Scholarlink

<p>To address the challenge of separating and identifying composite fault characteristics in 3K-type rotary actuators, a diagnostic method integrating Ensemble Empirical Mode Decomposition (EEMD) and Multi-scale Fuzzy Entropy (MSE) is proposed. First, a nonlinear dynamic model of the system was developed, incorporating time-varying meshing stiffness, flank clearance, and manufacturing errors. By simulating root cracks and component fractures, transmission error responses under various health conditions were obtained using the fourth-order Runge–Kutta method. Analysis revealed that fault-induced impact modulation complicates the transmission error spectrum, rendering traditional spectral analysis inadequate for quantifying fault characteristics. To overcome this, EEMD was employed for adaptive signal decomposition. Key intrinsic mode function (IMF) components rich in fault information were selected based on Pearson correlation coefficients. Subsequently, MSE was introduced to quantify signal complexity and irregularity across multiple scales for each component, constructing a high-dimensional fault feature vector. Simulation results demonstrated the method’s effectiveness in distinguishing between healthy states and predefined composite fault modes. Further validation through fatigue tests of rotary actuators successfully identified early-stage faults in support rings, confirming consistency between the dynamic model and diagnostic approach. This research indicates that the proposed technique offers an effective new pathway for condition monitoring of rotary actuators under complex operating conditions.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Research on Fault Diagnosis Technology for Rotary Actuators Based on EEMD and Multi-scale Fuzzy Entropy

  • 期刊论文

  • 发表日期：

    2026年6月24日

查看全文
  • Guowei Li,
  • He Yu,
  • Wankai Shi,
  • Wei Chen

摘要

To address the challenge of separating and identifying composite fault characteristics in 3K-type rotary actuators, a diagnostic method integrating Ensemble Empirical Mode Decomposition (EEMD) and Multi-scale Fuzzy Entropy (MSE) is proposed. First, a nonlinear dynamic model of the system was developed, incorporating time-varying meshing stiffness, flank clearance, and manufacturing errors. By simulating root cracks and component fractures, transmission error responses under various health conditions were obtained using the fourth-order Runge–Kutta method. Analysis revealed that fault-induced impact modulation complicates the transmission error spectrum, rendering traditional spectral analysis inadequate for quantifying fault characteristics. To overcome this, EEMD was employed for adaptive signal decomposition. Key intrinsic mode function (IMF) components rich in fault information were selected based on Pearson correlation coefficients. Subsequently, MSE was introduced to quantify signal complexity and irregularity across multiple scales for each component, constructing a high-dimensional fault feature vector. Simulation results demonstrated the method’s effectiveness in distinguishing between healthy states and predefined composite fault modes. Further validation through fatigue tests of rotary actuators successfully identified early-stage faults in support rings, confirming consistency between the dynamic model and diagnostic approach. This research indicates that the proposed technique offers an effective new pathway for condition monitoring of rotary actuators under complex operating conditions.