing the system to be mounted along the perimeter without conflicting with the ceiling layout. The installation height was chosen so that the bottom of the TALS is above the typical occupant head height while aligning with the window head, thereby maintaining headroom clearance and avoiding direct glare for both seated and standing occupants. Aluminum was adopted for the bottom surface because of its high thermal conductivity and emissivity, consistent with conventional radiant panel design practice, whereas the upper surface was finished with a high-reflectance sheet (approximately 95%) to maximize daylight redirection. A 50 mm insulation layer between the upper and lower parts was included to suppress upward heat loss and to concentrate the heating and cooling output toward the occupied zone. The present prototype was therefore designed to represent a realistic, retrofittable module, although a detailed numerical optimization of its geometry and material properties is left for future work.

The TALS uses a heating and cooling method that involves the direct transfer of heat, and it has the heat dissipation properties of natural convection and radiation. The radiant heating and cooling sources used by the TALS for the mock-up experiments were generated by an air source heat pump, and the pump supplied the heating source (45 °C) and cooling source (15 °C) with a split supply pipe that was connected to the FCU. The supply temperature of the cooling source was set to be higher than the dewpoint temperature (14.8 °C) to prevent condensation, and the flow rate supplied to the TALS was 7 L/min. The HVAC system was controlled to maintain an indoor temperature of 20 °C in winter and 26 °C in summer. The indoor temperature in the test laboratory was regulated by a 1-way FCU with a capacity of 3.5 kW (8.3 L/min), which satisfies the office design load per unit area (150 kcal/h·m²) specified in KS C 9306 [16]. The heat source was provided by a 42 kW heat pump (120 L/min), a heat-source circulation pump (120 L/min), and a chilled/hot water circulation pump (60 L/min). A buffer tank and an expansion tank were installed to keep the piping pressure and the temperature of the heat source constant. Figure 3 shows the bottom surface of the light shelf being heated and cooled by activating the TALS’ radiant heating and cooling function in the winter and summer, respectively. Analysis of the thermal imaging results indicated that heat loss occurred as the heated and cooled thermal sources passed from the air source heat pump through the buffer tank and supply pipes, and there was a small temperature difference due to radiant exchange via the surface of the bottom radiant panel. The heat transfer that occurred in the experimental environment included radiant heat exchange between the TALS’ radiant heating and cooling panel and its surroundings, natural convection due to the indoor vertical temperature difference, and forced convection heat exchange between the heated/cooled air produced by the air conditioning system and the surrounding air. The bottom surface is made of aluminum, which has high thermal conductivity and emissivity, so that radiant heating and cooling can be effectively delivered to the indoor space.

2.2. Specifications of Test Bed

Full-scale test bed experiments were performed to evaluate the proposed TALS system’s indoor thermal environment improvements and heating and cooling energy saving effects (Figure 4). The test bed was located in an education research facility building at C University of Gwangju Metropolitan City in Republic of Korea, and it consisted of test cells of the same size. It was on a plot of land that was not affected by sun shadows due to the surrounding buildings or trees, and it provided building envelope performance that satisfied the passive house thermal transmittance standard presented in the Ministry of Land, Infrastructure, and Transport’s “Building Energy Saving Design Standards” [17,18]; thus, the indoor/outdoor thermal load variation was minimized (

Table 1. Characteristics of Test bed.

Property		Value
Test bed size		10.35 m (W) × 12.4 m (L) × 2.7 m (H)
Test cell size		2.75 m (W) × 6.2 m (L) × 2.7 m (H)
Latitude, longitude, azimuth		35.13°, 126.92°, 23°
Heating and cooling system		EHP (electric, 1-way ceiling type)
Heating set point temperature		20 °C (dead band 0.5 °C)
Cooling set point temperature		26 °C (dead band 0.5 °C)
U-value	Wall	0.32 W/m2K
	Roof	0.26 W/m2K
	Slab	0.35 W/m2K
	Window	1.80 W/m2K, THK26 (5LE + 16AR + 5LE)
Operation schedule		09:00–18:00

). The two test cells were the same size and had building envelopes with the same thermal transmittance (U-values), and active equipment systems with the same specifications were installed in the test room where the TALS was installed (hereinafter referred to as “TALS”) and the reference room where the TALS was not used (hereinafter referred to as “non-TALS”). In addition, the data monitoring systems for capturing the experiment data consisted of the same types of sensors and equipment and were installed at the same locations to perform the experiments under identical conditions (Figure 5). Two TALS units were installed horizontally at a height of 2.2 m from the floor below the perimeter skylights, and the indoor heating and cooling set temperatures of the electric heat pump (EHP) were set as 20 °C in the winter and 26 °C in the summer during office building operating hours (9:00 to 18:00) to perform the experiments.

The local discomfort and predicted mean vote (PMV) values that resulted from the thermal stratification of the indoor space were compared and analyzed to physically verify that the indoor thermal environment was improved by the use of the TALS. For thermal stratification, the percentage of discomfort (PD) caused by the horizontal temperature differences and vertical temperature differences at head height (1.1 m) and ankle height (0.1 m) in the perimeter zone and the center of the room were calculated using the standard measurement methods presented ISO 7730’s “Ergonomics of the thermal environment” [19] and ASHRAE Standard 55’s “Thermal environmental conditions for human occupancy” [20,21,22]. Perimeter zone measurements were performed 0.5 m from the window toward the interior to examine the TALS’ radiant heat diffusion and transfer, and the center of the room was measured with a non-contact temperature sensor at the center of the test cell. In addition, to evaluate the thermal comfort properties of the TALS’ radiant heating and cooling, the thermal environment index PMV was measured at the work surface height (0.85 m) at the center of the room. The Sensor information for PMV measuring was as

Table 2. Sensor information for PMV.

Device	Feature	Explanation
Globe thermometer	Range	0–120 °C
	Application	ISO 7243 [23], ISO 7726 [24], DIN EN 27726 [25], DIN 33403 [26]
Turbulence probe	Temperature range (accuracy)	0–50 °C (±0.5 °C)
	Absolute pressure range (accuracy)	700~1100 hpa (±3 hpa)
	Wind speed range (accuracy)	0–5 m/s (±0.03 m/s)
CO2 probe	Temperature range (accuracy)	0–50 °C (±0.5 °C)
	Humidity range (accuracy)	5–95% (±3%)
	Absolute pressure range (accuracy)	700–1100 hpa (±3 hpa)
	CO2 range (accuracy)	0–10,000 ppm (±50 ppm)

.

