The　efficiency　and　dynamic　response　of　air　compressors　are　crucial　for　stability　and　lifespan　of　hydrogen　fuel　cells.　A　predictive　control　scheme　with　surge-　and　choke-constrained　awareness　is　proposed　to　ensure　safe　and　efficient　operation　of　air　compressors　in　this　study.　The　proposed　scheme　consists　of　an　efficiency　enhancement　model　predictive　control　(EE-MPC),　and　an　improved　active　disturbance　rejection　control　(IADRC).　Surge-　and　choke-constrained　awareness　is　achieved　by　comparing　predicted　air　flow　with　surge　and　choke　limitations.　Simultaneously,　the　EE-MPC　is　constrained　with　oxygen　excess　ratio　(OER)　and　obtains　optimal　solution　by　searching　active　set.　The　reference　flow　and　supply　manifold　pressure　trajectories　for　IADRC　are　generated　by　EE-MPC.　A　designed　piecewise　differentiable　nonlinear　smoothing　function　is　embedded　in　IADRC.　The　disturbances　are　estimated　for　coordinating　flow　and　pressure　control.　Under　China　heavy-duty　commercial　vehicle　test　cycle　for　bus　conditions,　root-mean-squared　errors　(RMSEs)　of　flow　and　pressure　are　3.27　g　s-1　and　1.88　x　103　Pa,　respectively,　and　the　mean　efficiency　can　be　enhanced　by　13.4%　compared　to　the　MPC　with　fixed　OER.　Finally,　a　controller　hardware-in-the-loop　test　is　conducted,　with　flow　and　pressure　RMSEs　of　2.48　g　s-1　and　4.28　x　103　Pa　between　the　test　and　simulation,　respectively.　This　study　proposes　a　predictive　control　scheme　with　surge-　and　choke-constrained　awareness　to　guarantee　safety　and　efficiency　of　air　compressors.　The　reference　flow　and　pressure　trajectories　are　formulated　by　efficiency　enhancement　model　predictive　control,　and　further　tracked　by　improved　active　disturbance　rejection　control.　The　proposed　scheme　can　efficiently　improve　fuel　cell　air　compressor　isentropic　efficiency　and　avoid　surge　and　choke.　image