Adaptive Objective Function Design for Genetic Algorithm-Based PID Control of Brushed DC Motors

Abstract

This study presents an advanced methodology for improving the performance of brushed DC motors through an adaptive objective function design for Genetic Algorithm (GA)-based PID controllers. Conventional GA-PID tuning methods often rely on static objective functions, which limit their ability to simultaneously optimize multiple performance criteria, including rise time, overshoot, settling time, steady-state error, and energy consumption. To address these limitations, the proposed framework introduces a dynamic weighting mechanism within the GA fitness function, allowing the controller to prioritize different performance objectives adaptively during various phases of motor operation. The research employs both simulation and real-time hardware validation using a 24V brushed DC motor controlled via FPGA, ensuring that theoretical results translate effectively into practical applications. Simulation results demonstrate that the adaptive GA-PID significantly outperforms conventional GA-PID controllers, reducing rise time by approximately 43%, overshoot by nearly 68%, and settling time by 44%, while maintaining steady-state errors below 0.3%. Energy efficiency also improved by 6–7% due to optimized torque allocation. Hardware experiments corroborate these findings, showing robust speed regulation under varying load torques from 0.05 to 0.6 Nm and consistent real-time adaptation of PID parameters without destabilizing the system. Comparative analysis with literature reveals that the adaptive objective function approach effectively integrates multi-objective optimization in a computationally efficient manner, avoiding the complexities of fuzzy logic or multi-heuristic GA methods while maintaining high precision and robustness. The framework's practical implications extend to industrial automation, robotics, and precision electromechanical actuation, where both performance and energy efficiency are critical. Limitations related to computational overhead and adaptation rate tuning are addressed, and strategies for mitigating these challenges in resource-constrained systems are discussed. Overall, this research provides a novel, systematic, and experimentally validated approach to GA-based PID optimization, establishing a foundation for future work in adaptive control, predictive optimization, and multi-motor coordinated systems. The methodology demonstrates that dynamic fitness adaptation not only improves transient and steady-state performance but also enables energy-conscious, scalable, and robust control of electromechanical systems, thereby contributing to both theoretical advancements and practical engineering solutions.