3. Results

3.1. Indoor Thermal Environment Performance Evaluation

In an indoor thermal environment, occupants feel thermal sensations as heat is supplied or removed [27,28], and creating a suitable thermal environment is important for improving occupant satisfaction, increasing work efficiency, and maintaining bodily health [29,30]. To verify the thermal environment improvement effects of the TALS’ radiant heating and cooling functions, experiments were performed in the winter and summer in two identically configured test spaces, and the results are compared and analyzed in this chapter.

The two test cells have identical rectangular plans, uniform envelope constructions, and no interior partitions or large furniture, so that the thermal field in each room can be regarded as predominantly one-dimensional from the window toward the interior. Therefore, two vertical measurement lines were selected: a perimeter line located 0.5 m from the window to capture the zone with the highest envelope load and a center line to represent the interior occupied zone. The winter heating environment experiments were performed over the course of 7 days, and the analysis was performed on data from three typical days during which the data were stable and TALS and the FCU operated simultaneously (Figure 6). In the figure, the green line indicates the supplied heating source temperature, and small temperature changes due to piping heat loss are observed. The red and blue lines indicate the indoor temperatures of the two test cells. The gray dotted line indicates the outdoor temperature, and the orange bar graph shows the global radiation. The summer cooling environment experiments were performed under the same conditions as the winter heating environment experiments, and data from three typical days with the highest global radiation levels were used for analyzing the results (Figure 7).

The winter experiments were carried out over seven consecutive days during the heating season, when outdoor air temperatures remained below approximately 10 °C, and three representative days with stable system operation were selected for detailed analysis. The summer experiments were similarly conducted over seven days during hot summer conditions, with outdoor temperatures exceeding 30 °C and strong global solar radiation, and three days with the highest solar loads were chosen as typical days. Figure 6 and Figure 7 present the measured outdoor temperature and global solar radiation during these periods together with the indoor temperatures and supply/return water temperatures, thereby characterizing the seasonal boundary conditions under which the TALS performance was evaluated.

3.1.1. Vertical Temperature Distribution

(1)

Winter Vertical Temperature Measurement Results

Because the building envelopes of office buildings mainly consist of windows and curtain walls, local temperature differences can occur because of winter window surface cooling, which can lead to phenomena such as radiative imbalances and cold drafts. The TALS can mitigate such phenomena by inducing radiant heating and cooling in the perimeter zone. In this section, the improvement effect of the TALS is analyzed by comparing the indoor vertical temperature distributions when the TALS was used and was not used. The measurement points were at head (1.7 m), chest (1.1 m), and ankle (0.1 m) height in the perimeter zone and at the center of the room, which were the indoor occupied areas. Figure 8 shows the TALS (test room) and non-TALS (reference room) data.

The two test cells were found to be slightly above the set temperature of 20 °C during occupancy hours, and the indoor temperature of the TALS cell was higher than that of the non-TALS cell owing to the radiant heating. The radiant heat increased toward the ceiling owing to the difference in air density. The largest temperature increase occurred at the head measurement point, and the temperature increased at the torso and ankle measurement points as well. At the perimeter head-height point, where a significant envelope load and heat occurred, the temperature increased by approximately 6.66 °C (from 15.94 to 22.60 °C) because of the TALS’ radiant heating. At the center of the room, the temperature increased by 5.24 °C (from 16.92 to 22.17 °C), and the radiant heat was transferred evenly throughout the indoor space. In contrast, the torso and ankle temperatures increased by small amounts, which is attributed to the fact that the radiant energy that was emitted from the TALS had insufficient driving force to reach the floor and disperse effectively. To increase the heating efficiency of the TALS, it is necessary to perform supplementary research on the use of radiating fans on the bottom surface of the light shelf or the adjustment of the installation height of the light shelf. The maximum indoor temperature increase caused by the installation of the TALS was 1.76 °C in the perimeter zone and 0.8 °C at the center of the room. These results indicate that building energy savings can be achieved through a reduction in the winter heating load.

When the criteria in

Table 3. Thermal environment comfort category evaluation indices.

Category	General Comfort	Local Discomfort
	PMV	PD (%)	Vertical Temperature Difference (°C)
A	−0.2 < PMV < +0.2	<3	<2
B	−0.5 < PMV < +0.5	<5	<3
C	−0.7 < PMV < +0.7	<10	<4

are applied to the vertical temperature differences in

Table 4. Measurement results for the winter vertical temperature distribution (°C) and PD.

Time	TALS								Non-TALS
	Perimeter Zone				Center Zone				Perimeter Zone				Center Zone
	1.1 m	0.1 m	ΔT	PD (%)	1.1 m	0.1 m	ΔT	PD (%)	1.1 m	0.1 m	ΔT	PD (%)	1.1 m	0.1 m	ΔT	PD (%)
9:00	18.57	17.13	1.44	1.07	19.43	18.83	0.60	0.52	18.28	17.55	0.72	0.58	19.56	18.74	0.82	0.63
10:00	22.41	19.52	2.89 *	3.60 *	22.41	20.39	2.02 *	1.75	21.55	20.28	1.27	0.92	21.51	19.77	1.73	1.37
11:00	24.87	21.58	3.29 **	5.00 **	24.42	21.85	2.56 *	2.75	24.21	22.84	1.36	1.00	23.92	21.36	2.57 *	2.76
12:00	26.52	23.37	3.16 **	4.49 *	25.89	23.12	2.77 *	3.27 *	25.82	24.64	1.19	0.86	25.46	22.45	3.01 **	3.98 *
13:00	27.93	25.10	2.83 *	3.44 *	27.11	24.81	2.30 *	2.21	27.14	26.06	1.08	0.79	26.54	24.02	2.52 *	2.66
14:00	28.75	26.40	2.35 *	2.30	27.81	24.53	3.28 **	4.95 *	27.94	26.91	1.03	0.75	27.31	23.82	3.49 **	5.90 **
15:00	29.66	27.33	2.33 *	2.27	28.69	25.48	3.21 **	4.68 *	27.58	26.72	0.86	0.65	26.88	23.21	3.66 **	6.75 **
16:00	27.73	26.39	1.34	0.98	26.97	23.82	3.15 **	4.46 *	26.46	25.79	0.66	0.55	26.20	23.04	3.16 **	4.50 *
17:00	24.65	24.31	0.34	0.42	24.59	22.85	1.74	1.37	23.23	23.10	0.13	0.35	23.80	22.28	1.53	1.15

* Comfort level category B: vertical temperature difference < 3 °C, PD < 5%. ** comfort level category C: vertical temperature difference < 4 °C, PD < 10%.

, it can be seen that the center zone of both rooms remains within ISO 7730 comfort Categories A–B and within the head–ankle temperature difference of approximately 3.0 °C recommended by ASHRAE Standard 55 for most occupied hours. In contrast, the perimeter zone occasionally reaches Category C in both rooms, with maximum vertical temperature differences of about 3.0–3.3 °C during mid-day hours, indicating locally stronger stratification near the window. This suggests that, while the overall room remains within acceptable limits, further improvement of the TALS design is required to mitigate vertical temperature differences in the immediate perimeter zone.

