L R AS Published on Sunday 12 December 2021 - n° 386 - Categories:various issues around PV

Artificial intelligence to determine the best combination of tandem cells

American scientists at MIT have developed a computer simulator capable of calculating the conversion efficiency of different materials and solar cell configurations,

This helps to guide the research and optimisation of new cell designs.

In fact, the number of materials with the potential to be used in each of the many layers of a solar cell is enormous. And even once they have chosen one to work with, scientists need to understand its interactions with the other materials present, as well as the effects of changing parameters such as layer thickness, dopant concentration and a host of others in order to get the best out of the cells they are working on.

With so many possibilities, this process can take a long time. Today, scientists are increasingly using artificial intelligence to guide them through the next steps in their hands-on lab work.

Using data on the configuration of a solar cell, it produces a predicted efficiency and shows which of the input parameters most affects the prediction. It can evaluate several variables for each layer, including doping concentration, dielectric constant and band gap. And its researchers point out that layer thickness is a particularly important variable.

https://www.pv-magazine.com/2021/12/09/google-lends-a-hand-in-the-search-for-new-solar-cell-designs-with-open-source-tool/

PV Magazine of 9 December 2021

Subscribe to the newsletter "Le Fil de l'Actu"...

Most read articles in the last 10 days

Most read articles in the last month