Inverse design of semiconductor laser parameters based on deep learning and particle swarm optimization method

Zihao Ma; Pei Feng; Yu Li

doi:10.1117/12.2548701

20 December 2019 Inverse design of semiconductor laser parameters based on deep learning and particle swarm optimization method

Zihao Ma, Pei Feng, Yu Li

Author Affiliations +

Proceedings Volume 11209, Eleventh International Conference on Information Optics and Photonics (CIOP 2019); 112092X (2019) https://doi.org/10.1117/12.2548701
Event: Eleventh International Conference on Information Optics and Photonics (CIOP 2019), 2019, Xi'an, China

Abstract

Deducing the design parameters of semiconductor lasers from a desired spectrum and light-current (L-I) curve, etc., i.e. the inverse design technique, is highly demanded both academically and in industry. Here, we propose an approach to obtain the inverse design of a laser by combining the deep learning algorithm and the particle swarm optimization (PSO) method. The deep-learning neural network (NN) is trained by the traveling-wave model (TWM) calculated database, and is used to predict the output power for any given new set of design parameters. The standard deviation of the NN approximation can be as low as 0.31mW, and the CPU time as fast as 0.07s, which is much more efficient compared with the TWM numerical algorithm (of CPU time 125.57s), for the same L-I curve calculation. By combining NN with the PSO algorithm, laser parameters can be inversely designed and optimized according to the given/desired L-I curve. Speed of the this process can be improved by about 17,500 times, and the designed parameters are found to be close to their preset values in the test, which indicates its possibility to solve the nonlinear problem for the semiconductor laser process.

Citation Download Citation

Zihao Ma, Pei Feng, and Yu Li "Inverse design of semiconductor laser parameters based on deep learning and particle swarm optimization method", Proc. SPIE 11209, Eleventh International Conference on Information Optics and Photonics (CIOP 2019), 112092X (20 December 2019); https://doi.org/10.1117/12.2548701

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available