High order perturbation method based statistical model updating using modal data
Finite element(FE)model updating is to modify the FE model with reliablemeasurement results to minimize the difference between simulated responses andmeasurement results.The modified FE model can better describe the mechanical behavior ofthe true structure,and can be used for structural health monitoring,performance evaluation,and the prediction of response of structure in various working environments.In practice,thedata often tested are the modal data of structures.However,the measurement results areusually uncertain more or less,and the uncertainty of the measurement results will inevitablylead to the uncertainty of the modified structure model.In this paper,the measurement errorsare assumed to be of beta distributions in the FE model updating.The structural updatingfactors are expressed by multi-variables non-orthogonal polynomials expansions.Afterwardsa stochastic FE updating equation is established with respest to the updating factors.Furtherthe high order perturbation method is applied to solve the equation so as to obtain thestatistical characteristics of the updating factors.Meanwhile the direct Monte Carlosimulation method is used to verify the correctness of the calculation results.Numericalexamples show that,using the first three orders of displacement modal informations,themodified results by the proposed method are consistent with the measurement results underdifferent noise levels.In the end,the experiments of a reinforced concrete structure areconducted to testify the effectiveness of the proposed method.
Model updating Updating factor Uncertainty High order perturbation Non-orthogonal polynomial
Hui Chen Bin Huang Youssouf Mohamed Witti
School of Civil Engineering and Architecture,Wuhan University of Technology,Wuhan 430070,China
国际会议
The 7th World Conference on Structural Control and Monitoring(7WCSCM)(第七届结构控制与监测世界大会)
青岛
英文
1959-1969
2018-07-22（万方平台首次上网日期，不代表论文的发表时间）