Study On The Strategy Of Improving The Resilience Of Distribution Network With Energy Micro Grid

In recent years,extreme weather events occur frequently all over the world,causing largescale power outages and economic losses,which makes the resilience of distribution network receive extensive attention.Extreme natural disasters are events with small probability and high risk,so the research on the ability of distribution networks to deal with extreme natural disasters is lack of data support.Compared with the reliability research under normal condition,it is more difficult to study the resilience under natural disasters.At the same time,with the development of controllable distributed power sources such as gas turbine and fuel cell,their supporting role for important loads in extreme events is also gradually discovered.The main work completed in this paper is as follows:(1)From aspects of source,network and load,the initial element models of distribution network were given and a fault scenario generation method suitable for high risk and low probability disaster events was proposed.The types of supporting power sources in the elastic distribution network under the condition of extreme disasters were studied,which can be used to support the important loads in extreme disasters.The typhoon disturbance model and the component fault probability model were established by using the data-driven and model-driven methods.Based on these,the Monte Carlo simulation method was used to generate random fault scenarios,which were used as the original topologies for the follow-up research to make up for the lack of disaster data.In terms of load,the Back Propagation(BP)neural network algorithm considering meteorological factors was used to form the forecast model of electric and heating load,which was used as the load demand in the follow-up research.(2)A distribution network resilience improvement operation strategy considering the thermal-electrical load correlation,complementarity and different operating modes of micro grid was proposed.Firstly,in the rated condition,combining with the current fault state of line topology,the distribution network was divided according to the optimization goal of maximizing the power value of supporting loads.Secondly,the energy micro grid was used as the energy source to support the important load of the distribution network in disaster.On each isolated island,the energy micro grid provided power to the load within the divided range,and the optimization strategy under different operation modes was studied.Finally,after the repair conditions met the requirement,repair resources were arranged reasonably to repair the faulty lines and minimized the economic losses in the repair process.All the above problems were transformed into mixed integer linear programming problems by piecewise linearization and solved by branch-and-bound method and ε-constraint method.The obtained results can provide guidance for the operation of the distribution network in extreme disasters,and the simulation results can also be used as the basis for dynamic indicators in the resilience evaluation.(3)A distribution network resilience evaluation system including static indexes and dynamic indexes was proposed.The indexes were respectively used to measure the nature of the distribution network during normal operation and the actual performance under extreme natural disasters.The same method was used for evaluation.Firstly,starting from the static and dynamic indexes that affect the resilience of distribution network,the core index system was screened according to the accessibility,the correlation and the contribution degree of indexes.Then the indexes were standardized,and the entropy weight method and the analytic hierarchy process were used to calculate the objective weight and the subjective weight respectively,which forming the comprehensive weight.Finally,the fuzzy comprehensive evaluation of the indexes was carried out by using the ascending half trapezoidal function.The lines need to be transformed were determined by dynamic index scoring,and the factors that need to be improved were determined by static index scoring,so as to guide the resilience upgrading of the distribution network.