Research On Identification Strategy Of Abrupt Power Change Of Provincial Power Grid

With the development of new energy and smart grid technologies,higher requirements have been placed on the monitoring of power system dispatch centers.Power system state estimation plays an important role as an important data processing link.The state estimation of the new information graph has the characteristics of low data redundancy,fast calculation speed,and good identification effect.The sudden cut of the generator and the access of new energy sources such as wind power make the power generation fluctuations larger.In order to further improve the effect of state estimation to identify the power mutation of the generator,this paper based on the identification theory of load mutation in the new information graph theory,by analyzing the power of the generator When the mutation occurs,the characteristics of the branch's new spread in the new spread chart,and the corresponding identification strategy is proposed.To supplement the theory of load mutation identification,the main research work is as follows:The identification strategy of generator power mutation is proposed.Aiming at the feature that the generator mutation node has the largest injection of new information,a method of searching for a path based on the branch's new interest difference is proposed from the generator power mutation node.For the sudden change of generator power,there is no corresponding load power change,and the end node cannot be determined by the power change of the load node.Moreover,with the feature that the absolute value of branch new spreads gradually decreases and tends to zero as the catastrophe network continues to extend,a method of setting branch new spread thresholds is proposed.After identifying the generator power mutation network,according to the characteristics of the new spread difference of the mutation path in the generator power mutation network,the new interest spread of each branch is corrected one by one.Studied the problem of identification of generator power mutation when there is bad data.By setting tree branch branch bad data and continuous branch bad data,analyze the impact of bad data on generator power mutation identification,and according to the characteristics of bad data in generator mutation network.Combine the new information graph theory to solve the identification problem when bad data and generator power mutation coexist.Studied the problem of identification of generator power mutation when there is topology change.Aiming at the situation that the topology change and generator power mutation in the province's power grid instance are not closely overlapped,two abnormal events are simultaneously identified according to the new information graph identification theory.The IEEE-30 node system simulates the situation where the two closely overlap.According to the information injected into the node,the generator power mutation node is searched,and the mutation path is searched based on the characteristics of the generator power mutation network to indirectly determine the generator power mutation.Revise the continuous support calculation in the generator power mutation network,identify the topology change,and realize the identification of the two.The simulation system model for simulating the power mutation of the actual system generator is established.The method of using the power flow to calculate the simulated forecast value in the original model is improved,and the state estimation is used to calculate the simulated forecast value,which can better simulate the sudden change of the generator power of the actual system.The simulation model can be used to study the identification of generator power mutation and the occurrence of overlapping of generator power mutation and bad data topology changes.The IEEE-30 node system is used to simulate the situation of generator power mutation,generator power mutation and topological change close overlapping.The calculation results verify the effectiveness of the generator power mutation identification strategy.