Performance improvement of vertically installed bifacial solar panels with adjustable reflectors optimized using the Taguchi method
The global deployment of bifacial photovoltaic (PV) modules has accelerated due to their ability to capture both direct and reflected sunlight, offering higher energy yields than traditional monofacial panels. However, their performance—especially in vertical installations—remains limited by subopti...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-09-01
|
| Series: | Energy Nexus |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2772427125001378 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | The global deployment of bifacial photovoltaic (PV) modules has accelerated due to their ability to capture both direct and reflected sunlight, offering higher energy yields than traditional monofacial panels. However, their performance—especially in vertical installations—remains limited by suboptimal rear-side irradiance during early morning and late afternoon periods, and by fixed or semi-passive reflector configurations that fail to respond to dynamic environmental conditions such as changing solar altitude angles and wind exposure. Most existing systems use static reflectors or high-albedo surfaces, which cannot actively adapt to maximize sunlight collection throughout the day or across seasons. Moreover, in dense urban or rooftop environments common in East Asia, space constraints and wind load risks further complicate the deployment of large, fixed reflectors.To address these limitations, this study proposes a novel adjustable reflector system for bifacial PV modules. The system is capable of automatically modifying both tilt angle and effective length on an hourly basis, based on real-time solar altitude and wind speed data. This adaptive configuration enhances solar irradiance capture and ensures structural safety throughout daily and seasonal variations. To optimize the system’s operational parameters with minimal experimental cost, the Taguchi method is employed for experimental planning, including control factor selection and orthogonal array design, enabling the identification of key parameter interactions and performance trends. Reflector-induced irradiance enhancement is quantitatively evaluated using reflector theory, with solar radiation fields adjusted according to Taiwan’s Typical Meteorological Year 2 (TMY2) data. These parameters are analyzed using the TRNSYS simulation platform to estimate annual energy gains and system performance under realistic climatic conditions.Simulation and experimental results show that performance deviations remained within 0.3 %, demonstrating high predictive accuracy. The optimized configuration includes aluminum reflectors, front and rear reflector angles set to half the solar altitude angle, reflector surface areas larger than the module surface, and a module azimuth angle of 110°, which improved efficiency by approximately 11 % compared to standard bifacial PV systems and by 3.19 % over non-optimized reflector configurations. Annual simulations showed a total power output increase of 71.32 % compared to conventional monofacial modules. Furthermore, when compared to commonly installed rooftop and ground-mounted PV systems in Taiwan, the proposed system achieves an annual generation of 599 MJ/year—significantly surpassing the 350 MJ/year output of traditional installations—thereby confirming its practical feasibility. Structural robustness was validated through ANSYS simulations under typhoon-level wind speeds (55 m/s), confirming that the maximum stress remained below material yield strength. Additionally, a wind speed sensor and automated retraction mechanism were incorporated to ensure structural safety under extreme weather conditions. This work demonstrates the feasibility and reliability of integrating TRNSYS simulation with Taguchi-based optimization to develop high-efficiency, sun-tracking PV systems suited for scalable, real-world deployment. |
|---|---|
| ISSN: | 2772-4271 |