In　this　paper,　we　applied　a　decentralized　model　predictive　control　(DMPC)　scheme　on　the　Polymer　electrolyte　membrane　fuel　cell　(PEMFC)　system.　The　DMPC　controller　is　used　to　realize　the　tracking　of　the　maximum　PEMFC　net　output　power.　The　scheme　of　DMPC　is　adapt　from　Alessio　et　al.　(2011).　In　the　proposed　DMPC　control　strategy,　two　DMPC　controllers　are　designed　to　manipulate　two　key　component　of　the　PEMFC　system,　which　are　the　compressor　and　the　stack　cathode.　Each　DMPC　controller　receives　information　from　the　local　states,　and　calculate　the　optimal　system　input　using　an　explicit　feedback　control　law.　The　simulation　results　show　the　DMPC　control　scheme　can　successfully　manipulate　the　PEMFC　system　to　the　optimal　working　condition,　with　acceptable　performance　loss　compared　to　centralized　model　predictive　control　(CMPC)　scheme.　(C)　2018,　IFAC　(International　Federation　of　Automatic　Control)　Hosting　by　Elsevier　Ltd.　All　rights　reserved.