Towards adaptive and efficient power electronics in industry 4.0: Fuzzy logic control of single-ended primary-inductor converter systems

Abstract

Reliable voltage regulation under source and load variations is essential for embedded power converters used in Industry 4.0 and renewable-energy systems. This study presents the design, fabrication, and common-platform comparison of a Proportional–Integral–Derivative (PID) controller and a fixed-rule Mamdani Fuzzy Logic Controller (FLC) for a Single-Ended Primary-Inductor Converter (SEPIC). Both controllers were developed in MATLAB/Simulink and implemented on the same Arduino Mega platform using a 50 kHz pulse-width-modulation signal. Their performances were evaluated through simulation and experiments involving reference-voltage, load-resistance, and input-voltage variations. The results reveal a controller-dependent trade-off rather than universal superiority. At a 7.5 V reference, the FLC reduced the absolute steady-state error from 0.096 to 0.003 V and the recovery interval from 640 to 590 ms, although its ripple increased from 0.68% to 1.25%. For a 100–80 Ω load transition, the FLC shortened recovery from 500 to 350 ms but increased the maximum output voltage from 18.0 to 21.7 V. Under input-voltage variations, the PID recovered within approximately 100 ms, whereas the FLC required 500–610 ms. These findings show that the PID provides better transient damping, while the FLC generally improves steady-state tracking accuracy. The study establishes an experimental basis for selecting or combining both controllers in embedded SEPIC applications.