Security analysis of large power grid based on complex network theory and Bayesian network model

Interconnection of large area, E/UHV electric transmission are the main solutions to solve the imbalance between energy resources and productivity in China. Due to that the transmission lines are often exposed in the outdoor environment, their reliability is influenced by multiple factors. Therefore, accurate evaluation of transmission status is crucial to the security of the grid. Traditional statistic based grid reliability analyses largely depend on the historical data, and has shortcomings in the time domain, since it does not reflect the changes of grid state and the surrounding environment. While the event-triggered analyses method is able to quantize the role of certain transmission line within the system, it still has the N-K analyses bottleneck. Hence, in this paper, we consider the information from the surrounding environment and the system internal operational status, and propose a novel status evaluation model of electricity transmission line based on Bayesian network and complex network theory. Bayesian network is used to quantitatively analyze the possibility of faults under various external conditions; the system geometric parameter from complex network theory is introduced to evaluate the different contributions of the transmission lines. Simulation result shows that the proposed model has good capability of identifying the high-risk lines, which play important roles in the system. Comparing with previous status evaluation methods which only using the information from surround environment, the accuracy for identifying critical lines is improved as it covers the system internal operational information. The method proposed requires low computational complexity and can be referenced in the maintenance work of large power grid.