Multi-parameter optimisation of dual-pump NALM fibre laser using machine-learning approaches

Sonia Boscolo, Ilya Gukov, Sergei K. Turitsyn, Christophe Finot

Research output: Contribution to conferencePaper

Abstract

Recently, a new design of a model-locked all-fibre Figure-8 laser employing a nonlinear amplifying loop mirror (NALM) with two active fibre segments and two independently controlled pump-power modules has been proposed and experimentally demonstrated. This laser layout combines the reliability and robustness of conventional Figure-8 lasers with the flexibility of nonlinear-polarisation-evolution (NPE) lasers, providing access to a variety of generation regimes with a relatively wide adjustment range of the pulse parameters. Moreover, it enables reliable and
reproducible live electronic adjustment of the lasing regimes, which is practically impossible to do by adjusting fibre-based polarisation controllers in NPE lasers.
The general issue of reaching a target mode-locked laser regime with a setup featuring many adjustable parameters can be intelligently addressed by using machine-learning techniques. Here, we apply predictive regression to find
optimum operating regimes in the NALM laser that are accessible through independent control of the pump powers in the gain segments, Pp,1, Pp,2, and the laser output coupling ratio β. We use a piece-wise propagation model for
generating data that characterises the laser. In the fibres, propagation follows a standard modified nonlinear-Schrödinger equation including gain saturation and spectral response for the active segments. The gain coefficient amplitude is dependent on the average signal and pump powers, the average power dynamics being described by standard rate equations. We have trained a gradient boosted tree algorithm on our dataset to identify high-energy, stable mode-locked solutions across the full variation range of the total pump level delivered to the active fibres, Pp,tot, the ratio Pp1/Pp,tot, and β (tens of thousands of points). The algorithm has quickly handled the whole parameter space.
Our approach paves the way for alternative approaches to the optimisation of nonlinear cavity dynamics, and can be generalised to other complex systems and higher degrees of freedom.
Original languageEnglish
Pages46
Publication statusPublished - Sep 2018
EventPhoton 2018 - Aston University, Birmingham, United Kingdom
Duration: 3 Sep 20186 Sep 2018
http://www.photon.org.uk/home

Conference

ConferencePhoton 2018
CountryUnited Kingdom
CityBirmingham
Period3/09/186/09/18
Internet address

Fingerprint

machine learning
fiber lasers
pumps
mirrors
optimization
lasers
fibers
adjusting
polarization
propagation
laser outputs
complex systems
spectral sensitivity
layouts
nonlinear equations
lasing
regression analysis
controllers
flexibility
degrees of freedom

Bibliographical note

© The Authors

Cite this

Boscolo, S., Gukov, I., Turitsyn, S. K., & Finot, C. (2018). Multi-parameter optimisation of dual-pump NALM fibre laser using machine-learning approaches. 46. Paper presented at Photon 2018, Birmingham, United Kingdom.
Boscolo, Sonia ; Gukov, Ilya ; Turitsyn, Sergei K. ; Finot, Christophe. / Multi-parameter optimisation of dual-pump NALM fibre laser using machine-learning approaches. Paper presented at Photon 2018, Birmingham, United Kingdom.
@conference{e15c17ccb6604ebf950c0a470d8a2167,
title = "Multi-parameter optimisation of dual-pump NALM fibre laser using machine-learning approaches",
abstract = "Recently, a new design of a model-locked all-fibre Figure-8 laser employing a nonlinear amplifying loop mirror (NALM) with two active fibre segments and two independently controlled pump-power modules has been proposed and experimentally demonstrated. This laser layout combines the reliability and robustness of conventional Figure-8 lasers with the flexibility of nonlinear-polarisation-evolution (NPE) lasers, providing access to a variety of generation regimes with a relatively wide adjustment range of the pulse parameters. Moreover, it enables reliable andreproducible live electronic adjustment of the lasing regimes, which is practically impossible to do by adjusting fibre-based polarisation controllers in NPE lasers.The general issue of reaching a target mode-locked laser regime with a setup featuring many adjustable parameters can be intelligently addressed by using machine-learning techniques. Here, we apply predictive regression to findoptimum operating regimes in the NALM laser that are accessible through independent control of the pump powers in the gain segments, Pp,1, Pp,2, and the laser output coupling ratio β. We use a piece-wise propagation model forgenerating data that characterises the laser. In the fibres, propagation follows a standard modified nonlinear-Schr{\"o}dinger equation including gain saturation and spectral response for the active segments. The gain coefficient amplitude is dependent on the average signal and pump powers, the average power dynamics being described by standard rate equations. We have trained a gradient boosted tree algorithm on our dataset to identify high-energy, stable mode-locked solutions across the full variation range of the total pump level delivered to the active fibres, Pp,tot, the ratio Pp1/Pp,tot, and β (tens of thousands of points). The algorithm has quickly handled the whole parameter space. Our approach paves the way for alternative approaches to the optimisation of nonlinear cavity dynamics, and can be generalised to other complex systems and higher degrees of freedom.",
author = "Sonia Boscolo and Ilya Gukov and Turitsyn, {Sergei K.} and Christophe Finot",
note = "{\circledC} The Authors; Photon 2018 ; Conference date: 03-09-2018 Through 06-09-2018",
year = "2018",
month = "9",
language = "English",
pages = "46",
url = "http://www.photon.org.uk/home",

}

Boscolo, S, Gukov, I, Turitsyn, SK & Finot, C 2018, 'Multi-parameter optimisation of dual-pump NALM fibre laser using machine-learning approaches' Paper presented at Photon 2018, Birmingham, United Kingdom, 3/09/18 - 6/09/18, pp. 46.

