An Efficient Method for Distribution System Restoration Based on Fuzzy Data Mining

This paper presents an efficient algorithm for distribution system restoration based on fuzzy data mining. The computation time can be decreased greatly by dividing the computation into two parts: one is the part of off-line calculation, in which by using the historical FTU measured records the selecting rules of optimal disconnection switches in all loops existed in distributed networks are calculated by fuzzy data mining. The other is the part of on-line calculation, in which just the tie switches are optimized by some optimization algorithms, and the corresponding optimal disconnection switches in each loop created by two or more closed tie switches can be estimated according to the selecting rules and the actual data measured by FTU. Simulation result indicates that the proposed method is of high calculation speed,it can guarantee the quality of solution at the same time.