Energy-Saving Performance and Optimization Study of Adaptive Shading System—A Case Study

Abstract

In the context of global energy challenges, adaptive shading systems have emerged as pivotal components in building energy efficiency research. This study systematically evaluates critical performance factors influencing energy efficiency in adaptive shading systems for buildings located in hot summer and cold winter climate zones, with a focus on parametric optimization of shading panel configurations. Through field measurements, orthogonal experimental design, and numerical simulations, this investigation centers on the adaptive shading system of a nearly zero energy building (NZEB). Four critical parameters—shading panel width, panel-to-window clearance, window-to-wall ratio (WWR), and surface reflectance—were rigorously analyzed through orthogonal experimental methodology and DesignBuilder^® simulations. This study identifies WWR and shading panel reflectance as the key factors for optimizing adaptive shading systems. Among the scenarios evaluated, the highest energy efficiency was achieved with horizontal shading devices on the south façade, featuring a panel width of 500 mm, a minimum clearance of 150 mm, a WWR of 55%, and a surface reflectance of 0.4. Under this configuration, the annual energy consumption was reduced to 8312.37 kWh, corresponding to a 2.1% decrease (8.31 MWh) in total site energy consumption (TSEC). This research provides valuable insights for energy-efficient building design in hot summer and cold winter regions, and supports the broader adoption of adaptive shading systems.

1. Introduction

Adaptive shading systems (ASSs), characterized by their capacity to autonomously modulate shading device configurations, represent a critical energy efficiency strategy in contemporary building design. As a secondary outer shell of buildings, this system enhances energy efficiency while regulating the indoor thermal and optical environment. Through real-time modulation of shading based on solar irradiance, ASSs reduce indoor heat gain and active cooling demand, enhancing the performance of nearly zero energy buildings (NZEB) and overall energy efficiency. Additionally, by lowering operational greenhouse gas emissions, ASSs contribute to climate mitigation efforts aimed at addressing global warming and the increasing frequency of extreme weather events, while improving energy efficiency and reducing environmental impact [1,2]. A bibliometric analysis was conducted using VOSviewer 1.6.20 (Centre for Science and Technology Studies, Leiden University, Leiden, The Netherlands) based on data from the Web of Science Core Collection (Clarivate Analytics, Philadelphia, PA, USA), focusing on the co-occurrence of keywords such as “Adaptive shading,” “Dynamic shading,” and other related technical terms. The results identified approximately 100 relevant studies published in the past five years (see Figure 1), reflecting the current research trends in the field. The analysis identified four predominant research clusters: (1) parametric performance optimization, (2) energy efficiency quantification, (3) computational modeling approaches, and (4) intelligent control algorithms. From these, it can be found that adaptive shading has become a significant focus in building energy-saving research, including technologies like dynamic and climate-adaptive shading.

2. Research Progress on Adaptive Shading Systems

2.1. Energy-Saving Studies on Adaptive Shading Systems

In studies on the application of adaptive shading in buildings, the main goal is to improve energy-saving efficiency. For most existing studies, the primary focus lies in demonstrating the substantial potential for improving energy efficiency, particularly in reducing summer cooling loads and lighting energy consumption. Investigations conducted by Kunwar et al. [3], Chandrasekaran et al. [4], and Wang et al. [5] have validated the superior performance of adaptive shading systems over conventional fixed solutions, with reported energy savings of 20% to 48.5% in cooling, heating, and lighting energy use. However, several studies have employed alternative parameters and comparative analyses with fixed systems to validate the performance improvements of adaptive shading systems. Researchers like Bazazzadeh et al. [6], Wu et al. [7], and Norouziasas et al. [8] have highlighted improvements in energy efficiency, visual comfort, and daylight utilization, with reported gains ranging from 10% to 91.5%.

In addition to the existing overall energy efficiency of the system, some studies focus on the integration of advanced materials and technologies in adaptive shading systems, aiming to further enhance building energy savings. Krarti’s innovative work revealed that smart glazing technologies could decrease annual building energy demand by 22% [9], while the static and dynamic photovoltaic (PV)-integrated shading devices were shown to achieve energy conservation potentials of 45–100% depending on different operational parameters [10]. Wu et al. [11] found that Thermochromic Adaptive Windows (TAWs) could also exhibit significant annual energy-saving performance of 8–14.5 kWh/m².

While the existing research has extensively explored the energy efficiency benefits of the integrated systems, numerous investigations have specifically focused on quantifying the energy-saving contributions of individual components: control processors, environmental sensors, and electromechanical actuators. Regarding the control unit development, numerous investigations have concentrated on optimizing its control system. Bi et al. [12] and Yang et al. [13] focused on control model optimization, achieving substantial energy savings of 38.3% and up to 20.7%, respectively, alongside significant improvements in daylight utilization. Tabadkani et al. [14,15] developed initiative control algorithms, including an hourly transmittance schedule and an open-loop strategy targeting pre-irradiance periods, which exhibited high accuracy with minimal deviation. In terms of the sensor location within adaptive shading systems, Dong et al. [16] studied the impact of sensor placement on energy consumption, finding a 23% reduction in peak cooling demand and 28–33% savings in total cooling and lighting demand. When referring to the actuators, Louis et al. [17] and Stelzmann et al. [18] proposed passive actuators—solar-powered, thermo-pneumatic actuators and self-regulating solar shading actuators—that reduce reliance on electronic components and improve system autonomy. Despite the diversity of approaches, these studies share a common goal: enhancing the intelligence and responsiveness of adaptive shading systems.

Beyond conventional studies, contemporary energy efficiency research actively integrates integrated emerging technologies. Kunwar et al. [19] observed that different modeling methods could lead to 20% variation in cooling and lighting energy consumption. Therefore, systems employing different technologies may result in varying levels of energy-saving efficiency. Dehwah et al. [20] found that Adaptive Envelope Technologies (AETs) could achieve 51% saving in cooling energy. Liu et al. [21] revealed that optimized PV Shading Devices (PVSDs) could reduce cooling and lighting demand by 48.7%. Zheng et al. [22] proposed a complex multiple-layered adjustable semi-shading system which decreased total energy consumption by 100 kWh. Wang et al. [23] developed a photovoltaic-integrated adaptive light shelf system that reduced net energy consumption by 3.56%. While these studies differ in terms of complexity and technologies, they share a common trajectory toward hybrid, multifunctional shading solutions that couple passive and active strategies.

However, despite the improvements in energy efficiency, many studies still lack standardized boundary conditions and broadly applicable evaluation frameworks. Moreover, the comparative analyses often emphasize fixed versus dynamic shading performance, without fully exploring variations across different climate zones, shading configurations, or control algorithms—factors that could significantly influence outcomes.

