As　people＇s　living　standards　improve,　the　demand　for　resources　is　also　increasing,　especially　power　resources.　This　also　puts　high　demands　on　the　power　sector:　how　to　save　power　generation　costs　while　effectively　meeting　people＇s　electricity　demand.　This　paper　takes　the　gas-steam　combined　cycle　power　plant　(CCPP)　as　the　research　object　and　uses　the　machine　learning　method　to　analyze　the　historical　data　of　the　power　plant　to　find　out　the　impact　of　the　environment　on　the　power　generation　efficiency.　And　establish　a　machine　learning　model　to　predict　the　net　power　generated　by　the　power　plant　to　help　its　intelligent　work.　The　experimental　results　show　that　the　machine　learning　model　established　in　this　paper　can　effectively　predict　the　net　electricity　generated,　and　at　the　same　time　find　the　main　factors　affecting　power　generation,　which　can　promote　the　improved　production　of　power　plants.　©　2020　IEEE.