Energy harvest and daylighting performance of dynamic photovoltaic shading system-a comparison study in various climate zones of China

Abstract

Dynamic photovoltaic shading systems (PVSDs) that can automatically adjust PV angles according to solar altitude and azimuth to enhance solar energy utilization, have become an important research focus in building-integrated photovoltaics (BIPV). The geographical and climatic characteristics of different cities have a significant impact on the optical performance and power generation efficiency of PVSDs, yet prior studies generally lack systematic cross-climate evaluations and regional correlation analyses. To address this gap, this study assessed the daylighting performance of dynamic PVSDs in six representative Chinese cities under diverse climatic conditions and conducted a regional correlation analysis. The results showed that, regardless of dynamic or static control strategies, south-facing PVSDs consistently achieved the highest annual electricity generation, followed by west-facing PVSDs. Compared with horizontal static systems, dynamic systems significantly increased annual power generation by 16.1%—29.9%. Using the south-facing configuration as an example, Beijing’s dynamic PVSDs yielded the highest annual power generation at 240 kWh, while Chengdu recorded the lowest at 110.54 kWh. In addition, during summer, dynamic PVSDs effectively met indoor lighting demands and improved illuminance uniformity; for instance, in Harbin, illuminance uniformity increased by 1.1%. Overall, dynamic PVSDs balanced power generation and daylighting objectives across different urban contexts, demonstrating practical value and broad applicability for building energy conservation and indoor daylighting environment optimization.

Graphical abstract

Keywords

Building integrated photovoltaics / Daylighting / Solar energy / Dynamic PV shading / Cross-climate Simulation

1 Introduction

In recent years, the rapid growth of the global population, accelerated urbanization, and advancements in industrialization have driven a continuous increase in energy consumption. According to predictions by the International Energy Agency, global energy demand is expected to rise by 30% between 2017 and 2040 (IEA, 2020a). Photovoltaics (PV) is a technology that employs solar cell semiconductor materials to directly convert solar energy into electrical power, representing a clean, highly efficient, and sustainable form of renewable energy (Ali et al., 2025). As shown in Fig. 1, the cumulative installed capacity of global PV systems exhibits a consistent growth trend, with China being the largest market for installed capacity (IEA, 2020b; Hossain et al., 2024).

To achieve the national 2060 carbon neutrality target, solar energy is expected to occupy a pivotal position within China’s energy structure. Furthermore, BIPV represents an innovative technology that integrates PV power generation systems with architectural envelope (Eder et al., 2019; Shukla et al., 2017; Saretta et al., 2019; Batista et al., 2025). These systems can be integrated into various building components, including facades, rooftops, and windows (Shukla et al., 2017; IEA, 2021; Ghosh, 2022). Among these applications, PVSDs, as a significant implementation of BIPV, simultaneously provide shading functions and efficiently harness solar energy (Chen et al., 2024; Kirimtat et al., 2022; Stamatakis et al., 2016; Zhang et al., 2017). Study has shown that, around 17%—34% total energy consumption could be saved with PVSDs in ideal yearly fixed titling angles in 5 different European cities (Baghoolizadeh et al., 2022). This integration offers a crucial pathway for promoting the synergy between green building development and energy transition. Compared to fixed PVSDs, dynamic PVSDs have showed more promising energy output in many scenarios. A study in Australia context showed that, the dynamic PV overhangs could support the office space to achieve even zero energy performance in certain climate zone (such as where Perth locates) (Krarti and Karrech, 2024). Wang et al. (2025b) showed that, in the context of Hong Kong, monthly and hourly adjusted dynamic PVSDs could generate around 9% and 15% more electricity than PVSDs fixed at yearly optimized angle. Lee et al. (2025) developed an airflow PVSDs that achieved higher daily power generation and lower annual net energy consumption compared to non-airflow PVSDs. Taveres-Cachat et al. (2019) improved the performance of PVSDs by optimizing the number and angle of louver slats. Jayathissa et al. (2017) evaluated dynamic PVSDs, and the results demonstrated that they could achieve net energy savings of 20%—80% compared to static PVSDs. Asfour (2018) simulated vertical and horizontal PVSDs at various tilt angles and found that PVSDs with a 45° tilt angle achieved the highest annual total solar radiation while providing an average of 96% effective window shading during summer. Han et al. (2024) investigated the impact of installation and architectural parameters on power generation in bifacial PVSDs, revealing that south-oriented configurations with smaller tilt angles and wider shading louvers achieved greater energy-saving benefits. Apart from pure energy performance, there is a growing trend of exploring multi-objective optimization of PVSDs, such as including daylight and visual comfort, payback time, etc. (Taveres-Cachat et al., 2019; Mandalaki et al., 2014; Wang et al., 2025). In summary, existing research on PVSDs generally lacks systematic evaluations across climate zones and analyses of regional correlations. Accordingly, this study focuses on the daylighting performance of dynamic PVSDs under cross-climate conditions and conducts a correlation analysis based on six representative cities. The innovation of this work lies in the integrated application of solar tracking for dynamic angle optimization, combined with assessments of daylight uniformity and illuminance compliance, to evaluate the daylighting performance of dynamic PVSDs across different climates. The findings not only provide theoretical support for balancing power generation and visual comfort in dynamic PVSDs but also offer a reference for their design and practical application.