2.2. Structure and Parameter Studies on Adaptive Shading Systems

The pursuit of performance enhancement in adaptive shading systems—primarily through structure refinements and parameter optimization—has emerged as a dominant research trend in architectural science over the past five years. In contemporary studies, structural innovation predominantly centers on two aspects: structural innovation and smart material integration. Recent research on advanced shading mechanisms goes beyond simple adjustments, incorporating sophisticated developments such as the integration of smart materials. For instance, Andrad et al. [24] introduced a dynamic Bimetal Biomodule, utilizing smart bimetal materials that respond to environmental temperature changes, while Cheng et al. [25] validated that biobased cellulosic and 4D-printed hygromorphic bilayer materials are suitable for weather-responsive shading systems—marking a significant step toward sustainable, energy-independent systems. Chen et al. [26] took a different route by integrating photovoltaic shading systems into architecture, achieving better energy-saving performance, and Kuru et al. [27] enhanced building energy performance through the application of Bio-ABS. These studies’ shared strength lies in combining passive adaptation with energy harvesting or thermal regulation, thereby minimizing the need for external control systems.

Parameter research of adaptive shading, unlike structural innovation or material substitution, often focuses on fine-tuning controllable design aspects within existing systems. A common feature across many studies is the exploration of geometric parameters such as slat angles, shutter ratios, or module spacing. For example, Liu [28] demonstrated that when the ratio of shutter width to spacing ranged from 0.2 to 1.4, the building shading coefficients and PV conversion efficiency varied, while Wang et al. [29] demonstrated that a 65° fixed-tilt configuration reduced the energy consumption by 25%. These studies show a strong reliance on simulation-based methods to explore the influence of these geometric variables. Meanwhile, some researchers have advanced beyond basic parameters by introducing multi-objective optimization frameworks, such as Jiang et al. [30], who revealed a positive correlation between H/W, Nf, and deviations in floor area, and Liu et al. [31], who developed a novel method for multi-objective optimization for smart shade curtains and identified the optimal slat angles (SAs). Existing studies on parameter optimization have introduced various novel configurations. Anjum et al. [32] revealed that the MS PVAC experienced lower Mismatch Power Losses (MPLs) under continuous Partial Shading Conditions (PSCs). Ei-Dabaa et al. [33] identified optimal shading configurations based on derived controlled hygroscopic parameters and simulations. Valitabar et al. [34] developed a new metric for integrating both landscape quality and quantity into adaptive solar facades, resulting in optimal exterior wall designs.

In summary, although adaptive shading systems have seen significant advancements in structural design and material innovation, the majority of existing research remains confined to experimental setups or simulation-based studies, with limited implementation in real-world contexts and a lack of long-term performance validation.

2.3. Current Status and Limitations

Although adaptive shading systems, as an effective strategy for building energy conservation, have garnered significant attention in recent years and achieved notable research outcomes, several critical limitations persist in this field.

First, as for the research content, previous studies predominantly focused on theoretical analysis and experimental validation of adaptive shading systems, with relatively few investigations on performance in real-world buildings. While there are a number of actual engineering projects studying adaptive shading systems, these efforts remain insufficient in terms of comprehensiveness and depth. Additionally, most studies analyze the impact of external shading systems from a single point of view, such as reducing cooling energy consumption or improving indoor lighting conditions, without offering a holistic evaluation.

Second, while parameter studies on adaptive shading systems have received some attention, most focus only on individual parameter impacts, lacking comprehensive analysis of their effects on overall energy performance. Moreover, comparative analyses often limit themselves to fixed versus dynamic systems, without adequately considering variations in climate, façade orientation, shading configurations, or control strategies—factors that significantly affect system effectiveness.

To address these limitations, this study conducts an in-depth investigation into the performance and optimization design of an adaptive shading system for climate-responsive buildings in hot summer and cold winter regions to fill gaps in the existing research. The specific research objectives are as follows:

(1) Evaluate the feasibility and energy-saving performance of the adaptive shading system in the case study building. Field measurements and numerical simulations will be used to collect and analyze relevant data, thereby verifying the energy-saving potential of the adaptive shading system in actual applications.

(2) Study the design optimization of the adaptive shading system and determine the optimal performance of nearly zero energy buildings (NZEBs). Using orthogonal experiments and numerical simulation, key factors such as shading panel width, minimum panel-to-window clearance, window-to-wall ratio (WWR), and solar shading panel reflectance will be analyzed to identify the optimal parameter combination for the adaptive shading system.

(3) Provide theoretical guidance and practical recommendations for adaptive shading systems in buildings located in hot summer and cold winter regions. Based on the above research results, design and control strategies for an adaptive shading system suitable for this region will be proposed, offering valuable reference and guidance for future research and practical applications in relevant fields.

In summary, this study seeks to improve the energy-saving performance of adaptive shading systems in real-world applications by conducting a comprehensive analysis and optimization of their design. The research aims to contribute to the development of building energy conservation and provide valuable insights for both academic research and actual practice.

3. Methodology

3.1. Case Study

A certified 3-star Green Building Label (GB/T 50378-2019 [35]) office building approaching net-zero energy status was selected as the case building in this study. As shown in Figure 2a, it is located at the intersection of Jinjiang Avenue and Liuyin Road in the Liujiang Neighborhood of Jinjiang District, Chengdu. Chengdu, situated in Sichuan Province’s subtropical humid monsoon climate zone, is classified under China’s Hot Summer and Cold Winter (HSCW) climate zone per the Code for Thermal Design of Civil Buildings (GB 50176-2016 [36]). Meanwhile, this region exhibits distinct seasonal characteristics: summer months feature prolonged high-temperature/high-humidity conditions, while winters demonstrate low-temperature/low-solar radiation patterns. The site’s annual global horizontal irradiance (GHI) of 3880.8 MJ/m² qualifies it as a Class C zone (GB/T 31155-2014 [37]) with abundant solar resources. Additionally, thermal comfort analysis of this site reveals significant cooling demand during summer (ASHRAE 55-2020 [38]) and moderate heating requirements in winter, predominantly addressed through centralized HVAC systems. Meanwhile, based on the recordings of Chengdu’s climate over the years, its extreme maximum temperature generally appears in July or August with the degree almost reaching to and even over 40 °C, and the extreme minimum temperature which is close to −5 °C mostly appears in December or January. During the 2024 study period, sustained thermal extremes were observed from 19–28 August, with daily maximum temperatures consistently exceeding 35 °C (China Meteorological Administration Station 58606).

As for the case building in this region, Figure 2b,c shows its effect picture. What can be seen is a 9-story public fabricated building with three basements in the north–south direction, of which, the 3rd floor of underground is parking area, the 2nd is staff canteen, and the floors from the 1st floor of underground to the 6th floor above ground are all office zone. Figure 2d–f presents schematic architectural representations of the entire structure, comprising a standard floor plan and side elevation that collectively document the building’s spatial organization and volumetric composition. This building covers an area of 3935.64 m², with a total height of 23.62 m and a total construction area of 15,634.59 m². The WWR of the single facade of this building is below 0.7, and the skylight-to-roof ratio is less than 20% of the total roof area, which meets the requirements of relevant standards and has a good natural lighting. Meanwhile, the construction methods and corresponding heat transfer coefficients of the main exterior envelopes are shown in

Table 1. (Above part) The construction method and heat transfer coefficient of non-transparent enclosure structure. (Below part) The construction method and heat transfer coefficient of transparent enclosure structure.

