Research On Fault Location And Power Supply Restoration In Distribution Network

Fault location and recovery in distribution network is one of the main functions of distribution automation and the basis of improving the reliability of distribution network power supply.The medium-voltage(MV)and low-voltage(LV)distribution network is oriented to end users,with complex topology,harsh operating environment and high failure probability.The accuracy of fault location and the effectiveness of supply restoration are directly related to the stability and reliability of power supply and the satisfaction of users to power companies.Aiming at the problems of fault location and recovery,this paper studies from five aspects: phase to phase fault location,single-phase grounding fault location,power supply recovery strategy,topology identification and fault location of LV distribution network,and intelligent terminal unit design.The main research work and innovative achievements of this paper are summarized as follows:1)This dissertation proposes a fault location method for active distribution network(ADN)based on FTU and distribution transformer alarm information.Aiming at the problem that single fault information source is insufficient in accuracy and integrity,the fault location method for ADN based on multi-source information fusion is proposed.Firstly,considering the different characteristics of FTU alarm and distribution transformer alarm information after fault,the correlation matrix corresponding to different alarm information reflecting the relationship between fault section and alarm information is dynamically generated according to the ADN’s topology,and a new switching function is constructed based on this,which effectively avoids the tedious logic operation.Then,according to the actual alarm information received by the two information sources,two optimization models are established respectively,and the genetic algorithm is used to calculate the fault probability of each section under the corresponding information sources.Finally,the results of fault location by the two information sources are fused by using the evidence fusion theory,and the final fault location results are obtained.The simulation results of the modified ieee-33 bus distribution network show that the proposed method can accurately locate single point faults and multiple faults,and has high fault tolerance.2)This dissertation proposes a method of single-phase grounding fault line selection and location based on phase reactive power fault component(RPFC).Firstly,in order to solve the problems of low accuracy of small current grounding fault location and poor sensitivity of high resistance fault detection,the phase RPFC is defined and its characteristics under small current grounding fault are analyzed.Then,the difference coefficient of phase RPFC is defined to measure the difference between three-phase RPFC of feeder,and its nature is analyzed.For normal feeder and downstream feeder section of fault point,the difference coefficient of phase RPFC is a positive number close to 0;whle for the upstream feeder section at the fault point,the phase difference coefficient of phase RPFC is approximately-1.Finally,the fault diagnosis is carried out by using the characteristics of the large discrimination of the difference coefficient in the above two cases,and the implementation scheme of fault line selection and location is given.The feeder terminal unit developed based on this method has passed the digital simulation test,the digital analog test and the physical model test,and has been used in the actual distribution network,where the accuracy of fault line selection and location is improved.3)This dissertation proposes a mixed integer programming model for ADN fault reconfiguration and recovery.The model aims to minimize the sum of user outage loss,switch operation aging cost,distributed generation and energy storage device discharge aging costs.Virtual 0 nodes and virtual branches are introduced to ensure the maximum connectivity and radiation of the independent power supply area.Through the linearization capacity constraint of the square term inscribed polygon,the load boundary switch set is defined to determine the lower limit of the load outage time,and then the fault reconstruction and recovery scheme is obtained.Based on IEEE 33 bus ADN and PG&E 69 bus ADN,a fault recovery example is formed.The results verify the effectiveness of the proposed model and algorithm.4)This dissertation proposes two topology identification methods and a fault location method for LV distribution network(LVDN).Firstly,a topology identification and monitoring method for LVDN based on multi granularity clustering and multivariate feature statistics is proposed.In this method,the data of intelligent electricity meters are used to identify topology by two stages.In the first stage,multi granularity clustering is used to identify and verify the transformer relationship and phase relationship of all relevant users in LVDNs.In the second stage,the dynamic topology changes are monitored in real time through multivariate feature statistical analysis.Secondly,a topology identification method based on the signal injected by the LV monitoring unit is proposed.In this method,the average current amplitude is used as the sequence basis for the LV monitoring unit to inject signals,and the loop matrix is used to judge whether the topology identification is completed,which reduces the times of signal injection and the impact of the injected signals on the LVDN.Finally,an optimization model and algorithm of fault location in LVDN is proposed,where the overcurrent signals captured by the LV monitoring unit is used.5)The intelligent feeder terminal unit(IFTU)and the integration transformer terminal unit(ITTU)are developed and applied to MV and LV distribution networks.The IFTU embeds the small current grounding fault protection function,adopts capacitive voltage sensor with excellent high-frequency transformation characteristics and current sensor with wide range and high precision,meets the requirements of signal measurement bandwidth and precision for grounding fault detection.The IFTU integrates with the circuit breaker to form an intelligent switchgear,which has been mass produced and put into operation.The ITTU developed with Internet of things technology adopts platform hardware design and distributed edge computing architecture,and has the functions of station area data local collection,local analysis and processing,which can improve the intelligent monitoring ability and lean management level of LV station area.The research of this paper can provide practical solutions and automation equipment for fault location and restoration in MV and LV distribution network,and promotes the development of fault analysis and handling technology.