A summary of the key information of existing PVSDs studies in recent 5 years was listed in Table 1. Prior studies mainly focused on the aspects of static PVSDs, a few of them explored the dynamic PVSDs, but there is a lack of investigation of dynamic PVSDs in the aspects of both energy and daylight and including the comparison of different climate zones. This research aims to fill the research gaps of understanding the overall performance of both energy and daylight of dynamic PVSDs in various climate zones, to provide a more general understanding of the benefits of this novel system, as well as the potential trade-off strategies.

2 Research methodology

This study employs a three-step research process. In the first step, six cities in China (Beijing, Chengdu, Guangzhou, Harbin, Kunming, and Shanghai) with different representative climatic zones were selected as the simulation regions to ensure the scientific rigor and practical applicability of the results across the country (as shown in Table 2). Then high-precision solar tracking simulations were conducted using Rhino and ClimateStudio to analyze solar radiation under various shading angles, thereby determining the optimal angles for PVSDs’ energy generation and providing critical parameters for subsequent indoor lighting performance evaluations (as shown in Table 3).

The corresponding energy generation for different tilting scenarios in each city was also calculated. In the second step, a digital 3D model of an office room was created in DesignBuilder for in-depth interior daylight analysis. The optimal shading angles for various cities obtained in the first step were employed for simulation, with a focus on assessing the impact on indoor light distribution and illuminance uniformity across different times, seasons, and weather conditions. Finally, a correlation analysis was conducted by applying the Spearman correlation method to explore the potential relationships between environmental factors and PVSDs’ energy and daylight performance. This three-step approach comprehensively validates the application effectiveness of PVSDs in green building design within typical cities. The research flow was illustrated in Fig. 2.

2.1 Optimal angular determination for PVSDs through solar tracking simulation

To evaluate the performance of the dynamic PVSDs under varying climatic conditions, several typical cities across different climate zones in China were selected as simulation regions. These included Beijing, Harbin, Shanghai, Chengdu, Guangzhou, and Kunming. The selection of these cities aimed to comprehensively cover a range of climatic types from cold to hot, enabling analysis of the dynamic PVSDs’ impact on indoor lighting environments under diverse environmental conditions.

For each city’s scenario, two mounting strategies for PVSDs were compared: fixed horizontal systems, in which the panel tilt remains constant throughout the year, and a monthly-adjusted configuration (ß_m for m = 1, 2, …, 12), in which tilt is reset each month in 5° increments. We adopted 5° increments on the basis of prior findings that finer adjustments yield <2% additional annual energy (Wang et al., 2025). Monthly adjustment thus offers a practical compromise between performance gain and maintenance effort.

The tilting angle simulation workflow comprises two sequential phases. In phase 1, detailed geometric models of the PVSDs and building surfaces using Rhino/Grasshopper were generated. In phase 2, these models were imported into ClimateStudio to perform an hourly radiation analysis over a full meteorological year (8760 h). The simulation draws on Typical Meteorological Year (TMY) data―direct normal irradiance, diffuse horizontal irradiance, ambient temperature, relative humidity, and wind speed based on different cities’ databases. The output provided plane-irradiance values (G_h(β), kWh/m²) for each hour across all candidate tilt angles (β = 0°, 5°, 10°, …, 90°) (IEA, 2020b). For example, Fig. 3 demonstrated the hourly irradiance values for south-facing PV shading with horizontal tilt in Beijing.

