Research On Resilience-based Recovery Decision Of Electric Emergency Management

Electric power systems(EPSs),which are critical infrastructure systems,provide essential services to support the economic development of a nation and ensure the wellbeings of its citizens.However,natural disasters that tends to become more frequent seriously threaten the security of EPSs.Electricity interruption may lead to the failures of other infrastructure systems(such as transportation and communication system)due to their interdependencies,resulting in major economic losses and serious social influence.To mitigate the impact of natural disasters on EPSs,it is of great urgency to strengthen electric power emergency management(EPEM)capability.Emergencies are usually characteristic of suddenness and extensiveness and existing disaster prevention and mitigation capacity are limited,which may cause power outages inevitable.Therefore,it is quite important to investigate the post-disaster restoration decision-making problem and make an efficient recovery plan,so the power supply services can be restored as quickly as possible.Many researchers have studied how to make resilience-based recovery plans,and developed different types of decision-making models and methods.However,existing methods are not very efficient,and few studies have considered uncertainties during in the restoration phase.Hence,this thesis summarizes the existing problems of current emergency recovery phase.Then,this thesis combines the knowledge of emergency management and the resilience theory,and proposes two decision-making methods for the resilience-driven restoration problem of EPSs.One method addresses the deterministic scenario,and the other method considerers the scenario in which the repair time of each damaged components is uncertain.These two methods provide useful tools for governments and electric utilities to quantify the efficiency of different restoration strategies,and design efficient recovery plans under limited restoration resources.Finally,this thesis summarizes the shortcomings of current EPEM,and puts forward a series of suggestions for its improvement.Main research efforts of this thesis are listed as follows:(1)From the perspective of resilience,this thesis points out the problems existing in EPEM,and analyzes the root of these problems,including: weak attention to recovery plan in EPEM,insufficient collaborative ability of different departments and institutions and low participation of social forces,and lack of intelligent recovery decision support platform to deal with emergencies.(2)Given the assumption that the input parameters of the post-disaster restoration phase decision-making model are deterministic,this thesis develops a deterministic resilience-driven repair sequencing decision model,and proposes two heuristic algorithms to solve it.Numerical experiments on EPSs are conducted to demonstrate the efficiency of the proposed algorithms.(3)Given that the repair time of each damaged component is uncertain,a two-stage stochastic resilience-driven repair sequencing decision model is proposed by extending the deterministic model.A heuristic algorithm is proposed to solve this model and its efficiency is also tested.In addition,this thesis further investigates the value of stochastic solutions based on above two decision models.Results show that reducing the uncertainty of the repair time of damaged components can help designing a more efficient recovery plan.(4)Based on the findings from applying the above two models to realistic systems,several suggestions are put forward to improve the post-disaster recovery capacity of governments and electric utilities,including: adding a post-disaster recovery decisionmaking procedure during the existing emergency management;increasing the participation of government and social forces and coordinating their efforts to increase the restoration resources and reduce the uncertainties in the post-disaster recovery phase;developing an intelligent recovery decision support platform to provide real-time and efficient recovery plans and visualize the recovery progress.