Reconstruction of instantaneous flow fields from transient velocity snapshots using physics-informed neural networks (Applications to pulsatile blood flow behind a stenosis)

Physics-informed neural networks (PINNs) offer a promising framework by embedding partial differential equations (PDEs) into the loss function together with measurement data, making them well-suited for inverse problems. However, plain PINNs face challenges with time-dependent PDEs due to the high c...

Бүрэн тодорхойлолт

Номзүйн дэлгэрэнгүй

Үндсэн зохиолчид:	Kakeru UEDA, Hiro WAKIMURA, Satoshi II
Формат:	Artigo
Хэл сонгох:	Inglês
Хэвлэсэн:	The Japan Society of Mechanical Engineers 2026-05-01
Цуврал:	Journal of Biomechanical Science and Engineering
Нөхцлүүд:	physics-informed neural networks (pinns) magnetic resonance imaging (mri) inverse problem incompressible navier-stokes equations acceleration term pressure gradient transient flow blood flow stenosis
Онлайн хандалт:	https://www.jstage.jst.go.jp/article/jbse/21/2/21_26-00093/_pdf/-char/en
Шошгууд:	Шошго нэмэх Шошго байхгүй, Энэхүү баримтыг шошголох эхний хүн болох!