One-year performance and loss analysis of vertical east-west bifacial PV modules benchmarked against tilted configurations
Resumen
Vertical east–west bifacial photovoltaic systems offer potential advantages, including reduced soiling, land-use compatibility, and morning–afternoon generation profiles. However, their performance under low-latitude coastal climates remains insufficiently documented. This study presents a one-year field assessment of a vertical east–west bifacial PV module installed in Lima, Peru, and compares it with north-facing tilted bifacial and monofacial modules. Current–voltage curves, front- and rear-side irradiances, and module temperatures were monitored at 5 min resolution. DC energy yield and module performance ratio were analyzed using successive corrections for temperature, bifaciality, and single-diode-model-based I–V reconstruction to identify dominant loss mechanisms. The tilted bifacial module achieved the highest annual DC yield, 3.79 kWh/kW/day, exceeding the monofacial module by 6.5%. In contrast, the vertical bifacial module reached 2.82 kWh/kW/day, despite receiving a total incident reference yield comparable to that of the monofacial configuration. Temperature-related losses were similar across modules, approximately 3–5%, and did not explain this gap. Instead, the vertical configuration was mainly limited by bifaciality-related losses of 11.6%, due to the high fraction of incident irradiance on the less efficient rear side, and structural self-shading losses of approximately 4.5% annually. After all corrections, performance ratios converged to 92–93%, indicating comparable intrinsic module behavior. These results provide experimental evidence and a replicable loss-decomposition framework for assessing vertical bifacial PV systems and guiding module design. This framework is particularly relevant where urban integration, reduced maintenance, land-use constraints, or adapted morning–afternoon generation profiles may justify vertical deployment despite lower annual energy yield.
