A novel damage evaluation method for timber utility poles based on wavelet packet transform and support vector machine

Publication Type:
Conference Proceeding
Citation:
Proceedings of 7th International Conference on Structural Health Monitoring of Intelligent Infrastructure, 2015
Issue Date:
2015-07-03
Full metadata record
Files in This Item:
In this paper, a novel damage evaluation approach based on wavelet package transform (WPT) and support vector machine (SVM) for the multi-sensor GW-based damage assessment of in-situ timber utility poles. First, WPT is utilized to extract energy features of guided wave signals. Then, to eliminate the multicollinearity between extracted features, principle component analysis (PCA) is adopted and energy features are replaced by a few principle components. Finally, a classifier model base on SVM is constructed to assess the pole condition. To improve the estimation accuracy of the model, particle swarm optimization (PSO) is used to optimize the parameters in SVM. The new method is validated on a number of laboratory timber specimen (undamaged and damaged) that are experimentally tested using an impact hammer for wave excitation and a multi-sensor array is utilised to capture transversal response wave signals. WPT-based energy feature extraction and PCA is subsequently applied to the recorded wave signals, and the health condition of the timber specimen is identified by the pre-trained classifier. The experimental results verify that the proposed method is effective achieving a high identification accuracy of up to 95%.
Please use this identifier to cite or link to this item: