具有通量限制扩散和自适应网格加密的多群辐射流体动力学
Multigroup radiation hydrodynamics with flux-limited diffusion and adaptive mesh refinement
摘要：
辐射转移在恒星形成过程中起着关键作用。由于计算成本高，迄今为止，辐射流体动力学模拟主要是在灰色近似下进行的。近年来，多频率辐射流体动力学模型开始出现，试图更好地解释不透明度随频率的变化。我们希望开发一个有效的多组算法的自适应网格加密程序RAMSES，这是适合于沉重的原恒星坍缩计算。由于显式辐射转移方法的时间步长限制，我们构建了一个时间隐式求解器的基础上稳定的双共轭梯度算法，并实现在通量限制扩散近似下的RAMSES。我们提出了一系列的测试，证明了我们的计划在处理与频率相关的辐射流体动力学流的高性能。我们还提出了一个初步的模拟三维原恒星崩溃使用20个频率组。简要讨论了灰色和多群分析结果的差异，并说明了这种新方法所带来的大量信息。我们已经实现了一个多群通量限制扩散算法的RAMSES代码。该方法对标准的辐射流体动力学测试表现良好，也被证明是成熟的计算恒星形成的背景下开发。
英文摘要：
Radiative transfer plays a key role in the star formation process. Due to a high computational cost, radiation-hydrodynamics simulations performed up to now have mainly been carried out in the grey approximation. In recent years, multi-frequency radiation-hydrodynamics models have started to emerge, in an attempt to better account for the large variations of opacities as a function of frequency. We wish to develop an efficient multigroup algorithm for the adaptive mesh refinement code RAMSES which is suited to heavy proto-stellar collapse calculations. Due to prohibitive timestep constraints of an explicit radiative transfer method, we constructed a time-implicit solver based on a stabilised bi-conjugate gradient algorithm, and implemented it in RAMSES under the flux-limited diffusion approximation. We present a series of tests which demonstrate the high performance of our scheme in dealing with frequency-dependent radiation-hydrodynamic flows. We also present a preliminary simulation of a three-dimensional proto-stellar collapse using 20 frequency groups. Differences between grey and multigroup results are briefly discussed, and the large amount of information this new method brings us is also illustrated. We have implemented a multigroup flux-limited diffusion algorithm in the RAMSES code. The method performed well against standard radiation-hydrodynamics tests, and was also shown to be ripe for exploitation in the computational star formation context.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
