In recent years, proton exchange membrane fuel cell (PEMFC) systems have emerged as a promising sustainable energy source, recognized for their high power density and zero-emission characteristics. However, these systems are highly sensitive to variations in the oxygen-excess ratio (OER), which directly impacts their efficiency and lifespan. Controlling the OER presents significant challenges due to the system’s inherent nonlinearities, model uncertainties, and susceptibility to external disturbances. To address these challenges, we propose an adaptive dynamic surface control with a disturbance observer (ADSCDOB) method to mitigate adverse effects and enhance control accuracy. The disturbance observer is designed using a nonlinear approach to effectively estimate uncertainties and external disturbances. The ADSCDOB method integrates adaptive dynamic surface control technology, enabling robust control while overcoming the "explosion of complexity" problem typical of traditional backstepping methods. The stability of the closed-loop system is rigorously established through theoretical analysis. Simulations validate the effectiveness of the ADSCDOB method, demonstrating its rapid response, minimal error, and robust stability across various operating conditions.