针对规则波中船模波浪力测量信号非线性非平稳的特点，提出基于互补集合经验模态分解（CEEMD）的信号分析方法来降低非平稳性并准确提取各阶波浪力组分。首先，对实测信号低通滤波预处理以过滤高频噪声；其次，对预处理后的信号进行CEEMD分析并获得不同频段的本征模态函数（IMF）；然后，求取各IMF的归一化自相关函数及其方差，设定方差阈值，根据阈值判定随机噪声占主导的噪声模态和有效模态；最后，采用FFT-FS算法计算各IMF的特征频率，根据各特征频率依次确定波浪力的各阶组分。通过对实测信号的分析表明该方法能够有效地提取信号中的各阶波浪力组分。
Aiming at how to reduce the impact of non-stationary on the test signals of non-linear wave exciting forces on a model ship, a signal analysis method based on complementary ensemble empirical mode decomposition (CEEMD) is proposed, to make extraction accuracy of each order wave force component possible. First of all, the non-linear wave exciting force signal is preprocessed by low-pass filtering method to eliminate high frequency noises before CEEMD. Secondly, CEEMD signal processing method is used for the signal obtained in the first step, and different frequency bands of intrinsic mode functions (IMFs) can be adaptively obtained according to the signal characteristics. Thirdly, normalized autocorrelation functions together with their variances for the IMFs are evaluated, and a variance threshold is chosen to identify the dominant modes of the random noises and valid modes of the signal from the IMFs. Finally, characteristic frequency of each valid mode is obtained by using the FFT-FS method, and this frequency can help to determine the relative mode belonging to which order wave force component. Test signals of non-linear wave exciting force on a model ship measured in an experiment have been analyzed by using the analysis method presented above, and the results show it is effective and can be applied to extract each order wave force component from a measured signal.
2018,40(9): 6-12 收稿日期：2018-05-14
DOI：10.3404/j.issn.1672-7649.2018.09.002
分类号：U661.32
基金项目：教育部重点实验室开放基金资助项目（2015121201）
作者简介：王启兴(1968-),男,高级工程师,主要从事舰船总体性能及舰船建造合同履行监督管理研究
参考文献：
[1] 魏泽, 赵战华, 刘家瑞. 二阶波浪力数值计算与试验方法研究[J]. 船舶力学, 2017(s1):251-257
[2] 刘应中, 缪国平. 二维物体上的二阶波浪力[J]. 中国造船, 1985(3):3-16
[3] 许勇, 欧勇鹏, 董文才. 基于低通滤波和经验模态分解的舰船耐波性试验信号分析方法研究[J]. 船舶力学, 2009, 13(5):712-717
[4] HUANG N E, SHEN Z, LONG S R, et al. The empirical mode decomposition method and the Hibeert Spectrum for non-stationary time series analysis[J]. Proc. Royal. Soc. London A, 1998, 454(A):903-995
[5] 钱荣荣. 基于经验模态分解的动态变形数据分析模型研究[D]. 徐州:中国矿业大学, 2016.
[6] WU Zhao-hua, HUANG N E. Enstemble empirical mode decomposition:a noise-assisted data analysis method[J]. Advances in Adaptive Data Analysis, 2011, 1(1):1-41
[7] YEH J R, SHIEH J S, HUANG N E. Complementary enstemble empirical mode decomposition:a noval noise enhanced data analysis method[J]. Advances in Adaptive Data Analysis, 2010(2):135-156
[8] 张守成, 张玉洁, 刘海生. 一种改进的EMD硬阈值去噪算法[J]. 计算机测量与控制, 2014, 22(11):3659-3661
[9] 王亚萍, 匡宇麒, 葛江华, 等. CEEMD和小波半软阈值相结合的滚动轴承降噪[J]. 振动、测试与诊断, 2018, 38(1):80-86
[10] 许勇, 董文才, 欧勇鹏. 基于FFT-FS频谱细化技术的船模耐波性试验测量信号分析方法研究[J]. 船舶力学, 2012, 16(5):497-503
[11] 杨涛, 乐友喜, 曾贤德, 等. 基于自相关函数特性的CEEMD全局阈值去噪方法研究[EB/OL]. http://kns.cnki.net/kcms/detail/11.2982.P.20180124.1129.032.html.
[12] XU Yong, DONG Wen-cai. Numerical study on wave loads and motions of two ships advancing in waves by using 3-D translating-pulsating source[J]. Acta Mechanica Sinica, 2013, 29(4):494-502
[13] 许勇. 波浪中近距离航行的多船水动力干扰机理研究[D]. 武汉:海军工程大学, 2012.
