Aiming　at　the　large　amount　of　data　and　complex　structure　of　the　heating　pipe　network　system,　this　paper　studies　the　fault　diagnosis　method　of　heating　pipe　network　leakage　based　on　deep　belief　network.　Firstly,　using　the　theory　of　graph　theory　to　establish　the　hydraulic　calculation　model　of　the　leakage　condition　of　the　heating　pipe　network,　then　according　to　the　pressure　change　value　of　the　pressure　monitoring　point　in　the　pipe　network;　the　deep　belief　network　is　used　to　establish　the　fault　diagnosis　model　of　the　heating　pipe　network　leakage,　and　the　feasibility　of　the　method　is　verified　by　experiments.　The　experimental　results　show　that　the　proposed　method　has　higher　accuracy　for　the　prediction　of　leakage　pipe　segments,　and　is　superior　to　BP　(Back　Propagation　Neural　Network)　neural　network　and　support　vector　machine　traditional　fault　diagnosis　method.　©　2019　IEEE.