Font Size: a A A

Research On Emergency Situation Evolution And Emergency Decision Making For Flood Disaster Chain Of Hydropower Project Construction

Posted on:2018-05-07Degree:MasterType:Thesis
Country:ChinaCandidate:T T HuangFull Text:PDF
GTID:2382330566951584Subject:Systems Engineering
Abstract/Summary:PDF Full Text Request
Due to the dynamism,uncertainty and complexity of the secondary disasters of flood disaster chain of hydropower project,decision-maker is not clear enough to understand the emergency evolution process,so that there exits short decision time,less contingency plans and huge complexity of decision-making in decision-making process.Therefore,it is a key issue how to use the multi-source information obtained during the process of flood disaster,then discover the evolution trend and law of flood disaster chain,and provide theoretical basis for emergency decision-making.This paper takes the emergencies of flood disaster chain of hydropower project as research object,and establishes the model of emergency situation information fusion,emergency situation estimation,emergency situation deduction and emergency decision.Aiming at the uncertainty of multi-sensor information fusion and the fusion ability of D-S evidence theory for uncertain information and its shortcomings,the conflict coefficient and combination rule of classical D-S evidence theory are improved.Time factor is introduced in the integration of evidence according to situation evolution changing over time.On the basis of the information fusion and situation estimation,the model of emergency situation evolution is established.Firstly,Bayesian network(BN)is used to establish single disaster model.Bayesian structure learning and parameter learning are carried out by using little amount of sample data and expert domain knowledge.The concept of intuitionistic fuzzy sets is introduced taking into account the ambiguity of expert experience.Then,the situation evolution model is established by using dynamic Bayesian network(DBN)based on single disaster models and the field sensor data.On the basis of the evolution model,a dynamic decision model based on multi-attribute utility function is established.In the model,considering the uncertainty of situation deduction,the interval number and cloud theory are introduced,and the influence of secondary disaster is taken into account.The model established in this paper is an effective attempt to study the situation information fusion,situation evolution and emergency decision making of flood disaster chain,especially considering the uncertainty and dynamic of disaster chain,which further enriches the research contents in related fields.In the main chapter,the paper analyzes the corresponding cases,and validates the rationality and applicability of the model through the case analysis.It provides strong support for the relevant decision making departments to make reasonable decision when deal with the specific flood disaster chain emergency events.
Keywords/Search Tags:Flood disaster chain, Information fusion, Situation evolution, Emergency decision making, D-S evidence theory, Bayesian network, Dynamic Bayesian network
PDF Full Text Request
Related items