Regulating　the　air　supply　is　crucial　for　high　efficiency　and　reliable　operation　of　proton　exchange　membrane　fuel　cell　systems　(PEMFCs).　In　this　study,　a　fused　multi-model　predictive　control　(FM-MPC)　with　an　adaptive　compensation　is　proposed　for　the　oxygen　excess　ratio　(OER)　regulation　of　the　air　supply　system.　The　FM-MPC　is　designed　based　on　the　linearized　PEMFC　model　at　low　and　high　power　phases,　with　two　linear　MPCs　imple-mented　and　combined　using　adaptive　featured　weights.　An　adaptive　compensation　strategy　is　created　to　address　the　imbalance　of　the　two　MPCs　and　external　load　disturbances.　The　stability　of　the　proposed　control　is　analyzed　using　Lyapunov＇s　second　law.　Simulation　results　demonstrate　that　the　proposed　method　exhibits　less　overshoot　and　faster　response　than　conventional　MPCs,　with　the　OER　total　sum-of-squares　error　(TSSE)　reduced　by　59.4%　and　87.7%　for　New　European　Driving　Cycle　(NEDC)　and　Urban　Dynamometer　Driving　Schedule　(UDDS)　con-ditions,　respectively.　Finally,　a　Hardware-In-the-Loop　(HIL)　experiment　verifies　the　real-time　application　po-tential　of　the　proposed　controller,　with　a　mean　relative　error　(MRE)　of　1.12%　between　experiment　and　simulation.