Optimized design of graphene waveguide composite structure pressure sensor based on machine learning

Aiming for high sensitivity and satisfying transmission loss, intelligent optimization methods involving machine learning are worthwhile to apply for optical pressure sensors development since machine learning is versatile to correlate the complex and nonlinear relationship between device performanc...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on instrumentation and measurement 2024-09, p.1-1
Main Authors: Li, Yan, Zhao, Le, Zhou, Huaxu, Zhu, Kehui, Ning, Zijun, Yang, Fuling
Format: Article
Language:English
Subjects:
High sensitivity
Machine learning
NSGA-II
Optical pressure sensor
Optical sensors
Optical waveguides
Optimization
Pareto
Pressure sensors
Propagation losses
Sensitivity
Sensors
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Aiming for high sensitivity and satisfying transmission loss, intelligent optimization methods involving machine learning are worthwhile to apply for optical pressure sensors development since machine learning is versatile to correlate the complex and nonlinear relationship between device performance and sensor structural parameters. This paper proposes a three-stage framework to conduct optimal design of an optical pressure sensor facing multiple optimization objectives. Firstly, a dataset consisting of 1600 samples with combinations of varying waveguide lengths and core widths is used to train and establish a prediction model, which also acts as the objective functions for subsequent optimization. Secondly, a genetic algorithm is applied to find out the Pareto front within the sample space according to optimization objectives. Thirdly, by inverse design based on the Pareto optimal set, the multi-objective optimization for sensor with optimal structural parameters is achieved. Experimental studies show that the optimized design of the graphene waveguide composite structure pressure sensor based on machine learning gives a sensitivity increase of 63.7% and a transmission loss reduction of 18.5% compared to the similar sensors designed in previous work. This effectively verifies the feasibility of the proposed optimization design method in this paper. Without large number of sensor fabrications and trial-and-error experiments, the reverse design of sensors with various performance requirements could be achieved automatically, which provides a new approach to improve the performance of sensors in the future.
ISSN:0018-9456
1557-9662
DOI:10.1109/TIM.2024.3457932