Font Size: a A A

Bridge Health Monitoring System Based On Multi-sensor Information Fusion Research And Implementation

Posted on:2018-02-22Degree:MasterType:Thesis
Country:ChinaCandidate:J ShenFull Text:PDF
GTID:2322330536979870Subject:Electronic and communication engineering
Abstract/Summary:PDF Full Text Request
With the development of bridge construction technology in China,many kinds of large-scale Bridges have been built,bridge health directly affect the transportation,economic and social life,longterm,real-time condition monitoring is critical.Information fusion means: fusion processing on multisensor information,to evaluate bridge structure conditions,provide the basis for bridge maintenance,decision.Based on the analysis of bridge monitoring system research present situation and the basis of information fusion technology,this paper proposed fusion bridge monitoring system solutions based on multi-sensor information.Based on the research of the monitoring content and characteristics of the bridge structure,determine the composition of the sensor network.Analysis of the deflection Angle conversion algorithm was deduced,the light wavelength and strain transformation algorithm and a curve fitting method,Design of the sensor data acquisition based on LabVIEW PC program,and the procedures of JavaEE based on MVC pattern.In addition,due to the large number sensor,and measuring time cycle is long,design based on SQL SERVER database system,provide data basis for fusion processing,as well as the database optimization problems are analyzed,The advantages of using neural network in data recognition,designed a three layers BP neural network,at the same time to avoid falling into local minimum,using the method of additional momentum training function,based on Matlab for Angle data and acceleration data,multi-sensor analysis and single sensor data fusion are compared.Simulation results show that the network data showed a faster convergence speed and higher recognition accuracy,specification design of three layer BP neural network can meet the demand of fusion and has high reliability,At the same time to verify the multisensor data fusion classification accuracy is higher than single sensor data fusion.In this paper,the application of object of tianjin haihe river bridge,by means of hardware module,software module and fusion algorithm research and design,Has realized the monitoring system based on multi-sensor information fusion [1] the function such as data acquisition,transmission,calculation and display,and the practical application in tianjin haihe river bridge site.
Keywords/Search Tags:health monitoring, sensor network, data collection, neural network, information fusion
PDF Full Text Request
Related items