Then the hourly irradiance values were aggregated to compute annual and monthly energy yields. For the fixed horizontal system, annual irradiation per unit panel area is

\begin{array}{}\text{(1)}& R_{\text{fixed}}(\beta )={\displaystyle \sum _{h=1}^{8760}{G}_h(\beta )}\mathrm{,}\beta =0^{\circ }\mathrm{.}\end{array}

For the monthly-adjusted system, we partition the year into twelve calendar months M_m (m = 1, …, 12). Monthly irradiation at tilt ß is

\begin{array}{}\text{(2)}& R_m(\beta )={\displaystyle \sum _{h\in M_m}{G}_h(\beta )}\mathrm{.}\end{array}

We use a traversal algorithm to identify the optimal tilt angle for each month

\begin{array}{}\text{(3)}& {\beta }_m^{*}=argma{x}_{\beta \in \mathrm{0,5,}\dots \mathrm{,90}}{R}_m(\beta )\mathrm{,}\end{array}

and the dynamic system’s annual yield is

\begin{array}{}\text{(4)}& R_{dyn}={\displaystyle \sum _{m=1}^{12}{R}_m({\beta }_m^{*})}\end{array}

Figure 4 showed the pseudocode for calculating the annual solar radiation of monthly dynamic PV shadings.

A simple PV model is utilized to simulate PV energy output by considering four primary factors: area of surface (A_surf), module conversion efficiency (η_cell), DC to AC conversion efficiency (η_invert), and the proportion of the surface area containing active solar cells (f_active). These parameters are assumed to remain constant and are not adjusted for variations in temperature and cities. The annual electrical power produced by PV shadings is calculated using

\begin{array}{}\text{(5)}& E=A_{\text{surf}}*f_{\text{active}}*{\eta }_{\text{cell}}*{\eta }_{\text{invert}}*R\mathrm{.}\end{array}

2.2 Indoor daylight simulation setting

A digital office model was defined and created in the professional software DesignBuilder for the simulation in this study. The office dimensions were set at 3 m (length) × 3 m (width) × 3 m (height), with a window centrally positioned on the façade. The window dimensions were 1.7 m in width × 1.6 m in height, and the bottom edge of the window was 0.9 m above the floor. A dynamic PVSDs were installed above the window, with dimensions of 0.7 m in width × 1.7 m in length, consistent with the width of the window to achieve optimal shading coverage (the digital model as shown in Fig. 5). The dynamic PVSDs were made of monocrystalline silicon. This choice was made due to the technology’s maturity and economic feasibility. The authors have done an outdoor experimental test of PVSDs, which were installed on a real temporary building and measured the electrical data continuously for days under different weather conditions (including sunny days and cloudy days (Wang et al., 2025). The prior results showed the monocrystalline silicon PVSDs have promising energy output capacity and presented an average daily energy conversion efficiency of nearly 16%. In this study, for annual simulations, we set the PV conversion efficiency at 15%, which should present reliable results referring to the real field measurements. In this study, illuminance level and illuminance uniformity were selected as the evaluation indices for daylight performance.

When creating a comfortable indoor lighting environment, it is essential to ensure adequate illuminance while minimizing uneven distribution of light, which could lead to negative impacts on humans’ visual comfort and work efficiency (Xiang and Matusiak, 2022; Shishegar and Boubekri, 2016; Aries et al., 2015). The standards for illuminance vary depending on the type of visual task, the intended use of the space, and users’ visual needs, with relevant regulations differing across countries and regions. In this study, based on the Chinese national standard GB/T 50034—2024, the target illuminance for the task area of a standard office is set at 300 lx at the height of desktop. Illuminance uniformity is a key parameter for assessing how evenly light is distributed within a specific area. This study uses the ratio of the minimum illuminance to the average illuminance as the evaluation method. According to GB/T 50034—2024, this ratio should not be lower than 0.6 to ensure a comfortable and efficient indoor lighting environment.

During the daylighting simulation, the lighting power density was set at 0.8 W/m ², and the measurement point was positioned at a height of 0.8 m, corresponding to the typical height of an office desk. The illuminance measurement point was placed at the center of the office space. The lighting system was scheduled to operate from Monday to Friday, between 7:00 a.m. and 7:00 p.m. The reflectance values of the interior surfaces were set as follows: 80% for the walls, 50% for the ceiling, and 35% for the floor. In addition, for all simulation scenarios, the sky condition was set as a CIE clear day and four representative solar terms―Spring Equinox (March 20), Summer Solstice (June 21), Autumn Equinox (September 22), and Winter Solstice (December 21)―were selected as evaluation points. This approach allowed for a comprehensive analysis of the impact of seasonal variations on the indoor lighting environment.

