摘要
模态参数是体现海上风电结构运行安全状态的关键指标,然而复杂多变的海洋环境会导致实测振动信号中混有大量低频高能噪声,严重影响模态识别精度。为实现海上风电结构模态参数的准确识别,提出一种能够抵抗低频高能噪声影响的模态参数识别方法(CEEMDAN-VMD-SSI,CVS)。首先,利用完全自适应噪声集合经验模态分解法(complementary ensemble empirical mode decomposition with adaptive noise, CEEMDAN)滤除原始信号中的高频噪声;随后,通过麻雀优化算法(sparrow’s optimization algorithm, SSA)以最小包络熵作为适应度函数迭代计算自适应确定变分模态分解法(variational mode decomposition, VMD)的信号分解层数K和惩罚因子α,实现信号的VMD自适应优化分解以剔除低频高能噪声影响;最后,再采用随机子空间方法实现信号中模态参数的识别提取。研究分别针对构造仿真含噪信号和原型观测信号开展了识别效果对比验证。结果表明:相比于传统模态识别方法,CVS方法在信噪比、波形相似系数、相对误差等参数方面具有更好的有效性和精确性;同时,该方法对实测信号的处理能力强,降噪效果好,能够准确识别结构固有频率、叶轮转动频率(1P)和叶片扫掠频率(3P),具有良好的工程适用性,为后续基于实测数据开展海上风电结构模态参数识别与运行安全评价提供了新思路。
Modal parameters are key indexes to reflect safe operation status of offshore wind power structures.However,complex and ever-changing marine environment can cause a large amount of low-frequency high-energy noise mixed in measured vibration signals to seriously affect the accuracy of modal identification.Here,to realize correct identification of modal parameters of offshore wind power structures,a modal parametric identification method was proposed to resist effects of low-frequency high-energy noise.Firstly,the complete ensemble empirical mode decomposition with adaptive noise(CEEMDAN)method was used to filter out high-frequency noise in original signals.Then,the sparrow optimization algorithm(SOA)was used to iteratively calculate signal decomposition layer number K and penalty factorαof the adaptive deterministic variational mode decomposition(VMD)method taking the minimum envelope entropy as the fitness function,and VMD adaptive optimization decomposition of signals was realized to eliminate effects of low-frequency high-energy noise.Finally,the stochastic subspace identification(SSI)method was used to identify and extract modal parameters in signals.The contrastive verification of recognition effects was performed for simulated noisy signals and prototype observation signals.The results showed that compared with traditional modal recognition method,the proposed method has better effectiveness and accuracy in parameters of signal-to-noise ratio,waveform similarity coefficient and relative error;meanwhile,the proposed method has strong processing ability for measured signals and good denoising effect,and it can accurately identify structure natural frequencies,impeller rotating frequency of 1 P and blade sweeping frequency of 3 P;the proposed method has good engineering applicability and it can provide new ideas for subsequent modal parametric identification and operational safety evaluation of offshore wind power structures based on measured data.
作者
董霄峰
时泽坤
彭泓浩
DONG Xiaofeng;SHI Zekun;PENG Honghao(State Key Lab of Hydraulic Engineering Intelligent Construction and Operation,Tianjin University,Tianjin 300350,China;School of Civil Engineering,Tianjin University,Tianjin 300350,China)
出处
《振动与冲击》
北大核心
2025年第9期214-222,265,共10页
Journal of Vibration and Shock
基金
国家自然科学基金资助项目(52471297)。
