Orchestrating TRANSP Simulations for Interpretative and Predictive Tokamak Modeling with OMFIT
Orchestrating TRANSP Simulations for Interpretative and Predictive Tokamak Modeling with OMFIT
复制标题
使用 OMFIT 协调 TRANSP 模拟以进行解释性和预测性托卡马克建模
DOI：
10.1080/15361055.2017.1398585
复制
发表时间：
2018
影响因子：
0.9
通讯作者：
F. Poli
中科院分区：
文献类型：
--
作者：
B. Grierson;X. Yuan;M. Gorelenkova;S. Kaye;N. Logan;O. Meneghini;S. Haskey;J. Buchanan;M. Fitzgerald;S. Smith;L. Cui;R. Budny;F. Poli
摘要 OMFIT 工作流程管理器中使用 TRANSP 模拟，以在 DIII-D、NSTX、JET 和 C-MOD 托卡马克上实现独立于机器的实验分析、事后验证和预测时间相关模拟。用于准备来自等离子体轮廓诊断和平衡重建的输入数据的程序，以及处理时间相关的加热和电流驱动源以及有关中性回收的假设的程序，在不同的机器上有所不同，但通过使用通用的工作流程管理器进行了简化。 TRANSP 模拟保真度设置已纳入 OMFIT 框架，对比镜头间分析、功率平衡和快速粒子模拟。计算先前建立的一系列数据一致性度量，例如实验与计算的中子速率、平衡存储能量与轮廓和快离子压力的总存储能量的比较，以及实验与计算的表面环路电压的比较。数据一致性指标之间的差异可能表明输入量的错误，例如电子密度分布或 ，或表明异常的快速粒子输运。 OMFIT 提供了评估验证指标对输入量敏感性的措施，包括输入配置文件扫描和标准化后处理可视化。对于预测模拟，TRANSP 使用 GLF23 或 TGLF 来预测核心等离子体剖面，并在等离子体外部区域中使用用户定义的边界条件。在后处理中提供国际托卡马克物理活动（ITPA）验证指标，以评估输运模型的有效性。通过使用 OMFIT 来编排实验数据准备、操作模式选择、提交、后处理和可视化的步骤，我们简化并标准化了 TRANSP 的使用。
Abstract TRANSP simulations are being used in the OMFIT workflow manager to enable a machine-independent means of experimental analysis, postdictive validation, and predictive time-dependent simulations on the DIII-D, NSTX, JET, and C-MOD tokamaks. The procedures for preparing input data from plasma profile diagnostics and equilibrium reconstruction, as well as processing of the time-dependent heating and current drive sources and assumptions about the neutral recycling, vary across machines, but are streamlined by using a common workflow manager. Settings for TRANSP simulation fidelity are incorporated into the OMFIT framework, contrasting between-shot analysis, power balance, and fast-particle simulations. A previously established series of data consistency metrics are computed such as comparison of experimental versus calculated neutron rate, equilibrium stored energy versus total stored energy from profile and fast-ion pressure, and experimental versus computed surface loop voltage. Discrepancies between data consistency metrics can indicate errors in input quantities such as electron density profile or , or indicate anomalous fast-particle transport. Measures to assess the sensitivity of the verification metrics to input quantities are provided by OMFIT, including scans of the input profiles and standardized postprocessing visualizations. For predictive simulations, TRANSP uses GLF23 or TGLF to predict core plasma profiles, with user-defined boundary conditions in the outer region of the plasma. International Tokamak Physics Activity (ITPA) validation metrics are provided in postprocessing to assess the transport model validity. By using OMFIT to orchestrate the steps for experimental data preparation, selection of operating mode, submission, postprocessing, and visualization, we have streamlined and standardized the usage of TRANSP.