AI method learns from experienced personnel on solar perovskite production lines – pv magazine International

Academics from MIT and Stanford who have proposed a new production method for perovskite solar cells have also developed a machine-learning system that benefits from the experience of seasoned workers – and they’ve put it online so that everyone can use it.

There is no substitute for experience and US researchers have recognized this by proposing a machine learning method for producing perovskite solar cells that can incorporate observations from experienced production line personnel.

Academics from the US institutions MIT and Stanford – as well as the Singapore-MIT Alliance for Science and Technology – have developed an artificial intelligence (AI) system that can include input from experienced workers as well as data experiments conducted by other researchers, to help control the many variables that can affect the mass production of perovskite cells.

Rapid spray approach

The researchers, whose article was published in the latest issue of Joule, developed the AI ​​system to improve the rapid sputtering plasma processing (RSPP) production method they suggested for the commercial production of perovskite solar cells. An article published Wednesday on the MIT News website about the research said: “The [RSPP] the manufacturing process would involve a roll-to-roll moving surface, or a series of sheets onto which the precursor solutions of the perovskite compound would be sprayed or ink-jetted as the sheet rolled. The material would then go through a hardening stage, providing fast and continuous output.

The machine learning code developed by the academics has been placed on the open-source software development site GitHub, from which solar perovskite makers can download it and MIT mechanical engineering professor Tonio Buonassisi said that the researchers contacted the photovoltaic companies to inform them of the code. and their RSPP approach to perovskite fabrication.

The AI ​​method was directed by the researchers to optimize the power output of perovskite cells, but Buonassisi said his colleagues strive to direct it to maximize cost and durability, as well as other potential characteristics. , the same time.

Former MIT research assistant Zhe Liu, one of the scholars who worked on the project and is now at Northwestern Polytechnic University in the Chinese city of Xi-an, told MIT News that Chinese perovskite solar companies are focusing on small products such as solar rooftops. tiles, but three manufacturers are under pressure from investors, or are already on track, to produce full-size perovskite solar panels “within two years.”

The research, supported by US government agencies, the Department of Energy and the National Science Foundation, also involved Stanford scholars Reinhold Dauskardt, professor of materials science and engineering; Ph.D. Nicholas Rolston; and Austin Flick and Thomas Colburn; and Zekun Ren of the Singapore-MIT Alliance for Science and Technology.

This content is copyrighted and may not be reused. If you wish to cooperate with us and wish to reuse some of our content, please contact: editors@pv-magazine.com.

Source link

About Donald P. Hooten

Check Also

The 369 Manifestations Method: How to Make Your Therapy Sessions Work

Psychologist Nancy Sokarno walks us through the 369 Manifestations Method and how to get the …