Font Size: a A A

Design Of Microstructured Fiber Based On Machine Learning Optimization And Its Nonlinear Scattering Effect And Sensing Performance

Posted on:2023-08-24Degree:MasterType:Thesis
Country:ChinaCandidate:Z JinFull Text:PDF
GTID:2530306914464744Subject:Electronic and communication engineering
Abstract/Summary:
In recent years,the flexible design of microstructured fiber has attracted the attention of many researchers and made progress in many fields.Microstructured fibers have significant advantages in some optical properties,such as large mode field area,high birefringence,low loss and high nonlinearity.In particular,the nonlinear performance of microstructured fiber can be several orders of magnitude higher than that of ordinary fiber.There are many nonlinear effects,among which stimulated Brillouin scattering(SBS)is widely used in optical fiber communication.The main work of this paper is the structure design of microstructured fiber and the study of nonlinear scattering effect and sensing performance.The main research contents are as follows:(1)A micro-structure polarization-preserving few-mode fiber structure based on an elliptic ring core is designed.The fiber structure has an elliptic ring core and four circular pores arranged in the center diamond.It Can support 14 polarization modes,and can effectively separate the adjacent modes.Then,the properties of the structure fiber,such as broadband characteristics,dispersion and mode field area,are analyzed.The minimum effective refractive index difference of each mode is increased to 1.49×10-4 by neural network reverse optimizatiol,which is 38%higher than that without optimization.Then,a high birefringence and high nonlinear photonic crystal fiber(PCF)structure is designed.The air holes in the fiber cladding are arranged in rectangular pattern.Through neural network optimization,the birefringence value of 1550nm wavelength is increased from 2.17×10-2 to 3.81×10-2,with an increase of 75.6%,and the nonlinear coefficient is increased from 61.1W-1km-1 to 78.9W-1km-1,with an increase of 29.1%.The performance improvement effect is significant.(2)In order to study the influence of structural parameters of PCF on The Brillouin scattering characteristics,the effects of the number and spacing of pores in the outer layer and the long axis radius of elliptic pores in the inner layer on the Brillouin scattering characteristics of PCF with high birefringence and high nonlinear were simulated by finite element method.The changes of Brillouin frequency shift and acoustoptic coupling coefficient are analyzed.The results show that the acoustoptic coupling coefficient of the PCF designed in this paper can reach 96.3%,which can effectively enhance the intensity of Brillouin scattering spectrum.(3)In order to study the temperature sensing characteristics of PCF,two different filling methods were selected to fill the air hole of PCF with temperature sensitive liquid ethanol.The influence of different filling methods on the temperature sensitivity performance of photonic crystal fiber was analyzed.By comparison,it was found that at the wavelength of 1550nm,the high temperature sensitivity of 1.58×10-5℃-1 can be obtained by the high birefringence and high nonlinear photonic crystal fiber.The optical fiber has potential application value in the field of temperature sensor.The two microstructured fiber structures designed in this paper can be reverse-optimized by using neural network,which provides a new and more convenient method for fiber optimization.In addition,the designed PCF structure achieves both high birefringence and high nonlinearity,which can be applied in the sensing system and can effectively improve the intensity of Brillouin scattering spectrum.
Keywords/Search Tags:brillouin scattering, microstructured fiber, high birefringence, the high nonlinear, neural network
Related items