GIS Based Stability Monitoring And Topological Analysis On Cascading Failure Of Power Grid

Electric power is the basic industry of national economic development. The development of electric power directly affects the development of national economy. However, burning coal is still the main way to generate power. The rapid development of electric industry has caused great harm to environment. For example, the shortage of non-renewable energe and the emission of lots of harmful gas which will have bad effect on human's healthy. To resolve this conflict, traditional power grid should be reformed to smart power grid. Smart power grid allows the access to large-scale clean energy and achieves the optimal allocation of power resources, so the power grid will be more economical, efficient and environmentally friendly. But the complex structure of smart power grid challenges the security of the grid.This paper studies two security issues in power grid based on GIS. The one is the stability monitoring of power grid and the other one is cascading failure. The former focuses on the optimal placement of PMU(OPP). PMU is the core monitoring device of wide area measurement system. To make the power grid full observability with placing the least number of PMUs, the paper proposes a hybrid algorithm which combines genetic algorithm(GA) and minimum spanning tree(MST) algorithm to solve OPP problem. In particular, the GA adopted in this paper has improved the mutation operator by taking the topology of grid into consideration. The latter mainly discusses two aspects. One is studying the evolution of cascading failure and analyzing the structure vulnerability of grid, while the other one is putting forward mitigation strategy for cascading failure. In the paper, two kinds of topology model base on complex theory have been proposed to study cascading failure. They are unrecoverable model and recoverable model. Based on these two models, we can study the evolution of cascading failure and make analysis for power grid's structural vulnerability. The mitigation strategy is inspired by load-transfer schemes in real-world network. This paper proposed the mitigation model for cascading failure based on unrecoverable model.The paper presents example analysis and comparison analysis of the hybrid algorithm. The example analysis adopts two provincial power grids to simulate. One is 57-node power grid and the other one is 135-node power grid. The simulation results show that the 57-node power grid needs to place 9 PMUs to make the power grid full observability while the 135-node power grid needs 17 PMUs. We conduct experiments for the same two grids used MST algorithm and the results show that the 57-node needs 9 PMUs and 135-node needs 18 PMUs. The comparison analysis adopts IEEE 30-bus and IEEE 39-bus power grid to simulate. We discuss three other algorithms and compare them with the hybrid algorithm proposed by this paper. The simulation results demonstrate the h ybrid algorithm has advantages over the other three in quality and variety of optimum solutions.The simulation experiments of the two topology models of cascading failure are based on the two provincial power grids mentioned above. The result of unrecoverable model indicates that the required redundancy(RED) is good at measuring the importance of nodes. The relationship diagram of RED and initial load shows that load-based attack strategy is not that efficient in unrecoverable model. In recoverable model, the difference between the initial network efficiency and the final network efficiency is used to measure one node's importance and simulation result shows the load-based strategy is still efficient in this model. In order to verify the effectiveness of the mitigation strategy for cascading failure, we conduct experiments based on unrecoverable model and mitigation model for cascading failure and we find the average percentage of failure in mitigation model is obviously lower than the recoverable model.This paper has two innovation points. The first one is proposing a hybrid algorithm which combines GA and MST algorithm to solve OPP problem. In this algorithm, we take the topology of power grid into consideration. The second one is designing a visual simulation analytical system based on GIS to show the configuration of PMU and the evolution process of cascading failure.