| With the formation and expansion of the new AC/DC hybrid grid pattern in China,and the interlocking faults of AC/DC transmission system have become the fault normal under the new pattern.In this paper,based on the analysis of the evolution mechanism of AC/DC transmission system interlocking faults,the second line of defense of power system against interlocking faults is strengthened.Firstly,the interlocking fault of AC/DC transmission system is analyzed.Based on software PSCAD/EMTDC,based on three modules of IEEE39 node system,CIGRE(International Council on Large Electric systems)HVDC test model,and the control and protection system of the HVDC project in operation,the model of the AC/HVDC transmission system is built.Then,based on the model,multiple DC simultaneous commutation failure links in the interlocking fault chains of AC/DC transmission systems are explored.Finally,the simulation analysis of power flow transfer after interlocking faults is carried out on the model of AC/DC transmission system.Secondly,based on Bayesian probability,a Bayesian network for interlocking faults of AC/DC transmission system is established by considering the key fault nodes.The risk index of key fault nodes in the network nodes is put forward.The assessment level of risk index is determined according to the prescribed value of three lines of defense in power system.Finally,the Bayesian network of interlocking faults of AC/DC transmission system is visualized by using the simulation software Genie.In the GENIE simulation,the interlocking fault of AC/DC transmission system is set to occur 100%,and the occurrence probability of each fault node is backward pushed back.The occurrence probability of interlocking faults in AC/DC transmission system is analyzed qualitatively and quantitatively,and the fault nodes with high occurrence probability should be paid special attention and improved in the early construction and later operation of the power network,so as to improve the safe and stable operation of the AC/DC transmission system.Finally,a real-time online monitoring and early warning system based on swarm intelligent optimization algorithm(slime mold algorithm)and support vector machine regression machine learning method is proposed for the interlocking faults of AC/DC transmission system.Firstly,the two regression prediction methods and their accuracy are compared.Then,the slime mold intelligent optimization algorithm was used to optimize the algorithm with high precision.The early warning method of interlocking fault of AC/DC transmission system is put forward.Finally,the validity of the proposed early warning system is verified on the model of AC/DC transmission system. |