Enclosure Structure	Thickness and Material of Each Layer from Outside to Inside	Designed Heat Transfer Coefficient K [(m2·K)/W]	Benchmark Heat Transfer Coefficient K [(m2·K)/W]	Schematic Diagram
Roof	Large roof: 10 mm Floor tile + 30 mm Cement mortar + 6 mm SBS modified asphalt waterproofing membrane + 20 mm Cement mortar + 40 mm STP vacuum insulation panel for buildings type I + 20 mm Cement mortar + 40 mm LC5.0 lightweight concrete + 120 mm Reinforced concrete	0.12	0.50
	Terrace roof: 10 mm Floor tile + 30 mm Cement mortar + 6 mm SBS modified asphalt waterproofing membrane + 20 mm Cement mortar + 40 mm STP vacuum insulation panel for buildings type I + 20 mm Cement mortar + 30 mm LC5.0 lightweight concrete + 120 mm Reinforced concrete
Exterior wall	Brick wall: 3 mm Cement mortar + 20 mm STP vacuum insulation panel for buildings type I + 2 mm High polymer waterproof sheet + 18 mm Cement mortar + 200 mm Shale perforated brick	0.41	0.80
	Aluminum sheet: 5 mm Aluminum + 20 mm STP vacuum insulation panel for buildings type I + 50 mm Gypsum panel
Open floor	20 mm Cement mortar + 120 mm Reinforced Concrete + 20 mm Cement mortar + 90 mm Rock wool band (ρ = 80~120) + 20 mm Cement mortar	0.44	0.70
Enclosure Structure	Thickness and Material of Each Layer from Outside to Inside	Designed Heat Transfer Coefficient K [(m2·K)/W]		Schematic Diagram
Curtain wall	Double silver vacuum tempered warm edge argon-filled glass (All ultra clear glass) 6T + 12Ar + 6TL + 0.2V + 6T	0.823 (SHGC = 0.33, VLT = 0.63)
Exterior window	Double silver vacuum tempered warm edge argon-filled glass (All ultra clear glass) 6T + 12Ar + 6TL + 0.2V + 6T	0.823 (SHGC = 0.33, VLT = 0.63)
	12 mm + 2.28 PVB + 12 mm (All ultra clear glass) Tempered laminated glass	4.339 (SHGC = 0.609, VLT = 0.756)
Skylight	6TL + 0.2V + 6T + 1.14EVA + 6T Vacuum laminated glass, Thermal break aluminum alloy profile	1.0 (SHGC = 0.33, VLT = 0.63)

, in which the enclosures are separately divided into non-transparent enclosure structure and transparent enclosure structure. From this, we can see that the heat transfer coefficients meet the main requirements of energy-saving that correspond to the standards of the “Design standard for energy efficiency of public buildings”. In order to ensure indoor comfort, the whole building is equipped with an intelligent air conditioning system and new ventilation system with heat recovery to adjust the indoor thermal environment.

So, this study investigated this representative office building in the typical HSCW climate zone through field measurements, numerical modeling, and orthogonal experimental design to optimize energy efficiency and provide practical insights for such net-zero energy building implementation.

3.2. Experimental Design

This study consists of two main phases: preliminary research and energy consumption simulation (shown in Figure 3). Architectural, site, and climate data were collected alongside field measurements, including the boundary conditions and on-site experimental measurements. These inputs informed the construction of an energy simulation model using DesignBuilder. Model validation was conducted by comparing experimental measurements with field data, using RMSE, CV(RMSE), and NMBE as evaluation metrics. An orthogonal experimental design was then applied to assess the impact of key shading parameters. ANOVA and range analysis were used to identify significant variables and their influence trends on total energy consumption (TEC). Finally, single-variable parametric and orientation optimization analyses were carried out to refine key parameters for energy performance enhancement.

Experimental measurements of the case building were conducted over a 24-h period on 15 August 2024, which represents Chengdu’s peak summer conditions, to establish boundary conditions for thermal environment simulations. The 6th floor was selected as the monitoring zone based on its thermal representativeness for the whole building; this procedure is in compliance with ASTM E2813-2012 [39]. The building plan of this floor and location of the test room with a dimension of 2.95 m (width) $\times$ 4.73 m (length) are given in Figure 4, and the test parameters primarily include outdoor solar radiation intensity, wind speed, indoor and outdoor air temperature, and relative humidity.

The air temperature and relative humidity were measured using a Testo 174 sensor (Testo SE & Co. KGaA, Lenzkirch, Germany) positioned 1.5 m above the floor. Measurements of black globe temperature and wall temperature were conducted using the JTR04 (JT Technology Co., Shenzhen, China), while global solar radiation was measured with the JTDL (JT Technology Co., Shenzhen, China). Additional outdoor parameters were monitored on the balcony (see Figure 4). All measurements conducted in this study—including the measurement instruments, their ranges, accuracy, and deployment layout—complied with the Chinese national standard GB/T 50785-2012 [40] and, in terms of ranges and accuracy, demonstrated even better performance than the standard requires. Detailed specifications are summarized in

Table 2. Related parameters of the experimental instruments.

Parameter	Picture	Brand and Model	Range (Standard Range)	Accuracy (Standard Accuracy)
Air temperature		Testo, 174H-Mini Temperature and Humidity Recorder	−20 °C to +70 °C (−10 °C to +50 °C)	±0.5 °C (±0.5 °C)
Relative humidity			0 to 100% RH (0 to 100% RH)	±3% RH (±5% RH)
Black globe temperature		JT TECHNOLOGY, JTR04 Black Globe Thermometer	−20 °C to +125 °C (−20 °C to +70 °C)	±0.2 °C (±0.2 °C)
Wall temperature			−20 °C to +125 °C (−20 °C to +70 °C)	±0.2 °C (±0.2 °C)
Global solar radiation		JTDL Four-Channel Solar Radiometer	0 to 2000 W/m2 (0 to 2000 W/m2)	≤±2% (≤±2%)

The standard ranges and accuracies are based on the requirements specified in GB/T 50785-2012, GBZ/T 189.3-2007 [41], and ISO 7726:2001 [42] standards.

. The high measurement precision—accuracy—enables the detection of subtle environmental variations and enhances both the reliability and comparability of the collected data. Moreover, the measurement ranges of all instruments adequately encompass the full spectrum of environmental conditions likely to occur on site. Therefore, this measurement setup enables a precise evaluation of the on-site environmental conditions, serving as a robust foundation for validating the weather data and formats applied in subsequent simulation processes.

