【摘要】 针对旋转矢量(rotary vector, RV)减速器多源耦合严重，行星齿轮局部故障所引起的冲击易被其他干扰分量所淹没，故障特征提取困难的问题，结合编码器信号的优势提出了一种基于自适应最大二阶循环平稳盲解卷积(adaptive maximum second order cyclostationarity blind deconvolution, ACYCBD)的RV减速器行星齿轮局部故障检测方法。首先，拾取伺服电机内置光编码器信号，并利用向前差分计算获得瞬时角速度(instantaneous angular speed, IAS)信号；然后，依据特征评价指标(characteristic evaluation indicator, CEI)最大化原则自适应确定ACYCBD优化滤波器长度，并对IAS信号进行增强；最后，通过识别时域中与故障冲击周期相匹配的理论齿数实现RV减速器故障检测。通过试验数据分析，并将所提方法与现有的稀疏低秩分解算法和增强CYCBD算法对比，验证了所提方法的有效性。更多还原

【Abstract】 Here, aiming at the problem of due to severe multi-source coupling of rotary vector(RV) reducer, impacts caused by planetary gears local faults being easily overwhelmed by other interference components to result in difficulty to extract fault features, combined with advantages of encoder signals, a local fault detection method for RV reducer planetary gears based on adaptive maximum second-order cyclo-stationary blind deconvolution(ACYCBD) was proposed. Firstly, built-in optical encoder signals of servo motor were picked up and the forward difference calculation was used to obtain instantaneous angular speed(IAS) signals. Then, according to the principle of maximizing the characteristic evaluation index(CEI), the length of ACYCBD optimized filter was adaptively determined and IAS signals were enhanced. Finally, RV reducer fault detection was realized by identifying the theoretical number of gear teeth in time domain to match fault impact period. The effectiveness of the proposed method was verified by analyzing experimental data and comparing the proposed method with existing sparse low rank decomposition algorithms and the enhanced CYCBD algorithm.更多还原