The　uncertainty　of　renewable　distributed　generation　(RDG)　has　a　great　impact　on　the　planning　of　distribution　networks.　To　reduce　this　adverse　effect,　firstly,　the　power　model　of　RDG　with　temporal　interval　is　established　according　to　the　typical　power　output　scenario　of　RDG　in　this　paper.　Besides,　to　improve　convergence　and　iterative　efficiency　of　the　interval　power　flow　algorithm,　a　modified　interval　Krawczyk　iterative　algorithm　is　proposed　by　improving　the　interval　expansion　of　the　iterative　operator　and　descending　the　dimension　of　the　iterative　process.　Furthermore,　a　multi-objective　optimal　configuration　model　of　RDG　in　the　distribution　network　is　established,　to　minimize　the　annual　investment　and　operating　costs　of　RDG,　and　maximize　the　annual　energy　saving　and　emission　reduction,　and　minimize　the　annual　node　voltage　deviation.　Moreover,　a　modified　interval　Krawczyk　iterative　algorithm　is　adopted　to　satisfy　the　voltage　constraints,　and　the　model　is　solved　by　a　multi-objective　genetic　algorithm　with　an　elite　strategy.　Finally,　in　terms　of　the　fuzzy　closeness,　the　optimal　configuration　for　different　applications　is　screened　from　the　obtained　Pareto　optimal　solutions.　The　effectiveness　of　the　proposed　method　and　the　modified　interval　Krawczyk　iterative　algorithm　are　verified　by　simulations　on　the　IEEE-33　bus　system.　©　2021,　Power　System　Technology　Press.　All　right　reserved.