基于工作电流分析的回转副摩擦状态识别实验研究

回转副摩擦状态的变化对机械设备的运行有着重要影响. 针对回转副摩擦状态的识别问题, 设计了以回转副工作电流为摩擦特征来源的实验方案. 通过采集摩擦过程的回转副驱动电机工作电流信号, 分析实验信号的频域和时频域特征, 建立工作电流的摩擦特征库, 用于训练多分类支持向量机分类器, 并在分类器上完成摩擦状态的识别. 实验结果表明, 经特征提取与降维处理后建立的多分类支持向量机分类器具有近90%的识别正确率, 所提出的实验方案有利于实现远距离信号的采集, 适用实际生产加工.

Experiment on friction state identification based on working current analysis

The change of the friction state of the rotary pair has an important influence on the operation of the mechanical equipment. Aiming at the problem of recognizing the friction state of the rotary pair, an experimental scheme is designed with the working current of the rotary pair being taken as the source of friction characteristics. By collecting the working current signal of the rotary pair drive motor in the friction process, the frequency and time-frequency domain characteristics of the experimental signal are analyzed, and the friction feature library of the working current is established to train the multi-class support vector machine classifier, and finally the identification of friction status is completed on such classifier. The experimental results show that the established multi-class support vector machine classifier established after feature extraction and dimensionality reduction has a recognition accuracy of nearly 90%. The proposed experimental scheme is also conducive to the realization of long-distance signal acquisition and is suitable for actual production and processing.