【摘要】 机械传动系统中摩擦力等非线性因素的影响会导致系统频率特性辨识结果毛刺多、精度低。为了提高机械传动系统频率特性辨识精度，提出一种基于复合激励信号的系统频率特性辨识方法。在对数扫频信号上叠加一个恒值信号，将复合后的信号作为辨识所需的激励信号输入到系统的指定位置；采集相应的输入输出数据，基于快速傅里叶变换的数据处理方法对采集的数据进行处理，绘制系统的频域响应曲线。仿真和实验对比结果表明：辨识结果中毛刺产生的主要原因包括机床传动系统中的非线性摩擦力、噪声信号、采样时间和快速傅里叶变换的离散性等，非线性摩擦是产生毛刺的主要原因，对辨识结果的影响最大。相较于传统的单一激励信号辨识方法，基于复合激励信号的系统频率特性辨识方法可以显著减小非线性摩擦力造成的辨识误差，有效提高机械传动系统的频率特性辨识精度，进而为伺服进给系统的控制参数整定和高性能运动控制提供数据支持。更多还原

【Abstract】 The influence of nonlinear factors such as friction in the mechanical transmission system will lead to many burrs and low accuracy of the system frequency characteristic identification results.In order to improve the accuracy of frequency characteristic identification of mechanical transmission system, a system frequency characteristic identification method was proposed based on composite excitation signal.A constant value signal was superimposed on the logarithmic swept signal, and the composite signal as the excitation signal required for identification was input into the specified position of the system.The corresponding input and output data were collected, and then the collected data were processed through data processing method based on fast Fourier transform, and the frequency domain response curve of the system was plotted.The simulation and experimental comparison results show that, the main causes of burrs in the identification results include the nonlinear friction in the machine tool′s transmission system, noise signals, sampling time and the discreteness of the fast Fourier transform.Nonlinear friction is the primary cause of burrs and has the greatest impact on the identification results.Compared with the traditional single excitation signal identification method, this method can greatly reduce the identification error caused by nonlinear friction, effectively improve the frequency characteristic identification accuracy of the mechanical transmission system, and provide data support for the control parameter adjustment and high-performance motion control of servo-feeding system.更多还原