Research On Fault Location For Distribution Network Based On Self-adapting Artificial Bee Colony Algorithm

In recent years,the research for fault location applied in transmission network has been relatively mature.However,due to the difference between distribution network with transmission network,such as much more node numbers,more discrete load distribution,more complicated structure,and shorter line length,smaller fault current,all these lead to the failure when the same methods are directly applied to distribution network.Moreover,the introduction of distributed power also increases the difficulty of fault location for the distribution network.On this basis,this article has done the following research work on the fault location in distribution network with DGs:In modeling and location analysis,most traditional fault location methods do not consider about the existence of practical nonlinear model in distribution network,such as the nonlinearity of load and grounding resistor.This makes the adopted model can't approach the practical network,so the location results will lose practical meaning.Aiming at the question above,this thesis firstly proposes a fault area location method for distribution network with DG which takes account of the nonlinearity,the load adopts quadric polynomial model,the grounding resistor adopts the experience formula model,the nonlinear equation constructed by nodal voltage equation will be solved out through quasi-Newton algorithm,then the fault area can be located according to the definition of fault feature value.After obtaining the fault area result,this thesis then proposes a precise fault location method based on self-adapting artificial bee colony algorithm.The solution problem of precise fault location is transformed into the problem of the optimal solution of the fitness function.The fault feature of some position is set as the fitness function,then self-adapting artificial bee colony algorithm is adopted to obtain the precise location.The simulation results in 10 kV distribution network model of the SINPOPEC West-Northern oil field suggest the validity of the proposed fault location method.Lastly,in order to further verify the accuracy of the proposed method and to combine with the practical engineering in future,the software has been written based on COMTRADE standard recorded fault files to realize complete fault diagnosis in this thesis.This software simply achieves the read,display and analysis of the recorded fault data,the identification of fault time,fault type and fault line section.It can be adopted as the follow-up main tool to verify the algorithm in spot environment.