Research And Application Of Large-span Steel Structure Buildings Health Monitoring System Based On WSN

A large number of large-span steel structure buildings have novel appearance,superb craftsmanship,intricate structure,strong sense of sense of science and technology,and has the characteristics of light weight,high rigidity and high seismic performance.Therefore,they are usually widely used in various fields and become landmark buildings in cities or regions.However,due to long-term exposure to the natural environment,under the combined action of various factors such as high temperature,wind and snow,it is easy to cause structural fatigue and structural damage due to factors such as material aging and complex load fatigue effects.Therefore,it is essential to provide these building structural health monitoring.By keeping a close eye to the dynamic changes of related parameters data,we can understand the health status of the building in real time and prevent accidents.This article takes a large-span steel structure exhibition hall of a museum in Shaanxi Province as the research object and studies the deployment of wireless sensor nodes,the collection of information and data,the short-term prediction of micro-strains,and the assessment of health status.The main content of the paper is as follows:(1)Aiming at the WSN coverage problem of the sensing layer of the Structural health monitoring system,this paper uses the improved firefly algorithm to optimize the deployment of sensor nodes.From the optimization results,the improved firefly algorithm has better performance than artificial fish swarm algorithm,hybrid seed particle swarm optimization and chaos glowworm swarm optimization.The coverage rate of IFA is 14%,13.81% and 1.13% higher than that of AFSA,HSPSO and CGSO,respectively.Moreover,the improved Firefly algorithm can improve the quality of collected data and reduce equipment costs.(2)Establish a short-term prediction model of micro-strain based on VMD-GRU,and perform short-term prediction of the micro-strain of the steel structure at the third monitoring point YB9 of the museum exhibition hall.Experiments show that the VMDGRU prediction model has higher accuracy,smaller errors,and can be better prediction of the short-term micro-strain change trend of the monitoring point.(3)Designed a health monitoring information management and evaluation system,in which the data of the relevant parameters of the monitoring points can be visualized,so that users can obtain the data of different parameters of the monitoring points in a timely manner.Predictive analysis predicts the health status of the location of the monitoring point and determine the level of safety of the building.Based on the WSN,the paper optimizes the deployment of sensor nodes,improves the network service quality,and realizes the collection,transmission,storage and use of health monitoring data;Meanwhile,the VMD-GRU prediction model is used to shortterm the micro-strain of the monitoring point Forecast: When the predicted value exceeds the corresponding safety level,the user are able to know the change in the measured value of the micro-strain at the monitoring point in a short period time.If the measured value also exceeds the corresponding safety level in a short time,the SHM system needs to quickly issue an early warning message to remind the management personnel take timely measures to maintain and strengthen the museum exhibition hall.