文摘
In this article, we present a graph theoretic method to visualize and analyze the system behaviorunder different operating conditions. The system attributes (or variables) are the nodes in thegraphs, and partial correlation between a pair of attributes defines the distance betweencorresponding nodes, resulting in a fully connected graph. Then, the redundant links are reducedusing Pathfinder Network Scaling technique to uncover the latent network structure. We use asimulated biological reactor dataset in normal and faulty operation to validate our method. Themethod is general and can be used to analyze several different systems.