Dye-sensitized solar cells (DSSCs) are known for their excellent low-light performance, cost-effectiveness, and flexibility. The photoanode has a crucial role in enhancing the overall performance of DSSCs and can be modified with different nanostructures. This study explores the impact of photoanode structure on the power conversion efficiency (PCE) of DSSCs, where four configurations of freestanding TiO₂ nanotube arrays (f-TNAs), closed-up, closed-down, open-up, and open-down, were employed as photoanodes. Performance was evaluated based on current density (J_sc), open-circuit voltage (V_oc), fill factor (FF), and PCE concerning dye adsorption, electrolyte diffusion, electron transport, and barrier layer. DSSCs based on open configurations, open-up and open-down f-TNAs, demonstrated superior performance, achieving PCE of 7.73% and 7.71%, respectively. The primary distinction between the DSSCs based on open-up f-TNAs and those based on open-down f-TNAs lies in the dye adsorption time and electron diffusion characteristics. The PCE for DSSCs with closed-down f-TNAs was measured at 6.78%, while DSSCs with closed-up f-TNAs showed a lower PCE of 5.52%. The presence of a barrier layer under the bottom of f-TNAs impacted the PCE for DSSCs with closed-down f-TNAs, whereas for DSSCs with closed-up f-TNAs, insufficient dye loading, poor electrolyte diffusion and barrier layer reduced the performance.

Perovskite solar cells (PSCs) have emerged as a promising technology for high-efficiency, low-cost energy production. However, their path to commercialization has been impeded by inefficient and time-consuming manufacturing processes. This research leverages machine learning (ML) models and open data to optimize PSC fabrication, focusing on devices incorporating light-scattering materials and metal nanoparticles (NPs) in the electron transport layer (ETL). The study employed two ML models for predictions: the Extra Tree Regressor (ETR) for PSCs with scattering materials, and the Ridge Regression (RR) model for PSCs with metal NPs. The ETR model predicted 15.960% efficiency for PSCs with scattering materials, identifying short-circuit current density (Jsc) as the critical factor. The RR model forecasted 18.122% efficiency for PSCs with metal NPs, highlighting open-circuit voltage (Voc) and fill factor (ff) as key contributors. Experimental validation closely aligned with these predictions. PSCs fabricated with TiO2 nanotube array fragments as scattering materials achieved an average efficiency of 15.819% (Jsc = 25.089 mA/cm², Voc = 0.902 V, ff = 0.699), deviating only 0.88% from the prediction. PSCs incorporating silver NPs as plasmonic agents reached an average efficiency of 17.887% (Jsc = 25.905 mA/cm², Voc = 0.942 V, ff = 0.733), with a 1.30% difference from the predicted value. This research demonstrates the power of combining ML with experimental validation to enhance PSC performance. The approach shows potential for application in other energy technologies, offering a data-driven pathway to accelerate the development and commercialization of advanced energy solutions.Perovskite solar cells (PSCs) have emerged as a promising technology for high‑efficiency, low‑cost energy production, yet their path to commercialization is hindered by inefficient and time‑consuming manufacturing processes. This work leverages machine learning (ML) models and open data to guide PSC fabrication, focusing on devices incorporating light‑scattering materials and metal nanoparticles (NPs) in the electron transport layer (ETL). Among 18 regression algorithms benchmarked on open datasets, the Extra Trees Regressor (ETR) was identified as the best performer for PSCs with scattering materials, while Ridge Regression (RR) showed the highest accuracy for PSCs with metal NPs. SHapley Additive exPlanations (SHAP) applied to these models revealed that efficiency gains in scattering‑based PSCs are primarily governed by short‑circuit current density (Jsc), whereas in metal‑NP‑based PSCs they are dominated by open‑circuit voltage (Voc) and fill factor (ff).Guided by these insights, PSCs were fabricated using TiO2 nanotube array (TNT) fragments as scattering centers and TiO2‑coated Ag NPs as plasmonic and charging agents in the ETL. For devices with scattering materials, the ETR model predicted an efficiency of 15.960%, identifying Jsc as the critical factor, while experimentally fabricated cells achieved an average efficiency of 15.819% (Jsc = 25.089 mA/cm², Voc = 0.902 V, ff = 0.699). For devices with Ag NPs, the RR model forecasted an efficiency of 18.122%, highlighting Voc and ff as key contributors, and the corresponding experimental PSCs reached an average efficiency of 17.887% (Jsc = 25.905 mA/cm², Voc = 0.942 V, ff = 0.733). The close agreement between prediction and experiment validates the ML‑guided design rules and confirms that scattering layers mainly enha