Robust Differentiator-Based High-Order Sliding Mode Control for PMLSM with Predefined Performance

Authors:

Abstract
To achieve the collaborative optimization of both steady-state and transient performance in the PMLSM system, a prescribed performance-based high order sliding mode control method is proposed. First, a Prescribed Performance Function (PPF) is constructed based on the hyperbolic cotangent function, which inherently relaxes the initial condition constraints, thereby eliminating the need for precise initial error information. The position tracking error is transformed into a mapped error. By constraining the dynamic boundary of this transformed error, we can quantitatively regulate the system's steady-state precision and transient response indicators, theoretically ensuring that the tracking error converges within the prescribed performance envelope. Second, a dynamic fractional-order terminal sliding mode surface is combined with the dynamic super-twisting algorithm. The online adjustment of the control gain enables real-time and optimal self-tuning of the disturbance compensation strength across the entire operating range, thereby significantly enhancing the system's robustness against uncertainties. Additionally, to address the inherent limitation of potentially slow convergence in certain phases of the classical super-twisting algorithm, a linear correction term is introduced, which accelerates the finite-time convergence process of the sliding mode variable and effectively shortens the system's transition time to steady-state. Furthermore, a robust differentiator is employed to provide noise-immune estimation of the acceleration signal. Finally, experimental results demonstrate improved transient performance while maintaining steady-state performance.

This publication has 11 references indexed in Scilit: