Hits:

Indexed by:期刊论文

Date of Publication:2018-07-01

Journal:APPLIED OCEAN RESEARCH

Included Journals:SCIE、Scopus

Volume:76

Page Number:139-147

ISSN No.:0141-1187

Key Words:Damage identification; Random decrement signature; Autocorrelation function; Soft yoke single point mooring (SPM) tower system; Support Vector Machine (SVM)

Abstract:In this research, to identify the damage of the nonlinearity system under ambient loads, an intelligent damage identification method based on long-term monitoring data is proposed. The random decrement technique and the autocorrelation function algorithm are used to extract free decay of the structure from long-term monitoring data. The random decrement signatures, autocorrelation function, the frequency of free response and the peak points of the frequency spectrum are used as the features of the structure. These features are then input into the Support Vector Machine (SVM) to classify the current state of the system and their identification accuracy is compared. The simulation experiments results show the extracted features are capable of representing the changes of the system inherent characteristics. Finally, the proposed method is applied to the data analysis of the soft yoke single point mooring (SPM) tower system, and provide the reference for the damage identification of the soft yoke SPM tower system.