Accelerating the Design of Double-Absorber Solar Cells: From Surrogate Model-Assisted Reinforcement Learning and Multi-Algorithm Optimization Comparison to Transfer Learning

Engineering notes

Engineering notes will be added by the Power for Solar editorial team.

Chinese explanation / 中文解读

中文解读待补充：本站会优先为光伏效率、钙钛矿太阳能电池、储能技术、太阳能热利用、BIPV、并网技术等高价值论文补充中文说明。

Original abstract

Lead-free double-absorber perovskite solar cells offer broad-spectrum absorption and environmental benefits, but their multilayer heterostructure creates computational challenges for conventional design optimization. This study introduces an automated framework integrating SCAPS-1D simulation, multilayer perceptron (MLP) surrogate modeling, metaheuristic algorithms, and reinforcement learning (RL). Using FTO/ZnO/Cs2TiBr6/RbGeI3/CuI/Au cells, the MLP model trained on Latin hypercube sampling data achieved high accuracy (R2 > 0.95). The proximal policy optimization (PPO) RL agent converged to 27.41% power conversion efficiency (PCE) in approximately 20 steps. For direct 15-dimensional optimization, simulated annealing and particle swarm optimization reached 98% target PCE with 138 and 111 function evaluations, respectively, while Grey Wolf Optimizer (GWO) yielded the highest average PCE. Transfer learning successfully adapted the pretrained model to a novel FASnI3/Sb2S3 structure, improving the prediction accuracy of PCE, JSC, and FF. This work systematically optimizes Cs2TiBr6/RbGeI3 solar cells while establishing an efficient, generalizable paradigm for intelligent photovoltaic device design, validation, and material discovery.

Links and sources

• PDF from original source

Comments