G. Kane, P. Drobniak, S. Kazamias, V. Kubytskyi, M. Lenivenko, B. Lucas, J. Serhal, K. Cassou, A. Beck, A. Specka, F. Massimo
{"title":"通过数值优化进行激光等离子加速器电子源设计的代用模型研究","authors":"G. Kane, P. Drobniak, S. Kazamias, V. Kubytskyi, M. Lenivenko, B. Lucas, J. Serhal, K. Cassou, A. Beck, A. Specka, F. Massimo","doi":"arxiv-2408.15845","DOIUrl":null,"url":null,"abstract":"The optimisation of the plasma target design for high quality beam\nlaser-driven plasma injector electron source relies on numerical parametric\nstudies using Particle in Cell (PIC) codes. The common input parameters to\nexplore are laser characteristics and plasma density profiles extracted from\ncomputational fluid dynamic studies compatible with experimental measurements\nof target plasma density profiles. We demonstrate the construction of surrogate\nmodels using machine learning technique for a laser-plasma injector (LPI)\nelectron source based on more than 12000 simulations of a laser wakefield\nacceleration performed for sparsely spaced input parameters [1]. Surrogate\nmodels are very interesting for LPI design and optimisation because they are\nmuch faster than PIC simulations. We develop and compare the performance of\nthree surrogate models, namely, Gaussian processes (GP), multilayer perceptron\n(MLP), and decision trees (DT). We then use the best surrogate model to quickly\nfind optimal working points to get a selected electron beam energy, charge and\nenergy spread using different methods, namely random search, Bayesian\noptimisation and multi-objective Bayesian optimisation","PeriodicalId":501274,"journal":{"name":"arXiv - PHYS - Plasma Physics","volume":"26 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2024-08-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Surrogate Models studies for laser-plasma accelerator electron source design through numerical optimisation\",\"authors\":\"G. Kane, P. Drobniak, S. Kazamias, V. Kubytskyi, M. Lenivenko, B. Lucas, J. Serhal, K. Cassou, A. Beck, A. Specka, F. Massimo\",\"doi\":\"arxiv-2408.15845\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"The optimisation of the plasma target design for high quality beam\\nlaser-driven plasma injector electron source relies on numerical parametric\\nstudies using Particle in Cell (PIC) codes. The common input parameters to\\nexplore are laser characteristics and plasma density profiles extracted from\\ncomputational fluid dynamic studies compatible with experimental measurements\\nof target plasma density profiles. We demonstrate the construction of surrogate\\nmodels using machine learning technique for a laser-plasma injector (LPI)\\nelectron source based on more than 12000 simulations of a laser wakefield\\nacceleration performed for sparsely spaced input parameters [1]. Surrogate\\nmodels are very interesting for LPI design and optimisation because they are\\nmuch faster than PIC simulations. We develop and compare the performance of\\nthree surrogate models, namely, Gaussian processes (GP), multilayer perceptron\\n(MLP), and decision trees (DT). We then use the best surrogate model to quickly\\nfind optimal working points to get a selected electron beam energy, charge and\\nenergy spread using different methods, namely random search, Bayesian\\noptimisation and multi-objective Bayesian optimisation\",\"PeriodicalId\":501274,\"journal\":{\"name\":\"arXiv - PHYS - Plasma Physics\",\"volume\":\"26 1\",\"pages\":\"\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-08-28\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"arXiv - PHYS - Plasma Physics\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/arxiv-2408.15845\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - PHYS - Plasma Physics","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2408.15845","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Surrogate Models studies for laser-plasma accelerator electron source design through numerical optimisation
The optimisation of the plasma target design for high quality beam
laser-driven plasma injector electron source relies on numerical parametric
studies using Particle in Cell (PIC) codes. The common input parameters to
explore are laser characteristics and plasma density profiles extracted from
computational fluid dynamic studies compatible with experimental measurements
of target plasma density profiles. We demonstrate the construction of surrogate
models using machine learning technique for a laser-plasma injector (LPI)
electron source based on more than 12000 simulations of a laser wakefield
acceleration performed for sparsely spaced input parameters [1]. Surrogate
models are very interesting for LPI design and optimisation because they are
much faster than PIC simulations. We develop and compare the performance of
three surrogate models, namely, Gaussian processes (GP), multilayer perceptron
(MLP), and decision trees (DT). We then use the best surrogate model to quickly
find optimal working points to get a selected electron beam energy, charge and
energy spread using different methods, namely random search, Bayesian
optimisation and multi-objective Bayesian optimisation