With　the　current　power　system　being　electronicized　nowadays,　synchrophasor　estimation　(SE)　plays　an　increasingly　critical　role　in　the　monitoring,　analysis,　and　protection　of　the　state　of　the　power　grid.　One　proposal　has　been　brought　to　enhance　the　commonly　used　discrete　Fourier　transform　(DFT)　algorithm　that　has　a　large　error　due　to　the　inherent　spectral　leakage　and　fence　effect　in　asynchronous　sampling　by　interpolations.　This　article　proposes　an　SE　algorithm　based　on　an　optimized　combined　cosine　self-convolution　(OCCSC)　window.　Using　the　sidelobe　characteristics　of　the　window　functions　as　the　evaluation　index,　an　optimized　combined　cosine　window　is　obtained　by　the　accelerated　particle　swarm　optimization　(APSO)　algorithm.　The　window　function＇s　self-convolution　is　used　to　further　expand　its　excellent　characteristics.　Based　on　these　characteristics,　combined　with　interpolated　DFT,　simulations　and　experimental　verifications　compared　with　other　algorithms　show　the　effectiveness　of　the　proposed　algorithm.