|
Continuous adjustable rigid frame bridge built , but subject to some uncertainty and the impact of the uncertainties in the construction process , coupled with continuous rigid frame bridge cantilever casting construction more construction phase , the construction process there the conversion of the structural system , it is very necessary to control the construction process of continuous rigid frame bridge . For continuous rigid frame bridge , elevation control is the most important because the actual project , stress fluctuations , and elevation measurements can be within control accuracy . View from the practical experience of the past , if the elevation reached the target state , the structure of the internal forces often will reach the appropriate within the control . In this thesis, the following : first, the development of continuous rigid frame bridge construction control are summarized on the main factors in the construction process , and comparative analysis of continuous rigid frame bridge several error prediction prediction method , the basic theory and BP neural network are analyzed in detail . Second , through the establishment of the finite element model of the example project continuous rigid frame bridge construction control parameter sensitivity analysis . Parameter sensitivity analysis , parameter changes in the scope of control of about 10% , and the selected control objectives , such as the the maximum cantilever state of the main beam deflection values ??to calculate the change in the amplitude of the main beam control objectives , based on various parameters on the The control objectives of the sensitivity parameters determine the extent of non - sensitive parameters . Third, the detailed discussion of the method for determining continuous rigid frame bridge formwork elevation , vertical control programs to develop the main bridge in Meishan , closed and the T configuration state deflection monitoring results and Closure Segment pushing fine ocean analysis . Fourth , the use of BP nerve network , establish elevation prediction model , BP nerve network predicted results with the measured results and other prediction methods to predict the results were compared , that BP nerve network model in the continuous rigid frame bridge vertical control good results . To Meishan main bridge left pieces , for example , Meishan Bridge left pieces of folded segments at both ends of the height difference of only 2mm, the adjacent beam segment elevation error difference is no more than 15mm , to reach the the beam bottom line shaped smooth purpose . Small and target elevation difference Overall , the biggest at 26mm, the the bridge elevation control achieved good results .
|