Research On Monitoring Techniques And Structural Performance For Steel Spatial Structures

The risk probability of large-span steel structure is higher than other buildings during construction stage because of the complicated construction process and construction technology.As the construction stage is time-variant,the real structure is not exactly as the same as the design,in terms of internal force and other aspects.And field monitoring is an efficient way to get the difference quantificationally.Moreover,in the long-term of service,the structure is inevitable to degenerate,due to environmental erosion,the aging behavior of materials and other factors.Once an accident of large-span steel structure occurs,it may cause inestimable loss of property and casualties.Thus,it is full of significance to carry out the long-term health monitoring for large-span steel structure.So that,we can grasp the changeable discipline of internal force,assess the structural state and estimate the remaining service life of structure.So far,the monitoring work of long-span spatial structure is mostly aimed at a certain stage,while the long-term monitoring practice is rare.To overcome the disadvantages of wired monitoring systems,our lab developed a field measuring system based on wireless sensing,and it was applied to the steel structure of the Shaoxing Stadium to continuously monitor the stress,temperature,displacement and acceleration from the very beginning of the construction.As the roof of the stadium is retractable,this paper preliminary designed a wireless network which can dynamically change the network topology during the movement of the retractable roof.Steel tension rods are the key components of the structure.This paper designed a special device to install sensors on the steel tension rods.The result of verification experiment showed that the performance of this method is quiet good,especially when the strain under 400??.And when the strain under 800??,the error is no more than 3%.Based on the field monitoring data,this paper focused on the performance of steel tension rods during four stages,namely,the pretension stage,the unloading stage,daily service stage and process of the movement of the retractable roof.Compared with the finite element analysis,the mechanical characteristics of steel tension rods are carefully discussed in both construction and service stage.In the long-term service process,the stress variation is mainly caused by the temperature change.Based on long-term data of stress and temperature,this paper analyzed the temperature distribution,and the relationship between temperature and stress in terms of different components,different seasons and different years.By means of the finite element simulation,the influence of the support stiffness on temperature stress of the structure was studied.Based on the correlation between the support stiffness and temperature response of the structure,a method using principal component analysis was proposed to detect the change of the support stiffness.Based on the filed monitoring data from 2014 to 2017,the result of the detection method showed that the stiffness did not have obvious difference during these four years,and the supports maintained a good working condition.Aiming at the data missing problem of long-term monitoring,a data reconstruction method is proposed based on the characteristics of long-term stress monitoring data of spatial steel structures.Neural network technology is used to establish the correlation between stress data of different measuring points,and the correlation between stress and temperature data.The accuracy and applicability of this method is studied,in terms of data missing types,effect of sunshine and the amount of training data.The method is more suitable for the missing data of long-term service stage rather than construction stage.Through the introduction of the long-term monitoring of Shaoxing stadium,this paper put forward some beneficial exploration of life-cycle monitoring and data analysis for large-span steel structure.And the author hopes that it may accumulate experience for assessing the structural state and estimating the remaining service life of structure.