Research On Fault Location Of Distribution Network Based On Double Criteria In Petri Network

With the expansion of power grid scale, the distribution management and system has been difficult to adapt to the scale of the power grid. At the same time, the fault location technology of power distribution system is also faced with many problems,such as weak generality, long operation time and poor fault tolerance. Therefore, the study of good performance fault location for distribution network is the basis to solve the above problems and it is the main research work of this thesis.In order to adapt to the scale of development of distribution network and improve the operation speed, fault tolerance and versatility of the distribution network fault location method, this thesis presents a fault location Petri net model of distribution system based on double fault criterion parallel reasoning. This method is based on two criteria, which is collected by FTU, from SCADA. The fault tolerance is improved by double fault criterion and probability information measure. Then,based on this, a fault location Petri net model is established. Combined with the probability information, it can improve the fault tolerance of the method.In this thesis, firstly, according to the method of decoupling and partitioning of distribution network topology, the complex distribution network is decoupled. Then it analyzes the selection basis and characteristics of the two kinds of criteria and establishes a generic Petri net model unit for fault location, on this basis. According to the set of network topology which is decoupled and partitioned, the fault location model of the target topology is obtained by superposition of the universal model unit.The construction process of distribution network fault location model is simplified.Finally, in order to verify the effectiveness of the method, this thesis takes the typical connection mode of distribution network in our country as an example,the simulation and verification of the method are carried out. It can be learnt from the example and simulation result that the method can construct an adaptive fault location Petri net model for a variety of topology of the distribution network. It can overcome some difficulties and miscarriage of justice, which caused by fault information and inconspicuous failure characteristics.