共轭梯度算法在振动信号处理中的应用

为了提高液体火箭发动机试验振动信号频域数据处理的精度,提出了一种基于共轭梯度和AR模型的谱估计算法。该算法计算复杂度低,估计出的谱分辨率高,可以克服传统的经典傅里叶变换功率谱估计算法在信号信噪比降低时不能有效区分相近频率点谱线的问题,解决了传统算法旁瓣泄漏严重的固有缺点。通过对算法在不同信噪比条件下的仿真实验分析与真实试验数据验证,充分表明了此算法在低信噪比条件下,估计的谱仍具有高分辨率的特点。

Application of conjugate gradient algorithm in vibration signal processing

In order to improve the accuracy of frequency-domain data processing of vibration signal in liquid rocket engine test,a spectral estimation algorithm based on conjugate gradient and AR model is proposed.The algorithm has low computational complexity and high estimated spectral resolution,which can overcome the problem that the traditional Fourier transform power spectrum estimation algorithm cannot effectively distinguish the spectral lines of similar frequency points when the signal-to-noise ratio is reduced.In addition,it also solves the inherent shortcoming of the traditional algorithm with serious side leakage.Based on the simulation analysis and test data verification of the algorithm under different signal-to-noise ratio conditions,the estimated spectrum of this algorithm still has the characteristics of high resolution under low signal-to-noise ratio conditions.