In　this　paper,　a　new　rough-probabilistic　neural　network　(RSPNN)　model,　whereby　rough　set　data　and　a　probabilistic　neural　network　(PNN)　are　integrated,　is　proposed.　This　model　is　used　for　structural　damage　detection,　particularly　for　cases　where　the　measurement　data　has　many　uncertainties.　To　verify　the　proposed　method,　an　example　is　presented　to　identify　both　single　and　multi-damage　case　patterns.　The　effects　of　measurement　noise　and　attribute　reduction　on　the　damage　detection　results　are　also　discussed.　The　results　show　that　the　proposed　model　not　only　has　good　damage　detection　capability　and　noise　tolerance,　but　also　reduces　data　storage　memory　requirements.