Research On Fault Segment Location Method Of Distribution Network With Distributed Generation

With a large number of distributed generation such as wind energy and solar energy have been connecting to the distribution network,the structure of distribution network has changed from a traditional radial type to a multi-power supply system,and the power flow of the distribution network system has also changed from unilateral power flow to bilateral power flow.Traditional relay protection schemes have difficulties to accurately determining the fault area.Therefore,study fault segment location method of distribution network with distributed generation is pressing.This paper analyzes in detail the specific classification of the distribution network fault location methods,the advantages and disadvantages of various location algorithms,focusing on the matrix algorithm and quantum genetic algorithm,and proposes a method for locating fault sections in distribution networks with distributed generations based on improved matrix algorithm and a method for locating fault sections in distribution networks with distributed generations based on improved quantum genetic algorithmsFirst of all,this paper analyzes and explains the classification of the current common segment location methods,and elaborates the advantages and disadvantages of hot arc search method,matrix algorithm,genetic algorithm,particle swarm algorithm,immune algorithm,and ant colony algorithm in the application of current distribution network fault section location in detail.This paper also analyzes the influence of distributed generations which are accessed to the distribution network on the protection of the traditional distribution network.In the second place,this paper analyzes the shortcomings of the direct algorithm for fault location of the distribution network.When the fault information is distorted,the fault area cannot be accurately located and when the fault occurs in the T-connected busbar area,the fault area cannot be accurately located.The original matrix algorithm is improved.The improved algorithm uses the real-time power flow direction to automatically update the network description matrix for the distribution network with distributed generation.The fault judgment matrix can be generated only by subtraction,which greatly reduces the complexity of matrix calculations and shortens the fault judgment time.Using the SIMULINK platform to build a simulation model to simulate the occurrence of busbar faults,load outgoing faults,line area faults and missing fault information in the distribution network.The simulation results verify that the algorithm can accurately and effectively locate the T-connected busbar area and has strong fault tolerance in the face of missing fault information.At last,this paper explains that the existing algorithms of the indirect algorithm for fault section location in the distribution network are easy to fall into the local optimum,which leads to inaccurate fault location,and the fault area cannot be accurately located when the fault information is distorted.The original quantum genetic algorithm is improved.The improved algorithm introduces the gradient descent method in the quantum rotation angle setting process,and combines the objective function with the quantum rotation angle,so that the new quantum rotation angle control strategy can adaptively change the size and direction of the quantum rotation angle.After the fault,according to the changes in the operation mode of the distribution network system,considering the leading role of the system power supply,new switch and feeder functions are constructed.Optimizing the original objective function,this algorithm makes the failure information false alarm as a random event consistent with the probability of occurrence of the fault information false alarm,and has nothing to do with the specific type of distortion at the same time.Use MATLAB software to build a simulation model,and simulate the occurrence of single faults,multiple faults and fault information distortion in the distribution network.The simulation results show that the algorithm can effectively locate the fault area and can well jump out of the local optimum.It has strong fault tolerance in the face of the lack of fault information.This paper compares the classical quantum genetic algorithm,the quantum ternary genetic algorithm and the improved quantum genetic algorithm,and draws a conclusion that the improved quantum genetic algorithm has fast calculation speed,strong convergence ability,and strong fault tolerance and applicability.