论文标题
部分可观测时空混沌系统的无模型预测
Hubble Space Telescope Observations of Active Asteroid P/2020 O1 (Lemmon-PANSTARRS)
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。
We present Hubble Space Telescope observations of active asteroid P/2020 O1 taken to examine its development for a year after perihelion. We find that the mass loss peaks <~1 kg/s in 2020 August and then declines to nearly zero over four months. Long-duration mass loss (~180 days) is consistent with a sublimation origin, indicating that this object is likely an ice-bearing main-belt comet. Equilibrium sublimation of water ice from an area as small as 1580 m^2 can supply the observed mass loss. Time-series photometry shows tentative evidence for extremely rapid rotation (double-peaked period < 2 hr) of the small nucleus (effective radius ~420 m). Ejection velocities of 0.1 mm particles are comparable to the 0.3 m/s gravitational escape speed from the nucleus, while larger particles are ejected at speeds less than the escape velocity. These properties are consistent with the sublimation of near-surface ice aided by centripetal acceleration. If water ice sublimation is confirmed, P/2020 O1 would be the icy asteroid with the smallest semimajor axis (highest temperature), setting new bounds on the distribution of ice in the asteroid belt.