Damage Identification of Concrete Arch Dams Based on Wavelet Packets and Neural Networks

A dam may be damaged by occasional extreme loads such as major earthquakes or terrorist attacks during its service. According to the needs of emergency assessment, this paper studies a rapid damage identification method for location and degree in concrete arch dams which is based on dynamic characteristics data, using wavelet transform, packet decomposition, BP neural network D-S evidence theory related experimental verification. The results show that relative difference curvature mode (δφk), coefficient (Wfk) energy (δKk) can effectively identify position dam, δφk first four modalities has best overall recognition effect; Wfk requires high number measurement points, should at least 64 close possible; δKk better effect than two same points. significantly improves reduces misjudgment single-damage method. trained data one measuring point when there single instance, points no fewer double damage. test verify feasibility paper, provide theoretical basis post-disaster assessment information system dams.