Bridge Structure Health Monitoring Mass Data Preprocessing And Mining Analysis

With the vigorous development of Bridges,the safety issue of bridges is getting more and more attention.A lot of bridge health monitoring system is established,which can get the information of structural response behavior through sensors in various environments.At present,most bridge structure health monitoring system has perfect data collection,transmission and storage function.Thus,the large amount of data was collected and it implies the rich information of the bridge's state and surrounding environment.The analysis and utilization of these data can help people to understand the working status of the bridge in real time and to find out the problems of the bridge in time.Therefore,it is crucial to deal with these data accurately and rationally.In this paper,the massive monitoring data collected from the Shanghai Wu-song bridge are analyzed and pretreated.The common data pretreatment problems are also analyzed and summarized.And based on the MATLAB software to compile the relevant procedures to deal with the temperature and deflection data of Wu-song bridge,getting the more reliable and complete data.On this basis,the temperature field analysis of the bridge and the correlation between temperature and deflection are analyzed.The specific work of the thesis is as follows:(1)This paper investigates the present situation of data preprocessing and mining analysis of the existing health monitoring system,and summarizes the methods of data preprocessing and the utilization of data mining.(2)This paper makes a theoretical analysis of various data preprocessing methods.And taking the health monitoring data of a pre-stressed concrete continuous beam bridge as an example,using MATLAB software to write program code program codes to performed the problem of abnormal value elimination,missing value completion,data trend extraction and data smoothing and standardized processing etc.(3)On the basis of data preprocessing,the analysis of temperature monitoring data was carried out,and the expression of the daily variation curve and the annual variation baseline of the box beam temperature were obtained.The article puts forward the influence of weather factors on the monitoring temperature distribution.And explored the vertical temperature distribution of the concrete box girder under the influence of the outside temperature of three kinds of weather,such as clear,rain and shade;(4)In this paper,the factors that cause the long-term deformation of the bridge deflection are analyzed.Next,the correlation between outside atmospheric temperature and mid-span deflection is analyzed.The correlation analysis of the outside atmospheric temperature and the mid-span deflection shows that there is a period of about 24 h between the temperature and deflection cross-correlation function.And the relationship between daily temperature and mid-span deflection was studied.