论文标题
部分可观测时空混沌系统的无模型预测
Unique identification and domination of edges in a graph: The vertex-edge dominant edge metric dimension
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。
Dominating sets and resolving sets have important applications in control theory and computer science. In this paper, we introduce an edge-analog of the classical dominant metric dimension of graphs. By combining the concepts of a vertex-edge dominating set and an edge resolving set, we introduce the notion of a vertex-edge dominant edge resolving set of a graph. We call the minimum cardinality of such a set in a graph $\G$, the vertex-edge dominant edge metric dimension $\g_{emd}(\G)$ of $\G$. The new parameter $\g_{emd}$ is calculated for some common families such as paths, cycles, complete bipartite graphs, wheel and fan graphs. We also calculate $\g_{emd}$ for some Cartesian products of path with path and path with cycle. Importantly, some general results and bounds are presented for this new parameter. We also conduct a comparative analysis of $\g_{emd}$ with the dominant metric dimension of graphs. Comparison shows that these two parameters are not comparable, in general. Upon considering the class of bipartite graphs, we show that $\g_{emd}(T_n)$ of a tree $T_n$ is always less than or equal to its dominant metric dimension. However, we show that for non-tree bipartite graphs, the parameter is not comparable just like general graphs. Based on the results in this paper, we propose some open problems at the end.