Home  >  Article  >  Technology peripherals  >  Artificial intelligence can identify solar cell defects

Artificial intelligence can identify solar cell defects

PHPz
PHPzforward
2023-09-23 22:45:09601browse

Artificial intelligence can identify solar cell defects

Reference News Network reported on September 23 According to TASS news agency reported on September 19, Russian scientists have developed a machine learning system that can identify defects in solar cell structural units.

The artificial intelligence (AI) technology is also able to identify errors in the production process that are most likely to lead to defects. According to a report from the Press Office of the Institute of Artificial Intelligence on the 19th, the software package has been tested in actual production. "The result of the joint efforts of the team was the successful industrial testing of the system during actual production at the Novchebok-Sarskhevel plant. When two or more images are detected during the sorting stage," the report said , when there are similar defects with similar positioning, the system will notify the node in the factory production line equipment that is most likely to be related to the discovered defect in real time."

The research team led by Semyon Budyonny worked with experts from Sol IT Technology Company and Herwell Company to develop this algorithm

The notice pointed out that manufacturers of solar panels often have product defects when manufacturing new solar cells. These defects can appear at different stages of the technology chain, leading to a significant reduction in the efficiency of solar panels. To identify these defects, factories can take images using special cameras that track the light emitted by solar cells when exposed to an electric field or current.

The researchers prepared a database of 68,000 images and used it to train the network they developed. The first tests show that the algorithm is able to identify defects and their possible sources with about 90% to 95% accuracy.

Bujoni said additional large-scale data labeling is currently underway to improve the accuracy of defect location. If the second phase of testing is successful, it is planned to put the software into industrial trial operation

The above is the detailed content of Artificial intelligence can identify solar cell defects. For more information, please follow other related articles on the PHP Chinese website!

Statement:
This article is reproduced at:sohu.com. If there is any infringement, please contact admin@php.cn delete