Answers are grounded in this paper and cite their sources.
This analysis shows the impact of specific process parameters on fill factor and efficiency, highlighting BP neural networks for predictive modeling.
The research and development of traditional perovskite solar cells requires significant time and cost for process parameter optimization. To address this issue, this study collected relevant data on perovskite solar cells efficiency and employed a back propagation (BP) neural network prediction method, selecting seven key process parameters of perovskite solar cells to predict their fill factor and power conversion efficiency. The prediction results demonstrate: the mean absolute error of the predicted value of the filling factor is 0.050, themean relative error is 0.072, and the mean square error is 0.005. The mean absolute error of the predicted value of the efficiency is 1.305, the mean relative error is 0.074, and the mean square error is 2.679. The results show good agreement between the predicted values and real measurements for both the fill factor and power conversion efficiency. This indicates that BP neural networks can effectively predict the fill factor and efficiency of perovskite solar cells, enabling process parameter optimization to ultimately reduce development time and costs.
No takes yet. Share an insight, caveat, or question.
Sun et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: