The　large-scale　integration　of　electric　vehicles　and　distributed　power　sources　will　further　exacerbate　the　peak　valley　load　difference　in　power　systems,　which　has　posed　a　serious　challenge　to　the　stability　and　economic　operation　of　the　hybrid　AC/DC　distribution　network.　A　hierarchical　decentralized　coordinated　scheduling　framework　for　the　hybrid　AC/DC　distribution　network　coupled　with　the　photovoltaic-storage-charging　integrated　stations　considering　multiple　uncertainties　is　firstly　proposed.　In　this　framework,　a　multi-objective　optimization　model　with　the　pursuit　of　the　minimum　day-ahead　operation　cost　and　peak　valley　difference　of　net　load　is　presented　for　the　distribution　network,　which　is　solved　by　the　normalized　normal　constraint　method.　On　the　other　hand,　a　two-stage　robust　optimization　model　considering　a　virtual　battery　model　and　source-load　uncertainty　set　is　established　for　the　integrated　station,　which　is　solved　by　the　column-and-constraint　generation　algorithm.　Furthermore,　an　improved　alternating　direction　method　of　multipliers　is　developed　to　solve　the　above　decentralized　optimization　problem.　Finally,　the　proposed　model　and　solving　method　are　validated　via　a　modified　33-bus　hybrid　AC/DC　distribution　network　coupled　with　3　integrated　stations.　Simulation　results　demonstrate　that　the　multi-objective　scheduling　strategy　can　achieve　prominent　effects　of　peak　shaving　and　valley　filling　compared　with　the　single-objective　strategy.　Specifically,　the　peak　valley　difference　of　net　load　has　reduced　by　70.1　%.　Moreover,　the　coordination　operation　of　various　controllable　devices　can　improve　the　operation　flexibility　of　power　systems　and　alleviate　the　influence　of　the　multiple　uncertainties.　In　addition,　the　solving　time　of　the　proposed　method　has　shortened　by　39.9　%　compared　with　the　traditional　algorithm.