基于小波变换和支持向量机缆力预测  

Prediction of Mooring Load based on Wavelet Transform and Support Vector Machine

作　　者：邱占芝[1] 吴婷婷 

机构地区：[1]大连交通大学软件学院,辽宁大连116028 [2]大连交通大学电气信息学院,辽宁大连116028

出　　处：《大连交通大学学报》2016年第5期109-112,共4页Journal of Dalian Jiaotong University

基　　金：大连市科技计划资助项目(2014A11GX006)

摘　　要：针对外海环境动力作用下缆力变化时变、非线性的特点,研究了系泊缆力预测方法.利用小波变换将缆力序列分解为不同频段上的低频细节子序列和高频近似子序列,利用支持向量机(SVM)分别对缆力子序列进行回归预测,结合各频段的输出结果得到缆力预测结果.仿真结果表明,小波变换能够反映缆力数据的变化特征,为SVM的学习、预测提供精确的训练样本,基于小波变换和支持向量机的预测方法精度高,优于直接应用SVM预测方法.A prediction of mooring load was studied according to the mooring load variation,which is charactered by time-varying and nonlinear under the offshore dynamic factors. Through the wavelet decomposition and single reconstruction,the original mooring load series is decomposed into a layer of approximation coefficients and several layers of detail coefficients. Each layer of the coefficients is used to regression and forecast by support vector machine （SVM）. After integrating layers of coefficients, the predictive value of the original time series is obtained. The simulation results show that the wavelet transform can reflect the characteristics of the mooring data change, providing accurate samples for SYM learning and prediction. The prediction based on wavelet transform and SVM method is better than the direct application of SYM prediction method in high-precision forecast results.

关 键 词：缆力预测 小波变换 支持向量机 非平稳时间序列 

分 类 号：U664.4[交通运输工程—船舶及航道工程] U653.2[交通运输工程—船舶与海洋工程]