Because　of　the　strong　nonlinearity　of　the　controlled　system,　it　cannot　be　expressed　in　the　form　of　state　space,　and　the　calculation　of　nonlinear　programming　is　large,　so　it　is　difficult　to　get　the　analytical　solution.　At　present,　it　is　generally　solved　by　numerical　method.　The　GPC　stability　theory　based　on　state　space　analysis　and　internal　model　control　is　difficult　to　be　applied　in　this　case.　In　this　paper,　a　fast　algorithm　of　online　optimization　of　nonlinear　model　predictive　control　based　on　the　field　of　hydropower　installation　is　presented.　A　fast　algorithm　control　system　of　online　optimization　of　nonlinear　model　predictive　control　is　established　and　simulated　by　MATLAB.　The　simulation　results　show　that　the　dimension　of　nonlinear　programming　solution　in　rolling　optimization　is　effectively　reduced　and　the　feasibility　of　MPC　is　improved.　Compared　with　single　machine　and　multi　machine　systems,　the　model　predictive　control　method　is　proved　to　have　strong　ability　of　generator　terminal　voltage　maintenance　and　electromechanical　oscillation　damping.　©　Content　from　this　work　may　be　used　under　the　terms　of　the　Creative　Commons　Attribution　3.0　licence.