|
基于DS-DBN多测点的工业机器人RV减速器故障诊断
储晓静，张文婷 2024/9/20 8:35:32
江苏省连云港工贸高等职业技术学校，机电工程系，江苏连云港 222061
摘要：RV减速器在工业机器人得到广泛应用，直接影响到工业机器人运行效率和成本预算。为了提高减速器故障识别能力，融合DS证据理论和深度信念网络(DBN)设计了一种基于DS-DBN多测点的故障诊断方法，并开展实验测试验证。研究结果表明：经过30代训练后达到收敛状态，训练四个测点的DS-DBN故障诊断模型分类误差均小于5%，表明采用本文方法能够快速获得全局最优参数。相比较单点测试，通过DS证据理论完成测点故障诊断融合处理后，可以将故障诊断准确率提升至接近100%的程度。该研究有助于排斥工业机器人RV减速器隐藏的问题，提高使用寿命和经济成本。
关键词：工业机器人；RV减速器；故障诊断；多测点；深度信念网络；DS证据理论
中图分类号：TH16 文献标识码：A 文章编号：
Fault diagnosis of RV reducer for industrial robot based on DS-DBN multi-point measurement
Chu Xiaojing, Zhang Wenting
Department of Mechanical and Electrical Engineering, Lianyungang Higher Vocational and Technical School of Industry and Trade, Lianyungang 222061, China
Abstract: RV reducer is widely used in industrial robots, which directly affects the operating efficiency and cost budget of industrial robots. In order to improve the fault identification ability of reducer, a fault diagnosis method based on DS-DBN was designed by integrating DS evidence theory and deep belief network (DBN). The results show that the classification error of the DS-DBN fault diagnosis model is less than 5% after 30 generations of training, indicating that the proposed method can quickly obtain the global optimal parameters. Compared with the single point test, the fault diagnosis accuracy can be improved to nearly 100% after the fusion processing of the detection point fault diagnosis by DS evidence theory. The research helps to exclude hidden problems of industrial robot RV reducer, improve service life and economic cost.
Key words: Industrial robot; RV reducer; Fault diagnosis; Multiple measuring points; Deep belief network; DS evidence theory
0 引言
RV减速器在工业机器人得到广泛应用，其运行稳定性直接影响到动作效率[1]。随着信息处理技术的进步，机电控制设备的运行自动化水平也获得了明显提升，要求工业机器人运行检测和故障诊断满足更高的标准[2-3]。针对上述情况，需引入深度学习算法，从而更加准确诊断具有减速器的设备故障信号[4]。
深度信念网络（DBN）具有时变特性，可以实现故障诊断准确率的显著提升[8]。相关方面的研究吸引了很多的学者，取得了一定的研究成果。张智禹等[5]提出融合注意力机制的改进深度置信网络故障诊断方法，引进余弦损失函数降低网络拟合负担，有效提高变工况下齿轮箱故障（未完，下一页）
附件下载：基于DS-DBN多测点的工业机器人RV减速器故障诊断
|