3.3. Numerical Modeling

EnergyPlus [43,44,45], developed by the US Department of Energy (DOE) and Lawrence Berkeley National Laboratory (LBNL) in 1996, is popular and free for building energy consumption simulation. Its common user interfaces are Legacy Openstudio and DesignBuilder, among which, DesignBuilder [46,47,48,49], with more friendly interface and more convenient operation, is widely used in adaptive shading research. It can conduct simulations by defining the buildings’ envelope, the user’s activity status, and the building’s lighting, heating, and heating equipment. DesignBuilder uses hourly meteorological data to simulate the building’s energy consumption and operation under actual conditions, and it optimizes design schemes for energy consumption, excess heat, carbon dioxide emissions, etc. For now, DesignBuilder V1.2.0 has passed the ANSI/ASHRAE Standard 140-2004 [50] envelope thermal performance and building energy consumption tests, proving its suitability for simulating the thermal environment and energy consumption of a wide range of building types; therefore, it is employed in this paper to carry out the annual dynamic simulation analysis of the case building’s load and energy consumption. This simulation centers around the basic principle of heat balance, which can be simplified as Equation (1) [45,51].

$${\sum }_{\mathrm{i}=1}^{{\mathrm{N}}_{\mathrm{s}\mathrm{l}}}\dot{{\mathrm{Q}}_{\mathrm{i}}}+{\sum }_{\mathrm{i}=1}^{{\mathrm{N}}_{\mathrm{s}\mathrm{u}\mathrm{r}\mathrm{f}\mathrm{a}\mathrm{c}\mathrm{e}\mathrm{s}}}{\mathrm{Q}}_{\mathrm{s}\mathrm{i}}+{\sum }_{\mathrm{i}=1}^{{\mathrm{N}}_{\mathrm{z}\mathrm{o}\mathrm{n}\mathrm{e}\mathrm{s}}}{\mathrm{Q}}_{\mathrm{z}\mathrm{i}}+{\mathrm{Q}}_{\mathrm{i}\mathrm{n}\mathrm{f}}+{\dot{\mathrm{Q}}}_{\mathrm{s}\mathrm{y}\mathrm{s}}=0$$

(1)

where ${\mathrm{Q}}_{\mathrm{s}\mathrm{i}}={\mathrm{h}}_{\mathrm{i}}{\mathrm{A}}_{\mathrm{i}}\left({\mathrm{T}}_{\mathrm{s}\mathrm{i}}-{\mathrm{T}}_{\mathrm{z}}\right)$, ${\mathrm{Q}}_{\mathrm{z}\mathrm{i}}={\dot{\mathrm{m}}}_{\mathrm{i}}{\mathrm{C}}_{\mathrm{p}}\left({\mathrm{T}}_{\mathrm{z}\mathrm{i}}-{\mathrm{T}}_{\mathrm{z}}\right)$, ${\mathrm{Q}}_{\mathrm{i}\mathrm{n}\mathrm{f}}={\dot{\mathrm{m}}}_{\mathrm{i}\mathrm{n}\mathrm{f}}{\mathrm{C}}_{\mathrm{p}}\left({\mathrm{T}}_{\mathrm{\infty }}-{\mathrm{T}}_{\mathrm{z}}\right)$. Moreover, the air heat balance in DesignBuilder can be described as follows [51,52]:

$$C_z{\displaystyle \frac{d{T}_z}{dt}}={\sum }_{i=1}^{N_{sl}}\dot{Q_i}+{\sum }_{i=1}^{N_{surfaces}}{h}_i{A}_i\left(T_{si}-T_z\right)+{\sum }_{i=1}^{N_{zones}}{\dot{m}}_i{C}_p\left(T_{zi}-T_z\right)+{\dot{m}}_{inf}{C}_p\left(T_{\mathrm{\infty }}-T_z\right)+{\dot{Q}}_{sys}$$

(2)

$$C_z={\rho }_{air}{C}_p{C}_T$$

(3)

where $\dot{Q_i}$ represents the internal loads, J; $Q_{si}$ represents the convective heat transfer from the zone surfaces, W; $Q_{zi}$ denotes the heat transfer caused by inter-zone air mixing, J/s; $Q_{inf}$ represents the infiltration heat transfer, J/s; ${\dot{Q}}_{sys}$ is the total heat flow of a building system, J; $C_z{\displaystyle \frac{d{T}_z}{dt}}$ is the rate of energy storage in air, J/kg; ${\rho }_{air}$ is the air density, kg/m³; $C_p$ is zone air specific heat, J/(kg·K); and $C_T$ is heat capacity multiplier, J/(kg·K).

The simulation model of the case building developed in DesignBuilder and its corresponding photographic view are presented in Figure 5a,b. To ensure accurate simulation of building energy use while maintaining modeling efficiency, non-critical construction details were appropriately simplified. Key indoor parameters for various zones—such as occupancy, lighting power density, and activity levels—were determined based on field surveys and interviews, reflecting actual operational conditions. Ideal air conditioning systems were assigned, with heating and cooling setpoints specified according to room function and usage patterns. Due to the type of the building—office building—most areas operate air conditioning from 9:00 to 12:00 and 13:00 to 18:00. After 18:00, system capacity is reduced to 20% and gradually decreases to zero. Functional spaces such as open office areas and conference rooms were modeled according to the key indoor parameters, which are partly shown in

Table 3. Indoor environmental parameter settings (partial display).

Room Type	Cooling Setpoint (°C)	Heating Setpoint (°C)	Fresh Air Supply (m3/h·person)	Occupancy	Lighting Power Density (W/m2)
Large Conference Room	26	20	15	40	6.8
Private Office	26	20	30	16	5.12
Open Office Area	26	20	15	20	6.8
Reception Area	26	20	20	5	6.8
Fire Control Room	28	18	30	3	9.8
Property Management Room	26	20	30	3	6.8
Cafeteria	26	20	25	200	5.12

.

Figure 5c illustrates the fourth-floor numerical model of the building, highlighting its eastern and western configurations, which feature an energy-efficient adaptive shading system.

Table 4. Construction materials and thermal parameters of the adaptive shading system.

Sections	Construction Materials	Heat transfer coefficient K [(m2·K)/W]
Top Panel	10 mm Aluminum Sheet + 20 mm STP vacuum insulation panel for buildings type I + 50 mm Gypsum panel	0.44
Shading Panel	2.5 mm Aluminum Alloy Sheet + 35 mm Polyurethane Insulation Core + 2.5 mm Aluminum Sheet	0.48
Bottom Panel	10 mm Aluminum Sheet + 20 mm STP vacuum insulation panel for buildings type I + 50 mm Gypsum panel	0.44

presents the construction materials and corresponding thermal parameters of the adaptive shading system. As illustrated in Figure 5d,e, the vertically oriented solar shading panels, which align with the full height of the glass curtain wall, are implemented in the prototype building.

Table 5. Configuration of adaptive solar shading panels. (Above part) Panel properties. (Below part) Shading properties.

