基于分数阶傅里叶变换的邻近阶比分离研究

提出了一种基于分数阶傅里叶变换（Fractional Fourier Transform, FRFT）的邻近阶比分离方法。根据变速器输入轴转速信号及传动比确定FRFT最佳阶次，对变速器升速过程振动信号进行最佳阶次FRFT，在该分数阶域分离邻近阶比分量，并对分离出的单分量信号进行阶比分析。试验结果表明，根据转速信号确定FRFT最佳阶次，准确、快速、鲁棒性好，并具有自适应性；最佳阶次的FRFT能准确分离提取邻近阶比分量，对分离出的目标阶比分量进行单分量分析，能有效解决邻近阶比胶合问题。

Adjacent order components separation based on fractional fourier transformation

An method to separate adjacent order components based on fractional fourier transformation(FRFT) was presented.The best FRFT order was calculated according to rotating speed signal of a gear box’s input axis and its gear ratio,and the vibration signal in the gear box’s accelerating process was analyzed using FRFT with the best FRFT order,the adjacent order components were separated into single component signals,and the separated single component signals were analyzed with order analysis,respectively.The experimental results showed that the best FRFT order gained from the rotating speed signal is correct,quick,insusceptible and adaptive;the adjacent order components can be separated with FRFT accurately,and the agglutination problem of adjacent order components is solved effectively.