摘要
针对舰船耐波性实测信号中趋势项难以消除而影响滤波精度这一问题,采用经验模态分解方法消除趋势项的影响。根据舰船对波浪运动响应的低频特点,在经验模态分解前先进行低通滤波,可使实测运动信号中的谱峰频率处在相对高频位置,减少EMD迭代次数,并使有用信息包含在第一个IMF中,方便对有用模态的识别。还针对实船耐波性试验无法直接获得垂荡位移的实际问题,对垂向运动加速度联合采用低通滤波、数值积分和EMD去趋势项消除积分误差的方法获得垂荡,通过模型耐波性试验以及对实舰横摇角速度采用该方法求得的横摇与实测横摇的比较,验证了该方法在舰船耐波性实测信号分析中的有效性。
A method based on low-pass signal filtering and empirical mode decomposition was applied to eliminate high-frequency noises and low-frequency trends included in signals measured in seakeeping test. A signal is deposed by low-pass filtering before empirical mode decomposition, which can make the signal＇s peak frequency become relatively high frequency, simplify the iterating process of EMD and make it easy to identify the effective IMF.Owing to the heave of ship can not be directly measured in seakeeping test,a method combined with low-pass filtering,numerical integration and EMD was used to calculate it from the vertical motion acceleration signal. A large number of signals measured in the seakeeping tests of a high speed craft model and a ship were analyzed by this method and the results show this method is effective.
出处
《船舶力学》
EI
北大核心
2009年第5期712-717,共6页
Journal of Ship Mechanics
基金
全国优秀博士学位论文作者专项资金资助项目(200551)
国家"863"计划资助项目(2007AA11Z242)
