Human-Simulated Sliding Mode Intelligent Control Considering Partial Uncertainty of Magneto-Rheological Damper in Structural Vibration

Magneto-rheological (MR) dampers are widely used in vibration control systems due to their excellent controllability of output characteristics. Over prolonged operation, MR dampers tend to lose damping effectiveness, mainly due to temperature rise, MR fluid leakage, and magnetic coil failure. Notably, conventional control schemes for MR systems often overlook the inherent damping loss phenomenon. This study begins with an empirical investigation of the effect of temperature on the damping force of MR dampers. A comprehensive scaled three-story structural model incorporating an MR damper is then developed, accounting for temperature effects as a partial uncertainty factor. An innovative sliding mode observer is proposed to estimate the damping force loss caused by temperature increase. Using the acquired uncertainty data, a novel control strategy called Human Simulated Sliding Mode Control is introduced. The effectiveness of the proposed control strategy is evaluated through simulations using seismic excitations from the El Centro, Taft, and Kobe earthquakes. To further validate the proposed control strategy, a Hardware-in-the-Loop (HIL) test system using dSPACE is established. Control strategies tolerant to uncertainties, addressing partial uncertainty in MR dampers under different frequency sine wave excitations, are systematically examined. The results confirm that the proposed control scheme effectively reduces the negative impact of actuator uncertainty on the performance of MR semi-active three-story structural systems.