PNCDI III, Program 5.Subprogram 5.1 ELI-RO Project Type : ELI-RO Contract no. 03 / 01.10.2020 Project span: 01.10.2020 - 30.09.2023 Project leader: Dr. Valer Toşa
Project members
|
INCDTIM |
Project Summary
The main purpose of this project is to develop a fast numerical instrument which performs the complete temporal-spectral reconstruction of the ultrashort laser pulses at ELI-NP facility. In laser physics laboratories in general, and at ELI-NP in particular ultrashort pulses with duration from femtoseconds down to attoseconds are routinely generated. Their application ranges from exploring the fundamental constituents of matter from nuclear processes to electronic processes in atoms, molecules, solids to application even in medicine. The exact temporal and spectral shape of such pulses is crucially important, but difficult to measure. Experimental pulse characterization methods have been developed, like the frequency resolved optical gating (FROG) and the dispersion scan (D-scan). These methods work well experimentally, but as raw result they yield a complicated 2D image, so called trace. Very time consuming and slowly converging iterative retrieval algorithms are necessary to reconstruct the spectral amplitude and phase of the studied pulse from the measured trace. Here we propose to develop a software based on artificial neural networks (ANN) for the complete reconstruction of laser pulses generated at ELI-NP. ANNs are high performance and high speed deep learning algorithms, which have been successfully used in diverse nonlinear problems and applications like image and pattern recognition, classification, identification. Our research group at INCDTIM (Cluj-Napoca) has previous experiences in modeling ultrashort laser pulse propagation in ionizing medium, as well as exploring ANN for prediction purposes in nonlinear physical and chemical processes. We have part of code at basic level of development and we propose to further develop and extend the ANN code in order to be able to receive as input the 2D dispersion scan trace, and yield as predicted output the spectral amplitude and phase of the unknown pulse. Our goal is to demonstrate the method's scientific feasibility and critical properties through proof-of-concept experiments, i.e. teach the ANN to reconstruct the real laser pulses from the experimentally measured traces which will be provided by the partner group at ELI-NP. An adaptive mirror will be introduced in the diagnostics beamline with the purpose to correct the phase front distortions of the laser pulse which are inevitably present due to propagation effects and beam steering. The software will be extended to assess the effects of wavefront distortions on the focusing and pulse measurements. The project will directly contribute to the laser system development at ELI-NP, offering solutions for problems related to Alignment and Diagnostics, Pulse shaping, Adaptive optics and Quality of focus. |
No. |
Name |
Professional degree |
Profession |
Professional experience |
Project Coordinator INCDTIM |
||||
1 |
Toşa VALER |
SRI |
Physicist |
Numerical modeling |
2 |
Katalin KOVÁCS |
SRII |
Physicist |
Modeling pulse propagation |
3 |
Ana Maria Mihaela GHERMAN |
SR |
Chemical Engineer |
Artificial Neural Network specialist |
4 |
István Ferenc TÓTH |
SR |
Physicist |
ANN programming |
5 |
Attila BENDE |
SRI |
Physicist |
Modeling laser-atom interaction |
Partner IFIN-HH |
||||
6 |
Daniel URSESCU |
RSI |
Physicist |
Ultraintense lasers: diagnostics and experiments |
7 |
Dan G. MATEI |
RSIII |
Physicist |
Laser metrology and material science |
8 |
Anda Maria TALPOȘI |
PhD student |
Engineer |
Propagation of ultrashort pulses |
9 |
Vicentiu IANCU |
PhD student |
Physicist |
Wavefront metrology and corrections |
10 |
Aurelian COSTEA |
Engineer |
Electronics Engineer |
Electronics and systems engineering |
11 |
Andreea-Magdalena CĂLIN |
Technician |
Technician |
Equipment operator / programmer |