Aiming　at　the　plant　with　a　time　delay,　a　new　control　strategy　for　internal　model　control　was　proposed　to　set　up　an　inverse　model　of　the　minimal　phase　subsystem　by　using　extreme　learning　machine　with　in-out　system　datum.　Moreover,　the　stable　error　for　internal　model　control　system　with　a　delay　unit　was　proposed　to　evaluate　the　system　performance.　The　internal　model　control　system　based　on　extreme　learning　machine　was　employed　to　the　plant　with　a　delay　and　a　continuous　stirred　tank　reactor　respectively.　The　experimental　results　indicate　that　the　internal　model　control　system　based　on　extreme　learning　machine　is　realizable　with　a　small　stable　error　and　strong　robustness.　©　Copyright.