Font Size: a A A

Application Research Of Mas Theroy In The Liquid Rocket Testing Bed Fault Diagnosis

Posted on:2011-03-11Degree:MasterType:Thesis
Country:ChinaCandidate:Y ZhaoFull Text:PDF
GTID:2132330338479846Subject:Instrument Science and Technology
Abstract/Summary:PDF Full Text Request
The reliable running of liquid rocket's testing bed is the key factor of the ground test's success. If the testing bed has faults ,the whole test will fail and it will cause a great economy lose.In consequence,the research of testing bed's fault diagnosing is very important.As testing is a big system ,its structure is complicated,the symptom and type of fault is diverse.So traditional method as the method based on knowledge,method based on rough set can not diagnose fault exactly and rapidly.This paper present a diagnosing method based on multi-agent system(MAS),which is directed to the weakness of traditional intelligence diagnosing method.It researched the designing of individual Agent,the communication and coordination between Agents in system.The whole system software is designed under Labview platform. This paper verified the feasibility of method through simulational experiment,realized the intelligentization of complicated fault diagnose system,improved the efficiency of diagnose and accuracy of the result,optimized the structure of system.In order to the flow of system's designment,analysised the structure and function module of testing bed fault diagnose system first,and then built a three layer structure for the testing bed fault diagnose system based on MAS.Base on the research of individual Agent,adopted a method called "insert model"to signed them.Acommplished the designment of administrator Agent,user interact Agent,interenet interface Agent,fault simulate Agent,diagnose agent etc.These individual Agent are programmed independently,running autonomously and have characteristic such as sociability.They can run in MAS in parallel successfully.Key points of the research are communication between Agents,the function of coordination,collaboration and negotiation.By these function,system achieved the sociability and can accomplish some tast which individual Agent can not.Using black board and message transmition to realize communication.We break up black board to some diffirent areas.So every Agent can share black board source according to there own need.Using existing research of utility evaluation for reference,built a utility evaluation method based on time and precision.And built a negotiation method based on utility evaluation.In the end,this paper used real tesing data of testing bed,combining simulated data to verified this method's utility and effectiveness.
Keywords/Search Tags:Multi-agent system, Artificial Intelligence, test bed, fault diagnosis, negotiation method
PDF Full Text Request
Related items