(2)

Summer Vertical Temperature Measurement Results

Analyzing the summer vertical temperatures showed that, except for a short period around noon, the head-, chest-, and ankle-height temperatures in the TALS room were lower than those in the non-TALS room (Figure 9). The apparent “temperature reversal” observed around noon at ankle height in the perimeter zone of the TALS room was caused by the temperature sensor being exposed to direct solar radiation due to the high summer solar altitude, and should therefore be regarded as a measurement artifact rather than a physical effect of the TALS. Importantly, even during this period the corresponding vertical temperature differences in

Table 5. Summer vertical temperature distribution (°C) and PD measurement results.

Time	TALS								Non-TALS
	Perimeter Zone				Center Zone								Perimeter Zone
	1.1 m	0.1 m	ΔT	PD (%)	1.1 m	0.1 m	ΔT	PD (%)	1.1 m	0.1 m	ΔT	PD (%)	1.1 m	0.1 m	ΔT	PD (%)
9:00	26.07	25.56	0.51	0.49	25.70	25.24	0.46	0.47	26.28	26.30	0.02	0.31	25.85	26.02	0.17	0.27
10:00	26.40	26.00	0.40	0.44	25.66	25.14	0.52	0.49	26.61	26.78	0.17	0.27	25.86	25.97	0.11	0.29
11:00	26.77	26.78	0.01	0.31	25.69	25.02	0.67	0.55	26.80	27.32	0.52	0.20	25.68	25.86	0.18	0.27
12:00	27.23	27.68	0.46	0.21	25.67	24.98	0.69	0.57	27.20	27.81	0.61	0.19	25.73	25.86	0.13	0.28
13:00	27.41	28.37	0.97	0.14	25.70	25.00	0.70	0.57	27.30	28.05	0.75	0.17	25.74	25.86	0.12	0.28
14:00	27.40	28.59	1.19	0.11	25.69	24.93	0.76	0.60	27.29	28.05	0.76	0.16	25.75	25.85	0.10	0.29
15:00	27.10	28.14	1.05	0.13	25.69	25.01	0.68	0.56	27.13	27.74	0.60	0.19	25.67	25.80	0.13	0.28
16:00	26.63	27.38	0.75	0.16	25.70	25.08	0.62	0.53	26.87	27.44	0.58	0.19	25.85	25.89	0.04	0.30
17:00	26.36	26.99	0.63	0.18	25.66	25.15	0.50	0.48	26.53	26.96	0.43	0.23	25.80	25.89	0.08	0.29

remain below 2 °C and the PD values remain within the ISO 7730 Category A limits. At all other times, the TALS generally maintained lower temperatures throughout the room, and the indoor temperature decreased by up to 0.92 °C at the center of the room. This overall reduction in the occupied-zone temperature with height indicates that the TALS’ radiant cooling can slightly improve thermal comfort by promoting cooler air near the floor due to increased air density. We acknowledge that shielding or reorienting the perimeter ankle-height sensor to avoid direct solar radiation would have been preferable, and this will be implemented in future measurements.

As summarized in Table 5, the maximum vertical temperature difference in summer is approximately 1.2 °C in both rooms, which satisfies the ISO 7730 Category A limit (ΔT < 2 °C) and is well below the recommended 3 °C head–ankle temperature difference in ASHRAE Standard 55.

3.1.2. Local Discomfort

The developed TALS incorporates perimeter window surfaces to improve the indoor thermal environment. To evaluate the possibility that the TALS generates local air currents or causes imbalanced radiation when it is installed in the perimeter zone, the vertical temperature difference and the PD caused by the vertical temperature difference were used as evaluation indices. The vertical temperature difference is a physical index that evaluates discomfort due to thermal stratification in indoor spaces, and it assesses the temperature difference between chest height (1.1 m) and ankle height (0.1 m) when an occupant is sitting. The PD caused by the vertical temperature difference is calculated using Equation (1), and the indoor thermal distribution characteristics are classified into comfort level categories A–C, in accordance with ISO 7730. The comfort range classification standard is presented in Table 3. Because of the symmetry of the test cells and the absence of interior obstructions, the center measurement point can be considered representative of the average indoor thermal environment, whereas the perimeter point captures the local effects of the window and the TALS. Nevertheless, the present analysis is limited to these two locations, and a more detailed two-dimensional characterization of the thermal field (e.g., via denser sensor networks or computational fluid dynamics) is left for future work.

$$PD=100/[1+\exp (5.76-0.856\ast \mathsf{\Delta }Ta,v)\left]\right[\%]$$

(1)

Here, $PD$ represents the percentage of discomfort, and $\mathsf{\Delta }Ta,v$ represents the vertical temperature difference (the temperature at 1100 mm minus the temperature at 100 mm).

(1)

Winter Local Discomfort Analysis Results.

Table 4 presents the PDs with respect to the winter vertical temperature distributions and vertical temperature differences. The TALS’ average temperature was higher in all spaces, indicating that the TALS’ radiant heating was effective. In the perimeter zone of the TALS cell, the temperature was higher than the heating set temperature, which is attributed to the transfer of radiant heat from the TALS surface to the space around the TALS. The vertical temperature distribution results indicated that the temperature at chest height was higher than that at ankle height owing to natural convection of the radiant heat from the TALS bottom surface. However, at ankle height, the radiant heating effect was weaker than expected because the heat loss caused by heat transmission from the window exceeded the radiant heat arriving from the TALS.

In the perimeter zone, the indoor space thermal stratification phenomenon caused by TALS heating was more distinct. In the TALS cell, a large vertical temperature difference occurred from 10:00 to 15:00, which corresponded to comfort level categories B and C according to the ISO 7730 standard. In the perimeter zone, the TALS cell provided a less comfortable environment than the non-TALS cell, which was evaluated as being in comfort level category A. This is because a large amount of heat dispersal occurred around the height where the TALS was installed, and the effect of heat transfer was small at the floor; therefore, temperature differences occurred according to the vertical height. The measurement point at the center of the room was not directly reached by solar radiation from the sun, and it exhibited stable temperatures owing to heat transfer caused by the window’s Solar Heat Gain Coefficient and heat transmittance, as well as convection of the indoor air; therefore, it can be considered representative of the temperature of the entire room.

