プラント向けエッジ側状態監視処理システムの開発
概要
In this paper, we propose a remote monitoring system for field instruments in plant operation using SMEc, which combines a camera with deep learning. By the needle position estimation logic, a recognition rate of 99% or more could be obtained within the target accuracy of 1 scale error. It was also confirmed from the response evaluation test that the dynamic behavior had sufficient responsiveness. Therefore, we believe that the introduction of SMEc has given us the prospect of realizing a system capable of "Introduction of systems without modification of actual machines" "Collecting field data in narrow areas" and "Data transmission with low communication capacity". In addition, continuous monitoring of plant facilities and equipment can be realized with a small capital investment, and contribution to the safe and stable operation of the plant can be expected.
