Reliability Analysis Of Distribution Network CPS Based On Non-Sequential Monte-Carlo Simulation

With the development of smart grid technology,the coupling degree of information flow and energy flow in the distribution network is deepening.The intelligent level of switching devices is also continuously improved,which can actively shed load and reconstruct network after faults.As a result,the traditional distribution network gradually evolved into a Cyber Physical System(CPS).Traditional reliability analysis methods of distribution networks focus on analyzing the physical system and do not consider the fault reconstruction.These methods cannot cope with the wide application requirements of intelligent switching devices in the distribution network.This paper studies the power supply reliability from the perspective of information physical integration for medium voltage distribution networks which are the main cause of power failures of end users.The main research results obtained in this paper are as follows:Firstly,the reliability indices of distribution network power supply are classified and summarized.The basic assumptions of the structural composition and operational characteristics of medium voltage distribution networks are given.Combined with Non-Sequential Monte-Carlo Simulation(NSMCS),the calculation method of reliability indices and the method of formulating convergence conditions are established.Secondly,it is difficult to analyze the reliability by simulation methods via random sampling of the system state for the distribution network CPS.The technology of the intelligent switches in the active distribution network is analyzed and the reliability assessment model of distribution network CPS considering the impact of the failure rate of the information physical system is developed.The model proposes a method for mathematical representation of the distribution network CPS and random sampling of the system state.The physical actions of switching devices and the states of corresponding intelligent distribution terminals are implicitly embedded in the state of the electrical components rather than explicit modelling.Thanks to this,a random state sample of the network system does not rely on its history.Taking minimizing load shedding as the optimization objective and the electrical components capacity limitation as the constraint,the operating state and reliability indices of a single state sample are solved.Combined with the importance sampling,the reliability evaluation of the distribution network CPS power supply is simulated through NSMCS.The case study shows that the model can effectively evaluate the impact of information system failure rate and physical system failure rate on the reliability of distribution network CPS.Finally,it is necessary to maintain the open-loop operation and reconstruction of the medium voltage distribution network.Relevant assumptions and specific implementation methods for applying the radiality constraints in the random state sampling of the distribution network CPS are analyzed.The reliability evaluation model of distribution network CPS considering radiality constraints is proposed.The model takes maximizing power restoration as the optimization obj ective,and takes the electrical components capacity limitation and the radial network structure limitation as the constraints.It simulates the actual network structure after faults.The operating state and reliability indices of a single state sample are solved.Combined with the importance sampling,the reliability evaluation of the distribution network CPS is simulated through the NSMCS.The case study proves that the extended model can make a more accurate assessment of the power supply reliability of the active distribution network CPS with automatic fault reconstruction.Models established in this paper assume that all switching devices with working distribution terminals can automatically perform optimal reconstruction actions after faults according to optimization objectives and constraints.These models are suitable for grid planning problems of active distribution network CPS with widely deployed intelligent distribution terminals.