Fault Section Location In Power Distribution Network Based On Artificial Physics Algorithm And Rough Set Theory

Fault location is one of the research hot spots of distribution network, and its accuracy and positioning speed not only affect the production and living of the people, but also become one of the most important indicators of power supply enterprises. According to the existing automation level of distribution network, the distribution network is divided into the complete automation network and the incomplete automation network. We studied the difference between their main source of information failures, and took the appropriate ways to accomplish the fault location of the power distribution network.For the complete automation network, the distribution terminal device(FTU, RTU) installed at disconnect switch can collect and convey the real-time information, so a novel intelligent algorithm(artificial physics algorithm) is proposed to carry on the research of fault location. When the power distribution network fault occurs, the current of the disconnect switch will have a greater change, and can be detected by the terminal equipments, so the current information is used to abstract the fault location of the distribution network into a binary search process. We use chaotic mapping and reverse initialization to improve the artificial physics algorithm, and apply it to the distribution network fault location. The simulation results show that the improved artificial physics algorithm has better search ability and convergence precision, and it has better fault tolerance and accuracy in fault location of the power distribution network.For the incomplete automation network, only part of the disconnect switch have the switch alarm information and the fault indicator information, and the marketing 95598 system will receive the phone of outage users at the same time, so the fault location of distribution network is studied based on multi-source information. In this paper, we make use of the repair information and the switch alarm information to form a decision table based on rough set theory, and reduce redundancy to reduce the storage space in order to achieve the rapid location of the distribution network fault. In addition, considering smart meters are provided in the marketing system from positioning to the transformer power supply area, we adopt the measure of calling the voltage of smart meters to locate fault accurately. The simulation results show that the proposed scheme can eliminate information blind area, improve the positioning accuracy, and the calling of smart meters can locate the fault to the building or even users accurately.Finally, the MATLAB / GUI is used to design a simulated fault location software which integrates the proposed schemes in this paper. It can provide a better reference for related personnel.