Linkun Chen
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Linkun Chen has a total of 2 co-authors. Here are the top co-authors by number of shared publications:
- Chao Wang
College of Smart Materials and Future Energy, Fudan University, Shanghai, China
1 shared publication - Rui Jin
College of Smart Materials and Future Energy, Fudan University, Shanghai, China
1 shared publication
Latest Publications
- 9 June 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Quantum Electronics
- Vol. PP (00189197) , 1
Theoretical Modeling and Parameter Estimation of Erbium-doped Fiber Ring Laser Based on Physics-Informed Neural Network
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.
- 9 June 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Quantum Electronics
- Vol. PP (00189197) , 1
Theoretical Modeling and Parameter Estimation of Erbium-doped Fiber Ring Laser Based on Physics-Informed Neural Network
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.