An improved Sfa-optimized neuroendocrine PID controller for DC-DC buck-boost converter-inverter driven dc motor in a PV system

This paper introduces an advanced tuning method for neuroendocrine PID (NEPID) controllers in DC-DC buck-boost converter-inverter driving DC motor (DCBBCI-DC motor) systems, powered by photovoltaic (PV) sources. To achieve reliable tuning in this setting, the choice of optimization algorithm is critical. The conventional norm-limited smoothed functional algorithm (NL-SFA) operates in a memoryless manner, which makes it highly susceptible to being trapped in local optima and unable to preserve the best objective function accuracy across iterations. To overcome these limitations, we propose the norm-limited smoothed functional algorithm with remembrance tracking (NL-SFA-RT). This enhanced method introduces a remembrance function alongside a norm-limiting mechanism to stabilize convergence and mitigate the erratic updates observed in the traditional approach. Moreover, the remembrance function enables the algorithm to retain and exploit the best solution from each iteration, thereby improving optimization efficiency and ensuring higher objective function accuracy. The NEPID controller parameters are optimized through the NL-SFA-RT-based method to achieve precise tracking of converter voltage and motor angular velocity under dynamic PV source conditions. A comparative analysis against the safe experimentation dynamics (SED)-based method, adaptive SED (ASED)-based method, and standard NL-SFA-based method demonstrates the proposed method’s superiority in control accuracy, energy efficiency, and computational performance. Statistical evaluations and dynamic response analyses reveal significant reductions in objective function values, total norm errors, and control input energy while maintaining lower computational time. These results validate the proposed method as a practical and efficient tool for real-time control applications in renewable energy systems, particularly for DCBBCI-DC motor system.