Considering the measurement results for the center points of the rooms, both cells generally satisfied ISO 7730 comfort Categories A–B during most of the occupied period. However, the non-TALS room experienced a longer duration in Category C (e.g., around 12:00–16:00), whereas the TALS room showed slightly smaller PD values and a shorter time in Category C. This suggests a modest improvement in local discomfort performance at the center of the room when the TALS is operating. Although the TALS is installed in the perimeter zone, it still affects the thermal conditions at the center of the room by modifying the overall heat balance and air-flow pattern. In the non-TALS room, heating is provided only by the ceiling-mounted FCU and the cold window surfaces in winter tend to generate downdrafts, which can increase the vertical temperature difference even at the interior point. In the TALS room, the additional radiant heating near the façade increases the mean radiant temperature and warms the air in the perimeter zone, thereby reducing cold downdrafts and promoting convective mixing toward the interior. Given the small, unobstructed test cell geometry, this combined radiant–convective effect leads to a slightly more uniform vertical temperature distribution at the center point.

The experimental results indicated that the use of the TALS in the winter can improve thermal comfort and ensure the expected level of radiant heating. It is expected that the performance of the TALS can be improved by enhancing the radiant heat transfer performance and mitigating heat loss caused by overall thermal transmission from the windows to further improve the thermal comfort.

(2)

Summer Local Discomfort Analysis Results

For the summer local discomfort observation experiments, the indoor temperature level was maintained at 26 °C, and cooling experiments were performed under the same conditions used in the winter experiments presented in the previous section. EHP-type FCU cooling systems with the same specifications were operated in both the TALS and non-TALS test cells, and an additional radiant cooling effect was induced in the TALS cell only.

The results of the summer thermal environment experiment indicated that the average temperatures in the TALS test room were lower than those in the non-TALS test room at all the measurement points because of the TALS’ radiant cooling effect (Table 5). In the perimeter zone, under the two experimental conditions, the temperature at head height was higher than the set temperature owing to the solar radiant energy flowing through the opening. However, the indoor temperature of the TALS cell was lower than that of the non-TALS cell at the center of the room, which can be considered the representative point for the indoor temperature of the entire room because it exhibited the most stable temperature distribution owing to the convective dispersal of the indoor air. Considering this result, the TALS’ radiant cooling was effective.

In the perimeter zone, the TALS room had a larger average vertical temperature difference (1.19 °C) than the non-TALS room because of cold, dense air currents that flowed down to the floor as a result of the TALS’ radiant cooling, and there was a possibility of local discomfort at certain times. This difference in temperature distribution was not large overall; however, it implies that further calibration of the system will be needed in the future to maximize the TALS’ radiant cooling effect. The slightly larger average vertical temperature difference in the TALS perimeter zone (1.19 °C) is partly influenced by a short period around noon when the ankle-height sensor was exposed to direct solar radiation, which leads to a local overestimation of the ankle temperature. Nevertheless, the vertical temperature differences and PD values in Table 5 remain below the ISO 7730 Category A limits, so this measurement artifact does not change the overall conclusion that summer vertical stratification is small in both rooms.

3.1.3. Predicted Mean Vote

PMV is a theoretical evaluation index that considers the overall comfortableness of the indoor thermal environment as it is felt by the indoor occupants’ senses. To evaluate the thermal comfort performance improvement effect of the TALS, an experimental environment was created according to the evaluation standards of ISO 7730 and ASHRAE Standard 55, and the indoor thermal environment comfort range results were calculated for each of the experimental conditions. A PMV physical quantity closer to 0 indicates a more comfortable environment, and the range of −0.5 < PMV < +0.5 (categories A and B) is defined as a comfortable indoor thermal environment (Table 3). In this study, PMV is used as a standardized global index in accordance with ISO 7730 and ASHRAE Standard 55, and is interpreted in combination with the local discomfort indices (vertical temperature difference and PD) described in Section 3.1.1 and Section 3.1.2 to account for non-uniform thermal stratification in the perimeter zone.

(1)

Winter PMV Measurement Results

To evaluate the expected thermal sensation improvement effect of TALS’ radiant heating, measurements were performed for 7 days when the outdoor temperature was ≤10 °C, and PMV data were measured at the center of the room for 72 continuous hours during which the data were stable. The results are presented in Figure 10. In the experiment, an indoor EHP-type FCU heating system was operated from 9:00 to 18:00, and the indoor temperature was maintained at approximately 20 °C.

The PMV measurement results indicated that a comfortable indoor thermal environment was created in the test room by operating the TALS radiant heating system and the FCU air conditioning system simultaneously, and during the 3 days experiment data analysis period, the PMV comfort range corresponded to comfort level categories A and B as defined in the ISO 7730 standard. In contrast, in the non-TALS test room, which used only the EHP-type FCU heating system, the results were outside of the comfortable range for most timeslots, and the indoor temperature was lower than that in the TALS test room. Ultimately, the experimental results confirmed that the TALS was effective for saving heating energy, as the temperature difference between the TALS and non-TALS rooms yielded a quantitative difference in the building heating energy. Specifically, we examined the duration for which the category A and B comfort ranges were maintained. The TALS room achieved a duration of 18 h and 28 min, whereas the non-TALS room had a duration of 5 h and 25 min; thus, the thermal comfort maintenance time provided by the TALS’ radiant heating was approximately 3.4 times longer. In addition, we examined the continuous time spent in category A, which is the best thermal comfort range; this was 4 h and 42 s for the TALS room and 1 h and 16 s for the non-TALS room. The time for which the TALS room maintained a category A environment was approximately 3.7 times longer, confirming the effectiveness of the TALS.

(2)

Summer PMV Measurement Results

The summer experiments were performed over the course of 7 summer days during which outdoor temperatures of ≥30 °C were observed, and experiments were performed in each of the two test cells to examine the radiant cooling thermal sensation improvement caused by the use of the TALS. In the non-TALS test room, an EHP-type FCU cooling system was operated in isolation. In the TALS test room, an EHP-type FCU cooling system and the TALS’ radiant cooling were operated together.

The PMV was analyzed with respect to the TALS’ radiant cooling. The results indicated a range of +0.4 < PMV < +1.1 for the TALS room and a range of +0.4 < PMV < +1.2 for the non-TALS room, which does not confirm the significance of the thermal comfort improvement caused by the TALS’ radiant cooling (Figure 11). However, the cumulative time in comfort level category B (orange) according to the ISO 7730 standard was found to be 1 h and 33 min for the TALS room and 12 min for the non-TALS room, which confirms the effectiveness of the TALS, as the continuous time in the comfortable range was increased by a factor of approximately 7.8 via the installation of the TALS. In addition, the time spent in comfort level category C was 38 h and 20 min for the TALS room and 14 h and 53 min for the non-TALS room. Considering the experimental results comprehensively, the absolute values of the PMV comfort level improvements were not large, but the experiments confirmed that the use of the TALS improved the indoor comfort and that it is possible to provide building energy savings that correspond to the temperature difference.

