The　large-scale　integration　of　renewable　energy　results　in　frequent　changes　to　the　topology　of　the　distribution　network.　The　change　in　the　distribution　network　topology　requires　the　reconstruction　of　the　state　estimation　model,　which　will　significantly　impair　the　efficiency　of　the　state　estimation.　Moreover,　the　fusion　of　multi-source　measurement　data　brings　compatibility　challenges,　affecting　estimation　accuracy.　Therefore,　this　paper　proposes　an　interval　dynamic　state　estimation　approach　for　distribution　networks　based　on　the　fusion　of　partition　and　multi-source　measurement　data.　Firstly,　a　strategy　for　fusing　multi-source　measurement　data　is　proposed,　and　the　deviations　caused　by　asynchronous　sampling　times　are　considered.　The　distribution　network　is　divided　into　multiple　subareas　based　on　topology　information　interaction.　In　the　event　of　a　change　in　the　topology　of　the　distribution　network,　only　the　state　estimation　model　for　the　relevant　subarea　is　reconstructed,　after　which　the　topological　information　of　the　subarea　is　exchanged,　and　the　equivalent　injected　power　is　updated　following　the　new　topology.　This　process　ensures　the　accuracy　of　state　estimation　and　enhances　the　efficiency　of　the　estimation　procedure.　Then,　an　interval　dynamic　state　estimation　model　is　enveloped　to　account　for　the　uncertainty　of　non-synchronous　deviation.　The　improved　interval　Kalman　filter　algorithm　is　used　to　solve　the　model.　The　expansion　of　intervals　during　the　solution　process　is　suppressed　by　the　linearization　of　the　measurement　function　and　upper-bound　optimization.　Finally,　the　proposed　method　is　validated　using　simulations　on　the　IEEE　119-bus　distribution　system,　demonstrating　its　feasibility　and　effectiveness.　©　2025　Power　System　Technology　Press.　All　rights　reserved.