Adaptive Fault-Tolerant PID Control Synthesis for Buck Converters via Multi-Objective Genetic Algorithm Optimization

This paper presents a Genetic Algorithm (GA)-optimized PID control strat egy for DC–DC buck converters, developed to ensure robust voltage regulation under multiple fault and degradation conditions. Conventional PID controllers, typically tuned using heuristic methods, suffer from poor adaptability when passive components such as inductors and capacitors undergo aging or ther mal stress. The proposed GA-based tuning framework overcomes this limitation by optimizing the proportional (K _p ), integral (K _i ), and derivative (K _d ) gains using a multi-scenario fitness formulation that minimizes the Integral Absolute Error (IAE) and overshoot across diverse operating conditions. The methodol ogy explicitly accounts for simultaneous degradation of both the inductor and capacitor—a dual-fault scenario often neglected in literature—which significantly alters converter dynamics and stability margins. High-fidelity PLECS simula tions are carried out considering practical non-idealities including equivalent series resistance (ESR), diode voltage drops, switching dead-time, and measure ment noise. Stability is verified through eigenvalue and Lyapunov analyses of the closed-loop system. Simulation results confirm that the GA-optimized PID controller provides superior performance with minimal overshoot (< 5%), rapid settling time (< 0.6 ms), and effective rejection of load, input, and compo nent disturbances. The proposed approach offers a computationally efficient and hardware-compatible framework for robust converter control in safety-critical applications such as renewable energy systems and electric vehicles.