Identification of wind loads on super-tall buildings by Kalman filter

Abstract

An inverse method based on the discrete time Kalman filter is developed to simultaneously estimate the unknown wind-induced responses and wind loads of high-rise buildings using limited response measurements. The accuracy and effectiveness of the presented estimation method are assessed based on both wind tunnel test results and field measurements of wind effects on a super-tall building. The wind-induced responses obtained by wind tunnel test are used to identify the wind loads on the super-tall building and evaluate the effects of crucial factors, such as covariance matrix of noise, initial estimation error, noise in measurements, modelling errors of structural dynamic properties and number of vibration modes, on the identification results. In addition, the wind loads are also estimated based on the field measured displacements during Typhoon Neoguri. Comparisons between the identified wind loads from the field measured displacements and the wind tunnel test results are made to verify the applicability of the inverse technique. The results of this study illustrate that the proposed inverse method is an efficient tool for predicting the wind loads on super-tall buildings from limited measurements of structural responses.