Online estimation of the friction coefficient in sliding isolators

The identification of nonlinear structural dynamic systems is a common task in civil engineering, where it is applied for damage detection and health monitoring. In this paper, a structural dynamic identification procedure based on the extended Kalman filtering (EKF) is proposed for online estimation of the coefficient of friction of friction-based isolators, such as Friction Pendulum Systems or Curved Slider Surface Sliders, starting from the measured accelerations of the isolated superstructure. The suitability of the EKF estimator for the proposed application is demonstrated for two case studies. In the first case study, focused on the numerical analysis of a two degree of freedom system isolated with Curved Surface Sliders, the procedure is used to identify the material parameters of an exponential friction model. The second case study illustrates the application of the procedure to an experimental investigation carried out on a base-isolated structural prototype tested on a shake table. Both numerical and experimental results demonstrate that the proposed EKF observer provides reliable state estimation and correct identification of the coefficient of friction accounting for its dependence on the velocity, and can be applied to isolators characterized by different levels of friction. The methodology can be used for the determination of the friction coefficient of sliding isolators in real applications, as well as for the assessment of any modification of the coefficient of friction due to the aging, wear and contamination of sliding devices.