It should be noted that PMV was originally developed for relatively uniform thermal environments and may not fully capture the effects of radiant asymmetry and spatial non-uniformity associated with the TALS. In the present work, PMV is therefore used mainly as a comparative indicator between the TALS and non-TALS rooms under identical boundary conditions, while local non-uniformities are evaluated through vertical temperature differences and PD. A more comprehensive comfort assessment using alternative models that explicitly consider mean radiant temperature, operative temperature, or adaptive comfort (supported by additional measurements or simulations of surface temperatures and radiant exchange) is left for future research.

3.2. Heating and Cooling Energy Savings

To analyze the reductions in the perimeter zone load and the heating and cooling energy caused by the installation of the TALS, summer and winter empirical experiments were performed in the test bed that imitated an office building. As described in the previous section, the TALS radiant heating and cooling system and an EHP-type FCU were operated together in the test room, whereas in the reference room, only the FCU was operated. The indoor temperature was set as 20 °C for heating and 26 °C for cooling, with an FCU dead band of 0.5 °C. The TALS’ flow rate was fixed to 7 L/min, and the EHP-type FCU was a constant-air volume system in which operations were controlled by turning the system on and off.

In both test cells, the HVAC operation schedule was 09:00–18:00. In the TALS room, the hydronic loop supplying the TALS panel operated continuously during this period with a fixed flow rate of 7 L/min and a constant supply-water temperature in heating and cooling modes. The FCU in each room was controlled by the same room thermostat: it switched on when the measured indoor temperature moved outside the setpoint band (20 °C ± 0.5 °C in winter and 26 °C ± 0.5 °C in summer) and switched off again when the temperature returned within this band. Therefore, in the non-TALS room the FCU was the only device responsible for meeting the room load, whereas in the TALS room the TALS panel provided continuous radiant heating or cooling in the perimeter zone and the FCU acted as a backup terminal unit that operated only when the TALS alone could not maintain the indoor temperature within the ±0.5 °C dead band.

The heating source temperature provided by the TALS and the FCU was set as 45 °C, and the cooling source temperature was set as 10 °C. The heating and cooling energy savings were evaluated by comparing and analyzing the TALS’ radiant energy exchange heat values (the temperature differences between the heating and cooling water inlets and outlets) and the total power consumption due to FCU control. The energy used by the heat pump, which heated and cooled the heating source, and the circulation pump (conveyance energy) was the same between the test cells; therefore, it was excluded from the analysis.

The heating and cooling rates of the hydronic circuits were calculated from the measured supply and return water temperatures and the measured volumetric flow rates using the standard energy balance (Equation (2)). It should be emphasized that the above comparison focuses on the room-side thermal demand. The calculated heating and cooling energy values represent the sensible heat delivered to or removed from the rooms by the hydronic circuits and the FCU, together with the FCU fan electricity use. The same air-source heat pump and main circulation pumps served both test cells under identical control, and the additional electricity consumption of the TALS circulation loop was not separately metered. Consequently, the reported 39.2% (heating) and 7.7% (cooling) reductions represent reductions in delivered room-side demand and should be regarded as an upper bound on possible system-level energy savings.

$$Q=\rho c_p\dot{V}\mathsf{\Delta }T$$

(2)

where $\rho$ is the density of water, $c_p$ is the specific heat of water, $\dot{V}$ is the volumetric flow rate, and $\mathsf{\Delta }T$ is the temperature difference between the inlet and outlet. The hourly ‘heat values’ (kWh) listed in

Table 6. Winter heating energy usage results.

Time	TALS					Non-TALS
	FCU		Source			FCU		Source
	Operating Time	Electricity Usage (kWh)	Inlet (°C)	Outlet (°C)	Heat Value (kWh)	Operating Time	Electricity Usage (kWh)	Inlet (°C)	Outlet (°C)	Heat Value (kWh)
9:00	24 min 5 s	0.011	43.47	40.70	1.333	39 min 40 s	0.019	43.50	40.83	1.277
10:00	5 min 2 s	0.002	43.33	42.50	0.397	27 min 35 s	0.013	43.37	41.45	0.917
11:00	2 min 59 s	0.001	43.76	43.14	0.299	21 min 8 s	0.010	43.80	42.34	0.699
12:00	-	-	43.83	43.36	0.227	18 min 6 s	0.008	43.87	42.60	0.611
13:00	-	-	44.09	43.62	0.228	16 min 4 s	0.007	44.14	43.01	0.542
14:00	-	-	44.16	43.66	0.237	13 min 4 s	0.006	44.19	43.26	0.456
15:00	-	-	44.07	43.61	0.222	5 min 35 s	0.003	44.11	43.81	0.155
16:00	-	-	43.99	43.53	0.219	9 min 29 s	0.004	44.01	43.26	0.369
17:00	-	-	44.07	43.60	0.228	15 min 4 s	0.007	44.11	43.07	0.500
Total	FCU: 0.015 kWh		TALS: 3.391 kWh			FCU: 0.077 kWh		Non-TALS: 5.527 kWh

and

Table 7. Summer cooling energy usage results.

Time	TALS					Non-TALS
	FCU		Source			FCU		Source
	Operating Time	Electricity Usage (kWh)	Inlet (°C)	Outlet (°C)	Heat Value (kWh)	Operating Time	Electricity Usage (kWh)	Inlet (°C)	Outlet (°C)	Heat Value (kWh)
9:00	14 min 2 s	0.007	14.60	15.68	0.501	20 min 5 s	0.009	14.74	15.79	0.489
10:00	19 min 43 s	0.009	15.16	16.21	0.491	31 min 6 s	0.015	15.31	16.30	0.465
11:00	24 min 25 s	0.011	15.11	16.42	0.613	44 min 1 s	0.021	15.25	16.69	0.673
12:00	24 min 6 s	0.011	15.23	16.59	0.635	58 min 1 s	0.027	15.37	17.08	0.801
13:00	29 min 44 s	0.014	15.09	16.53	0.674	55 min 1 s	0.026	15.23	16.97	0.812
14:00	27 min 23 s	0.013	15.19	16.62	0.670	55 min 2 s	0.026	15.33	17.04	0.800
15:00	29 min 1 s	0.014	15.18	16.58	0.654	46 min 1 s	0.021	15.33	16.79	0.686
16:00	19 min 3 s	0.009	15.11	16.31	0.558	32 min 1 s	0.015	15.26	16.33	0.502
17:00	15 min 4 s	0.007	15.23	16.23	0.470	24 min 4 s	0.011	15.38	16.25	0.407
Total	FCU: 0.095 kWh		TALS: 5.266 kWh			FCU: 0.171 kWh		Non-TALS: 5.635 kWh

were obtained by integrating the calculated rates over each 1 h interval and converting from kJ to kWh. The electricity consumption of the FCU was determined by multiplying the measured electric power of the fan (28 W) by the cumulative operating time in each hour. The same air-source heat pump and circulation pumps supplied both test cells under identical control; therefore, their electricity consumption and the corresponding coefficient of performance (COP) affect both rooms equally and were not included in the comparison. The analysis in this chapter thus focuses on the room-side heating and cooling demand and the relative reduction in delivered energy achieved by the TALS.

