中文摘要：

摘要：风电机组齿轮箱振动信号的在线频谱分析对信号处理算法的快速性要求很高，提出采用稀疏快速傅里叶变换（SFFT）算法进行风机齿轮箱的频谱分析。SFFT算法主要利用窗函数过滤信号，然后散列傅里叶系数，最后进行定位与估值运算，能快速地计算出信号频谱中k（信号的稀疏度）个拥有最大值的傅里叶系数。该算法结构简单，运行时间相对于信号长度n呈亚线性。通过对风电机组齿轮箱的实际振动信号分析，验证了SFFT算法较之FFT算法运行速度快，非常适合振动信号的在线频谱分析。

英文摘要：

The on-line spectrum analysis for vibration signal of wind turbine gearbox demands fast algorithms. In order to improve the speed of the algorithm, the spectrum analysis with Sparse Fast Fourier Transform （SFFT） algorithm is proposed in this paper. First, the SFFT algorithm is used to leverage the window function to filter vibration signal. Then, after the Fourier coefficients being hashed, the k largest （in magnitude） coefficients of the Fourier Transform of vibration signal can be estimated by locating and estimating methods. The SFFT algorithm has a simple structure and a sub-linear runtime in the signal size n. It has been verified through analyzing the vibration signal of wind turbine gearbox that the SFFT algorithm has a faster performance than FFT algorithm in on-line spectrum analysis.