摘要
为满足近海导管架平台实时监测和损伤识别的需求,提出一种基于小波包分析、主成分分析与支持向量机的损伤识别方法。对平台甲板上安装的加速度传感器信号进行小波包变换,得到信号的敏感频带特征;利用敏感频带中的小波包系数进行主成分分析,剔除其中对波高敏感的主体参数,得到对损伤敏感的残差参数;将残差参数输入支持向量机进行训练,实现对损伤信号的识别。将该方法应用于渤海某平台的损伤识别中,结果表明该方法对海洋平台信号的识别具有较高的精度。
For the implementation of real time monitoring and damage identification of offshore platform,a damage identification method based on wavelet analysis,principal component analysis and Support Vector Machine(SVM)is proposed.The wavelet packet transform of the acceleration signal of sensors installed on the platform deck is performed to obtain the sensitive frequency band characteristics of the signal.The wavelet packet coefficients in the sensitive frequency band are used for principal component analysis,the main parameters sensitive to the wave height are eliminated,and the residual parameters sensitive to the damage are reserved.The residual parameters are input into the SVM to realize the identification of the damage signal.The method is applied to the damage identification of a platform in Bohai Sea.The results show that the method achieves high accuracy for the identification of the platform signals.
作者
徐鑫
黄焱
田育丰
XU Xin;HUANG Yan;TIAN Yufeng(State Key Laboratory of Hydraulic Engineering Simulation and Safety,Tianjin University,Tianjin 300072,China)
出处
《中国海洋平台》
2019年第6期23-30,39,共9页
China offshore Platform
基金
国家科技重大专项(编号：2016ZX05058-004-002)
关键词
海洋平台
损伤识别
小波包
主成分分析
支持向量机
offshore platform
damage identification
wavelet packet
principal component analysis
Support Vector Machine(SVM)
