Using surrogate models to optimise designs for laser-driven fusion power production

使用替代模型优化激光驱动聚变发电的设计

• 批准号: 2588325
• 金额: --
• 依托单位: University of Warwick
• 依托单位国家: 英国
• 项目类别: Studentship
• 财政年份: 2021
• 资助国家: 英国
• 起止时间: 2021 至 无数据
• 项目状态: 未结题
• 来源: https://gtr.ukri.org/projects?ref=studentship-2588325
• 关键词: Using; surrogate; models; optimise; designs

Laser-plasma experiments in high-energy density physics (HEDP) and fusion research trigger kinetic scale instabilities whose effects must be included in large-scale fluid simulations. The difference in time and spatial scales between the kinetic and fluid models, along with the cost of the kinetic modelling, have hindered the full inclusion of important kinetic processes in laser-driven fusion simulations. It is critical for the design and interpretation of experiments that this is solved. This project aims to develop a surrogate model for the kinetic processes suitable for including in fluid scale simulations.Laser-plasma experiments in high-energy density physics (HEDP) and fusion research trigger kinetic scale instabilities whose effects must be included in large-scale fluid simulations. The kinetic instabilities are three-wave parametric couplings which generate significant non-thermal electron populations. These so called hot-electrons can propagate deeper into the target than the driving laser and alter the hydrodynamic compression. This can hinder the interpretation of HEDP experiments and possibly prevent the level of compression required for a successful laser-fusion experiment. Three-wave laser-plasma interactions (LPI) have been studied for over 50 years but only recently have realistic simulations been possible which match observations.Warwick is the lead developer of the EPOCH code which is currently being used to perform (world leading) LPI simulations but this can only be run for pico-second time-scales and requires large National HPC resources. The fluid time-scale is on nano-seconds and thus a coupling of these processes requires deriving a suitable surrogate model for the LPI appropriate for specific experimental configurations which can be included in radiation hydrodynamics simulations. Warwick is also the developer of the UK Odin code which is an arbitrary Lagrangian-Eulerian radiation hydrodynamics code for HEDP and fusion. This project will develop the multi-scale modelling required for coupled EPOCH-Odin simulations able to interpret experiments which are dominated by two-plasmon decay and stimulated Raman scattering LPIs.

激光等离子体在高能密度物理(HEDP)和聚变研究中的实验引发了动力学尺度不稳定性，这种不稳定性的影响必须包含在大规模流体模拟中。动力学模型和流体模型在时间和空间尺度上的差异，以及动力学建模的成本，阻碍了在激光驱动聚变模拟中充分纳入重要的动力学过程。这个问题的解决对于实验的设计和解释是至关重要的。该项目旨在开发一种适用于流体尺度模拟的动力学过程代理模型。激光-等离子体在高能密度物理(HEDP)和聚变研究中的实验触发了动力学尺度不稳定性，其影响必须包含在大规模流体模拟中。动力学不稳定性是产生大量非热电子布居的三波参数耦合。这些所谓的超热电子可以比驱动激光传播到目标更深的地方，并改变流体动力压缩。这可能会阻碍对HEDP实验的解释，并可能阻止成功的激光聚变实验所需的压缩水平。三波激光-等离子体相互作用(LPI)已经被研究了50多年，但直到最近才有可能进行与观测相匹配的真实模拟。Warwick是EPOCH代码的主要开发者，该代码目前被用于执行(世界领先的)LPI模拟，但这只能在皮秒时间尺度上运行，并且需要大量的国家HPC资源。流体的时间尺度是纳秒的，因此这些过程的耦合需要推导出适合于特定实验构型的LPI的合适的替代模型，该模型可以包括在辐射流体动力学模拟中。沃里克也是英国ODIN程序的开发者，这是一种用于HEDP和聚变的任意拉格朗日-欧拉辐射流体动力学程序。该项目将发展多尺度模拟所需的耦合EPOCH-ODIN模拟，能够解释以双等离激元衰变和受激拉曼散射LPI为主的实验。