Resilient ANFIS Framework for Robust Boost Converter Stability in Real World Conditions

The control of Boost converters under varying operating conditions poses a significant challenge due to nonlinear dynamics and external disturbances, often leading to poor regulation and stability issues with traditional controllers. This study introduces a novel Adaptive Neuro-Fuzzy Inference System (ANFIS) controller for Boost converters, evaluated across an unprecedented 27 diverse test scenarios, including load variations, parameter uncertainties, voltage fluctuations, noise, temperature drifts, switching frequency changes, and combined disturbances. Unlike prior studies with limited test scopes, this work establishes a comprehensive resilience framework, achieving up to 56% lower overshoot and 90% faster settling time compared to open-loop (OLBC) and PI-based closed-loop (CLBC) controllers in 25 of 27 cases (average overshoot: 29.82%, settling time: 0.0105 s). Limitations in extreme transients (210–225% overshoot) are transparently analyzed. Validated via MATLAB Simulink, the ANFIS controller offers robust control for applications like electric vehicles and renewable energy systems, setting a new benchmark for power electronics resilience.