| In the power system,the distribution network is the last link to protect the vital infrastructure of the people's livelihood and serve the residents.The safe and stable operation is directly related to the electricity consumption,power quality and reliability of each user.However,in the daily operation and maintenance process of distribution network,in order to reduce network loss,balance load and flexible and reliable power supply,it is necessary to constantly adjust the dis tribution network topology structure.As a result,the topology connection informati on in the distribution network GIS system of power company can not be updated in real time,which causes many errors,and it has been unable to solve effectively.With the development of smart grid strategy and the improvement of power system measurement technology,it provides a new idea for distribution network topology identification and makes it possible to accurately estimate the impedance parameters of distribution lines.Therefore,this paper presents a method of topology identification and parameter estimation based on measurement data analysis of distribution network.Firstly,aiming at the topological chaos of medium voltage distribution network,a method for identifying the topology structure of medium voltage distribution network based on spectral clustering and maximum spanning tree algorithm is proposed.The method is based on the theoretical similarity of the distribution transformer voltage on the same feeder.For the asymmetry of the transformer outlet voltage caused by the three-phase load imbalance,the reduction method of the lowvoltage outlet side of the distribution transformer is proposed.The characteristic data of each distribution transformer is extracted,and the spectral relationship analysis of the distribution network is realized by spectral clustering analysis.On the basis of this,the mutual information value is used as the similarity measure between different nodes,and the maximum spanning tree algorithm is used to reconstruct the topology of the distribution network.The method is validated in both the measured data of the distribution network and the IEEE 33 node power distribution system.Then,based on the identification of the topological structure of medium-voltage distribution network,a parameter estimation method of m edium-voltage distribution network based on sequential quadratic programming algorithm is proposed.Different from previous studies,this paper chooses a more accurate mat hematical model of line parameters and considers line losses.Using the measured data of distribution transformers,the voltage matrix of upstream nodes is deduced.According to the two scenarios that upstream nodes are not measurable and measurable,the o ptimal model of minimum voltage variance of upstream nodes is established,and the pa rameters of power lines in medium-voltage distribution network are estimated.The problem is transformed into an optimization problem.Aiming at this non-linear optimization problem,a sequential quadratic programming(SQP)method is proposed to solve the problem.Finally,the resistance and reactance values of power lines are obtained.At last,the distribution network topology identification platform based on measurement data analysis is introduced.According to the specific requirements of the platform,the application architecture,data architecture and communication architecture are designed in detail.It has many functions such as data acquisition,storage,calculation and analysis,visualization and so on.The platform adopts the structure of front-end and back-end separation,which facilitates the function expansion and call embedding of the platform.It lays a solid foundation for power companies to carry out advanced functions such as distribution network state estimation,power flow calculation,load forecasting,network reconfiguration and so on.The practicability of this platform is proved by showing a specific case of distribution network topology identification. |
PDF Full Text Request