Research On Fault Recovery Of 750kV Substation

Northwest 750 k V interconnection project is an extra high voltage transmission project which is the largest project with the highest voltage level, the longest line and the maximum complexity in the developing history of northwest power grid. As the hub of 750 kV power grid, 750 k V substation contains three voltage levels which are 750 k V, 330 k V and 660 kV respectively. And the double protection configuration is used, which makes the network topology of 750 k V substation more complex. When substation goes wrong, operators need to restore the power supply of the failure areas quickly and accurately. Therefore, the research on substation fault recovery is highly significant to the safety and stable operation of power network.This thesis takes the 750 kV substation as the research object. And according to national relevant operating procedures, a reasonable strategy for substation fault recovery is formulated. Through the topology connection structure of substation main wiring and fault diagnosis, breadth-first search algorithm is used to search non-fault power failure area and fault areas are then isolated. Petri net and genetic algorithm fault recovery methods are used.Firstly, aimed at voltage level characteristics of 750 k V substation and connection characteristics of the main wiring topology, substation is partition processed. Graph theory knowledge is used to establish substation network topology connection graph and the dates is storage by adjacency matrix. When the substation goes wrong, the fault features are recorded by constructing substation fault information matrix and fault judgement matrix, then breadth-first search algorithm is used to search the power failure area of substation and provide the basis for the substation fault recovery.Next, aimed at the fault of bus, line and transformer which may probably occurs in 750 k V substation, two steps are adopted: insulate the fault point and restore power supply. Firstly, the basic module is built that is switches and circuit switches opening and closing which is used when we build lines and buses models of Petri net. Then Petri net models of fault recovery of transformers is given according to the way and the max amount of transformers. What's more, different conditions of Petri net models of buses and lines are discussed. Finally, we find the transitions of be fired depend on the reasoning rules of Petri net,and give the order of recovery of kinds of faults.Finally, according to national relevant operating procedures, we analyze and classify various operating states of the primary equipment of 750 k V substation, then establish the operation cost matrix to represent the change of the primary equipment condition caused by the failure, and lastly generate the fault recovery matrix combining with the actual connection structure of substation. On this basis, considering minimizing the non-fault power failure region and the operating cost of primary device, the multi objective optimization model is built. After that, the genetic algorithm is used to solve the model and complete the substation fault recovery operation.