Bridge monitoring abnormal data identification method and system based on fusion sequence features

The invention discloses a fusion sequence feature-based bridge monitoring abnormal data identification method and system, and the method comprises the steps: obtaining bridge health monitoring data, extracting a feature vector of the monitoring data as a data point, constructing a data abnormity identification tree, setting a segmentation threshold value, comparing the feature vector with the segmentation threshold value, and obtaining a data abnormity identification result. According to a comparison result, respectively dividing the feature vectors into a first sub-tree and a second sub-tree; according to a data exception identification tree, calculating an initial exception score of the data, calculating the total probability density of the initial exception score of the data, setting a data exception model, according to the total probability density, calculating a final exception score of the data, and according to the final exception score of the data, identifying exception data in the monitoring data; an