2.3 Correlation analysis

In order to explore the potential relationships between environmental factors (e.g., solar radiation, ground horizontal illuminance) and performance indicators such as PV power output and indoor visual comfort, correlation analysis plays a vital role. The most commonly employed methods are Pearson, Kendall, and Spearman correlation coefficients. The Pearson coefficient is widely used to assess linear association between continuous variables (Sedgwick, 2012). The Kendall coefficient measures rank correlation and is especially suitable for evaluating large datasets (Kendall, 1938; Xu et al., 2013). Spearman’s rank correlation coefficient is also used to capture relationships between ranked features and is well-suited for analyzing both linear and non-linear associations (Spearman, 2010). Compared to Kendall, Spearman method is generally more sensitive in detecting weak correlations within small dataset (Ye et al., 2023). Considering the potentially non-linear and weak relationships between environmental and performance variables, Spearman correlation analysis is applied in this study. Spearman’s rank correlation coefficient measures the strength and direction of a monotonic relationship, with values ranging from -1 (a completely negative correlation) to 1 (a completely positive correlation). The expression for calculating this coefficient is as follows:

\begin{array}{}\text{(6)}& r_s=1-\frac{6{\displaystyle \sum }{d}_i^2}{n(n^2-1)}\mathrm{.}\end{array}

d_i denotes the rank difference between each paired of observations, while n indicates the total number of paired data points.

3 Results and discussion

3.1 Optimal tilt angles of dynamic PVSDs for electricity production

The monthly optimal tilt angles of dynamic PVSDs for energy generation in different cities were obtained from the solar tracking simulation. For each city, three orientations including south, east, and west were analyzed. For all six cities, the optimal tilt angles of south-oriented PVSDs showed the largest angle variations. For instance, in Beijing, the optimal tilt angles of south-oriented PVSDs range from 20° in June to 70° in December (0° means horizontal and 90° refers to vertical). While the optimal tilt angles of east-oriented and west-oriented PVSDs were limited to a small range of around 40°—60°. For instance, Guangzhou, the optimal tilt angles of east-oriented PVSDs range from 45° to 50°, showed slightly angle changes. The optimal tilt angles of dynamic PVSDs in four representative solar terms (March, June, September and December) were listed in Table 4.

3.2 Annual energy production

Figure 6 illustrated the annual power generation distribution of dynamic PV shading systems in different cities. The results showed that, regardless of whether the PVSDs was dynamic or static, the south-facing orientation consistently achieved the highest annual power generation, followed by the west, with the east orientation being the lowest. Critically, across all three orientations (east, south, and west), the dynamic PVSDs consistently outperformed static PVSDs in power generation. Taking the south-facing orientation as an example, Beijing had the highest annual power generation with dynamic PVSDs (240.36 kWh), followed by Harbin (223.65 kWh), while Chengdu had the lowest (110.54 kWh). A further comparison of the annual power generation between dynamic and static horizontal PVSDs facing south (see Table 5) showed that in Harbin, the dynamic system achieved the largest increase over the static system, with an improvement of 29.9%, whereas in Chengdu, the increase was the smallest but still achieved 16.1%. These results indicated that northern cities typically possessed more benefits from optimal angles for PV power generation and received higher levels of solar radiation, thereby demonstrating a clear advantage in power output. In contrast, low-latitude cities such as Chengdu, Guangzhou which experienced more cloud cover, exhibited significantly lower energy production compared to high-latitude cities like Harbin and Beijing, which enjoyed more sunny days.

3.3 Daylighting performance

The daylight simulation results showed that, although the indoor illuminance under the static PVSDs was generally slightly higher than that of the dynamic PVSDs, in all studied cities, the dynamic PVSDs maintained indoor illuminance levels above 300 lux at all times except 18:00 (as shown in Table 6 and Table 7), thereby effectively meeting the requirements for daily indoor lighting.

Figures 7—12 presented indoor illuminance and uniformity of static and dynamic PVSDs across four seasons in different cities. Based on the indoor illuminance statistics, a comparative analysis was conducted for south-facing rooms equipped with static and dynamic PVSDs, respectively. It was observed that the two shading strategies exhibited significant differences in indoor daylighting performance. Firstly, under the fixed PVSDs (see Table 6), the indoor illuminance of south-facing rooms in various cities during the vernal equinox, summer solstice, autumnal equinox, and winter solstice showed that cities at higher latitudes generally had higher average indoor illuminance levels. For example, Beijing had an annual average indoor illuminance of 866.4 lux, the highest among all cities, followed by Harbin with 848.9 lux. In contrast, cities at lower latitudes, such as Guangzhou, Chengdu, and Kunming had lower annual average illuminance than Harbin and Shanghai. This phenomenon indicated that, under static PVSDs, high-latitude regions benefited from lower solar altitude angles and longer daylight durations, thus receiving more abundant sunlight. Even though the shading system reduced some incoming light, high-latitude cities still maintained superior indoor daylighting conditions.

