The　large-scale　access　of　EVs（Electric　Vehicles）　and　distributed　energy　has　posed　great　challenges　to　the　safe　and　economic　operation　of　distribution　networks.　Considering　that　HADDN（Hybrid　AC／DC　Distribution　Network）　has　the　characteristics　of　high　flexibility　and　controllability　and　significant　partitioning　features，a　distributed　cooperative　optimization　method　of　HADDN　active　and　reactive　power　considering　the　dual　uncertainties　of　source　side　and　load　side　is　proposed.　Firstly，the　EV　charging　load-wind　power　scenario　generation　method　is　proposed，which　comprehensively　considers　the　traffic　topology　structure，the　residents’travel　behavior　and　the　interval　difference　of　wind　power　prediction　error.　The　optimal　scenario　reduction　algorithm　based　on　Wasserstein　distance　is　used　to　generate　typical　probability　scenarios.　Secondly，the　stochastic　optimization　model　of　HADDN　is　established　to　minimize　the　network　loss　and　node　voltage　deviation.　Then，the　objective　decomposition　and　region　decoupling　of　the　proposed　multi-objective　optimization　model　are　performed，and　the　alternating　direction　method　of　multipliers　based　on　the　objective　value　exchange　principle　is　used　to　solve　the　above　model.　Finally，the　effectiveness　of　the　proposed　method　is　verified　by　a　numerical　example.　©　2022　Electric　Power　Automation　Equipment　Press.　All　rights　reserved.