Font Size: a A A

Researchon Intelligent Matching Method Of Traffic Emergency Preplan On Freeway

Posted on:2016-11-21Degree:MasterType:Thesis
Country:ChinaCandidate:M M HuangFull Text:PDF
GTID:2272330503977585Subject:Control theory and control engineering
Abstract/Summary:PDF Full Text Request
Intelligent matching of highway emergency rescue preplan has a crucial influence on emergency rescue decision system. So how to get the preplan which is appropriate to the current emergency rescue quickly and efficiently is the key issue of this paper. This paper presents the application of case-based reasoning to preplan intelligent matching process, based on case-based reasoning process 4R, focusing on the matching algorithm.The purpose of this paper is to get effective rescue preplan in the shortest possible time after freeway incident occurred which can help makers to decision making. Firstly, this paper analysis of highway emergencies carefully to obtain described attribute and proposes one kind of case representation based on XML; Secondly, by classifying described attribute of highway emergencies, this paper proposes confidence hybrid similarity matching algorithm which uses different similarity calculation method base on the types of Attributes. Preplan confidence is put forward in this algorithm which builds a confidence decision tree through analyzing confidence index and using C4.5 algorithm. It combines similarity and confidence on historical cases to judge the best preplan; Thirdly, with the consideration of event described attribute missing because that local roads has relatively low level of information collection, this paper proposes a matching algorithm based on Bayesian probability model. It is improved on the basis of the naive Bayes classifier to construct double Bayesian probability model, using only a small amount of Attribute acquired to realize matching. A learning algorithm is proposed to update the Bayesian probability model.Finally, this paper introduces the overall design framework and the implementation process of the software system. The system can select the appropriate matching algorithm automatically through Judging by the total number of non-null Attribute. It also uses the simulation examples to verify its correctness. The results are in line with the expected theoretical analysis. So this shows that the matching algorithm of traffic Emergency preplan on Freeway in this paper is feasible and effective.
Keywords/Search Tags:Freeway, Intelligent Matching of Preplan, Case-based Reasoning, Confidence Hybrid Similarity, Bayesian Probability Model
PDF Full Text Request
Related items