Research on an Adaptive Control Method Integrating Natural Logarithmic Sliding Mode and State Observation for Horizontal Vibration Suppression in High‐Speed Elevators
Abstract
To address the horizontal vibration problem of high‐speed elevator car systems induced by multi‐source excitations, including guide‐rail irregularities, traction machine disturbances, and shaft aerodynamic effects, this paper proposes an adaptive control method integrating natural logarithmic sliding mode control and state observation. First, the aforementioned multi‐source excitations and system uncertainties were categorized into three types: parameter variations, external disturbances, and state variations. Based on this, the State–Parameter–Uncertainty High‐Speed Elevator Vibration Model (SPU‐HEVM) was established. Second, a sliding‐mode observer was designed to reconstruct the system state, and unknown parameters were updated online using a projection‐adaptive law. Building on this, the natural logarithmic sliding surface was integrated with the sliding‐mode observer and the projection‐adaptive law into a unified control framework, enabling performance‐oriented parameter design and reducing reliance on exact models and trial‐and‐error tuning. Finally, the accuracy of the model was verified through real‐world elevator experiments and multi‐source composite simulations, and the proposed method was compared with adaptive sliding mode control (ASMC) and adaptive terminal sliding mode control (ATSMC). The results demonstrate that the proposed method in this paper can accurately estimate system states and parameters, with typical vibration acceleration values reduced by more than 25% compared to ASMC, significantly improving vibration suppression performance and system robustness.