Optimization speed control of linear induction motors using CSA algorithm and Nonlinear Disturbance Observer

Abstract

This paper investigates a high-performance robust control for linear induction motor (LIM) drives based on the combination of the Crow Search Algorithm (CSA) and a Nonlinear Disturbance Observer (NDO) within the Field-Oriented Control (FOC) framework. The dynamic behavior of LIMs is strongly affected by longitudinal end-effects, parameter variations, and external disturbances, which degrade the performance of conventional PI-based controllers. To overcome these drawbacks, the CSA metaheuristic is employed to optimally tune the speed PI controller gains by minimizing a transient performance index, ensuring a balanced trade-off between fast response and stability. In addition, a nonlinear disturbance observer is designed to estimate lumped disturbances arising from load variations, model uncertainties, and end-effect nonlinearities, allowing their compensation within the control loop. The proposed CSA-NDO-FOC control structure is developed using the LIM dynamic model expressed in the synchronous (d)-(q) reference frame while explicitly considering the end-effect phenomenon. The effectiveness of the proposed approach is validated through detailed simulations in the MATLAB/Simulink environment and compared with a conventional PI-FOC scheme. The obtained results demonstrate significant performance improvements under sudden load disturbances and parameter variations, including a reduction in overshoot from 9.33% to 4.33%, a decrease in settling time by approximately 37.5%, and an improvement in steady-state accuracy of nearly 76%. These results confirm that the combined CSA-based optimal tuning and NDO-based disturbance compensation provide a robust and effective solution for improving the dynamic performance and disturbance rejection capability of LIM drives in high-precision linear motion applications.