- Technique helps spot design problems in advance
- Search technology applied to the discovery of ‘rare critical events’
NEC and Japan's National Institute of Advanced Industrial Science and Technology (AIST) has announced a new efficient search technology for the discovery of ‘rare critical events’.
The idea is to get the technology to repeat simulations over and over and then to use artificial intelligence to learn and interpret the simulation results. This results in an efficient way of uncovering rare critical events that would be difficult to discover at the product design stage due to the extremely low probability of their occurrence.
As a result the new technology significantly shortens validation time (the time taken for highly-skilled experts to discover such rare events at the product design stage) and therefore reduces the risk of multiple critical events being overlooked.
To prove the concept, the technology was applied to a communications technology problem around the design of an optical device to detect stray light. The probability this occurring is very low at around 1 in 100 million, but critical because it causes significant performance degradation as a noise signal.
While skilled experts might have required around a week to track such a rare event down to determine its significance, the technology dramatically shortened search time: completing the search in around one day and successfully discovering several critical events without any being overlooked.
The application of the technology is quite broad. NEC and AIST say they plan to extend the field of applications for this technology from optics design to structural design, such as bridges and buildings, and fluid design such as engines, and will continue working in cooperation to contribute to the development of new technologies merging AI and simulations, and to their application in industry.
The technology is one of NEC's lineup of cutting-edge AI technologies "NEC the WISE"
Sign up to receive TelecomTV's top news and videos, plus exclusive subscriber-only content direct to your inbox.