Photovoltaic　power　generation　is　an　effective　way　to　use　solar　energy,　which　is　a　recognized　ideal　renewable　energy　source.　However,　photovoltaic　that　is　susceptible　to　weather　conditions　is　unstable,　and　will　adversely　affect　the　power　grid.　Therefore,　it　is　necessary　to　improve　the　accuracy　of　solar　power　generation.　This　paper　uses　the　LSTM　model　to　predict　solar　power　generation.　At　the　same　time,　the　data　is　reduced　by　using　PCA　to　reduce　the　training　duration　of　the　model　and　improve　the　generalization　ability　of　the　model.　Compared　with　other　models,　simulation　experiment　shows　that　the　LSTM　model　is　better.　©　2018　IEEE.