Performance Optimization of BLDC Motor Control Using Sand Cat Swarm Algorithm and Linear Quadratic Regulator

Mar 1, 2025

Journal Journal of Robotics and Control (JRC)

DOI DOI: 10.18196/jrc.v6i1.24958

Issue 1

Volume 6

Brushless Direct Current (BLDC) motors are widely utilized in industrial applications due to their precision, efficiency, and ease of control. This study optimizes BLDC motor performance by enhancing the linear quadratic regulator (LQR) using the Matlab program's Sand Cat Swarm Optimization (SCSO) algorithm. The research evaluates key performance metrics, including settling time, overshoot, and cost function, to demonstrate the advantages of the proposed approach. Additionally, a comparative analysis was conducted using the butterfly optimization algorithm (BOA) and conventional LQR to validate the superiority of SCSO. Simulation results show that the LQR-SCSO method significantly improves performance, achieving a 77.2% reduction in settling time, a 91% reduction in overshoot, and a cost function of 0.3376. In comparison, the BOA method achieves reductions of 68.54% in settling time, 67.37% in overshoot, and a cost function of 0.8736, while the conventional LQR achieves reductions of 68% in settling time, 62.3% in overshoot, and a cost function of 1.8393. SCSO has excellent convergence and adaptability; however, the implementation is explored further in terms of computational cost adopted for industrial use in real time. The data are so highly processed that better controls are implemented to repeat simulations across defined parameters. The proposed LQR-SCSO approach is practical and potent in enhancing motor performance, which is a significant advancement and can applied in various fields in the industry, such as robotics and automated systems. However, the proposed method may face obstacles related to the higher computational complexity of higher-order applications, which can be a subject of future studies.

Adaptive control of a DC servo motor using particle swarm and gray wolf optimization algorithms

Mar 5, 2025

Journal AIP Conference Proceedings

publisher AIP Publishing

DOI https://doi.org/10.1063/5.0260224