Temporal-spatial Correlation Analysis And Vulnerability Assessment Method For Transmission Network

With the rapid development of the global energy internet,grid interconnection is becoming a tendency among countries in future.The grid interconnection is conducive to the implementation of a globally integrated development strategy and the realization of a community for globally economic destiny.However,due to the large-scale interconnection,the uncertain threats(e.g.,terrorist attacks,extreme weather)to the power grid are increasing;therefore ensuring the efficient,secure and stable operation of the interconnected grid is one of the critical issues.To meet the above-mentioned challenges,transmission network vulnerability assessment is an effective method to identify the vulnerable points of the grid and then protect some points so that the operation of the interconnected grid can be improved.The complex network theory based topological vulnerability assessment(CNT-TVA)for the transmission network is a popular method.The main idea of the CNT-TVA is to construct the topological indices,particularly extended topological indices,based on CNT and then the indices are employed to identify the structural vulnerable points of the network.However,the substantive progress on the CNT based operational vulnerability assessment has not yet been achieved;therefore to provide a new idea for the electrical network vulnerability assessment,this dissertation takes the transmission network as the research object and focuses on the CNT based operational vulnerability assessment.For the perspective of forced faults of a system,a mapping graph of operational states,i.e.temporal-spatial correlation graph,is constructed comprehensively considering topological,physical and operational features of the system.First,we employ the fault chain theory to construct the set of fault chains to reveal the adjacent fault relationships among branches.Furthermore,a mapping function is employ to map the set to the temporal-spatial correlation graph.The temporal-spatial correlation graph is a scale-free network by analyzing the topological properties of the graph based on the CNT.The scale-free properties of the graph demonstrate that the original physical network has some vulnerable branches.Once these vulnerable branches are deliberately attacked,the physical network has high vulnerability but it is robust under random attacks.We employ the constructed force-faults-based temporal-spatial correlation graph to construct the degree,in-degree and out-degree indices with weights.Then the indices are applied to identify the vulnerable branches.The validity of suggested vulnerable branches are verified by test systems.Furthermore,because the rankings of vulnerable branches change with the length of fault chains according to the indices,the change law of the rankings are concluded.Meanwhile,aspired by the law,a segmented temporal-spatial correlation graph is proposed to identify the vulnerable branches at the different stages of fault propagation.Moreover,because a branch's vulnerability depends on not only the importance of itself but also the adjacent fault relationships between it and other branches,we employ load-capacity model of the CNT to construct the dynamic vulnerable indices considering both importance of the branch itself and adjacent fault relationships among branches abstracted by the temporal-spatial correlation graph.The proposed method,verified using the test systems,demonstrates that adding the adjacent fault relationships to the indices can improve the accuracy of identification of vulnerable branches.In the process of fault propagation,some branches easily spread a fault to the system while others are easily effected by a propagated fault;therefore to distinguish the vulnerable characteristics of a branch,two vulnerable characteristics,the impactable and susceptible vulnerability,are proposed.According to the load-capacity model of the CNT,we construct the dynamic indices of two vulnerable features based on the temporal-spatial correlation graph of the forced faults.The simulations demonstrate that compared to the susceptible branches,the impactable branches have greater impacts on network function.To further reveal the effects of two typed of vulnerable branches on fault propagation,we propose a symmetric entropy to construct the impactable and susceptible temporal-spatial correlation graphs,respectively.By analyzing the topological properties of the two types of graphs,the impactable temporal-spatial correlation graph has scale-free properties while the susceptible graph has small-world properties.The scale-free properties of the impactable graph demonstrates that the impactable branches can easily trigger faults leading to the high vulnerability of the transmission network;therefore the impactable branches decide the depth of fault propagation.Meanwhile,the small-world properties of the susceptible graph demonstrates that the susceptible branches can be mutually affected during fault propagation;therefore the susceptible branches decide the breath of fault propagation.From the perspective of spontaneous faults of the system,we construct a cascading failure model considering the hidden faults of protection devices and then employ the mapping function to map the cascading failure model to a spontaneous-fault-based temporal-spatial correlation graph.Because the different fault branches have the different impacts on network function or structure during fault propagation,we define three types of weights of edges from the perspective of load shedding,topological structure and propagation possibility.According to the three types of weights,we investigate the propagation paths with high vulnerability and their spatial distribution features.Moreover,based on the temporal-spatial correlation graph,three types of betweenness,frequency betweenness,functional betweenness and structural betweenness are constructed to identify different types of vulnerable branches.The simulations verify the validity of the proposed indices.This dissertation create a complex-network-based technical framework to evaluate the operational vulnerability for the transmission network,which provides a valuable reference for the vulnerable analysis of the large-scale complex interconnected grid.