3.2.1. Heating Energy Saving Results

In this study, the heating energy comparison focuses on the room-side demand, which is calculated from the measured supply/return water temperatures and flow rates of the hydronic circuits and the measured FCU fan electricity use. The same air-source heat pump and circulation pumps serve both test cells under identical control; therefore, their electricity consumption and coefficient of performance (COP) affect both rooms equally and are not included in the relative comparison presented in this section.

The winter heating energy consumption experiments were performed over the course of 7 days when the meteorological conditions were clear and partly clear, and Figure 12 shows the results for a typical day when the data were stable. In the non-TALS cell, the EHP-type FCU was controlled by repeatedly turning it on and off to maintain a constant set temperature of 20 °C in the experimental space. As such, there was a continuous fluctuation in the heat values from 9:00 to 18:00 as the experiments were performed within a fixed range of indoor temperature fluctuations.

According to the measurement results, the perimeter zone load was reduced in the test room owing to the radiant heating from the TALS system that was installed in the perimeter zone. Except for the start of the experiments (from 9:00 to 11:20), when the EHP-type FCU’s on/off control was operated intermittently, the indoor temperature remained constant during operating hours, even though the room’s EHP heating system was not running, and a large amount of heating energy was saved.

In Table 6, the heating energy consumption when the TALS was used and not used is divided into the heat values of the heating sources and the EHP-type FCU’s electricity consumption amounts. First, considering the heating source in the test room, there were high heat values at 9 a.m. for heating the TALS radiant panel, and in the afternoon, only the heat values corresponding to the differences in the TALS circulating water inlet/outlet temperatures were consumed. The daily heating energy usage amount of the test room (TALS) was 3.391 kWh, and that of the reference room (non-TALS) was 5.527 kWh, confirming that the use of the TALS reduced the energy consumption by approximately 38.6%. In addition, the FCU’s operating time was shortened because the TALS’ excellent radiant heating effect maintained an indoor temperature that was higher than the set temperature for a long time in the test room. The EHP-type FCU’s power consumption during the 9 h corresponding to the office building’s operating hours was 0.015 kWh in the TALS room and 0.077 kWh in the non-TALS room, and the energy-saving effect according to the operation of the air conditioner was found to be 80.5%. Here, the EHP-type FCU’s power consumption was calculated according to the power consumption (28 W) during the operating time. Regarding the daily heating energy consumption, i.e., the sum of the TALS heat values and the EHP-type FCU electricity consumption of the winter heating system, the TALS room consumed 3.406 kWh of heating energy, while the non-TALS room consumed 5.604 kWh, indicating that the TALS reduced the heating energy by 2.198 kWh/d. When these heating energy savings are interpreted together with the comfort results in Section 3.1, it becomes clear that the approximately 39.2% reduction in heating energy is accompanied by increased vertical temperature differences in the perimeter zone, where the TALS room reaches ISO 7730 comfort Categories B–C during some hours, while the center of the room maintains Categories A–B and shows better comfort than the non-TALS room. Thus, the present TALS prototype provides a net benefit in terms of overall room comfort and heating energy use, but at the cost of somewhat reduced local comfort for occupants seated very close to the façade.

3.2.2. Cooling Energy Usage Comparison and Analysis

To analyze the TALS’ cooling energy saving effect, experiments were performed over the course of 7 days during which the outdoor temperature was measured at ≥34 °C. A typical day with stable data was selected, and the data were analyzed (Figure 13). The TALS cooling source supplied circulated water at approximately 15 °C via an air source heat pump from 9:00 to 18:00, i.e., the office building occupancy hours, and the indoor temperature was maintained at 26 °C (dead band: 0.5 °C). In summer, the TALS is expected to be useful for reducing the air conditioning system’s overall cooling energy consumption throughout the room because it is effective at cooling the building perimeter by performing radiant cooling via its bottom surface and reducing the indoor cooling load. Figure 13 shows the changes in the thermal environment (indoor/outdoor temperatures, supply/return water temperatures) over time in the two test cells during summer. A constant indoor temperature of 26 °C was maintained in both spaces, and the cooling peak heat values were lower than those in winter, but in the afternoon hours when the outdoor temperatures increased, cooling energy was continuously consumed, and the absolute values were larger.

According to the results for the cooling supply heating sources in the experiments involving the TALS’ cooling energy saving effect (Table 7), the return water temperature was higher than the supply water temperature owing to the EHP-type FCU cooling system and the TALS’ radiant cooling thermal exchange. The consumed heat values for cooling exhibited a similar pattern to the FCU operating state, and the non-TALS room continuously consumed approximately 0.9 kW of cooling heat after noon. In the TALS room, the cooling peak heat value exceeded that of the non-TALS room, and the heat consumption was steady, similar to the base load, but the total consumed heat value was lower. The overall cooling energy saving effect of the TALS was excellent. Additionally, in the TALS’ radiant cooling system, the EHP-type FCU operated fewer times and for a shorter duration owing to the perimeter cooling effect. Therefore, it is expected that the TALS can provide sufficient cooling as well as building energy savings.

The electricity consumption for the FCU cooling system in the non-TALS room was 0.171 kWh, whereas the TALS’ electricity consumption was 0.095. The TALS provided cooling energy savings of approximately 44%. A comparison of the heat values used for cooling indicated that the cooling energy usage during the 9 h corresponding to office building operating hours was 5.635 and 5.266 kWh/d in the non-TALS and TALS rooms, respectively. An analysis of the daily energy consumption, which was the sum of the TALS’ cooling heat values and the FCU electricity consumption that were used for summer cooling, revealed that the non-TALS room consumed 5.806 kWh of energy and the TALS room consumed 5.361 kWh of energy, confirming that approximately 0.445 kWh/d of cooling energy can be saved on a typical day. Unlike in winter, when the TALS mainly assists the heating system and improves the use of beneficial solar gains, in summer the same device serves to control and remove unwanted solar and transmission gains in the perimeter zone, so that deeper daylight penetration does not translate into higher cooling loads.

As discussed in Section 3.1.2 and Section 3.1.3, these cooling energy savings are achieved without a deterioration of thermal comfort: the vertical temperature differences remain within ISO 7730 Categories A–B, and the TALS room exhibits a longer cumulative duration within the comfortable PMV range than the non-TALS room. Therefore, in summer, the TALS improves or maintains comfort while reducing cooling energy use.

