Assessing the Influence of an Individual Event in Complex Fault Spreading Network Based on Dynamic Uncertain Causality Graph

作者：Dong, Chunling^*; Zhao, Yue; Zhang, Qin

来源：IEEE Transactions on Neural Networks and Learning Systems, 2016, 27(8): 1615-1630.

DOI：10.1109/TNNLS.2016.2547339

• Causal influence strength
• dynamics
• fault spreading model
• probabilistic reasoning and prediction
• time-variant betweenness centrality (TVBC)
• uncertain causality representation

Identifying the pivotal causes and highly influential spreaders in fault propagation processes is crucial for improving the maintenance decision making for complex systems under abnormal and emergency situations. A dynamic uncertain causality graph-based method is introduced in this paper to explicitly model the uncertain causalities among system components, identify fault conditions, locate the fault origins, and predict the spreading tendency by means of probabilistic reasoning. A new algorithm is proposed to assess the impacts of an individual event by investigating the corresponding node's time-variant betweenness centrality and the strength of global causal influence in the fault propagation network. The algorithm does not depend on the whole original and static network but on the real-time spreading behaviors and dynamics, which makes the algorithm to be specifically targeted and more efficient. Experiments on both simulated networks and real-world systems demonstrate the accuracy, effectiveness, and comprehensibility of the proposed method for the fault management of power grids and other complex networked systems.

• 出版日期2016-8
• 单位清华大学; 山东师范大学