In comparison, under the dynamic PVSDs (as shown in Table 7), the annual average indoor illuminance values of each city were generally lower, and the distribution among cities was more uniform. For instance, the annual average indoor illuminance of Beijing and Shanghai was 672.2 lux and 673.58 lux, respectively, while those of Harbin, Chengdu, Guangzhou, and Kunming were also relatively close. This suggested that dynamic PVSDs could intelligently adjust the shading area according to the solar altitude and sunlight intensity, more effectively responding to different solar terms and weather conditions. As a result, indoor illuminance remained relatively consistent and stable throughout the year, avoiding extreme fluctuations in light distribution. At the same time, the dynamic shading system had a more pronounced effect on balancing indoor illuminance among cities at different latitudes, thus enhancing the uniformity of indoor lighting between northern and southern cities. In summary, under static PVSDs, the higher the city’s latitude, the better the average indoor illuminance, which presented a clear geographical differentiation. In contrast, the dynamic PVSDs effectively mitigated this difference, making the indoor daylighting environment more balanced across cities at different latitudes.

Table 8 showed the indoor uniformity increase (%) of static PVSDs relative to dynamic PVSDs across different cities. The results indicated that, under summer conditions, dynamic PVSDs exhibited a slightly better overall performance than static PVSDs across cities, with the maximum improvement in indoor illuminance uniformity reaching 2.2%. This suggested that, in most cities, dynamic PVSDs not only met the required indoor illuminance standards but also delivered higher lighting quality in summer, thereby providing a more balanced and comfortable visual environment. While in the other seasons, the indoor daylight uniformity of rooms with PVSDs may not be superior to the ones with fixed PVSDs.

3.4 Trade-off analysis of energy production and daylight performance

Based on the results of energy generation in Section 3.2 and the daylight performance in Section 3.3, an energy-daylight trade-off analysis was conducted. In all climate zones and all orientations, the dynamic PVSDs consistently outperformed static systems in energy generation, achieving increases in annual electricity production. While static PVSDs provided slightly higher indoor illuminance levels in some scenarios, the seasonal average illuminance of dynamic PVSDs in various cities maintained indoor illuminance above 600 lux (double that of the building standard required 300 lux) during most periods, effectively meeting interior daylight requirements. In general, dynamic systems were more attractive than fixed systems. However, in some specific scenarios, trade-off consideration and strategies could be considered.

In summer seasons, rooms with dynamic PVSDs showed the least daylight reduction (8.3%—12%) compared to the rooms with fixed PVSDs. While improved daylight uniformities were found in all cities. Therefore, dynamic mode was recommended for PVSDs in the summertime for all cities. On the contrary, dynamic PVSDs could lead to the largest interior daylight level reduction in wintertime (19.7%—22.8% less in comparison with fixed modes). Furthermore, the dynamic systems showed reduced interior daylight uniformity (6.7%—10.4%) than the static ones in most scenarios in winter. Based on that, dynamic PVSDs may not be recommended for wintertime if the users prioritize indoor daylight performance over energy harvesting. For the spring season, the indoor daylight uniformities of static PVSDs in Beijing, Harbin and Chengdu were slightly better than the dynamic systems (2%—4%), while the differences in Guangzhou and Shanghai were marginal. For autumn times, dynamic PVSDs brought an obvious indoor daylight uniformity improvement (12%) than that of static systems in Chengdu. While in other cities, the differences were small or marginal. Based on the analysis, two trade-off application strategies were suggested:

1) If the users prioritize energy production, all cities are suggested to deploy the dynamic PVSDs strategy throughout the year.

2) If the users prioritize indoor daylight performance while wanting to harvest more electricity, the dynamic PVSDs mode was recommended to be deployed for all cities during summer; for Guangzhou and Shanghai in spring; for Chengdu, Guangzhou, Kunming, and Shanghai in autumn.

