Predicting and Optimizing the Power Performance of Dye-Sensitized Solar Cells Using Machine Learning Techniques

Solar cells hold great potential for the future of energy production. Dye Sensitized Solar Cells (DSSCs) emerge as a promising contender among third-generation solar cells, due to their cost-effectiveness and flexibility. Ongoing experiments aim to further enhance their performance by exploring alternative materials, with a specific emphasis on refining the photoanode for increased efficiency. Within the photoanode, the semi-conducting oxides demonstrate excellent conductivity and porosity. However, despite their promise, testing various material variations in DSSCs is challenging due to time and resource constraints. This research builds on past studies by introducing machine learning to predict efficiencies for alternative DSSC configurations, specifically emphasizing the semiconducting oxide. This machine learning algorithm aims to be a cost-effective approach to identify low-cost, high-performance configurations for DSSCs and predict the system's performance across various parameters, including Power Conversion Efficiency (PCE) and Fill Factor (FF). Using machine learning, this study aims to simplify the evaluation process and speed up the identification of optimal materials and their weight percentages for the semi-conducting oxide, contributing to the advancement of sustainable and efficient DSSC technologies.