$3M Boosts Perovskite Solar Mfg.

• University of Michigan project backed by $3 million grant from NSF seeks to improve process of semiconductor materials manufacturing for perovskite solar cells. New cyber manufacturing technology may radically transform manufacturing, while protecting company info and worker interests.

A new project led by the University of Michigan and backed by a four-year, $3 million grant from the National Science Foundation seeks to incorporate “federated learning” into the process of layer-by-layer deposition of semiconductor materials during the manufacturing of perovskite solar cells. The goal is to improve upon the process through an information-sharing approach that boosts cooperation between companies while protecting proprietary information and worker interests. Researchers will take data collected from isolated experiments with perovskite semiconductors to build predictive models for forecasting product quality and performance, helping companies to narrow down key parameters such as optimal pressure and temperature during manufacturing. In addition, the team has Sarah Crane, research manager at U-M’s Economic Growth Institute, and Julie Hui, assistant professor at the School of Information, who will look at how the technology can help create jobs. This project is part of the NSF’s “cyber manufacturing,” which has the potential to “radically transform concepts of manufacturing.”

How Can Federated Learning Improve Perovskite Solar Cell Manufacturing?

• Federated learning can be used to build a network of computers that work together to improve the process of manufacturing perovskite solar cells.
• This information-sharing approach allows companies to cooperate while protecting proprietary information and worker interests.
• Predictive models can be used to determine product quality and performance, enabling companies to identify key parameters such as optimal pressure and temperature.
• The technology has the potential to create jobs, as well as to “radically transform concepts of manufacturing.”
• By incorporating the data from isolated experiments with perovskite semiconductors, the project could help reduce cost and increase efficiency for manufacturers.
• The research could also lead to better understanding of how perovskite materials work, allowing for further innovation and the development of new applications.