With　the　continuous　development　of　power　electronics　technology,　the　integration　scale　of　multiple　sources　and　loads　based　on　power　electronic　converters　is　expanding　increasingly　in　the　low-voltage　distribution　system.　Due　to　the　coupling　of　short-circuit　current　mutation　information　at　all　levels　and　branches　on　the　AC　side,　the　traditional　short-circuit　fault　identification　methods　are　no　longer　applicable.　In　this　paper,　a　low-voltage　AC/DC　short-circuit　fault　current　detection　method　based　on　the　Wavelet　Packet　Decomposition　(WPD)　algorithm　is　proposed,　and　the　influences　of　short-circuit　fault　phase　angle　and　permeability　on　the　short-circuit　fault　current　mutation　characteristics　of　the　AC　side　in　the　low-voltage　distribution　system　are　comprehensively　analyzed.　A　novel　AC　short-circuit　fault　identification　method　based　on　the　extreme　Gradient　Boost　(XGBoost)　algorithm　is　proposed.　The　research　results　show　that　this　scheme　can　accurately　identify　all-phaseangle　short-circuit　faults　under　different　permeabilities　in　the　low-voltage　distribution　system.　©　2025　IEEE.