PINN neural network method for solving the forward and inverse problem of time-fractional telegraph equation

Results in Engineering, Год журнала: 2025, Номер unknown, С. 103997 - 103997

Опубликована: Янв. 1, 2025

Язык: Английский

---

Approximate Solution of the Nonlinear Buckmaster Partial Differential Equation using Exponential Fourth-order Differentiable Functions

Lafia Journal of Scientific and Industrial Research, Год журнала: 2025, Номер unknown, С. 14 - 20

Опубликована: Май 9, 2025

In this paper, the nonlinear partial differential equation, Buckmaster equation is solved using exponential cubic B-spline collocation method (ECBSM) and approximate solutions from are compared with those of hybrid (HCBSM). order to solve linearization technique needed linearize terms equation. This done by Taylor’s expansion approach. Further, linearized discretized into fully implicit scheme Crank-Nicolson scheme. Three examples used test proposed schemes methods. The absolute errors methods calculated comparison between results ECBSM HCBSM carried out. analyze accuracy approximation. Both possess a free parameter which aids in determining accurate results. general, proved reliable approximating

Язык: Английский

---

A data-driven approach to solving the Allen–Cahn equation in varying dimensions using physics-informed neural networks (PINNs)

Physics of Fluids, Год журнала: 2025, Номер 37(5)

Опубликована: Май 1, 2025

This study investigates a phase-field model governed by the Allen–Cahn equation, incorporating spatially varying forcing that significantly influences system dynamics resulting in pattern formation, controlled interface motion, and novel steady-state solutions. Physics-informed neural networks (PINNs) are employed to solve this system, results demonstrate PINNs can generate highly accurate solutions with relatively few iterations, even under diverse initial conditions. The high accuracy of PINN simulating nonlinear complex systems was confirmed comparison typical numerical findings show conditions have significant impact on rate type phase evolution: higher amplitudes indicate multi-stage evolution, while lower least amount roughness. equation shown minimize interfacial energy over time, field toward equilibrium. mobility thickness evolution also investigated. Rapidly changing circumstances provide an exception, momentarily increasing complexity, but (L) speeds up migration, improving separation. effect changes starting profile. For smoother configurations, it provides uniform separation; however, when profile has abrupt fluctuations, effects uneven space. scope kinetic-controlled applications such as alloy solidification polymer separation is expanded these findings, which very useful tool for modeling accurately computationally simulate dynamic systems.

Язык: Английский

---

Face Recognition by Siamese Network Using a Novel Distance Function

Lecture notes in networks and systems, Год журнала: 2025, Номер unknown, С. 327 - 338

Опубликована: Янв. 1, 2025

Язык: Английский

---

Quantum algorithms for scientific computing

Reports on Progress in Physics, Год журнала: 2024, Номер 87(11), С. 116001 - 116001

Опубликована: Окт. 11, 2024

Abstract Quantum computing promises to provide the next step up in computational power for diverse application areas. In this review, we examine science behind quantum hype, and breakthroughs required achieve true advantage real world applications. Areas that are likely have greatest impact on high performance (HPC) include simulation of systems, optimization, machine learning. We draw our examples from electronic structure calculations fluid dynamics which account a large fraction current scientific engineering use HPC. Potential challenges encoding decoding classical data devices, mismatched clock speeds between processors. Even modest enhancement techniques would far-reaching impacts areas such as weather forecasting, aerospace engineering, design ‘green’ materials sustainable development. This requires significant effort science, communities working together.

Язык: Английский

---

CNNs for Optimal PDE Solvers

2022 13th International Conference on Computing Communication and Networking Technologies (ICCCNT), Год журнала: 2024, Номер unknown, С. 1 - 8

Опубликована: Июнь 24, 2024

Язык: Английский

---