研究目的
To use machine learning (ML) to optimize material composition, develop design strategies, and predict the performance of perovskite solar cells (PSCs).
研究成果
ML tools can be used for crafting perovskite materials and investigating the physics behind developing highly efficient PSCs. The findings show that ML is very promising not only for predicting the performance, but also for providing a deeper understanding of the physical phenomena associated with the PSCs.
研究不足
The complexity of the reported experimental values leads to not negligible RMSE of the predicted values of solar cell performances. The model does not consider some parameters that might affect the performances of PSCs, such as grain sizes and variations in fabrication techniques.
1：Experimental Design and Method Selection:
ML models are developed using 333 data points selected from about 2000 peer reviewed publications to guide the design of new perovskite materials and the development of high-performing solar cells.
2：Sample Selection and Data Sources:
Data sets are collected from published literature, focusing on perovskite material compositions and their bandgaps, as well as PSC device performances.
3：List of Experimental Equipment and Materials:
Perovskite materials are synthesized and characterized using UV–vis absorption spectra to measure bandgaps.
4：Experimental Procedures and Operational Workflow:
New perovskite compositions are experimentally synthesized based on ML predictions to test the model's practicability.
5：Data Analysis Methods:
ML algorithms (LR, KNN, SVR, RF, and ANN) are used to predict bandgaps and PSC performances, evaluated using RMSE and Pearson’s coefficient.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