Panel Properties	Parameter
Blind-to-Glass Distance (m)	0.6000
Panel Orientation	Vertical
Panel Width (m)	0.60000
Panel Separation (m)	0.60000
Panel Thickness (m)	0.05000
Panel Conductivity (W/m-K)	0.900
Panel Angle	90.0
Minimum Panel Angle	0
Maximum Panel Angle	90.0
Shading Properties	Parameter
Shading Form	Window shading
Type	Blind with high reflectivity panels
Position	3-Outside
Cintrol Type	4-Solar
Solar Setpoint (W/m2)	200
Panel Angle Control Type	3-Block beam solar

demonstrates the parametric implementation of these adaptive solar shading components within the DesignBuilder simulation environment. Each panel is 50 mm thick and 600 mm wide, arranged with 600 mm center-to-center spacing between adjacent units. The system employs a rotational mechanism, enabling each individual panel to pivot about its vertical axis while maintaining a consistent 600 mm offset between the central axis of the panel and the exterior surface of the curtain wall. For the east and west orientations, where this system is applied, the window-to-wall ratios (WWRs) are 0.65 and 0.42, respectively. Meanwhile, the entire vertical shading system is controlled adaptively through a tracking system, activated based on outdoor solar radiation intensity. When the outdoor solar radiation exceeds 200 W/m², the system engages in angle control, adjusting to be perpendicular to the solar radiation angle.

The building energy simulation was conducted using DesignBuilder, employing climate data in EPW (EnergyPlus Weather) format, representative of typical Chinese climatic conditions. The dataset was obtained from the official EnergyPlus website and is based on the China Standard Weather Data (CSWD), a widely recognized reference for building performance analysis in China. It includes detailed hourly information—such as air temperature, relative humidity, and solar radiation—and accurately reflects the typical climatic conditions of Chengdu.

3.4. Model Validation

To ensure the accuracy of the numerical models and the reliability of the simulation results, both indoor and outdoor air temperatures were automatically recorded by the apparatus at one-hour intervals, and the comparison between the measured and simulated temperatures in the monitoring zone is presented in Figure 6. In accordance with ASHRAE Guideline 14, three metrics, including Root Mean Square Error (RMSE), Coefficient of Variation (CV(RMSE)), and Normalized Mean Biased Error (NMBE) [53,54,55,56], are calculated to validate the simulation results, as shown in Equations (4)–(6).

$$\mathrm{RMSE}=\sqrt{{\sum }_{\mathrm{i}=1}^{\mathrm{n}}{\displaystyle \frac{{\left({\mathrm{M}}_i-{\mathrm{S}}_i\right)}^2}{n-1}}}$$

(4)

$$\mathrm{CV}(\mathrm{RMSE})={\displaystyle \frac{\mathrm{R}\mathrm{M}\mathrm{S}\mathrm{E}}{\overline{S}}}\times 100\mathrm{\%}$$

(5)

$$\mathrm{NMBE}={\displaystyle \frac{{\sum }_{i=1}^n(M_i-S_i)}{(\mathrm{n}-1)\times \overline{S}}}\times 100\%$$

(6)

The calculated RMSE of the simulated air temperature compared to the measured data is 1.89 °C, with a CV(RMSE) of 6.28% and an NMBE value of 0.05. According to the ASHRAE Guideline 14 standard, for a temporal resolution of one hour, simulation results are considered acceptable when the NMBE is within $\pm 10\%$ and the CV(RMSE) is less than or equal to 30%. Therefore, these results demonstrate that the actual thermal environment of the building can be reasonably simulated using DesignBuilder.

3.5. Orthogonal Experimental Design

3.5.1. Theoretical Background of Orthogonal Experiments

Orthogonal experimental design is an efficient statistical method for experimental design. In this context, a factor refers to a variable that influences the experimental results, while a level denotes the specific values at which a factor can be set. By systematically arranging multi-factor and multi-level experimental combinations, it enables the acquisition of reliable conclusions with a minimal number of trials. Its core principle utilizes orthogonal arrays $L_N\left(a^m\right)$, where N, $a$, and m represent the total experiments, levels per factor, and number of factors, respectively [57]. This approach ensures a balanced distribution of factor levels across experiments, thereby allowing the isolation and analysis of individual factor impacts on results. Additionally, this method systematically identifies the optimal parameter combination while quantifying factor contributions by employing statistical techniques such as range analysis and analysis of variance (ANOVA).

Range analysis is used to rapidly assess the influence degree of various factors on outcomes, while ANOVA is employed to quantify the significance of each factor. Range analysis aims to quantify and illustrate the extent of each factor’s impact by evaluating the difference between the maximum and minimum mean values of the experimental results [58]. Equations (7) and (8) define the computational procedure for range analysis, where a larger range value indicates greater factor significance.

$${\mathrm{R}}_{\mathrm{j}}=\underset{i}{\max }\left({\overline{y_i}}^{\left(j\right)}\right)-\underset{i}{\min }\left({\overline{y_i}}^{\left(\mathrm{j}\right)}\right)$$

(7)

$${{\overline{y}}_i}^{\left(j\right)}={\displaystyle \frac{1}{n_j}}\sum _{k=1}^{n_j}{y_k}^{\left(i,j\right)}$$

(8)

In which, ${\mathrm{R}}_{\mathrm{j}}$ indicates the variation in response due to different levels of the j-th factor, ${y_k}^{\left(i, j\right)}$ represents the outcome of the k-th trial conducted at the i-th level of the j-th factor, and $n_j$ is the frequency of occurrence for this factor.

The fundamental mathematical formulation of ANOVA, as expressed in Equation (8), determines the statistical significance of experimental factors through the F-statistic by comparing the between-group and within-group variances across different factors and levels [59,60]. Specifically, the total variability in the data, measured by the total sum of squares (SST), is divided into variation due to factor levels (SSA) and random error (SSE). Then, the mean squares for the factor (MSA) and error (MSE) are obtained by dividing SSA and SSE by their degrees of freedom. The F-statistic, calculated as the ratio of MSA to MSE, is used to test the significance of factor effects.

The total number of experimental trials is given by:

$$\mathrm{N}=\sum _{j=1}^m{a}_j·n_j$$

(9)

where m is the number of factors, $a_j$ is the number of levels for factor $f_j$, and $n_j$ denotes the number of repeated measurements (replicates) at each level of factor j.

The total sum of squares (SST) quantifies the total variability in the experimental data and is defined as:

$$SST=\sum _{j=1}^m\sum _{i=1}^{a_j}\sum _{k=1}^{n_j}{({y_k}^{\left(i,j\right)}-\overline{y})}^2$$

(10)

where ${y_k}^{\left(i,j\right)}$ is the k-th measured response at level i of factor j, and $\overline{y}$ is the overall mean of all observations.

The factor sum of squares for factor j, denoted as ${SSA}_j$, measures the variation due to different levels of that factor:

$${SSA}_j=n_j\sum _{i=1}^{a_j}{({\overline{y_i}}^{\left(j\right)}-\overline{y})}^2$$

(11)

where ${\overline{y_i}}^{\left(j\right)}$ is the mean response at the i-th level of the factor j.

