How　to　make　full　use　of　the　redundant　and　complementarg　information　and　thus　assess　on　structural　work　conditions　has　been　a　difficult　problem　for　researchers　home　and　abroad.　In　order　to　efficiently　solve　this　problem,　a　structural　damage　identification　method　based　on　rough　and　probabilistic　neural　network　(PNN)　was　proposed　in　this　paper.　In　this　method,　rough　set　was　used　to　reduce　attributes　so　as　to　decrease　spatial　dimensions　of　data　and　extract　effective　features;　then　PNN　was　utilized　to　process　uncertain　data　and　proceed　probabilistic　reasoning　by　PNN;　and　analysis　and　damage　identification　were　achieved　finally.　To　validate　the　efficiency　of　the　proposed　method,　multi-damage　patterns　from　a　12-story　reinforced　concrete　frame　were　identified,　with　all　IA　above　85%,　and　a　comparison　was　made　between　the　proposed　method　and　a　PNN　classifier　without　data　processing　by　rough　set.　The　results　show　that　the　proposed　method　can　not　only　reduce　data　spatial　dimension,　redundant　attributes　and　uncertainty　but　also　improve　damage　identification　accuracy.