Supei Zheng

Dr. Supei Zheng

orcid

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications

1 November 2024 in Physics of Fluids
Verified
Journal Article
Closed access
PDF file
JLJiahao Liu
Jiahao Liu
SZSupei Zheng
Supei Zheng
XSXueli Song
Xueli Song
DXDoudou Xu
Doudou Xu
In this work, we demonstrate that physics-informed neural networks (PINNs) tend to propagate predicted shock wave information bidirectionally in time, which does not align with the actual evolution direction of solutions to hyperbolic conservation laws. This mismatch results in instability and hinders the reduction of the loss of governing equations, as well as the initial condition loss by the deep neural network. In order to tackle this problem, we simplify the complexity of the problem by constructing equivalent linear transport equations in the region of shock wave generation. The speeds of these linearized waves are governed by the Rankine–Hugoniot relations of conservation laws. This approach is termed the Locally Linearized PINNs method. Specifically, an appropriate shock wave detector is initially designed to identify domains where shock waves occur. Near shock waves, the original nonlinear equations are transformed into their linearized forms, thereby modifying the residual terms of the partial differential equations. Additionally, an equilibrium factor is introduced in fluid compression regions to reduce prediction errors and stabilize the training of deep neural networks. Numerical examples illustrate that Locally Linearized PINNs effectively address the challenge of predicting global solutions in PINNs and significantly improve shock-capturing performance for hyperbolic conservation laws.

1 January 2020 in Advances in Applied Mathematics and Mechanics
Verified
Journal Article
Closed access
XCXiaohan Cheng
Xiaohan Cheng
JFJianhu Feng
Jianhu Feng
SZSupei Zheng
Supei Zheng
Advances in Applied Mathematics and Mechanics (AAMM) provides a fast communication platform among researchers using mathematics as a tool for solving problems in mechanics and engineering, with particular emphasis in the integration of theory and applications. To cover as wide audiences as possible, abstract or axiomatic mathematics is not encouraged. Innovative numerical analysis, numerical methods, and interdisciplinary applications are particularly welcome.

16 February 2024 in Journal of Scientific Computing
Verified
Journal Article
Closed access
PDF file
MZMengqing Zhai
Mengqing Zhai
SZSupei Zheng
Supei Zheng
CZChengzhi Zhang
Chengzhi Zhang
MJMangmang Jian
Mangmang Jian

11 January 2025 in International Journal of Structural Stability and Dynamics
Verified
Journal Article
Closed access
LMLingjuan Ma
Lingjuan Ma
WYWen Yi
Wen Yi
XSXueli Song
Xueli Song
RLRongpeng Li
Rongpeng Li
YXYuzhu Xiao
Yuzhu Xiao
KWKaiming Wang
Kaiming Wang

1 January 2022 in 应用数学和力学
Verified
Journal Article
Closed access
PDF file
SZSupei Zheng
Supei Zheng
LXLi Xiao
Li Xiao
ZQZhao Qingyu
Zhao Qingyu
FJFeng Jianhu
Feng Jianhu
长安长安大学理学院,西安710064
长安大学理学院,西安710064