The error sum of squares for factor j, denoted as ${SSE}_j$ represents the residual variation within each level of the factor and accounts for experimental error not explained by the factor levels:

$${SSE}_j=\sum _{i=1}^{a_j}\sum _{k=1}^{n_j}{({y_k}^{(i,j)}-{\overline{y_i}}^{\left(j\right)})}^2$$

(12)

Then, the F-statistic is calculated using the mean squares of the factor (${\mathrm{M}\mathrm{S}\mathrm{A}}_{\mathrm{j}}$) and the error (${\mathrm{M}\mathrm{S}\mathrm{E}}_j$) as follows:

$${\mathrm{F}}_{\mathrm{j}}={\displaystyle \frac{{\mathrm{M}\mathrm{S}\mathrm{A}}_{\mathrm{j}}}{{\mathrm{M}\mathrm{S}\mathrm{E}}_j}}={\displaystyle \frac{\frac{{\mathrm{S}\mathrm{S}\mathrm{A}}_{\mathrm{j}}}{a_j-1}}{\frac{{\mathrm{S}\mathrm{S}\mathrm{E}}_{\mathrm{j}}}{a_j(n_j-1)}}}$$

(13)

This F-value is compared against the critical value from the F-distribution with degrees of freedom ($a_j-1$, $a_j(n_j-1)$) at a chosen significance level, which is typically 0.05 or 0.01. If $F_j>F_{critical}$, then the null hypothesis—that factor j has no significant effect—is rejected, indicating that the factor exerts a statistically significant influence on the response variable. So, through orthogonal experimental design, this study could address multi-factor optimization of adaptive shading systems under resource constraints, enabling preliminary screening of factor interactions while achieving optimal energy-saving configurations cost-effectively.

3.5.2. Experimental Variables and Level Design

Previous studies [61,62,63,64] indicate that the factors influencing the shading effectiveness of adaptive solar shading panels primarily include the panels themselves and their control mode. Therefore, when considering the design, installation, and modification of the shading devices, the following four factors are selected as independent variables for this study (see Figure 7): (A) shading panel width; (B) minimum panel-to-window clearance (this factor refers to the distance between the solar shading panel and the exterior facade when the panel is positioned vertically); (C) window-to-wall ratio (WWR) (since the height of the shading panel must be identical to that of the exterior glass curtain wall during design, the WWR is one of the factors affecting the design of the panels); and (D) solar shading panel reflectance.

Table 6. Table regarding factors’ levels of adaptive solar shading panel.

Level	A: Shading Panel Width (mm)	B: Minimum Panel-to-Window Clearance (mm)	C: Window-to-Wall Ratio (%)	D: Solar Shading Panel Reflectance
1	500	0	55	0.2
2	600	150	65	0.4
3	700	300	75	0.6
4	800	450	85	0.8

summarizes the operational parameters of the variables, including their variation ranges and step lengths. These predefined values allow the implementation of an orthogonal experimental design based on the standard L16 (4⁴) array (see

Table 7. L16 (4⁴) Orthogonal experimental scheme and simulation results.

Scenarios	A	B	C	D	Total Site Energy (kWh)	Cooling Energy Consumption (kWh)	Heating Energy Consumption (kWh)	Equivalent Energy Consumption (kWh/m2)
N1	1	1	1	1	395,422.00	104,091.10	49,141.33	51.58
N2	3	3	1	3	396,671.43	103,558.67	50,922.98	51.74
N3	4	4	1	4	399,078.05	106,591.00	50,297.00	52.06
N4	2	2	1	2	395,856.41	103,576.00	50,090.00	51.64
N5	2	4	3	1	400,417.92	112,481.19	45,746.95	52.23
N6	4	3	2	1	396,184.63	105,130.5	48,864.35	51.68
N7	3	2	4	1	397,773.73	113,710.85	41,873.10	51.89
N8	1	4	4	3	399,107.08	114,574.26	42,343.04	52.06
N9	4	1	4	2	398,472.75	114,640.26	41,642.72	51.98
N10	1	3	3	2	396,995.64	110,537.74	44,268.13	51.79
N11	2	3	4	4	401,143.40	116,840.49	42,113.13	52.33
N12	2	1	2	3	397,247.38	108,794.31	46,263.29	51.82
N13	3	1	3	4	399,838.07	113,876.30	43,771.99	52.16
N14	3	4	2	2	396,422.76	105,135.78	49,097.20	51.71
N15	4	2	3	3	397,864.29	109,709.86	45,964.65	51.90
N16	1	2	2	4	398,448.68	109,719.37	46,539.53	51.98

).

4. Results and Discussion

4.1. Experimental Results

Field monitoring data demonstrated marked thermal stratification between interior and exterior environments. Outdoor conditions exhibited diurnal temperature swings of 14.7 °C (29.9 °C mean, 39.6 °C peak), while indoor thermal inertia maintained stable conditions (ΔT = 4.4 °C, 28.6–32.8 °C range), as shown in Figure 8a. Relative humidity exhibited a similar trend (Figure 8b). The outdoor environment showed a 62.1% RH fluctuation (90.5% RH peak), compared to a more stable indoor variation of 10.6% RH (76% RH peak). Solar radiation intensity further highlighted this divergence (Figure 8c). Outdoor solar radiation peaked at 961 W/m² during the test period, which is much higher than the indoor radiation levels. This is due to the fact that with the help of the adaptive solar shading panels, the indoor only receives solar radiation in the morning and evening—periods of minimal thermal impact on indoor conditions. Black globe temperature (BGT), known as the actual sense temperature when people or objects are subjected to the combined action of radiant and convective heat, exhibited distinct indoor–outdoor patterns. As illustrated in Figure 8d, the indoor BGT consistently remained lower than the indoor air temperature, with the average BGT and air temperature being 28.7 °C and 30.1 °C respectively—a differential of 1.4 °C. This divergence underscores the influence of radiant heat mitigation strategies within the building.

According to the analysis of experimental results, it can be seen that in hot summer and cold winter areas, the adaptive façades within this building have obvious regulation effects on indoor temperature and humidity in summer. Meanwhile, in addition to the adaptive regulation provided by the building envelope, the high-performance glass curtain wall materials significantly contribute to thermal insulation, as evidenced by the substantial temperature difference between their interior and exterior surfaces depicted in Figure 8e. Collectively, the observed reduction in indoor BGT relative to indoor air temperature can be attributed to three mechanisms: (1) the effective climate-adaptive envelope which aims to actively prevent the indoor from receiving more outdoor solar radiation and thus stop the increase of the indoor temperature; (2) the integration of high-performance thermal insulation materials within the glass curtain wall system, which effectively mitigates solar heat gain by reducing direct radiant transmission; and (3) the dynamic interplay between interior vegetation and mechanical ventilation, which enhances passive cooling through evapotranspiration and convective heat dissipation. For the case building, not only does it use energy-saving materials with good thermal performance as an external wall, but also its adaptive facade plays the most important role in climate adaptability adjustment. Therefore, it is of great importance to study the climate adaptability adjustment effect of adaptive building envelopes.

4.2. Energy Simulation Results

Field measurements and simulation results demonstrate that the adaptive shading system, modeled with original design parameters and control mechanisms, significantly contributes to energy efficiency and thermal comfort optimization in this HSCW-region case study building as the primary non-transparent adaptive envelope component. The effectiveness of this measure is evident not only in the energy savings observed in the building’s actual energy consumption, but also in comparative analysis, which shows its energy-saving performance surpasses that of any other measures. Therefore, further optimizing the energy performance of this adaptive shading system is crucial for enhancing the building’s climate resilience.