3.5 Spearman correlation analysis

A Spearman correlation analysis was conducted to investigate the relationship between various features and PV power output across six cities. As shown in Fig. 13, Global Horizontal Radiation (GHR) and Global Horizontal Illuminance consistently exhibit strong positive correlations with PV power generation, indicating their critical roles as primary drivers of solar energy generation. The highest correlations are observed in Kunming (GHR: 0.91, GHI: 0.90), Chengdu (both 0.86), and Shanghai (both 0.85), which suggests a strong and stable irradiance-to-power relationship in these regions. Additionally, Average Illuminance and Average Daylight Factor show higher correlations with PV power output in northern cities like Beijing and Harbin. These metrics effectively reflect average solar radiation levels received by the PVSDs, particularly under clearer sky conditions and lower sun angles characteristic of higher latitudes.

Figure 14 further showed that south-facing PVSDs consistently achieve the highest correlation between GHR/GHI and power output across all cities. This indicates that, from an energy generation perspective, south orientation remains the most reliable choice regardless of geographic location.

To deepen the exploration of the dynamic shading effects of PV orientations, the Spearman correlation analysis was performed between PV azimuth angles and Average Illuminance at 9:00, 12:00, 15:00, and 18:00 across various cities. For this analysis, PV azimuth angles were represented numerically, with -90° indicating east-facing, 0° south-facing, and 90° west-facing orientations. As shown in Fig. 15, the results exhibited a clear time-dependent pattern: strong negative correlations at 9:00 indicate that east-facing PVSDs significantly improve morning mean illuminance. In contrast, high positive correlations at 15:00 suggest west-facing PVSDs enhance afternoon indoor average illuminance. By 18:00, moderate positive correlations indicate continued benefits from west-facing panels in late afternoon. These findings highlight the necessity of time-specific orientation optimization in designing dynamic PVSDs to balance both energy harvesting and indoor visual comfort.

From a design standpoint, these findings provide practical guidance:

Orientation: South-facing PVSDs should be prioritized for maximizing annual energy generation. East-facing systems can be considered in applications where morning daylight enhancement is crucial (e.g., classrooms or east-facing offices), while west-facing systems are advantageous for improving afternoon illumination in spaces with late operational hours.

Latitude: High-latitude cities benefit more from large seasonal tilt adjustments to extend sunlight capture during shorter winter days. In contrast, in low-latitude cities with higher solar altitude angles, moderate tilt angles are recommended to avoid excessive shading losses during midday.

Time-of-day optimization: For buildings with fixed operational patterns, PVSD orientation and tilt can be dynamically adjusted to match critical periods of occupancy (e.g., east-oriented tilt in the morning, west-oriented tilt in the afternoon), balancing daylight quality with energy output.

4 Limitation

The nature of this study is a country-scale study, the study focused primarily on energy generation and indoor illuminance. Therefore, real tests on real buildings in different representative cities were not conducted due to the limited resources and budget, also, other factors such as economic feasibility, user preferences, and long-term maintenance costs were not included in this study’s scope. Another limitation of this study was the simplification of PV energy model for calculation, which didn’t specifically reflect the temperature-induced efficiency fluctuations across the days, seasons in different climate zones. In this study, the energy conversion efficiency of mono-Si PVSDs was set as 15%, which was referenced by the average prior real outdoor tests. Current mono-Si PV panels can achieve an energy conversion efficiency of 23% under standard conditions (1000W/m2, 25°C), the temperature-induced efficiency loss is around-0.3% to -0.5% per degree Celsius. However, due to the complexity of temperature variations in different representative cities across the climate zones in China, and the lack of real measured PV efficiency-temperature data in these cities, we use the simplified PV energy model in this study. In future studies, it could be very beneficial to collaborate with scholars in other climate zones (also in other countries) to get more detailed and long-term outdoor registered PV efficiency-temperature data, to develop a more holistic and accurate PV energy model.

5 Conclusion

This study systematically analyzed the optical and power generation performance of dynamic PVSDs across different cities in various climate zones, and the main conclusions were as follows: (1) Regardless of whether the PVSDs adopted a dynamic adjustment or static fixed mode, the annual electricity generation was consistently highest for the south-facing orientation, followed by the west, with the east-facing orientation achieving the lowest output, indicating that azimuth angle had a significant impact on system efficiency; (2) In low-latitude cities with higher cloud cover, the potential advantages of PV power generation are significantly lower than those in high-latitude cities with abundant sunshine; (3) In all the studied cities, the annual electricity generation of dynamic PVSDs was significantly higher than that of static systems, with an increase ranging from 16.1% to 29.9%. The dynamic adjustment mechanism enabled more effective solar tracking and optimized light utilization, thereby markedly enhancing the overall power generation performance of the system; (4) Dynamic PVSDs effectively met indoor lighting requirements, maintaining indoor illuminance above 300 lux during most periods and thus adequately supporting the illumination standards necessary for daily indoor activities; (5) Moreover, dynamic PVSDs improved the uniformity and quality of indoor illuminance, especially under summer conditions and in most cities, providing a more balanced and comfortable visual environment, and were able to maintain relatively stable and consistent indoor illuminance throughout the year, significantly alleviating the illuminance disparity among cities at different latitudes and enhancing the overall uniformity and adaptability of the indoor light environment.

