The application of BP Neural Network in continuous girder bridge construction control

Deflection control of construction is a crucial step of bridge construction monitor control in cantilever construction process. The control results of which will directly influence the quality of closure construction of bridge structure. The deflection is influenced by so many factors which made the calculation and analysis process complicated. This paper use BP Artificial Neural Network technology to comprehensively discuss the influences on bridge linear cased by many various factors. In addition, the theoretical data of former several construction stages was used for network training, so that the construction deflections of subsequent construction stages will be forecasted. Through the comparison between predicted values and measured value, the reliability of Neural Network forecast method is verified. It´s concluded that: in the construction control, it is beneficial t o improve the control precision by means of combining preliminary prediction with later prediction, and there´ll be certain reference value in similar projects.