To　improve　the　diagnosis　rate,　it　is　proposed　in　this　paper　a　new　consensus　data　fusion　algorithm　and　structural　novelty　detection,　which　is　used　for　structural　health　monitoring　and　novelty　detection,　particularly　for　cases　when　the　important　data　are　missing　due　to　the　facts　of　disabled　sensor　or　transmission　network.　Firstly,　the　incomplete　data　is　supplemented　by　using　of　weighting-average　fusion　algorithm,　and　then　the　effect　of　variable　accuracy　in　measurement　and　environmental　effect　on　the　covariance　in　measurement　is　taken　into　account,　consequently　a　new　consensus　data　fusion　algorithm　is　proposed.　By　integrating　the　new　consensus　fusion　algorithm　and　wavelet　decomposition,　structural　novelty　detection　is　also　implemented.　To　validate　the　proposed　method,　two　numerical　examples　are　simulated　and　a　comparison　is　made　with　other　consensus　fusion　algorithms.　The　results　show　that　the　proposed　method　not　only　effectively　supplements　the　missing　important　data　but　also　has　preferable　novelty　detection　and　anti-noisy　capabilities　and　robustness.　This　implies　that　the　proposed　detection　method　is　feasible　and　effective.