A Wear Particle Detection Method Based on Coaxial Array Electrostatic Sensor and VMD-SR-DTW Model

作者：Yibing Yin, Long Feng, Q. Zhang, Yang Wang, Lei Song, Huanlei Chi, Zhenhua Wen · 发表于：IEEE Transactions on Instrumentation and Measurement · 年份：2024 · DOI：10.1109/tim.2024.3413158 · 被引用次数：7 · 研究领域：Electrical and Bioimpedance Tomography、Non-Destructive Testing Techniques、Mineral Processing and Grinding

As a wear particle detection technology for machinery, electrostatic monitoring has broad application prospects. However, there is currently a lack of a comprehensive method for achieving precise wear particle detection through electrostatic signals. This study proposes a wear particle detection method based on coaxial array electrostatic probes and a VMD-SR-DTW algorithm model. First, a joint signal processing model is constructed using variational mode decomposition (VMD) and sparse representation (SR) to reduce noise interference and enhance the signals triggered by wear particles. Furthermore, an algorithm model based on sliding windows and dynamic time warping (DTW) is developed by utilizing the multichannel signals from the coaxial array electrostatic sensor. The overall algorithm flow is analyzed in detail. Simulations and experiments are conducted to validate the effectiveness of the proposed method. The results demonstrate that the method effectively detects the signal pulses triggered by wear particles and provides a reliable approach for the engineering application of online monitoring.