Study On Power Grid Disturbance Identification Of AC/DC Parallel System

With the development of China's power system, the regional power systems will be interconnected gradually. For the long distance bulk power transmission from the West to the East, DC transmission scheme is widely used of its own advantages. However, as the dynamic behaviors of AC/DC parallel system are much more complex, disturbance in such a large system may also spread to a wide area rapidly and even cause blackout. Therefore, disturbance identification is necessary for the security and reliability of the system. In this dissertation, studies focus on DC single-pole block. Comparing with the features of some typical AC disturbances (short circuit, loss of generation and load shedding), a proper criterion for disturbance identification is presented. And also, Decision Tree of data mining approach is proposed to generate a set of classification rules by adopting disturbance database. Finally, with the simulation results on a 2010 planning grid, the proposed disturbances identification method is proved to be technically feasible.