论文标题
部分可观测时空混沌系统的无模型预测
Mapping Progenitors of Binary Black Holes and Neutron Stars with Binary Population Synthesis
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。
The first directly observed gravitational wave event, GW150914, featuring the merger of two massive black holes, highlighted the need to determine how these systems of compact remnant binaries are formed. We use the binary population synthesis code COSMIC (Compact Object Synthesis and Monte Carlo Investigation Code) to predict the types of massive stars that will show significant radial velocity variations, indicative of a potential compact object (i.e. a black hole or neutron star) orbiting the star. We "observe" the binaries generated in the populations with a similar number of epochs and RV accuracy as planned for the Milky Way Mapper. In this analysis, we are especially interested in systems where a compact remnant is orbiting a massive O or B star as these systems survived the first supernova and neutron star kick. We test the ability of the Milky Way Mapper observing strategy to distinguish among different mass loss and kick prescriptions. We find that Wolf-Rayet stars or hot subdwarfs in binaries could be detectable (i.e. luminous, high delta RV max), viable progenitors of such objects, while the different prescriptions primarily affect the number of sources.
