Optoelectronic Materials Informatics: Utilizing Random-forest Machine Learning in Optimizing the Harvesting Capabilities of Mesostructured-based Solar Cells

摘要

Materials informatics is a study field that focuses on investigating and applying the techniques of informatics to materials science and engineering. Machine Learning (ML) refers to data analysis techniques and algorithms take fall under the concept of artificial intelligence (AI). This paper targets applying both concepts to the renewable energy field study. More specifically, to analyze, enhance and optimize the performance of Mesostructured-based solar cells, as a booming light harvester in low-power applications. Herein, we investigate both dye-sensitized solar cells (DSSCs) and perovskite solar cells (PSCs). The power conversion efficiency of both cells is optimized in terms of the active mesoporous layer thickness as well as compact layer. A Random-forest algorithm is utilized for model regression, prediction, and optimization.