The substation changes high-voltage power which is produced by the plant into low-voltage power and supplies it to the customer. If a fault occurs at a substation, there is a high probability of a spread fault, so it is necessary to quickly determine the fault and reduce the time to restoration from the fault. Now SOP(Standard Operation Procedure) is fault restoration method applied by the KEPCO provides switch operation and failure restoration procedures etc for 15 types faults. However, SOP is insufficient responses to other than 15 types, and errors may occur because of the operator determination of the fault through work knowledge and experience. Furthermore, traditional SOP can not fault recovery considering transformer capacities.
Therefore, in this paper, we propose an algorithm that can support fault restoration by using optimization techniques GA(Genetic Algorithm) of AI. And reducing possibility of human error, it complements the SOP of fault recovery taking into account the transformer spare capacity, builds automation base, and provide a GUI for convenient.
The proposed GA for supporting fault restoration of 154kV substations is a method of generating chromosomes, and then finding the optimal solution of the fault restoration path through evolution. First, the choice target is the 4Bank 154kV substation, and SOP are collected and analyzed to perform the pre-processing through the corresponding one-line diagram. Next, we design and realize this algorithm by selecting constraints, fitness functions, and chromosome generation. In addition, performance verification and simulation were performed while changing weights to determine the final fitness function. In other words, it was pre-processed by receiving information on the operation of CB, DS, and M.Tr of substation in SOP, and programmed to indicate the optimal fault restoration path for the remaining capacity on the fault situation of the transformer. Finally, it can be seen that the proposed GA provides an optimal switch path to ensure that power is supplied regardless of the fault location of the M.Tr.
In conclusion, this paper can be supply the fault restoration support for the transformer spare capacity that is not in the SOP. And can plan to support fault restoration and reduce the possibility of operator error, and make it possible to conveniently, quickly, and accurately determine the fault recovery situation.