Efficient calculation of optical properties of suspended-core fiber via a machine learning algorithm

Shuyu Yuan; Shengchao Chen; Jianli Yang; Qian Yang; Sufen Ren; Guanjun Wang; Guanjun Wang; Guanjun Wang; Benguo Yu; Benguo Yu

doi:10.1364/AO.458048

Applied Optics
Vol. 61,
Issue 19,
pp. 5714-5721
(2022)
•https://doi.org/10.1364/AO.458048

Efficient calculation of optical properties of suspended-core fiber via a machine learning algorithm

Shuyu Yuan, Shengchao Chen, Jianli Yang, Qian Yang, Sufen Ren, Guanjun Wang, and Benguo Yu

Not Accessible

Your library or personal account may give you access

Get PDF
Email
Share
Get Citation
Copy Citation Text
Shuyu Yuan, Shengchao Chen, Jianli Yang, Qian Yang, Sufen Ren, Guanjun Wang, and Benguo Yu, "Efficient calculation of optical properties of suspended-core fiber via a machine learning algorithm," Appl. Opt. 61, 5714-5721 (2022)

Export Citation
- BibTex
- Endnote (RIS)
- HTML
- Plain Text
Citation alert
Save article

Check for updates

Abstract

Growing nonlinearity demands in mid-infrared applications place more outstanding requirements on fiber structure design. Chalcogenide suspended-core fibers (SCFs) are considered excellent candidates for mid-infrared applications due to their significant advantages in nonlinearity and dispersion management. However, traditional numerical methods for accurate modeling and optimization of SCFs often rely on the performance of computing devices and have many limitations when dealing with complex models. A machine learning algorithm is applied to calculate the optical properties of chalcogenide SCFs, including effective mode area, nonlinear coefficient, and dispersion. The established artificial neural network (ANN) model enables accurate prediction of the above optical properties of ${\rm As_2}{\rm S_3}$ SCF, for which in the wavelength range of 1.0 to 4.0 µm, the radius of the fiber core is 0.4 to 0.6 µm, and width of the cantilever is 0.06 to 0.09 µm. We demonstrate that this simple ANN model has considerable advantages over the traditional numerical calculation model in computational speed and resource utilization. In summary, the proposed model can quickly provide more accurate optical property predictions, providing a cost-effective solution for precise modeling and optimization of chalcogenide SCFs.

Full Article | PDF Article

More Like This

Machine learning approach for computing optical properties of a photonic crystal fiber

Sunny Chugh, Aamir Gulistan, Souvik Ghosh, and B. M. A. Rahman
Opt. Express 27(25) 36414-36425 (2019)

Resonance prediction and inverse design of multi-core selective couplers based on neural networks

Junjie Fan, Wei Huang, Ran Zhang, Zhiwei Gu, Binbin Song, and Shengyong Chen
Appl. Opt. 61(32) 9350-9359 (2022)

Broadband mid-infrared fiber optical parametric oscillator based on a three-hole suspended-core chalcogenide fiber

Hangyu Bai, Xiong Yang, Yizhen Wei, and Shiming Gao
Appl. Opt. 55(3) 515-521 (2016)

Previous Article Next Article

Data availability

Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.

Cited By

You do not have subscription access to this journal. Cited by links are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Figures (9)

You do not have subscription access to this journal. Figure files are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Tables (8)

You do not have subscription access to this journal. Article tables are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Equations (18)

You do not have subscription access to this journal. Equations are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Optimizer	LR	MSE (Training)	MSE (Validation)
SGD	0.001	0.0218	0.0217
	0.005	0.0264	0.0249
	0.01	0.0156	0.0151
RMSprop	0.001	0.0125	0.0123
	0.005	0.0121	0.0119
	0.01	0.0123	0.0121
Adagrad	0.001	0.0023	0.0022
	0.005	0.0020	0.0022
	0.01	0.0018	0.0014
Adam	0.001	0.0010	0.0013
	0.005	0.0017	0.0013
	0.01	0.0008	0.0011

Optimizer	LR	MSE	RMSE	MAPE	$R^{2}$
SGD	0.001	0.018	0.128	116.539	0.623
	0.005	0.018	0.128	107.244	0.624
	0.01	0.017	0.125	78.4533	0.648
RMSprop	0.001	0.019	0.134	60.6486	0.546
	0.005	0.016	0.121	33.8022	0.669
	0.01	0.016	0.123	31.49	0.685
Adagrad	0.001	0.014	0.114	30.6853	0.73
	0.005	0.014	0.113	29.9399	0.735
	0.01	0.013	0.112	28.4939	0.737
Adam	0.001	0.013	0.111	28.5548	0.743
	0.005	0.014	0.115	35.5825	0.73
	0.01	0.011	0.104	23.2043	0.777

Optimizer	MSE	RMSE	MAPE	$R^{2}$
SGD	0.018	0.133	79.565	0.623
RMSprop	0.019	0.138	33.665	0.673
Adagrad	0.014	0.120	30.509	0.726
Adam	0.014	0.118	29.276	0.737

Optimizer	LR	MSE	RMSE	MAPE	$R^{2}$
SGD	0.001	0.020	0.138	126.250	0.588
	0.005	0.020	0.139	116.181	0.589
	0.01	0.019	0.135	84.991	0.612
RMSprop	0.001	0.022	0.145	65.703	0.515
	0.005	0.018	0.131	36.619	0.631
	0.01	0.018	0.133	34.114	0.646
Adagrad	0.001	0.016	0.123	33.242	0.688
	0.005	0.015	0.122	32.435	0.693
	0.01	0.015	0.121	30.868	0.695
Adam	0.001	0.015	0.120	30.934	0.701
	0.005	0.016	0.125	38.548	0.688
	0.01	0.013	0.112	25.138	0.733

Optimizer	MSE	RMSE	MAPE	$R^{2}$
SGD	0.020	0.140	65.868	0.616
RMSprop	0.021	0.144	34.304	0.646
Adagrad	0.015	0.122	30.508	0.687
Adam	0.013	0.113	24.845	0.724

Optimizer	LR	MSE	RMSE	MAPE	$R^{2}$
SGD	0.001	0.023	0.160	145.673	0.570
	0.005	0.023	0.160	134.055	0.571
	0.01	0.022	0.156	98.067	0.593
RMSprop	0.001	0.025	0.167	75.811	0.499
	0.005	0.020	0.151	42.253	0.612
	0.01	0.021	0.154	39.362	0.627
Adagrad	0.001	0.018	0.142	38.357	0.668
	0.005	0.018	0.141	37.425	0.672
	0.01	0.017	0.140	35.617	0.674
Adam	0.001	0.017	0.139	35.693	0.680
	0.005	0.018	0.144	44.478	0.667
	0.01	0.015	0.130	29.005	0.711

Optimizer	MSE	RMSE	MAPE	$R^{2}$
SGD	0.021	0.146	96.923	0.586
RMSprop	0.021	0.144	38.903	0.619
Adagrad	0.017	0.131	35.202	0.666
Adam	0.015	0.122	28.667	0.703

	COMSOL Numerical Calculation Model (Number of Mesh Elements)			ANN Model (Our Proposed)
	8293	32,928	131,712	Training	Testing
Time consumption (second)	9.75	52.21	231.32	19	$< 1$
RAM utilization (MB)	384.4	1200.0	3056.6	1056

Abstract

Data availability

Cited By

Figures (9)

Tables (8)

Equations (18)

Applied Optics