论文部分内容阅读
提出一种利用小波包变换的随机信号功率谱分析的新方法。采用该法首次实现了背景响应与共振响应的分离。通过一个双频率信号的谱分析,考察了该方法的频率检测精度和分辨率。然后将该法应用于一个大跨屋盖风振响应的功率谱计算,检验了该法对随机信号的谱估计精度。计算结果表明:该方法速度快、精度高、分辨率也高,是一种适用于大型屋盖各种响应信号的谱估计算法;脉动响应信号的小波包分解是获得背景和共振响应时程的有效途径。
A new method of power spectrum analysis based on wavelet packet transform is proposed. The method is used for the first time to achieve the background response and resonance response separation. Through the spectral analysis of a dual frequency signal, the frequency detection accuracy and resolution of the method are investigated. Then the method is applied to the power spectrum calculation of wind-induced response of a long-span roof, and the accuracy of the method for estimating the spectrum of stochastic signals is tested. The results show that the proposed method is fast, accurate and has high resolution. It is a spectral estimation algorithm suitable for various kinds of response signals of large roofs. The wavelet packet decomposition of the pulsatile response signal is a method to obtain background and resonant response time-history Effective Ways.