In　order　to　realize　effective　noise　reduction　of　wind　power　under　the　influence　of　complex　power　quality　disturbance(PQD),　an　application　fusion　method　of　improved　smoothing　and　noise　reduction　strategy　and　start-stop　energy　storage　element　is　proposed　in　this　paper.　Firstly,　the　complex　power　quality　disturbance　is　filtered　and　de-noised　based　on　Wiener　filtering(WF)　algorithm　to　obtain　stable　disturbance　output.　Then,　a　filtering　self-built　library(FSBL)　is　established　to　perform　parallel　operations　of　multiple　filtering　algorithms　and　obtain　the　optimal　output　power　according　to　the　constraints　of　time　scale.　After　that,　a　hybrid　energy　storage　system(HESS)　with　start-stop　standby　energy　storage　element　is　proposed.　According　to　the　maximum　entropy　difference　drop　point　and　permutation　entropy　test,　the　start-stop　instructions　of　standby　energy　storage　elements　can　be　achieved.　Finally,　the　state　of　charge(SOC)　is　optimized　according　to　fuzzy　control　constraints.　In　order　to　verify　the　effectiveness　of　the　proposed　method,　this　paper　constructs　the　wind　power　output　based　on　the　hardware　experimental　platform.　The　power　quality　disturbance　database　is　further　constructed　to　verify　and　analyze　power　quality　disturbance　types　from　simple　to　complex.　The　results　show　that　the　proposed　method　is　reliable.　And　the　hybrid　energy　storage　system　with　start-stop　standby　energy　storage　elements　can　be　widely　used　in　many　fields　and　scenarios.