Hilbert-Huang变换在大坝监测资料分析中的应用

运用Hilbert-Huang变换（HHT）从时频域的角度对大坝监测资料进行了分析。通过经验模态分解（empirical mode decomposition,EMD）,测值序列被分解为以不同频率波动的内在模态分量（IMF）和代表序列变化趋势性的余量,其分解所得的各IMF分量的变化具有调幅调频特征,余量则可以反映测值趋势性变化的局部特征。效应量和环境量测值的Hilbert谱存在一定的相似性,测值变化的调幅调频特征反映了大坝、基岩和周围环境组成的复杂系统的非线性和非平稳性。

Application of Hilbert-Huang Transform in Analysis on Monitoring Data of Dam

By using the Hilbert-Huang transform(HHT),a newly developed time-frequency analysis method for nonlinear and nonstationary time series is proposed,the monitoring data is analyzed in time-frequency domain.The monitoring data is decomposed into several intrinsic mode function(IMF) components fluctuating at different frequencies and a residual representing the trend of the data through the empirical mode decomposition(EMD).IMF components exhibit amplitude modulation and frequency modulation.The residual can reflect the local feature of the trend.The Hilbert spectrums of the monitoring data of environment variables and effect variables are similar.The amplitude modulation and frequency modulation of monitoring data reflect the nonlinearity and nonstationarity of the complex system composed of dam,rock and environments.