Machine Learning with Knowledge Constraints for Process Optimization of Open-Air Perovskite Solar Cell Manufacturing

Perovskite photovoltaics (PV) have achieved rapid development in the past decade in terms of power conversion efficiency of lab-scale small-area devices; however, successful commercialization still requires further development of low-cost, scalable, and high-throughput manufacturing techniques. One of the key challenges to the development of a new fabrication technique is the high-dimensional parameter space, and machine learning (ML) can be used to accelerate perovskite PV scaling. Here, we present a ML framework of sequential learning for manufacturing process optimization. We apply our methodology to the Rapid Spray Plasma Processing (RSPP) technique for perovskite thin films in ambient conditions. With a limited experimental budget of screening 100 conditions process conditions, we demonstrated an efficiency improvement to 18.5% for the best performing device, and found 10 conditions to produce the top performing devices of >17% efficiency. Our model is enabled by three innovations: (a) flexible knowledge transfer between experimental processes by incorporating data from prior experimental data as a soft constraint; (b) incorporation of both subjective human observations and ML insights when selecting next experiments; (c) adaptive strategy of locating the region of interest using Bayesian optimization first, and then conducting local exploration for better performing devices. In virtual benchmarking, our framework achieves faster improvements with limited experimental budgets than traditional design-of-experiments methods (e.g., one-variable-at-a-time sampling). In addition, this framework is shown to enable researchers’ domain knowledge in MLguided optimization loop; therefore, it has the potential to facilitate the wider adoption of ML in scaling to perovskite PV manufacturing.