For　GPC　with　constraints,　these　shortcomings　are　made　up　on　the　basis　of　nonlinear　programming　algorithm.　Because　of　the　strong　local　search　ability　of　nonlinear　programming　and　the　strong　global　search　ability　of　the　algorithm,　combined　with　the　advantages　of　the　algorithm　and　introduced　into　the　rolling　optimization　process　of　GPC,　the　optimal　control　law　is　obtained.　In　this　paper,　a　constrained　generalized　predictive　identification　algorithm　based　on　the　control　field　of　hydropower　industry　is　proposed,　and　a　model　of　the　constrained　generalized　predictive　identification　algorithm　based　on　the　control　field　of　hydropower　industry　is　established.　The　simulation　results　verify　the　applicability　and　effectiveness　of　the　combined　algorithm　in　the　study　of　GPC,　improve　the　accuracy　and　speed　of　the　optimization　results,　and　obtain　satisfactory　control　effect.　©　Published　under　licence　by　IOP　Publishing　Ltd.