Study On The Algorithm For Feeder Section Fault Location In Distribution Network

Distribution network is the core area of the power system that is directly facing the power users, The quality of power supply in distribution network is related to not only the quality of people’s daily life but also the ability to carry out normal social production activities.When a fault occurs in distribution network,the efficiency of fault location is the prerequisite for isolating the fault section, reducing the scope of power outages and restoring clients’electricity supply in time.As the method for fault location in distribution network is been invented one after another,the method is correct or not can directly affect the efficiency of fault location.Only if the method can meet the needs of these following three conditions:fast locating the fault sections, suitable for every topology, having a certain fault tolerance,then the method can ensure the efficiency and practicality of fault location.Traditional matrix algorithm operations for fault location in distribution network are complicated. When part of fault information is missing or fault information distortion,the incorrect information will cause the fault location error. This paper presents a new algorithm for fault location in distribution network,which is based on the principle of net-shaped matrix.The algorithm can accurately locate the fault sections in complex grid for not only single fault or multiple faults,but also feeder ends faults.Meanwhile,the method is quite fit for the changing complicated topology distribution network fault location.It uses the direction of trend as a positive direction while the network is in normal operation, which can make the process of fault location much easier.Although lack of swich information from the FTU or the information becomes distort, it can still fix the problems.The simulation result shows that the algorithm reduces the location’s searching times effectively,and it has a good ability in fault tolerance.Traditional fault location methods are changing the multi-objective optimization problems into single-objective optimization problems to solve them by adding the weighted values,but this will cause the algorithm over-reliance the weight values.Moreover,the weight values obtained by subjective decisions sometimes are not accurate.In this paper,a multi-objective fault location model is built to solve the fault location problems in distribution network.By using the NSGA-Ⅱ algorithm based on Pareto optimization theory,the method can avoid the affect to the fault location results caused by weight values.There is an improve to NSGA-Ⅱ algorithm which can enhance the efficiency of the algorithm too. Finally,an optimal set processing method that fit for distribution network is put forward to filter out the best solution from optimal solution set which is the best match to the actual situation. The simulation results show that the imporved NSGA-Ⅱ algorithm based on Pareto optimization theory can locate the fault in distribution network in every complicated situation,and the algorithm also has a significant advantage in terms of locating efficiency,locating universality and fault tolerance.