Research On Resilience Improvement And Evaluation Method Of Distribution Network Under Extreme Weather

In recent years,the frequency of extreme weather events with small probability has become higher and higher,which has a huge impact on the power system.The distribution network is more vulnerable to disasters due to low automation.Resilience was introduced into the power industry as a new concept to express the ability of power grid to withstand extreme internal and external events,reduce losses,and recovery as soon as possible.This thesis mainly focuses on the improvement measures of the distribution network’s resilience and its evaluation methods.The main contents generally include the following aspects.First,analyze the impact of extreme weather events on the distribution network,establish a vulnerability model of distribution network components.Propose distribution network resilience boosting strategy from the perspective of pre-disaster prevention measures,which conclude component strength improvement,vegetation management,and comprehensive resilience improvement.On this basis,a three-level optimization model with the lowest total cost as the goal is established,and it is reconstructed into a two-level optimization problem through a greedy search algorithm.The higher level selects the best resilience improvement strategy for important lines,and the lower level selects the lines that are most susceptible to extreme weather and have a significant impact on load shedding.In order to verify the effectiveness of the proposed strategy for resilience improvement,simulation analysis was carried out in consideration of various aspects in the improved EPRI test circuit.Secondly,a resilience improvement method for recovering important loads with the microgrid was proposed from the perspective of disaster recovery.Microgrid can be disconnected from the grid when natural disasters occur,and it acts as a "virtual feeder" in the distribution network to maintain the power supply of important internal loads or even external nearby important loads.Based on this,the concept of microgrid continuous operating time is proposed and the utility of microgrid for load recovery is analyzed.A dual-objective chance constraint programming model is established to maximize the power supply that can be used for important load recovery under the premise of considering multiple constraints.And the voltage change is minimal.Considering the uncertainty of distributed power output and load demand,a microgrid operation model based on Markov chain is established,and the optimal load recovery path is found through a two-layer heuristic recovery scheme.Finally,a resilience assessment method based on load importance is proposed aiming at most of the current assessment methods that consider the resilience of all loads.Quantify extreme weather and use Monte Carlo method to simulate failure scenarios.According to the national standard,the importance of the load is graded and a temporary load shedding strategy is formulated.The comprehensive load loss is introduced as an indicator to quantitatively evaluate the resilience of the distribution network,and simulation analysis is performed in the IEEE-33 bus system.