摘要
针对当前系统故障诊断研究忽略非线性特性问题,本文旨在分析大型工业机器人系统故障非线性频谱特性。以机器人驱动系统故障引发的整个机器人执行器故障为例,建立大型六关节工业机器人模型,采集电机转速和定子电流分别作为系统输入和输出,利用自适应辨识算法得到系统前4阶NOFRF频谱,分析了不同状态下NOFRF频谱变化规律。实验结果表明,与传统故障特征表征方法相比,NOFRF频谱对状态变化更敏感,而且频谱包含的非线性故障特征更丰富,将其作为故障特征对系统进行诊断可以有效提高诊断准确率。
Aiming at the problem that the current research on system fault diagnosis ignores nonlinear characteristics,this paper analyzes the nonlinear spectral characteristics of faults in large industrial robot systems.Taking the failure of the entire robot actuator caused by the failure of the robot drive system as an example,a large six-joint industrial robot model is established.The motor speed and stator current collected serve as the system input and output respectively.The first four-order NOFRF spectra of the system are obtained by using the adaptive identification algorithm,and variation rules of the NOFRF spectra under different states are analyzed.The experimental results show that compared with traditional methods for characterizing fault features,the NOFRF spectra are more sensitive to state changes,and the spectra contain richer nonlinear fault features.Using them as fault features for system diagnosis can effectively improve the diagnostic accuracy.
作者
张娓娓
陈乐瑞
Zhang Weiwei;Chen Lerui(Henan Polytechnic Institute,Nanyang473000,Henan,China;School of Intelligent Sensing and Instruments,Zhongyuan University of Technology,Zhengzhou 451191,Henan,China)
出处
《船电技术》
2025年第9期30-34,共5页
Marine Electric & Electronic Engineering
基金
河南省自然科学基金面上项目(NO.242300421417)。
作者简介
张娓娓(1985-),女,讲师,研究方向:自动化控制。E-mail:zww801231@163.com。
