Research On State Estimation And False Dataattack Detection Of Smartdistribution System

With the high-permeability distributed generations intermittent connected,large-scale electric vehicles random charging and the measurement errors in large number of intelligent measurement equipment,the fluctuation and randomness of distribution system are greatly increased,which brings difficulties to accurately perceive the operation state of distribution system.At the same time,the intellectualization of distribution system makes data and information flow bidirectionally,which lead to system suffering from the threat of false data attacks.It is difficult to characterize all the information and potential risks of distribution system only by traditional deterministic state estimation.So,it brings many difficulties and challenges to the monitoring and control of the distribution system.In order to meet the demand of monitoring and management,this paper studies the state estimation algorithm based on the modeling and analysis of distribution network uncertainties.From the point of view of defenders,a defense strategy using interval state estimation to detect data attacks is proposed.The main research contents are as follows:(1)An interval pseudo-measurement data modeling method considering the uncertainties of distribution system is proposed.Three typical nodes: load node,wind power node and photovoltaic node,are selected.According to their history power data,the uncertainties are mined by wavelet neural network.And the interval pseudo-measurement data are generated to improve the redundancy,observability and estimation accuracy of distribution system.(2)The interval state estimation algorithm in distribution system is studied.The theoretical basis of interval modeling for distribution network parameters,distributed generations and measurement data is given.On the basis of traditional distribution system state estimation,a three-phase interval state estimation model which is more in line with the actual state of distribution system is established.Aiming at the conservatism problem in the process of solving,a new method using IGE solution as initial value and Krawczyk operator as iterative optimization algorithm is proposed.Case studies show that under the same conditions,the results of interval state estimation based on Krawczyk operator are much more accurate than those based on IGE algorithm,which provides more reliable and valuable boundary information for the power system control center.(3)A data attack detection mechanism based on interval state estimation is proposed.Firstly,from the attacker's point of view,the basic principle of false data injection attack and the construction of its general model are studied.With this as the theoretical basis and important guidance,a detection mechanism against false data injection attacks is proposed from the perspective of defenders.Guided by the results of interval state estimation,this defense mechanism can effectively detect most of the system state anomalies caused by false data attacks.The monitoring of operation state of distribution system is more effective,and the observable and controllable level of the system is higher.