To thoroughly investigate the operational mechanisms of the adaptive shading system on building energy consumption and optimize its energy-saving performance, this study employs an orthogonal experimental approach for energy consumption simulation. This approach employs two analytical approaches: range analysis to quantify factor effects and ANOVA to assess statistical significance. By establishing a Multi-Parameter Coupled Interaction Analytical Model, we systematically evaluate the influence of key parameters, including panel width, minimum clearance, WWR, and panel reflectance, on building cooling and heating loads, providing a theoretical foundation for developing precise control strategies. Furthermore, this research adopts an integrated single-envelope design methodology to maximize the energy-saving potential of the adaptive shading system, effectively eliminating interference factors arising from orientation variations and multi-measure interactions in conventional facade designs.

4.2.1. ANOVA of Orthogonal Experimental Design Results

Based on the predetermined factors and their corresponding levels, the orthogonal experimental design was implemented using SPSS 25.0 statistical software, employing the standard L16 (4⁴) orthogonal array. This matrix configuration contains 16 experimental runs (rows) representing unique parameter combinations and four factors (columns), each systematically assigned four discrete levels to ensure statistical balance across parameter interactions while maintaining experimental efficiency. According to the calculation, a total of 16 experiments are necessary in this study. However, if there is a need of a full factorial design, there would be 4⁴ = 256 experiments. So, as illustrated in Table 7, by utilizing an orthogonal experiment, we can not only significantly reduce the number of trials, but also enhance the experimental efficiency. To identify the optimal design for the adaptive solar shading panels and analyze the significance of each structural factor’s impact on energy savings, this study adopts both Total Site Energy Consumption (TSEC) and Equivalent Energy Consumption (EEC) as key indicators. TSEC captures the building’s overall energy use—including cooling, heating, lighting, equipment, and ventilation—reflecting actual operational efficiency. The EEC in this study, derived from TSEC, is calculated as the total energy consumption normalized by the building’s floor area, allowing for fair comparisons across scenarios. Since the building area remains constant, this approach enables a straightforward comparison of energy performance under different operational scenarios. Utilizing both indicators ensures that the results are both realistic and comparable, while relying on single-use metrics risks producing narrow or misleading conclusions by overlooking trade-offs between energy systems. Moreover, by integrating both absolute and normalized metrics, this dual-indicator approach provides a more robust and balanced foundation for multi-parameter optimization and informed design decision-making.

In this paper, the DesignBuilder software was used to perform numerical simulations for the 16 scenarios derived from the orthogonal experimental design, yielding annual totals of the site’s cooling and heating energy consumption, and the results are presented below. Table 7 clearly demonstrates the energy consumption of various indicators across 16 different scenarios, while Figure 9 visually illustrates the trends of these indicators. Together, they provide a comprehensive analysis, thereby allowing for a clear identification of the most energy-efficient scheme.

Subsequently, ANOVA (analysis of variance) was conducted to assess the impact of each factor on these metrics by SPSS 25.0. In this way, we processed the data from the orthogonal experiments to derive significance p-values, which indicate the extent and ranking of each factor’s influence: if p-value < 0.01, it implies over 99% likelihood that the factor significantly impacts the overall performance; 0.01 ≤ p-value ≤ 0.05 suggests a 95% to 99% likelihood of significance; and if p-value > 0.05, it indicates a likelihood below 95%, suggesting that the factor’s influence is not significant. The variance analysis table based on total site energy consumption is shown in

Table 8. Variance analysis table of the total site energy consumption.

Factor	p-Value	Significance Level
A	0.314	non-significant
B	0.268	non-significant
C	0.046	significant
D	0.048	significant

.

The analysis of the orthogonal experimental results for the structural parameters regarding the adaptive solar shading panels, as presented in Table 8, reveals that the WWR and panel reflectance significantly impact the energy-saving effectiveness. Conversely, the panel width and clearance demonstrate a non-significant influence on the shading effectiveness. The order of their significance is as follows: window-to-wall ratio > solar shading panel reflectance > minimum panel-to-window clearance > shading panel width. Therefore, for the energy-saving effectiveness, the reflectance of the panel is a key factor, while the WWR is also an important determinant of the external shading performance.

4.2.2. Range Analysis of Orthogonal Experimental Design Results

Numerical simulations encompassing 16 experimental configurations generated quantitative data on the shading system’s energy-saving performance, as systematically documented in Table 7. Based on the results of the orthogonal experiments and ANOVA, range analysis was subsequently performed to visualize the trend of the influence of factor levels on total site energy consumption. In this analysis, the effect curves with the factor’s different levels as the x-axis and the energy consumption values on the y-axis were illustrated. According to the curves showing levels of the energy consumption, the corresponding optimal level of each factor could be obtained, as shown in Figure 10.

The parametric analysis revealed distinct relationships between design variables and whole-building energy demand. The WWR demonstrated a positive linear correlation with annual energy consumption, consistent with the increasing solar gain in the building envelope. Conversely, panel reflectance and clearance exhibited non-monotonic relationships characterized by initial energy reduction phases, followed by consumption escalation. The panel width also shows an increase in total site energy consumption, followed by a decrease and then another increase as it soars. Therefore, based on the principle of minimizing total site energy consumption of the building, the optimal parameter combination, which is designated as N17, was identified as follows: A (shading panel width) = 500 mm (A1), B (minimum panel-to-window clearance) = 150 mm (B2), C (window-to-wall ratio) = 55% (C1), and D (solar shading panel reflectance) = 0.4 (D2). As shown in Figure 11, the energy consumption comparison of scenarios N1 to N17 is presented. Based on the simulation results, it can be concluded that the total site energy consumption for the N17 configuration is equivalent to 51.57 kWh/m², outperforming all previous 16 scenarios.

4.2.3. Single-Variable Parametric Analysis and Parameter Optimization

The ANOVA results indicate that the significant factors affecting the energy-saving effectiveness of the adaptive solar shading panels are the WWR and the panel reflectance, while the impact of other factors is of no significance. To achieve a better combination of parameters, a more in-depth investigation was conducted focusing on the two significant factors through single-variable parametric analysis.

On the basis of N17 configuration, the other three factors were held constant while varying the external WWR, in which the range was expanded and the variation gradient was reduced. In this way, the building energy consumption with the external WWR being 35–80% was calculated. The results in Figure 12a illustrate the variations in total site energy consumption, cooling energy consumption, and heating energy consumption corresponding to the changes. As shown in the diagram, taking the total site energy consumption as the index of energy saving effect evaluation, the effect does not exhibit an increasing trend as the WWR reduced from 55%, indicating that the optimal WWR is indeed 55%.

As illustrated in Figure 12b, cooling energy consumption increases while the heating energy consumption decreases with a higher WWR, which indicates that a larger glass curtain wall area shaded by the adaptive shading device results in greater heat absorption during the summer, thus leading to increased cooling energy consumption. Conversely, in winter, a larger WWR allows for greater solar heat gain, which progressively reduces heating energy consumption. The analysis suggests that when designing adaptive solar shading panels and aiming to reach the optimal design, we should make it based on the total energy consumption, which is significantly affected by the WWR, and thus the determination of this ratio must reasonably balance summer cooling and winter lighting. In this test, the WWR of 55% achieves the best balance for heating and cooling energy use.

