Performance and Robustness Analysis of Maiden Atom Search Optimizer Driven Tri-parametric PID-like Fractional Controller for Speed Control of DC Motor

Abstract In this work, the atom search optimization (ASO) technique is integrated with a fractional-order integral-derivative (FOID) controller to regulate the speed of a Direct Current (DC) motor. A notable feature of the FOID controller is its ability to emulate different classical control structures: by adjusting a single parameter, it can function as a PI controller, a PD controller, or as a true FOID controller. The ASO algorithm, inspired by atomic motion in nature, is computationally efficient, simple to implement, and has been successfully applied to a wide range of optimization problems. To validate the effectiveness of the proposed approach, the dynamic responses of FOID controllers tuned with different optimization algorithms were compared. The results show that ASO-based tuning achieves superior transient performance, motivating a further benchmark of the ASO-FOID controller against ASO-tuned fractional-order-proportional-integral-derivative (FOPID), proportionalintegral-derivative (PID), and proportional-integral (PI) controllers. In addition, several time-domain performance indices, including the Integral of Time-weighted Absolute Error (ITAE), Integral Absolute Error (IAE), and Integral of Squared Error (ISE), were computed to quantify the improvements. Frequency-domain (Bode) analysis further confirmed the enhanced stability margins of the proposed scheme. Numerical simulations demonstrate that the ASO-FOID controller offers superior performance with only three tuning parameters, similar to a conventional PID controller, while retaining the inherent advantages of fractional-order design such as improved robustness and flexibility.