针对实际工程中工业机器人末端抖动程度难以通过简单指标进行有效区分，且识别过程存在外部干扰大、抖动位置分布不均的问题，提出了变初位数据增强和深度特征提取的工业机器人末端抖动状态识别方法。首先，利用陷波滤波器滤除工业机器人末端振动信号中的工频干扰，并通过均方根阈值搜索抖动位置及改变搜索初始位置数据增强的方式，获取充足且能展现工业机器人运动状态的末端抖动数据；其次，采用连续小波变换对末端抖动数据进行分解以获得可充分保留末端抖动冲击与震荡特征的时频图；最后，为缓解特征降维及工况变化的影响，运用去除池化层和添加批归一化的卷积神经网络，对时频图进行深度特征提取和分类，从而实现工业机器人末端抖动状态识别。实验结果表明，所提方法在不同传感器采集方向识别准确率均达到90%以上，证明了该方法能够有效识别工业机器人末端抖动状态，并具有较好的泛化性和稳定性。
Abstract
For the actual engineering industrial robot end-jitter degree is difficult to be effectively distinguished by simple indicators, and the recognition process exists large external interference and uneven distribution of jitter position problems, an industrial robot end-jitter state recognition method with variable initial position data augmentation and deep feature extraction is proposed. Firstly, power-line interference filtering in industrial robot end vibration signals using trap filters, and by means of root mean square threshold search jitter location and changing the search initial position data enhancement, to obtain sufficient end-jitter data that can show the motion state of the industrial robot; Secondly, the end-jitter data is decomposed by continuous wavelet transform to obtain the time-frequency map that can fully retain the end jitter shock and vibration characteristics; Finally, to mitigate the effects of feature downscaling and work condition variation, a convolutional neural network with the pooling layer removed and batch normalization added is applied to perform deep feature extraction and classification of the time-frequency map, which enables industrial robot end-jitter state recognition. The experimental results show that the proposed method achieves an accuracy of more than 90% in different sensor acquisition directions, which proves that the method can effectively recognition the end jitter state of industrial robots and has good generalization and stability.
关键词
工业机器人 /
抖动状态识别 /
数据增强 /
连续小波变换 /
卷积神经网络
{{custom_keyword}} /
Key words
industrial robot /
jitter state recognition /
data augmentation /
continuous wavelet transform /
convolutional neural network
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 吴锦辉, 陶友瑞. 工业机器人定位精度可靠性研究现状综述[J]. 中国机械工程, 2020, 31(18): 2180-2188.
WU J H, TAO Y R. Review on research status of positioning accuracy reliability of industrial robots[J]. China Mechanical Engineering, 2020, 31(18): 2180-2188.
[2] 赵威, 王锴, 徐皑冬,等. 面向智能制造的工业机器人健康评估方法[J]. 机器人, 2020(4): 460-468.
ZHAO W, WANG K, XU A D, et al. An industrial robot health assessment method for intelligent manufacturing[J]. Robot, 2020(4): 460-468.
[3] 李本旺, 吴曾萍, 王茂林, 等. 工业机器人性能测评体系研究综述[J]. 现代制造工程, 2020(9): 149-155.
LI B W, WU Z P, WANG M L, et al. A review of the performance evaluation system for industrial robots[J]. Modern Manufacturing Engineering, 2020(9): 149-155.
[4] Jaber A A, Bicker R. Fault diagnosis of industrial robot gears based on discrete wavelet transform and artificial neural network[J]. Insight-Non-Destructive Testing and Condition Monitoring, 2016, 58(4): 179-186.
[5] Kim Y, Park J, Na K, et al. Phase-based time domain averaging (PTDA) for fault detection of a gearbox in an industrial robot using vibration signals[J]. Mechanical Systems and Signal Processing, 2020, 138: 106544.
[6] Nentwich C, Junker S, Reinhart G. Data-driven Models for Fault Classification and Prediction of Industrial Robots[J]. Procedia CIRP, 2020, 93: 1055-1060.
[7] 陈仁祥, 张勇, 杨黎霞, 等. 基于整周期数据和卷积神经网络的谐波减速器健康状态评估[J]. 仪器仪表学报, 2020, 41(2): 245-252.
CHEN R X, ZHANG Y, YANG L X, et al. Health condition assessment of harmonic reducer based on integer-period data and convolutional neural network[J]. Chinese Journal of Scientific Instrument, 2020, 41(2): 245-252.
[8] Jia F, Lei Y G, Lu N, et al. Deep normalized convolutional neural network for imbalanced fault classification of machinery and its understanding via visualization [J]. Mechanical Systems and Signal Processing, 2018, 110: 349-367.
[9] 胡茑庆, 陈徽鹏, 程哲, 等. 基于经验模态分解和深度卷积神经网络的行星齿轮箱故障诊断方法[J].机械工程学报, 2019, 55(7): 9-18.
HU N Q, CHEN H P, CHENG Z, et al. Fault diagnosis for planetary gearbox based on EMD and deep convolutional neural networks[J]. Journal of Mechanical Engineering, 2019, 55(7): 9-18.
[10] Wen L, Li X, Gao L, et al. A New Convolutional Neural Network-Based Data-Driven Fault Diagnosis Method[J]. IEEE Transactions on Industrial Electronics, 2018, 65(7): 5990-5998.
[11] 宫文峰, 陈辉, 张美玲, 等. 基于深度学习的电机轴承微小故障智能诊断方法[J]. 仪器仪表学报, 2020, 41(1): 195-205.
GONG W F, CHEN H, ZHANG M L, et al. Intelligent diagnosis method for incipient fault of motor bearing based on deep learning[J]. Chinese Journal of Scientific Instrument, 2020, 41(1): 195-205.
[12] Zhang T, Chen J, Li F, et al. Intelligent fault diagnosis of machines with small & imbalanced data: A state-of-the-art review and possible extensions[J]. ISA transactions, 2022, 119: 152-171.
[13] 杨蕊, 李宏坤, 贺长波, 等. 利用最优小波尺度循环谱的滚动轴承早期故障特征提取[J]. 机械工程学报, 2018, 54(17): 208-217.
YANG R, LI H K, HE C B, et al. Rolling Element Bearing Incipient Fault Feature Extraction Based on Optimal Wavelet Scales Cyclic Spectrum[J]. Journal of Mechanical Engineering, 2018, 54(17): 208-217.
[14] 吴晨芳, 杨世锡, 黄海舟, 等. 一种基于改进的LeNet-5模型滚动轴承故障诊断方法研究[J]. 振动与冲击, 2021, 40(12): 55-61.
WU C F, YANG S X, HUANG H Z, et al. An improved fault diagnosis method of rolling bearings based on LeNet-5[J]. Journal of Vibration and Shock, 2021, 40(12): 55-61.
[15] Ioffe S, Szegedy C. Batch normalization: accelerating deep network training by reducing internal covariate shift[C]//Proceedings of the 32nd International Conference on International Conference on Machine Learning. Lille, France. 2015(37): 448-456.
[16] Laurens V D M, Hinton G. Visualizing Data using t-SNE[J]. Journal of Machine Learning Research, 2008, 9(2605): 2579-2605.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
PDF(2868 KB)