This study indicated that dynamic PVSDs have significantly increased energy generation compared to static systems. Future research can focus on the design and implementation of intelligent control systems, utilizing artificial intelligence and machine learning algorithms to develop more precise dynamic adjustment strategies that can respond to environmental changes in real-time, optimizing solar energy capture and utilization efficiency. Additionally, research can investigate the impact of different dynamic adjustment mechanisms on system lifespan and maintenance costs, providing a more comprehensive economic assessment for practical applications. Dynamic PVSDs excel in meeting indoor daylighting demands, and future research can explore the integration of dynamic PVSDs with user needs to achieve human-machine interaction. Specifically, it can investigate how to use intelligent control systems to dynamically adjust the angle of PV shading based on user requirements, enhancing user comfort while reducing indoor energy consumption.

References

[1] Ali, A.O., Elgohr, A.T., El-Mahdy, M.H., Zohir, H.M., Emam, A.Z., Mostafa, M.G., et al., 2025. Advancements in photovoltaic technology: a comprehensive review of recent advances and future prospects. Energy Convers. Manag. X 26, 100952.

[2] Aries, M.B., Aarts, M.P., van Hoof, J., 2015. Daylight and health: a review of the evidence and consequences for the built environment. Light. Res. Technol. 47 (1), 6—27.

[3] Asfour, O.S., 2018. Solar and shading potential of different configurations of building integrated photovoltaics used as shading devices considering hot climatic conditions. Sustainability 10 (12), 4373.

[4] Baghoolizadeh, M., Nadooshan, A.A., Raisi, A., Malekshah, E.H., 2022. The effect of photovoltaic shading with ideal tilt angle on the energy cost optimization of a building model in European cities. Energy Sustain. Dev. 71, 505—516.

[5] Batista, F., Guimarães, A.S., Palmero-Marrero, A.I., 2025. Building integrated photovoltaics: a multi-level design review for optimized implementation. Renew. Sustain. Energy Rev. 220, 115837.

[6] Chen, X., Qiu, Y., Wang, X., 2024. A systematic review of research methods and economic feasibility of photovoltaic integrated shading device. Energy Build. 311, 114172.

[7] Custódio, I., Quevedo, T., Melo, A.P., Rüther, R., 2022. A holistic approach for assessing architectural integration quality of solar photovoltaic rooftops and shading devices. Sol. Energy 237, 432—446.

[8] Eder, G., Peharz, G., Trattnig, R., Bonomo, P., Saretta, E., Frontini, F., et al., 2019. Coloured BIPV: Market, Research and Development, vol. 15. IEA PVPS Task (Subtask E).

[9] Ghosh, A., 2022. Fenestration integrated BIPV (FIPV): a review. Sol. Energy 237, 213—230.

[10] Han, M., Lu, L., Sun, B., 2024. Overall energy performance of building-integrated bifacial photovoltaic sunshades with different installation and building parameters in hot and humid regions. Sol. Energy 275, 112619.

[11] Hossain, M.S., Wadi Al-Fatlawi, A., Kumar, L., Fang, Y.R., Assad, M.E.H., 2024. Solar PV high-penetration scenario: an overview of the global PV power status and future growth. Energy Systems 1—57.

[12] International Energy Agency, 2020a. IEA PVPS Report - Trends in Photovoltaic Applications 2020.

[13] International Energy Agency, 2020b. Trends in PV Applications 2020 (Paris, France).

[14] International Energy Agency, 2021. Categorization of BIPV Applications 2021 Task 15 Enabling Framework for the Development of BIPV.

[15] Ito, R., Lee, S., 2024. Performance enhancement of photovoltaic integrated shading devices with flexible solar panel using multi-objective optimization. Appl. Energy 373, 123866.

[16] Jayathissa, P., Luzzatto, M., Schmidli, J., Hofer, J., Nagy, Z., Schlueter, A., 2017. Optimising building net energy demand with dynamic BIPV shading. Appl. Energy 202, 726—735.

