The　complex　game　among　the　multi-market　stakeholders　and　the　uncertainty　of　high-permeable　renewable　energy　have　brought　great　challenges　to　the　planning　and　operation　of　the　distribution　systems.　Firstly,　since　the　micro-grid　and　the　distribution　network　belong　to　different　stakeholders,　we　present　a　bi-level　dispatch　model　for　the　active　distribution　network　with　multi-microgrids,　in　which　the　upper　model　and　the　lower　model　solve　the　optimization　problems　of　the　distribution　network　and　the　micro-grid　respectively.　In　addition,　to　handle　the　renewable　energy　uncertainty,　a　two-stage　stochastic　optimization　method　of　a　day-ahead　and　real-time　scheduling　based　on　the　above　bi-level　model　framework　is　adopted,　and　the　time-of-use　electricity　price　mechanism　is　introduced　to　fully　coordinate　the　energy　management　decisions　of　each　market　stakeholder.　Then,　the　two-layer　model　being　transformed　into　a　mix-integer　linear　programming　problem　with　the　Karush-Kuhn-Tucker　(KKT)　condition　and　the　second-order　cone　(SOC)　relaxation　technique,　the　optimal　scheduling　scheme　under　the　extreme　scenario　set　can　be　obtained.　Finally,　taking　the　IEEE-33　node　distribution　network　system　incorporated　in　the　microgrid　group　as　an　example,　it　is　verified　that　the　proposed　method　effectively　achieves　the　income　balance　of　the　distribution　network　and　the　micro-grid　group　considering　renewable　energy　uncertainty,　and　it　is　conducive　to　the　safe　and　stable　operation　of　the　distribution　network.　©　2021,　Power　System　Technology　Press.　All　right　reserved.