In addition to the daily energy use, Table 6 and Figure 12 also illustrate the temporal behavior of the heating demand. In the TALS room, the highest hourly heating rate occurs during the initial warm-up period at 9:00 when the radiant panel is first brought up to temperature; after this period, the required heating rate remains comparatively low and the FCU operation is minimal. In contrast, the non-TALS room requires more frequent FCU operation and exhibits higher heating rates during several afternoon hours to maintain the setpoint. These results indicate that the TALS reduces not only the total daily heating energy but also the magnitude and duration of the space heating demand that must be supplied by the FCU. It should be noted that the observed heating energy savings are primarily attributable to the targeted radiant heating in the perimeter zone and the resulting changes in the indoor thermal environment, with the improved use of winter solar gains playing a secondary role rather than being the sole cause of the reduction in heating demand.

As shown in Table 7 and Figure 13, the non-TALS room requires higher mid-day cooling rates than the TALS room, even though both rooms are controlled to the same indoor setpoint. Accordingly, the TALS reduces both total cooling energy use and peak hourly FCU demand by sharing the perimeter load through radiant cooling. The present energy analysis is based on hourly averaged heating and cooling rates; a more detailed peak-load and transient analysis of the coupled TALS–FCU system is left for future work using higher-resolution measurements and/or dynamic simulations. Although it may seem counterintuitive that a daylight-redirecting device lowers cooling demand, in summer the TALS primarily acts as a shading and radiant-cooling element: the shelf geometry and highly reflective upper surface partially shade the upper window and redirect high-angle solar radiation, while the cooled bottom surface directly offsets perimeter solar and transmission gains. Consequently, despite similar incident solar irradiance, the TALS room maintains the 26 °C setpoint with slightly lower cooling heat values and less FCU operation than the non-TALS room, leading to an approximately 7.7% reduction in daily room-side cooling energy demand under the present experimental conditions.

4. Discussion

This study experimentally evaluated a thermally activated light shelf (TALS) that combines a daylight-redirecting light shelf with a hydronic radiant heating and cooling panel in a full-scale twin test bed. The results presented in Section 3 indicate that the proposed system can improve indoor thermal comfort, particularly in terms of the cumulative time within the ISO 7730 comfort categories, while simultaneously reducing room-side heating and cooling energy demand under the tested winter and summer conditions. At the same time, the findings reveal important trade-offs between comfort and energy, as well as several methodological and practical limitations that should be considered when interpreting the results and applying the TALS concept in real buildings.

From a comfort perspective, the winter experiments highlight the dual role of the TALS. In the building perimeter zone, where envelope heat losses are concentrated, the radiant heating from the TALS leads to higher air temperatures and increased vertical temperature differences compared with the non-TALS room, and local thermal conditions occasionally fall into ISO 7730 Categories B–C. This indicates that the current prototype can create a less comfortable environment for occupants seated very close to the façade, primarily due to stronger stratification between chest and ankle heights. In contrast, at the center of the room, which can be regarded as representative of the overall room thermal environment, both rooms satisfy the ISO 7730 comfort criteria for most of the occupied period, but the TALS room shows slightly smaller percentages of dissatisfied and a shorter duration in Category C. The combined radiant and convective effects of the TALS thus lead to a modest improvement in whole-room comfort, even though perimeter comfort in winter still requires further refinement.

In summer, the TALS behaves primarily as a perimeter cooling and shading device. The vertical temperature profiles show small stratification in both rooms, with maximum head–ankle temperature differences of approximately 1.2 °C, which are well within the ISO 7730 Category A limit and the ASHRAE Standard 55 recommendation of a 3 °C maximum head–ankle difference. The TALS room maintains slightly lower average temperatures at most measurement points and achieves a longer cumulative duration within the comfortable PMV range than the non-TALS room, while avoiding a deterioration of vertical temperature differences. These results suggest that, under the tested summer conditions, radiant cooling integrated into the light shelf can support conventional FCU cooling without introducing significant additional local discomfort.

The energy analysis focuses on room-side heating and cooling demand derived from measured water temperatures and flow rates in the hydronic circuits, together with FCU fan electricity use. Under the selected clear winter day, the TALS room consumed approximately 39.2% less daily heating energy than the non-TALS room, corresponding to an absolute difference of 2.20 kWh/d. This relatively large percentage reduction reflects the small absolute loads of the highly insulated test cells and the dominant influence of the perimeter window zone. The savings arise from a combination of targeted radiant heating at the façade, reduced room-side perimeter heat loss due to the presence of the insulated light-shelf structure, and more effective use of beneficial winter solar gains rather than from the creation of additional external energy. In summer, the TALS reduces daily room-side cooling demand by about 7.7% under a representative hot day. Here, the system acts mainly as a shading and radiant cooling element: the shelf geometry and highly reflective upper surface partially shade the upper window and redirect high-angle solar radiation, while the cooled bottom surface directly offsets perimeter solar and transmission gains, allowing the FCU to operate less frequently. These findings show that the same device plays different roles in winter (assisting heating and mitigating perimeter losses) and summer (controlling and removing unwanted gains).

Several methodological and practical limitations must be acknowledged. First, the experiments were conducted in two identical, small, unobstructed office-like test cells in a single climate, and the analysis focused on seven winter days and seven summer days with representative clear and hot conditions. The reported heating and cooling energy savings should therefore be interpreted as prototype-level results under specific boundary conditions rather than universal figures. Second, the spatial characterization of the indoor thermal field is limited to two vertical measurement lines (perimeter and center), and the comfort analysis relies primarily on vertical temperature differences, PD, and PMV at a single interior location. Although the test cells were symmetric and free of interior obstructions, more detailed two-dimensional measurements or computational fluid dynamics simulations would provide a more complete picture of spatial non-uniformities. Third, the PMV index, while widely used in standards, has known limitations in radiant and non-uniform environments; in this study, PMV was used as a comparative indicator between the TALS and non-TALS rooms under identical conditions and was interpreted together with local discomfort indices, but future work should explore alternative comfort models that explicitly account for mean radiant temperature, radiant asymmetry, and adaptive comfort.

Instrumentation issues also introduce uncertainty. In particular, a small noon-time “temperature reversal” observed at ankle height in the perimeter zone of the TALS room during the summer experiment was identified as a measurement artifact caused by direct solar radiation on the sensor. Although the magnitude of this anomaly is less than 1 °C and does not change the comfort classification (vertical temperature differences and PD values remain within ISO 7730 Category A), it underscores the importance of shielding, orienting, and locating sensors carefully to avoid radiative bias. Future studies should adopt radiation shields or alternative sensor placements and may incorporate additional measurements of surface temperatures and radiant asymmetry to better quantify the radiant environment.