[17] Kendall, M.G., 1938. A new measure of rank correlation. Biometrika 30 (1—2), 81—93.

[18] Kirimtat, A., Tasgetiren, M.F., Brida, P., Krejcar, O., 2022. Control of PV integrated shading devices in buildings: a review. Build. Environ. 214, 108961.

[19] Krarti, M., Karrech, A., 2024. Evaluation of static and dynamic PV-Integrated shading systems for office spaces in Australia. Sol. Energy 277, 112736.

[20] Lee, S., Ito, R., 2025. Development and verification of an airflow-type photovoltaic-integrated shading device on building façades. Appl. Energy 383, 125292.

[21] Liu, J., Bi, G., Gao, G., Zhao, L., 2023. Optimal design method for photovoltaic shading devices (PVSDs) by combining geometric optimization and adaptive control model. J. Build. Eng. 69, 106101.

[22] Mandalaki, M., Tsoutsos, T., Papamanolis, N., 2014. Integrated PV in shading systems for mediterranean countries: balance between energy production and visual comfort. Energy Build. 77, 445—456.

[23] Saretta, E., Caputo, P., Frontini, F., 2019. A review study about energy renovation of building facades with BIPV in urban environment. Sustain. Cities Soc. 44, 343—355.

[24] Sedgwick, P., 2012. Pearson’s correlation coefficient. Bmj 345.

[25] Shishegar, N., Boubekri, M., 2016. Natural light and productivity: analyzing the impacts of daylighting on students’ and workers’ health and alertness. International Journal of Advances in Chemical Engineering and Biological Sciences 3 (1), 72—77.

[26] Shukla, A.K., Sudhakar, K., Baredar, P., 2017. Recent advancement in BIPV product technologies: a review. Energy Build. 140, 188—195.

[27] Skandalos, N., Karamanis, D., 2021. An optimization approach to photovoltaic building integration towards low energy buildings in different climate zones. Appl. Energy 295, 117017.

[28] Spearman, C., 2010. The proof and measurement of association between two things. Int. J. Epidemiol. 39 (5), 1137—1150.

[29] Stamatakis, A., Mandalaki, M., Tsoutsos, T., 2016. Multi-criteria analysis for PV integrated in shading devices for mediterranean region. Energy Build. 117, 128—137.

[30] Taveres-Cachat, E., Lobaccaro, G., Goia, F., Chaudhary, G., 2019. A methodology to improve the performance of PV integrated shading devices using multi-objective optimization. Applied energy 247, 731—744.

[31] Wang, M., Jia, Z., Xiang, C., 2025a. Machine learning-based adaptive control of PV shading for residential energy and visual comfort optimization. Build. Environ., 113359

[32] Wang, M., Jia, Z., Tao, L., Wang, W., Xiang, C., 2025b. Optimizing the tilt angle of kinetic photovoltaic shading devices considering energy consumption and power generation―Hong Kong case. Energy Build. 326, 115072.

[33] Wang, W., Xu, K., Xiang, C., 2025. Study on the application of a dynamic photovoltaic integrated light shelf for office buildings. Energy 314, 134205.

[34] Xiang, C., Matusiak, B.S., 2022. Façade integrated photovoltaics design for high-rise buildings with balconies, balancing daylight, aesthetic and energy productivity performance. J. Build. Eng. 57, 104950.

[35] Xu, W., Hou, Y., Hung, Y.S., Zou, Y., 2013. A comparative analysis of Spearman's rho and Kendall's tau in normal and contaminated normal models. Signal Process. 93 (1), 261—276.

[36] Ye, Y., Zhu, R., Yan, J., Lu, L., Wong, M.S., Luo, W., Qin, Z., 2023. Planning the installation of building-integrated photovoltaic shading devices: a GIS-based spatiotemporal analysis and optimization approach. Renew. Energy 216, 119084.

[37] Zhang, W., Lu, L., Peng, J., 2017. Evaluation of potential benefits of solar photovoltaic shadings in Hong Kong. Energy 137, 1152—1158.

[38] Zhang, L., Alizadeh, A.A., Baghoolizadeh, M., Salahshour, S., Ali, E., Escorcia-Gutierrez, J., 2025. Multi-objective optimization of vertical and horizontal solar shading in residential buildings to increase power output while reducing yearly electricity usage. Renew. Sustain. Energy Rev. 215, 115578.

Rights & permissions

2095-2635/2025 The Authors. Publishing services by Elsevier B.V. on behalf of KeAi Communications Co. Ltd.