Acta Mechanica Sinica, Journal Year: 2023, Volume and Issue: 40(1)
Published: Oct. 12, 2023
Language: Английский
Acta Mechanica Sinica, Journal Year: 2023, Volume and Issue: 40(1)
Published: Oct. 12, 2023
Language: Английский
Neural Networks, Journal Year: 2025, Volume and Issue: 185, P. 107165 - 107165
Published: Jan. 21, 2025
Language: Английский
Citations
3Chaos An Interdisciplinary Journal of Nonlinear Science, Journal Year: 2024, Volume and Issue: 34(3)
Published: March 1, 2024
Nonlinear dynamical systems with control parameters may not be well modeled by shallow neural networks. In this paper, the stable fixed-point solutions, periodic and chaotic solutions of parameter-dependent Lorenz system are learned simultaneously via a very deep network. The proposed learning model consists large number identical linear layers, which provide excellent nonlinear mapping capability. Residual connections applied to ease flow information training dataset is further utilized. Extensive numerical results show that can accurately forecasted for several Lyapunov times long-term predictions achieved solutions. Additionally, characteristics such as bifurcation diagrams largest exponents recovered from Finally, principal factors contributing high prediction accuracy discussed.
Language: Английский
Citations
7Communications in Nonlinear Science and Numerical Simulation, Journal Year: 2025, Volume and Issue: unknown, P. 108696 - 108696
Published: March 1, 2025
Language: Английский
Citations
0Mechanical Systems and Signal Processing, Journal Year: 2025, Volume and Issue: 232, P. 112751 - 112751
Published: April 15, 2025
Language: Английский
Citations
0Indian Journal of Pure and Applied Mathematics, Journal Year: 2024, Volume and Issue: unknown
Published: Jan. 31, 2024
Language: Английский
Citations
2Nonlinear Dynamics, Journal Year: 2024, Volume and Issue: unknown
Published: May 17, 2024
Abstract This paper presents a novel method for analyzing high-dimensional nonlinear stochastic dynamic systems. In particular, we attempt to obtain the solution of Fokker–Planck–Kolmogorov (FPK) equation governing response probability density system without using FPK directly. The consists several important components including radial basis function neural networks (RBFNN), Feynman–Kac formula and short-time Gaussian property process. area solving partial differential equations (PDEs) networks, known as physics-informed network (PINN), proposed an effective alternative obtaining solutions PDEs directly dealing with equation, thus avoids evaluating derivatives equation. approach has potential make network-based more efficient accurate. Several highly challenging examples systems are presented in illustrate effectiveness comparison equation-based RBFNN approach.
Language: Английский
Citations
2Chaos Solitons & Fractals, Journal Year: 2024, Volume and Issue: 185, P. 115134 - 115134
Published: June 13, 2024
Language: Английский
Citations
2Nonlinear Dynamics, Journal Year: 2024, Volume and Issue: unknown
Published: July 11, 2024
Language: Английский
Citations
2Mathematics and Computers in Simulation, Journal Year: 2024, Volume and Issue: 226, P. 645 - 662
Published: July 30, 2024
Language: Английский
Citations
2Nonlinear Dynamics, Journal Year: 2024, Volume and Issue: unknown
Published: Dec. 6, 2024
Language: Английский
Citations
2