Despite these limitations, the present study provides useful insights for the design and application of TALS systems. The results indicate that integrating radiant heating and cooling functions into a light shelf can enhance overall comfort and reduce room-side heating and cooling demand, but also reveal a winter trade-off between energy savings and local comfort near the façade. Designers considering TALS-like systems should therefore carefully select the installation height, water temperature setpoints, and control strategies, and may need to combine the system with gentle air mixing or other measures to mitigate perimeter stratification. At the building scale, the impact of TALS on lighting energy, occupant distribution, and whole-system performance (including heat pump COP and auxiliary pumps) should be evaluated using long-term monitoring and building energy simulations. In this context, the empirical data from the present study provide a useful benchmark and starting point for further development and optimization of façade-integrated radiant systems that jointly address daylighting, thermal comfort, and energy use.

To place the present results in the context of previous work, it is useful to briefly compare the magnitude of the observed energy savings with those reported for conventional light-shelf systems. Kim et al. evaluated a fixed interior light shelf based on the energy consumption for lighting and air conditioning in an office test bed and found that, in summer, the light shelf reduced lighting and cooling energy by approximately 0–10.5% and 6.9–9.3%, respectively, whereas in winter it increased lighting and heating energy use by up to about 25.3% and 3.2% [15]. Other studies on interior or PV-integrated light shelves have demonstrated improvements in daylight distribution and overall building energy performance, but they generally focus on lighting or total annual energy and do not incorporate an active radiant panel into the shelf itself. Under the specific full-scale mock-up conditions tested here, the proposed TALS prototype reduces the room-side heating and cooling energy demand by approximately 39.2% and 7.7%, respectively, while increasing the cumulative duration of ISO 7730 Category A–B comfort. These zone-level savings are broadly consistent with the significant HVAC energy and peak-load reductions reported for radiant ceiling panel systems compared with all-air systems, but are achieved through a façade-integrated device that simultaneously provides daylight redirection. At the same time, differences in climate, boundary conditions, and performance metrics across studies mean that our percentages should be interpreted as prototype-level benchmarks rather than universal values.

5. Conclusions

This paper proposes a TALS system in which radiant heating and cooling functions are incorporated into an existing light shelf. To examine the possibility of using this system in an actual building, empirical experiments were performed in a mock-up scale test bed to analyze the system’s indoor thermal environment improvement effects and heating and cooling energy saving effects. The following conclusions are drawn.

(1)

A hybrid light shelf system was developed that allows radiant heating and cooling on the bottom surface of a normal light shelf, and experiments were performed in a test bed that simulated an office space. The results indicated that the TALS was effective for improving the comfort of the indoor environment and saving heating and cooling energy.

(2)

The local discomfort caused by vertical temperature differences that can occur owing to TALS’ radiant heating and cooling was evaluated, and the results indicated that in the building perimeter zone during winter, the non-TALS room was in category A (−0.2 < PMV < +0.2, $\mathsf{\Delta }Ta,v$ < 2 °C) of the ISO 7730 comfort standard, whereas the TALS room was in categories B and C, suggesting that there is a need for improvement regarding the local discomfort caused by the TALS’ vertical temperature differences. In contrast, at the center of the room, which can be considered representative of the overall room thermal environment, both rooms generally satisfied the ISO 7730 comfort criteria, but the TALS room showed slightly smaller PD values and a shorter duration in Category C than the non-TALS room. This indicates a modest improvement in overall room comfort associated with the TALS, although the influence is weaker than in the perimeter zone and results from the combined radiant and convective effects of the system.

(3)

The results for the PMV, which is an evaluation index for indoor thermal environments as felt by the occupants, due to the TALS’ radiant heating and cooling were −2.9 < PMV < +0.4 for the TALS room and −2.9 < PMV < −0.1 for the non-TALS room in winter and +0.4 < PMV < +1.1 for the TALS room and +0.4 < PMV < +1.2 for the non-TALS room in summer. However, an analysis of 3 days of data corresponding to operating hours indicated that the TALS room was approximately 3.4 times more effective at heating and approximately 7.8 more effective at cooling than the non-TALS room with regard to the cumulative time for which the ISO 7730 comfortable range (categories A and B) was maintained. This suggested that the TALS can provide occupants with high levels of indoor thermal environment comfort.

(4)

Under the present full-scale mock-up test-bed conditions, the TALS reduced the room-side heating energy demand by approximately 39.2% on a representative clear winter day and the cooling energy demand by approximately 7.7% on a representative hot summer day, compared with the non-TALS configuration. In winter, the TALS mainly assists the heating system by providing radiant heating in the perimeter zone and modifying the perimeter heat losses, whereas in summer it primarily acts as a shading and radiant cooling element that helps to control unwanted solar and transmission gains. These results indicate that the proposed TALS has the potential to reduce both heating and cooling energy use; however, the reported percentages are specific to the tested prototype, climate, and control strategy, and a broader whole-building energy analysis should be conducted in future work.

(5)

The winter results reveal a trade-off between perimeter-zone comfort and heating energy savings: the TALS reduces total heating energy use and improves comfort at the center of the room, but it increases vertical temperature differences near the window, leading to Category B–C local discomfort during some hours. Future work should focus on design and control strategies (e.g., adjusting the installation height, supply-water temperature, or providing gentle air mixing) that mitigate this local discomfort while preserving the heating and cooling energy savings demonstrated in this study.

Compared with previous light-shelf studies, which typically report single-digit to low double-digit changes in cooling or total energy use and sometimes winter penalties when only the daylighting function is considered, the TALS prototype demonstrates a relatively strong reduction in zone-level heating demand while maintaining or improving summer comfort. However, these values are specific to the tested configuration and climate, and broader multi-climate and whole-building benchmarks are needed in future work. Future work should therefore address several specific issues that remain unresolved in this study. First, the present results are limited to a single mock-up test bed, façade orientation, and climate; additional experiments and simulations are needed to confirm the performance of TALS under different building geometries, orientations, internal loads, and climatic conditions. Second, the current prototype exhibited increased winter stratification and local discomfort near the façade, so future studies should systematically optimize the TALS geometry, installation height, water temperature and flow rate, and control strategy to enhance radiant capacity while reducing vertical temperature differences. Third, the comfort assessment relied mainly on PMV and a small number of measurement points; more detailed spatial measurements, alternative comfort indices that emphasize radiant and non-uniform effects, and subjective occupant surveys are required to better understand user responses. Finally, long-term field studies and whole-building analyses that include lighting energy, heat pump COP, auxiliary pumping power, and economic performance are needed to quantify the overall benefits and trade-offs of deploying TALS systems in real office buildings.