Inverse Parameters Estimation for Electro-Optic Frequency Combs Generation via Modified Physics-Informed Neural Network
To solve the prediction deterioration in the scenario of the all-fiber electro-optical frequency combs generation, we propose a modified physics-informed neural network for improved accuracy and efficiency. Simulations illustrate that it maximally reduces mean squared error by 82.4% in the inverse estimation and 42.0% in the forward propagation.
Authors
- Xihua Zou (ORCID: https://orcid.org/0000-0002-7240-4229)
- Hui Yang (ORCID: https://orcid.org/0000-0002-0973-8058)
- Jincong Hu (ORCID: https://orcid.org/0009-0005-8432-1681)
- Ying Wu (ORCID: https://orcid.org/0000-0002-7919-1107)
Institutions
- Southwest Jiaotong University (CN)
Publication Details
- Published
- 2024-11-02
- DOI
- https://doi.org/10.1109/acp/ipoc63121.2024.10810095
- Primary Topic
- Advanced Fiber Laser Technologies
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- National Natural Science Foundation of China
- National Key Research and Development Program of China