Spatial Steel Structure Response Estimation Method Based On Association Analysis

Structural responses are the basis for understanding structural behavior and assessing structural safety.However,due to the limited location and number of sensor arrangements,the monitoring data is incomplete and insufficient to fully evaluate the structure.The structural members are interconnected by nodes,and the structural responses have a certain correlation.However,the relationship between the responses changes with the change of working conditions.Therefore,the spatial steel structure response based on response correlation analysis under multiple working conditions is proposed.The estimation method realizes the multi-position multi-type structural response estimation by determining the association relationship and the mapping relationship between the measured points and the corresponding points to be estimated.The main research work of this paper is carried out in the following three aspects:Sensor placement optimization method based on response correlation analysis.By establishing a gray correlation degree matrix and a correlation coefficient matrix,a threshold is set and a response comprehensive correlation matrix is generated.Based on this,the association region of the component is determined by clustering,and the variation law is analyzed for the associated region under different working conditions,and the optimal arrangement method of the sensor is given.Taking the spatial reticulated shell structure as an example,the correlation region under different working conditions is analyzed,and the response change law is determined.Two kinds of load conditions under various load conditions and considering the load variation law are given.The sensor placement method considers the impact of different thresholds on the sensor placement and verifies the effectiveness of the proposed correlation analysis method.Structural response estimation method based on response mapping relationship.The response variable under multiple operating conditions is selected as the training data.The support vector machine based on particle swarm optimization algorithm is used to determine the mapping relationship between response variables.Based on the determined sensor placement points and corresponding points to be estimated,the mapping relationship is used to estimate the multi-type response under multiple operating conditions,the training data volume,the number of measured points and the influence of noise interference on the estimation accuracy of the response are verified.The proposed sensor placement method and the response map estimation under multi-case conditions are valid and good in anti-noise.Response estimation based on monitoring data of the Shenzhen Bay Stadium.Based on the wind pressure coefficient of Shenzhen Bay Stadium of the wind tunnel test,the wind loads of the equivalent winds at 0°,90°,180° and 270° are determined,and the responses of the structure under four working conditions are obtained through simulation..Correlate analysis of the structural responses under multiple operating conditions,determine the measured points and corresponding points to be estimated based on the existing measuring point arrangement,extract the circumferential rod as the analysis basis,and use the simulation data to estimate the response,to verify the effectiveness of the method on complex large-span spatial structures;and based on the monitoring data under daily and typhoon weather,the feasibility of the method in practical engineering is verified.