Coherent Generator Groups Recognizing Study Based On Fuzzy Clustering Theory

Research of recognizing coherent generator groups in power system is important to dynamic equivalence, transient stability calculation and provides essential data for improving system's behavior. Coherent generator groups recognition is a relatively old research problem and has numerous methods. With the fast development of computer technology and the improving requirement of the power system's stability, the more deeper research work has been done, therefore it has more reality meanings. In this paper, according to the fussy and uncertainty of the coherent generator groups dividing, fussy mathematics theory has been used to recognizing coherent generator groups in power system.Firstly, the basic conception, theory and method of fuzzy clustering has been imported. And the modeling method based on fuzzy clustering according to the fuzzy of recognizing coherent generator groups has been put forward.Secondly, two math models based on the fuzzy clustering in recognizing coherent generator groups has been deduced. The models based on the homology degree in generators account for the factors such as system construction, fault time, fault location. The merits and flaws of the two models has been compared through the case simulation.Finally, further research has been made to the transitive closure method in recognizing coherent generator groups based on fuzzy clustering theory and algorithms based on fuzzy ISODATA. Much simulation experiment has been made to choose the controlling parameters in different algorithms. Thus some experienced values of the optimized parameters have been given. The algorithm has been improved according to the features of recognizing coherent generator groups, therefore it is much more adaptive to engineering apply. The algorithm has been proved to be efficiency by using the case simulation test.