The　integration　of　high　proportion　renewable　energy　aggravates　the　net　load　fluctuation　and　affects　the　safe　and　stable　operation　of　the　distribution　network.　In　this　paper,　a　distributed　hybrid　energy　storage　capacity　optimization　configuration　method　based　on　double-layer　cluster　division　is　proposed　to　suppress　closest　net　load　fluctuation.　Firstly,　cluster　division　based　on　the　integrated　active/reactive-voltage　sensitivity　is　performed　to　initially　determine　the　hybrid　energy　storage　action　region.　According　to　the　result　of　partition,　the　power　flow　propagation　index　is　constructed　by　combining　the　power　flow　tracing　algorithm　and　the　topology　of　the　distribution　network.　Secondly,　by　setting　the　power　flow　propagation　index　threshold　to　divide　sub-clusters,　the　access　location　and　action　region　of　hybrid　energy　storage　are　determined.　Then,　based　on　different　power　flow　propagation　index　thresholds,　the　net　load　of　the　sub-cluster　is　decomposed　using　variational　modal　decomposition　(VMD)　to　determine　the　reference　power　of　energy-type　and　power-type　energy　storage.　The　NSGA-II　algorithm　is　used　to　solve　the　hybrid　energy　storage　capacity　optimization　configuration　model　with　the　rated　capacity　of　energy　storage　as　the　optimization　variables.　Finally,　corresponding　to　different　power　flow　propagation　index　thresholds,　the　optimal　configuration　scheme　is　determined　by　ranking　the　annual　comprehensive　configuration　cost　of　hybrid　energy　storage　and　the　tie　line　power　of　bordering　clusters.　Test　carried　out　on　the　IEEE　33-node　distribution　network　system　have　verified　the　effectiveness　and　feasibility　of　the　proposed　method.　©　2023　IEEE.