Machine Learning (ML)‐Assisted Design and Fabrication for Solar Cells

Abstract

Photovoltaic (PV) technologies have attracted great interest due to their capability of generating electricity directly from sunlight. Machine learning (ML) is a technique for computer to learn how to perform a specific task using known data. It can be used in many areas and has become a hot research topic recently due to the rapid accumulation of data and advancement of computer hardware. The application of ML techniques in the design and fabrication of solar cells started slowly but has recently gained tremendous momentum. An exhaustive compilation of the literatures indicates that all the major aspects in the research and development of solar cells can be effectively assisted by ML techniques. If combined with other tools and fed with additional theoretical and experimental data, more accurate and robust results can be achieved from ML techniques. The aspects can be grouped into four categories: prediction of material properties, optimization of device structures, optimization of fabrication processes, and reconstruction of measurement data. A statistical analysis of the literatures shows that artificial neural network (ANN) and genetic algorithm (GA) are the two most applied ML techniques and the topics in the optimization of device structures and optimization of fabrication processes are more popular. This article can be used as a reference by all PV researchers who are interested in ML techniques.

Designing and understanding light-harvesting devices with machine learning

“…Data-driven approaches are emerging as versatile and viable technology for light-harvesting research, connecting complete and comprehensive bottom-up theoretical models with Edisonian trial-and-error strategies. Indeed, light-harvesting research, notably for the design of OSCs and PSCs, increasingly integrates data-driven tools into their workflows at various levels 143 , including property screening 144 , 145 , candidate selection 76 , 136 , analysis 146 , and interpretation 89 , 147 . We have highlighted some of the ML models specifically designed to find new light-harvesting materials 76 , 94 , 135 , 136 .…”

Review on methods for improving the thermal and ambient stability of perovskite solar cells

“…The reversible degradation mechanism poses new challenges for assessing the performance and lifespan of PSCs. 38 In this section, we focus on the improvement of the humidity and thermal stability of the perovskite layer. We do not discuss reversible degradation mechanisms or bias-dependent degradation of perovskites.…”

Machine learning: Accelerating materials development for energy storage and conversion

“…Several early reviews have introduced the applications of ML to materials science, including materials discovery and design, catalysts, and structure prediction . Very recently, ML investigations on energy storage and conversion materials have rapidly increased, which have not been comprehensively summarized.…”