Prediction Of Health Monitoring Data Of Continuous Girder Bridges Based On Gray System And Time Series Theory

With the continuous development of traffic,the number of bridges is increasing and the use of bridges is getting longer and longer.The ensuing bridge safety problems also increase,how to use the bridge health monitoring system for real-time monitoring,timely access to structural information of the bridge,timely detection of safety problems,to take effective measures is particularly important,but also now bridge maintenance work Urgent need to discuss the issue.Up to now,the domestic and international health monitoring is mainly used in large-scale bridge projects with oblique girders.There are few health monitoring systems used in small and medium-sized prestressed concrete girder bridges,and the existing methods of bridge structure prediction can not Accurate and true prediction of it,which should arouse our attention.In this paper,a certain prestressed concrete continuous girder bridge in service is taken as the engineering background.Firstly,the health monitoring system of concrete continuous girder bridge is introduced in detail,including the safety monitoring system,sensor system,monitoring result and display system,and the system operation stability.Secondly,the gray theory system is introduced.The gray theory system is used to analyze the correlation of variables.The influence of temperature on the displacement and strain is judged according to the gray relational degree.The MATLAB software is used to realize the gray relational degree.The normalized data draw curve graphics,qualitative analysis,and compared with the actual.Finally,the correlation analysis of temperature,displacement,temperature and strain of concrete in real bridge shows that the gray relational degrees of displacement of the 23rd and 24th holes to the temperature are 0.8354 and 0.7963 respectively,and the 23th and 24th holes concrete strain Gray correlations to temperature were 0.8945 and 0.9299,respectively.Finally,using the time series theory method,the outliers of mass data are firstly processed for data smoothing to achieve the purpose of removing noise.The ARMA time series model is used to predict the concrete crack and strain of bridges.Due to the large amount of data,the fifth-order model is adopted to predict the accuracy of the prediction.The fifth-order prediction equation is obtained by plotting the fifth-order prediction curve and the prediction error autocorrelation function of the six measurement points.Predicting the data value of a certain point in time in August according to the prediction function equation and observing the law of the actual value and the predicted value.Because of the large error,the model of predictive function expression is modified and a correction coefficient is introduced to obtain a new set of Predict the functional formula,and then predict according to the new predictive formula and compare with the actual value,so that the corrected functional formula is closer to the actual value,which verifies the reliability of the revised model.This paper provides some advice for the health monitoring and data prediction of the prestressed concrete continuous girders in service,and provides a practical method for the structural prediction in the health monitoring system,which can guide the time node of the bridge operation and maintenance,The value of the project.