Abstract: Based on the analysis of noise source and noise characteristics in rapier loom’s tension-control system, the design method of the control structure and controller is proposed. On the base of the concept of the degree of control difficulty, a suitable analog low-pass filter is chosen for filtering the high-frequency signal and random noise. While in the periodical low-frequency noise processing，the frequency spectrum analysis is used to determine the period of the noise and the coherent mean method is used to obtain the signal. Then, the periodical vibration noise behind the analog low-pass filter can be estimated. Based on the estimation of the noise, a digital filter and feedback compensation are designed and employed to improve the control system’s performance. On the other hand, this method is also used to estimate the periodical vibration noise in front of the analog low-pass filter, and the digital filter and the feedforward compensation are designed to improve the control system’s performance. The effectiveness of the analysis and design methods is verified through matlab programming and simulink simulation. The analysis, methods and results proposed in this work provide some data for actual rapier loom’s tension control filter and controller design.

Key words: acoustics, feedforward compensation, filter, tension control, rapier loom, vibration estimation

摘要： 在分析剑杆织机张力控制中噪声源及噪声特性的基础上，提出控制结构和控制器的设计方法，选择合适的模拟低通滤波器，对检测通道的高频振动信号和随机噪声进行滤波。针对周期性低频振动噪声，一方面，采用频谱分析定周期+相干平均法定信号的方法估计模拟低通滤波器后的周期性振动噪声，并在反馈通道进行补偿，提高控制性能；另一方面，采用频谱分析定周期+相干平均法定信号的方法估计模拟低通滤波器前的周期性振动噪声，进行前馈补偿，提高控制性能。运用Matlab及仿真，验证分析和设计有效性，并为实际系统滤波器及控制器分析和设计提供依据。

关键词: 声学, 前馈补偿, 滤波器, 张力控制, 剑杆织机, 振动估计