Multi-parameter optimisation of dual-pump NALM fibre laser using machine-learning approaches. / Boscolo, Sonia; Gukov, Ilya; Turitsyn, Sergei K.; Finot, Christophe.

2018. 46 Paper presented at Photon 2018, Birmingham, United Kingdom.

Research output: Contribution to conferencePaper

TY - CONF

T1 - Multi-parameter optimisation of dual-pump NALM fibre laser using machine-learning approaches

AU - Boscolo, Sonia

AU - Gukov, Ilya

AU - Turitsyn, Sergei K.

AU - Finot, Christophe

N1 - © The Authors

PY - 2018/9

Y1 - 2018/9

N2 - Recently, a new design of a model-locked all-fibre Figure-8 laser employing a nonlinear amplifying loop mirror (NALM) with two active fibre segments and two independently controlled pump-power modules has been proposed and experimentally demonstrated. This laser layout combines the reliability and robustness of conventional Figure-8 lasers with the flexibility of nonlinear-polarisation-evolution (NPE) lasers, providing access to a variety of generation regimes with a relatively wide adjustment range of the pulse parameters. Moreover, it enables reliable andreproducible live electronic adjustment of the lasing regimes, which is practically impossible to do by adjusting fibre-based polarisation controllers in NPE lasers.The general issue of reaching a target mode-locked laser regime with a setup featuring many adjustable parameters can be intelligently addressed by using machine-learning techniques. Here, we apply predictive regression to findoptimum operating regimes in the NALM laser that are accessible through independent control of the pump powers in the gain segments, Pp,1, Pp,2, and the laser output coupling ratio β. We use a piece-wise propagation model forgenerating data that characterises the laser. In the fibres, propagation follows a standard modified nonlinear-Schrödinger equation including gain saturation and spectral response for the active segments. The gain coefficient amplitude is dependent on the average signal and pump powers, the average power dynamics being described by standard rate equations. We have trained a gradient boosted tree algorithm on our dataset to identify high-energy, stable mode-locked solutions across the full variation range of the total pump level delivered to the active fibres, Pp,tot, the ratio Pp1/Pp,tot, and β (tens of thousands of points). The algorithm has quickly handled the whole parameter space. Our approach paves the way for alternative approaches to the optimisation of nonlinear cavity dynamics, and can be generalised to other complex systems and higher degrees of freedom.

AB - Recently, a new design of a model-locked all-fibre Figure-8 laser employing a nonlinear amplifying loop mirror (NALM) with two active fibre segments and two independently controlled pump-power modules has been proposed and experimentally demonstrated. This laser layout combines the reliability and robustness of conventional Figure-8 lasers with the flexibility of nonlinear-polarisation-evolution (NPE) lasers, providing access to a variety of generation regimes with a relatively wide adjustment range of the pulse parameters. Moreover, it enables reliable andreproducible live electronic adjustment of the lasing regimes, which is practically impossible to do by adjusting fibre-based polarisation controllers in NPE lasers.The general issue of reaching a target mode-locked laser regime with a setup featuring many adjustable parameters can be intelligently addressed by using machine-learning techniques. Here, we apply predictive regression to findoptimum operating regimes in the NALM laser that are accessible through independent control of the pump powers in the gain segments, Pp,1, Pp,2, and the laser output coupling ratio β. We use a piece-wise propagation model forgenerating data that characterises the laser. In the fibres, propagation follows a standard modified nonlinear-Schrödinger equation including gain saturation and spectral response for the active segments. The gain coefficient amplitude is dependent on the average signal and pump powers, the average power dynamics being described by standard rate equations. We have trained a gradient boosted tree algorithm on our dataset to identify high-energy, stable mode-locked solutions across the full variation range of the total pump level delivered to the active fibres, Pp,tot, the ratio Pp1/Pp,tot, and β (tens of thousands of points). The algorithm has quickly handled the whole parameter space. Our approach paves the way for alternative approaches to the optimisation of nonlinear cavity dynamics, and can be generalised to other complex systems and higher degrees of freedom.

UR - http://www.photon.org.uk/home

M3 - Paper

SP - 46

ER -

Boscolo S, Gukov I, Turitsyn SK, Finot C. Multi-parameter optimisation of dual-pump NALM fibre laser using machine-learning approaches. 2018. Paper presented at Photon 2018, Birmingham, United Kingdom.