The Study Of Current Reckoning Method And The Combination With The Innovation Graph Approach

In modern power dispatching system, the result of the state estimation is the important basis to determine scheduling methods, plays an important role. The innovation graph approach, which applies the graph theory, can make the result visually clear and has obtained a certain of research results. It can identify the state estimation unnormal events and even can distinguish multiple abnormal events when they overlap with each other. There are more and more new energy resources connecting to grid. The solar energy and wind energy have the disadvantages of randomness and volatility, which influence the accuracy of power prediction, and then increase the difficulty of state estimation based on the forecast. The dissertation studies the estimation method for the grid loading new energy, which can get the running state of the real system more effectively, and provides basic data to the application of power system analysis and economic operation.Firstly, put forward the current reckoning method. The method adopts the graph theory on the running network to slove identification defects of the innovation graph approach in the new situation of power grid. Adopt the link branch reckoning technology can get the reckoning current set, which compares with the measured current. When the two sets are different, the difference of them, which is called current diffrence, can display particular features, and the features can be used to detect the bad data. And further more, study the influence of interacting bad data to the current reckoning method. Current reckoning method expresses the injection as branch current form, can detect and identify the injection bad data. Exclude the bad data of the measurement set and get the modified current set, which can be the information to distinguish topology changes. Recognize the bad data and topology changes orderly when the two events coexist. The method can realize the detection and identification of unnormal situation of state estimation. The example adopts the standard system.Secondly, combine innovation graph approph and current reckonning method to get complementary advantages, which is called combined identification method. The first combined manner adopts different main identification method according to the different identified events, that is, distinguish bad data relies the current reckoning method mainly, and the innovation graph approph plays a greater role in recognizing topology changes. The second manner adopts partitioned mode, along the branch longitudinal direction, which realizes the purpose of without reucing circuit numbers and keep the identification ability of the link branch reckoning technology. Then, choose estimation method appropritely based to the practical needs of subareas. The analyse example adopts the standard system.Lastly, study the network observability of current reckoning method accordin g to the innovation graph approph observation analysis. To the practicability of the current reckoning method, study the identification and locating of bad data and branch quitting operation on the double circuit transmission lines, the analyse example adopts the standard system. Adopt the example of the partial practical system of one province to verify the effectiveness of the new state estimation method.