Research on Sliding Mode Variable Structure Controller Based on Genetic Algorithm

Abstract

To address the nonlinearity and complexity of practical industrial plants, a novel control strategy is proposed based on genetic algorithm (GA). Different from existing adaptive control methods that heavily depend on system properties, the proposed variable structure control strategy effectively eliminates such reliance-induced defects. By utilizingGA for the optimal tuning of integral coefficients c, the controller achieves satisfactory performance, which is verified by numerical simulations. In practical DC motor speed control applications, the proposed method exhibited favorable control performance and superior robustness, achieving remarkable improvements in both dynamic response and steady-state characteristics. The comparative experimental results indicate that under time-varying complex instructions, the maximum position loop overshoot of the DC motor system adopting the improved sliding mode control strategy drops from 20% to zero, while the convergence speed rises by approximately 80%. With fast dynamic convergence, high steady-state tracking precision and strong robustness, the developed sliding-mode controller is applicable to the tracking of complex periodic speed commands.