Research On Fault Location And Power Restoration Of Distribution Network Under The Conditions Of Distributed Generation

When permanent fault occurrs in the distribution network with distributed generations, a fast and effective method to achieve the fault location and power restoration of power outage areas has an important significance to decrease the power outage loss and improve the reliability of power supply. This paper studies on islanding detection, fault location and network restoration and make the improvement and innovation on the existing algorithm to improve the reliability of distribution network. The main research is as follows:A method based on negative sequence current injection and voltage imbalance on the point of common coupling is proposed. The reasonable switching strategy of negative sequence current in the grid-connected inverter reduces the impact of injected current on the main power grid. The method, without non-detection zone, is also unaffected by asymmetric faults of the power grid. Case study shows that the method could detect islanding state accurately and quickly under the worst condition while the power of the PV matches the load power, and the voltage and the frequency of the PCC has small shift.A smallest error based fault location algorithm is proposed. First, the fault current is injected to each bus of the distribution network model and then the voltage variation value of the power source is calculated. Second, calculate the error between voltage variation measure value and calculation value and then search the fault current injection bus corresponding to the minimum error to determine the bus of the fault section. Finally, determine another bus of the fault section to complete fault section location. The algorithm have clear physical meaning, the calculation process is simple and has good fault location accuracy. The more distributed generation is connected to the distribution network the better fault location result may get.A multi-objective intelligent power restoration algorithm that considered islanding operation is proposed. It uses binary particle swarm algorithm to search Pareto non-dominated solutions to achieve the power restoration which is based on the analysis of the distributed system’s fault recovery process. According to different emphases of switching frequency and network losses this paper’s objectives are minimum load loss, switching frequency and minimum load loss, grid loss to obtain Pareto non-dominated solutions to reduce the number of Pareto non-dominated solutions. Power restoration is under the premise of minimizing loss of load to minimize the switching frequency and network losses. Method provides options for users to choose flexibly according to the situation.