论文标题
部分可观测时空混沌系统的无模型预测
Backward-Angle ($u$-channel) Production at an Electron-Ion Collider
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。
In backward photoproduction of mesons, $γp\rightarrow M p$, the target proton takes most of the photon momentum, while the produced meson recoils in the direction from which the photon came. Thus the Mandelstam $u$ is small, while the squared momentum transfer $t$ is typically large, near the kinematic limit. In a collider geometry, backward production transfers the struck baryon by many units of rapidity, in a striking similarity to baryon stopping. We explore this similarity, and point out the similarities between the Regge theories used to model baryon stopping with those that are used for backward production. We then explore how backward production can be explored at higher energies than are available at fixed target experiments, by studying production at an electron-ion collider. We calculate the expected $ep$ cross sections and rates, finding that the rate for backward $ω$ production is about 1/300 that of forward $ω$s. We discuss the kinematics of backward production and consider the detector requirements for experimental study.