Augmented Kalman filter based moving vehicle loads online identification

A novel moving vehicle dynamic load online identification method based on the augmented Kalman filter (AKF) was proposed. The vehicle load vector and the bridge structure state vector were batched together to form an augmented state vector, and the AKF algorithm was employed to yield the unbiased minimum variance estimate by using only a small amount of response measurement so as to make the vehicle load be identified in real time. Taking a simply supported beam-sprung mass vehicle-bridge coupling system as the object of numerical analysis, the feasibility and accuracy of the proposed method were examined, and the effects of road unevenness, vehicle speed, noise, sensor combination and sampling frequency on identification errors were investigated detailedly. The proposed method can accurately identify dynamic loads and is insensitive to measurement noises and vehicle speed. © 2022, Editorial Office of Journal of Vibration and Shock. All right reserved.