Abnormal Phenomena And Defence Of Cascading Failures In Complex Power Grids

Power system is becoming more and more complicated, with the operation of new generaters, transmission and distribution stations, and new transmission lines, as well as a growing number of gird-connection for photovoltaic power stations (PVs) and other new power generations. A minority of components’malfunction might cause more serious consequence than before. So it is necessary to analyze the mechanism of large blackout and the affects of PVs to power grids, and seek some solutions. The complex network theory is one of the most effective ways to study cascading failures in complex power grids. A large number of achievements, about cascading failures based on the complex network theory, show that the network robustness increases monotonically with its capacity (i.e., higher capacity is helpful for improving the robustness of complex networks). However, in the present work, it is found that the robustness non-monotonically varies with the capacity (i.e., the increase of capacity is not necessarily helpful for enhancing network robustness). The reasons for the non-monotonic phenomena are explored. And effective defense strategies of cascading failure in power grids are put forward. Furthermore, the effect on the power grids robustness of PV and the selection of the PV interconnection points are analyzed. The conclusion may be useful for the planning and designing of PV connected power grids. The main work is as follows,1) The homogeneity and the small world character of five IEEE power systems are analyzed, based on statistical characteristic indexes, such as degree, clustering coefficient, electrical distance and electrical betweenness. The results indicate that the IEEE-145power grid is more heterogeneous than others, and the IEEE-57, IEEE-118and IEEE-145power grids have the the small world character.2) The cascading failure model for complex power grid is put forword and the robustness of power grids to a single component malfunction is analyzed. The results indicate that the robustness of power grids may non-monotonically vary with the capacity. Here, we have divided the cascading failures into several sub-stages and have analyzed the number of overloaded nodes, the degree and electrical betweenness of overloaded nodes, and the average remaining load in each sub-stage. The results indicate that the increasing capacity protects several nodes at the beginning of malfunction, which lead to more loads remain in the power grid such that certain nodes cannot take the load and be removed subsequently, including certain nodes with many connections or large load. This eventually causes overloading of more nodes and a decline in the robustness of the power grid. These findings provide a hint on how to protect power grids during cascading failure by intentional removing some nodes and edges right after the initial malfunction.3) The intentional removals strategy (ⅠR), and the intentional islanding and removals strategy (Ⅱ&R) are put forward based on electrical betweenness, to defense the cascading failures in complex power grids. And the ⅠR, Ⅱ&R and the intentional islanding strategy (Ⅱ) are compared in the IEEE-39and IEEE-118power grids. The results indicate that all of the three strategies can reduce the load level at the beginning of malfunction, and thus suppress the spreading of cascading failure, improve the robustness eventually. Among them, Ⅱ&R is the best, and Ⅱ takes second place.4) The model of PV connected to complex power grid is built to analyze how the connection of a PV affects the robustness of power grids. The analysis results show that cascading failure may happen when a PV is connected to a power grid with relatively small capacity. And the robustness is closely related to the capacity, the PV output, and the interconnection point. Furthermore, we analyze the relationship between the characteristics of interconnection points and the smallest capacity that makes sure the system being safe when the PV output is injected into these points. The results indicate that the larger the electrical betweenness of the interconnection point is, the smaller the smallest capacity is. And the smallest capacity is relatively larger when the interconnection point is a transmission node. The smaller the capacity is, the lower the cost is. Consequently, the nodes with larger electrical betweenness should be a priority as PV interconnection points, and the generation nodes and distribution nodes may be rather than transmission ones.