Auer Sabine, Hellmann Frank, Krause Marie, Kurths Jürgen
Potsdam Institute for Climate Impact Research, 14412 Potsdam, Germany.
Institute of Mathematics, Technical University of Berlin, 10587 Berlin, Germany.
Chaos. 2017 Dec;27(12):127003. doi: 10.1063/1.5001818.
90% of all Renewable Energy Power in Germany is installed in tree-like distribution grids. Intermittent power fluctuations from such sources introduce new dynamics into the lower grid layers. At the same time, distributed resources will have to contribute to stabilize the grid against these fluctuations in the future. In this paper, we model a system of distributed resources as oscillators on a tree-like, lossy power grid and its ability to withstand desynchronization from localized intermittent renewable infeed. We find a remarkable interplay of the network structure and the position of the node at which the fluctuations are fed in. An important precondition for our findings is the presence of losses in distribution grids. Then, the most network central node splits the network into branches with different influence on network stability. Troublemakers, i.e., nodes at which fluctuations are especially exciting the grid, tend to be downstream branches with high net power outflow. For low coupling strength, we also find branches of nodes vulnerable to fluctuations anywhere in the network. These network regions can be predicted at high confidence using an eigenvector based network measure taking the turbulent nature of perturbations into account. While we focus here on tree-like networks, the observed effects also appear, albeit less pronounced, for weakly meshed grids. On the other hand, the observed effects disappear for lossless power grids often studied in the complex system literature.
德国90%的可再生能源电力都安装在树状配电网中。此类能源产生的间歇性功率波动给较低层级的电网引入了新的动态变化。与此同时，分布式资源在未来将必须有助于稳定电网以抵御这些波动。在本文中，我们将分布式资源系统建模为树状、有损耗电网中的振荡器，以及其抵御因局部间歇性可再生能源馈入而导致失步的能力。我们发现网络结构与波动馈入节点的位置之间存在显著的相互作用。我们研究结果的一个重要前提是配电网中存在损耗。然后，网络中最核心的节点会将网络分裂成对网络稳定性有不同影响的分支。制造麻烦的节点，即波动特别容易激发电网的节点，往往是净功率流出较高的下游分支。对于低耦合强度，我们还在网络中的任何位置发现了易受波动影响的节点分支。利用一种考虑了扰动湍流性质的基于特征向量的网络度量，可以高度准确地预测这些网络区域。虽然我们在此聚焦于树状网络，但对于弱网状电网也会出现观察到的这些效应，尽管不太明显。另一方面，对于复杂系统文献中经常研究的无损电网，观察到的这些效应则会消失。
