Research On Cable Force Adjustment Method Of Cable-stayed Bridge Cable Replacement Engineering Based On Gray Neural Network

For cable-stayed bridges which have been built for many years,the cable sheath is damaged and the line shape changes greatly due to various factors.From the 1970s to the early 1990s,China's cable-stayed bridges have been changed cables or strengthened by other measures.The remaining cable-stayed bridges also face the problem of cable replacement.In this paper,taking Dongfeng cable-stayed bridge as an example,studying the project of cable replacement.Firstly,according to the structure characteristics of cable-stayed bridges,the paper summarizes defects of stay cables,concrete girder and cable tower,pier foundation as well as ancillary facilities;then the paper analyzes reasons.In addition,the cable-stayed bridge cable replacement optimization theory at home and abroad is discussed.Through establishing a cable-stayed bridge numerical example,it compares with the advantages and disadvantages of the existing cable replacement theory.The paper adopts unknown load coefficient method and influence matrix method.Through MIDAS CIVIL,it analyzes construction process of replacing stay cable,compared with practical data.The internal force of the bridge structure is greatly improved by cable replacement,adjustment and reinforcement.Due to the presence of deflection and crack phenomenon at concrete girder mid-span and side span,different amount of external prestressed steel wires and steel bundles are arranged at upper and lower edges of the main girder before bridge deck is paved to effectively increase the reserve of compressive stress and reduce the negative bending moment of the main girder mid-span;then calculating it by theory method and numerical simulation.This reinforcement method has a significant effect on reducing the negative bending moment of the main beam.After reinforcement,it can effectively avoid the generation of new cracks and improve the durability of the main beam.The calculation of the strengthened pier foundation is carried out.The results show that the vertical reaction force of the pier increases a bit.After the bridge is strengthened,the bearing capacity of the bridge can be improved.At the same time,the bearing capacity of the bridge foundation can be met.For cable-stayed bridge replacement project,the grey theory combined with neural network prediction model is used.Grey relational analysis is used to get the data group with larger relational degree.The multi-variable grey GM(1,3)model is used to fit the measured cable force,and increased the relativity with other data.With the help of MATLAB neural network tool,the model of dynamic adjustment weight(threshold)is established to predict the next construction stage.The output value of the model is compared with the measured value of the project.The results show that the grey neural network can make better results and can basically meet the needs of cable-stayed bridge replacement project.