论文标题
部分可观测时空混沌系统的无模型预测
Simple method for determining asymptotic states of fast neutrino-flavor conversion
论文作者
论文摘要
储层计算是预测湍流的有力工具，其简单的架构具有处理大型系统的计算效率。然而，其实现通常需要完整的状态向量测量和系统非线性知识。我们使用非线性投影函数将系统测量扩展到高维空间，然后将其输入到储层中以获得预测。我们展示了这种储层计算网络在时空混沌系统上的应用，该系统模拟了湍流的若干特征。我们表明，使用径向基函数作为非线性投影器，即使只有部分观测并且不知道控制方程，也能稳健地捕捉复杂的系统非线性。最后，我们表明，当测量稀疏、不完整且带有噪声，甚至控制方程变得不准确时，我们的网络仍然可以产生相当准确的预测，从而为实际湍流系统的无模型预测铺平了道路。
Neutrino-neutrino forward scatterings potentially induce collective neutrino oscillation in dense neutrino gases in astrophysical sites such as core-collapse supernovae (CCSN) and binary neutron star mergers (BNSM). In this paper, we present a detailed study of fast neutrino-flavor conversion (FFC), paying special attention to asymptotic states, by linear stability analysis and local simulations with a periodic boundary condition. We find that asymptotic states can be characterized by two key properties of FFC: (1) the conservation of lepton number for each flavor of neutrinos and (2) the disappearance of ELN(electron neutrino-lepton number)-XLN(heavy-leptonic one) angular crossings in the spatial- or time-averaged distributions. The system which initially has the positive (negative) ELN-XLN density reaches a flavor equipartition in the negative (positive) ELN-XLN angular directions, and the other part compensates it to preserve the conservation laws. These properties of FFCs offer an approximate scheme determining the survival probability of neutrinos in asymptotic states without solving quantum kinetic equations. We also demonstrate that the total amount of flavor conversion can vary with species-dependent neutrino distributions for identical ELN-XLN ones. Our results suggest that even shallow or narrow ELN angular crossings have the ability to drive large flavor conversion, exhibiting the need for including the effects of FFCs in the modeling of CCSN and BNSM.
