Theoretical Modeling and Parameter Estimation of Erbium-doped Fiber Ring Laser Based on Physics-Informed Neural Network
- 9 June 2026
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Quantum Electronics
- Vol. PP (00189197) , 1
Abstract
A framework for the theoretical modeling and parameter estimation of erbium-doped fiber ring laser (EDFRL) is proposed based on physics-informed neural network (PINN). By introducing multi-dimensional parameterized input architecture, numerical transformation mapping and physics-consistent constraint strategy, this framework effectively addresses numerical stiffness and nonseparable boundary conditions in the closed-loop laser system. On this basis, forward prediction and inverse retrieval models were established. Within a multi-dimensional parameter space, the forward model exhibits a superior advantage in global generalization and computational efficiency, achieving high prediction fidelity (R2 > 0.99) and a maximum speedup by four orders of magnitude. Simultaneously, physical parameters were effectively retrieved from sparse data by the inverse model. Therefore, a collaborative workflow can be established, offering a novel and physically rigorous paradigm for the modeling, optimization, and calibration of nonlinear laser systems including EDFRL.Keywords
Funding Information
- Key research and development project of Xinjiang Uygur Autonomous Region (2024B03041)
This publication has 27 references indexed in Scilit:
- Physics‐Informed Neural Network for Nonlinear Dynamics in Fiber OpticsLaser & Photonics Reviews, 2022
- Estimates on the generalization error of physics-informed neural networks for approximating a class of inverse problems for PDEsIMA Journal of Numerical Analysis, 2021
- Learning nonlinear operators via DeepONet based on the universal approximation theorem of operatorsNature Machine Intelligence, 2021
- Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equationsJournal of Computational Physics, 2019
- A combined algorithm for simulating fiber lasers based on the shooting and relaxation methodsOptical Fiber Technology, 2012
- An efficient shooting method for fiber amplifiers and lasersOptics & Laser Technology, 2008
- A broadly tunable erbium-doped fiber ring laser: experimentation and modelingIEEE Journal of Selected Topics in Quantum Electronics, 2001
- Modeling of erbium doped fiber ring laserOptics Communications, 1997
- Analytical model for rare-earth-doped fiber amplifiers and lasersIEEE Journal of Quantum Electronics, 1994
- Modeling erbium-doped fiber amplifiersJournal of Lightwave Technology, 1991