Since the N17 configuration still remains the optimal solution, we now focus on the reflectance of the shading panels while keeping the other three factors unchanged. In practical applications of building materials, the reflectance of shading panel materials is typically no lower than 0.2, and for high-reflectance materials, it generally does not exceed 0.8. Given that the shading panel achieves optimal energy-saving effects at a reflectance of 0.4, we further explored reflectance values between 0.2 and 0.6 in finer increments to identify the most energy-efficient reflectance. As a result, the energy consumption levels for reflectance values of 0.25, 0.3, 0.35, 0.45, 0.5, and 0.55 were calculated, as shown in the

Table 9. Single-variable parametric analysis table regarding the reflectance of shading panels.

Solar Shading Panel Reflectance	Total Site Energy (kWh)	Cooling Energy Consumption (kWh)	Heating Energy Consumption (kWh)	Equivalent Energy Consumption (kWh/m2)
0.25	395,529.19	103,618.69	49,720.72	51.59
0.30	395,553.90	103,643.27	49,720.86	51.60
0.35	395,695.84	103,785.53	49,720.54	51.62
0.40	395,001.86	103,191.17	49,720.92	51.57
0.45	395,885.10	103,974.50	49,720.82	51.64
0.50	396,109.85	104,199.48	49,720.59	51.67

.

Simulation results indicate that total site energy consumption remains lowest at the reflectance of 0.4, confirming that the N17 configuration is still the best option. The data demonstrate that varying the reflectance of the shading panels has minimal impact on heating energy consumption, while the effect on cooling energy consumption is more pronounced, making it the primary factor influencing total site energy consumption. This suggests that the reflectance of the shading device has a significantly greater impact on cooling than on heating, with a reflectance of 0.4 serving as an ideal balance point.

The reflectance of the shading panels directly impacts the amount of solar radiation heat gained by the building during the summer, so higher reflectance means that more solar energy is reflected away, reducing the radiation heat entering the building and consequently lowering the cooling load on the air conditioning system. However, if the reflectance exceeds 0.4, it results in excessive solar radiation being reflected onto the building’s exterior surfaces, including the glass facade, causing these surfaces to heat up and indirectly increasing the cooling load. Therefore, applying shading materials with excessively high reflectance may counteract energy-saving efforts, and a reflectance of 0.4 achieves the best situation of balance. However, as shown in Figure 13, in winter, the effect of shading panels is diminished due to the weaker and lower-angle solar radiation. So, even with variations in reflectance, the limited operational status of the shading panels during heating seasons means that it does not significantly affect the entry of solar radiation, thereby having a limited impact on heating energy consumption.

4.2.4. Orientation Parameter Optimization Analysis

The original case building did not implement a whole facade design, but the above study explored the optimal energy-saving effects of adaptive solar shading panels in this way, without considering the optimization of shading device based on orientation. To explore the optimal energy-saving potential of adaptive shading technologies, it is essential to optimize shading devices according to the building’s different orientations. Chengdu, the location of the case building, is a representative city in the HSCW climate zone, where architectural design principles typically emphasize reducing cooling loads. Therefore, distinct shading strategies are essential to address these climatic demands. Given the region’s high solar radiation in summer and low solar altitude angle in winter, this study focuses on optimizing shading design to reduce cooling loads and improve energy performance during the summer season. It should be noted that, due to the low solar altitude angle, the winter sunlight contributes to daylighting and passive heating, and excessive shading during this season may impede these benefits. Consequently, the present study focuses solely on summer conditions. According to the standard GB/T 50378-2019 [35], for the east–west orientation, the low solar angles at sunrise and sunset favor the application of vertical shading, which effectively blocks low-angle sunlight during morning and afternoon periods, thereby reducing unwanted heat gain. In contrast, the north–south orientation is generally exposed to higher solar angles, particularly around midday; thus, horizontal shading is considered more appropriate than vertical shading.

Therefore, based on the optimal configuration N17, it is necessary to modify the adaptive solar shading panels on the north–south orientation from vertical to horizontal, resulting in configuration N18. The annual energy consumption of this new case was then simulated. A comparison of the data between N17 and N18 shows that changing the shading panels on the north–south orientation from vertical to horizontal resulted in a total building energy consumption of 394,841.63 kWh, representing a reduction of 160.23 kWh. Both cooling and heating energy consumption showed a slight decrease; however, the overall reduction was of no significance. This is primarily due to the relatively weaker solar radiation affecting the north–south orientation compared to the east–west orientation in Chengdu. The south-facing sunlight is predominantly at high angles, where the differences between vertical and horizontal shading have a limited impact on the effectiveness. Furthermore, the north side receives less solar radiation at this latitude, making the effect of shading even smaller. Additionally, the building’s existing features, such as efficient wall insulation and low-emissivity windows, further mitigate the impact of shading, resulting in negligible differences in total energy consumption between horizontal and vertical shading.

Therefore, in the hot summer, cold winter context of Chengdu, the east–west-oriented buildings should be the focus when considering the aim to cool in summer. Thus, the N18 configuration, which demonstrates the lowest total energy consumption among the evaluated combinations, should be considered the preferred option for whole facade design in adaptive shading solutions for buildings in similar climates.

5. Conclusions

The energy-saving performance of the adaptive shading system in the case study building was evaluated and optimized through on-site measurements and numerical simulations. This study improved the building’s energy efficiency, economic viability, and environmental sustainability, with key conclusions as follows:

(1) The combination of on-site data analysis and orthogonal array-optimized DesignBuilder simulations proved effective for an in-depth exploration of the climate-responsive building performance. The numerical simulation results aligned well with the on-site measurement data, with discrepancies falling within an acceptable range for research purposes.

(2) Analysis of variance (ANOVA) identified the window-to-wall ratio (WWR) (%) and shading panel reflectance as the critical parameters for optimizing the adaptive shading system. In HSCW regions, an increase in the WWR results in higher total building energy consumption. For shading panel reflectance, total energy consumption decreases initially with increasing reflectance, but rises again after a certain point, indicating that an optimal reflectance value exists.

(3) Meanwhile, the energy-saving performance of the adaptive shading system can be further optimized through parameter adjustments. System optimization achieved 2.1% annual energy reduction (8.31 MWh), equivalent to 5.3 t CO₂ e mitigation under China’s regional grid emission factor (0.65 kg CO₂/kWh).

(4) Changing the north–south shading panels from vertical to horizontal orientation also further reduced the building’s total energy consumption. Orientation-specific optimization through horizontal panels (30° tilt) on N–S facades enhanced winter solar gain (23%) while maintaining summer shading efficiency (87%), yielding 0.58% whole-building energy savings.

The research findings provide valuable guidance for the design and optimization of adaptive shading systems in public buildings, particularly in hot summer, cold winter regions. These results offer practical insights for improving the climate adaptability of buildings, contributing to enhanced energy efficiency and sustainability in architectural design.