基于次优网络深度学习的3D打印机故障诊断

doi:10.3901/JME.2019.07.073

机械工程学报 ›› 2019, Vol. 55 ›› Issue (7): 73-80.doi: 10.3901/JME.2019.07.073

• 基于深度学习的机械装备故障预测与健康管理 • 上一篇下一篇

基于次优网络深度学习的3D打印机故障诊断

李川^1,2,3, 张绍辉², José Valente de Oliveira³

1. 重庆工商大学国家智能制造服务国际科技合作基地重庆 400067;
2. 华南理工大学机械与汽车工程学院广州 510641;
3. 葡萄牙阿尔加维大学 CEOT 中心法鲁 8005-139 葡萄牙

收稿日期:2018-06-24 修回日期:2018-11-06 出版日期:2019-04-05 发布日期:2019-04-05
通讯作者: 李川(通信作者),男,1975年出生,博士,教授。主要研究方向为机械故障诊断和智能计算。E-mail:chuanli@21cn.com
作者简介:张绍辉,男,1985年出生,博士后。主要研究方向为机械故障诊断和智能计算。E-mail:350699269@qq.com;José Valente de Oliveira,男,1965年出生,博士,教授。主要研究方向为智能计算和故障诊断。E-mail:jvo@ualg.pt
基金资助:
国家自然科学基金（51775112，51605406）、国家重点研发计划（2016YFE0132200）和重庆高校优秀成果转化（KJZH17123）资助项目。

Fault Diagnosis for 3D Printers Using Suboptimal Networked Deep Learning

LI Chuan^1,2,3, ZHANG Shaohui², José Valente de Oliveira³

1. National Research Base of Intelligent Manufacturing Service, Chongqing Technology and Business University, Chongqing 400067;
2. School of Mechanical & Automobile Engineering, South China University of Technology, Guangzhou 510641;
3. CEOT, Universidade do Algarve, Faro 8005-139, Portugal

Received:2018-06-24 Revised:2018-11-06 Online:2019-04-05 Published:2019-04-05

摘要/Abstract

摘要： 为实现精度可靠且成本节约的3D打印机故障诊断，采用消费品级的姿态传感器采集打印机的健康状态数据，并提出次优网络深度学习以弥补低成本硬件精度的不足。次优网络深度学习在由预训练和精细调节组成的传统深度学习基础上，一方面提出预分类方法自适应确定次优的网络结构参数，另一方面采用精细分类方法进一步提高故障诊断分类的精度。试验中，将姿态传感器安装于并联臂3D打印传动链的末端即打印头上。传感器全部通道的运动信号作为输入信息，采用深度玻尔兹曼机构建了次优网络故障诊断算法进行大数据驱动的故障诊断。将所提出的次优网络深度学习故障诊断方法与其他方法相比较，其结果表明，所提出方法可以有效诊断3D打印机的传动故障。

关键词: 3D打印, 次优网络, 大数据, 故障诊断, 深度学习

Abstract: For reliable and cost-saving diagnosis of 3D printer faults, a customer-goods level attitude sensor is employed to collect healthy condition data of the printer. To make up for the inferior precision of the low-cost hardware, a suboptimal networked deep learning (SNDL) method is developed. In addition to conventional deep learning architecture composed of pre-training and fine-tuning, there are two new merits for SNDL. A pre-classifying strategy is first proposed to adaptively determine the suboptimal network structure. A fine-classifying approach is on the other hand adopted for further improving fault classification performance. In the experiments, the attitude sensor is installed on the end of the transmission chain (i.e., the printing head of the delta 3D printer). Motion signals collected from all channels of the sensor are used as the input vector of the suboptimal networks, which are developed by deep Boltzmann machines for the big data-driven fault diagnosis. The addressed SNDL method is compared with peer methods. The results show that the present SNDL is capable of effectively diagnosing transmission faults for the 3D printer.

Key words: 3D printing, big data, deep learning, fault diagnosis, suboptimal network

中图分类号:

TH17

李川, 张绍辉, José Valente de Oliveira. 基于次优网络深度学习的3D打印机故障诊断[J]. 机械工程学报, 2019, 55(7): 73-80.

LI Chuan, ZHANG Shaohui, José Valente de Oliveira. Fault Diagnosis for 3D Printers Using Suboptimal Networked Deep Learning[J]. Journal of Mechanical Engineering, 2019, 55(7): 73-80.

参考文献

[1] MOHAMMED M I, GIBSON I. Design of three-dimensional, triply periodic unit cell scaffold structures for additive manufacturing[J]. Journal of Mechanical Design, 2018, 140(7):071701.
[2] 常定勇,方跃法,叶伟. 双输出3D打印解耦并联机器人的设计与分析[J]. 机械工程学报, 2017, 53(7):39-46. CHANG Dingyong, FANG Yuefa, YE Wei. Design and analysis of a double output 3D printing decoupling parallel manipulator[J]. Journal of Mechanical Engineering, 2017, 53(7):39-46.
[3] MILLER D B, GLISSON W B, YAMPOLSKIY M, et al. Identifying 3D printer residual data via open-source documentation[J]. Computers & Security, 2018, 75:10-23.
[4] JIA F, LEI Y G, LIN J, et al. Deep neural networks:A promising tool for fault characteristic mining and intelligent diagnosis of rotating machinery with massive data[J]. Mechanical Systems and Signal Processing, 2016, 72-73:303-315.
[5] 李川,朱荣荣,杨帅. 基于多指标模糊融合的滚动轴承诊断的最优频带解调方法[J]. 机械工程学报, 2015, 51(7):107-114. LI Chuan, ZHU Rongrong, YANG Shuai. Optimal frequency band demodulation for fault diagnosis of rolling element bearings based on fuzzy fusion of multiple criteria[J]. Journal of Mechanical Engineering, 2015, 51(7):107-114.
[6] SHAO H, JIANG H, ZHAO H, et al. A novel deep autoencoder feature learning method for rotating machinery fault diagnosis[J]. Mechanical Systems and Signal Processing, 2017, 95:187-204.
[7] MAO Yongfang, QIN Shuren, QIN Yi. Demodulation based on harmonic wavelet and its application into rotary machinery fault diagnosis[J]. Chinese Journal of Mechanical Engineering, 2009, 23(3):419-425.
[8] DING X, HE Q. Energy-fluctuated multiscale feature learning with deep ConvNet for intelligent spindle bearing fault diagnosis[J]. IEEE Transactions on Instrumentation and Measurement, 2017, 66(8):1926-1935.
[9] JING Yabing, LIU Changwen, BI Fengrong, et al. Diesel engine valve clearance fault diagnosis based on features extraction techniques and FastICA-SVM[J]. Chinese Journal of Mechanical Engineering, 2017, 30(4):991-1007.
[10] SHEN C Q, WANG D, KONG F R, et al. Fault diagnosis of rotating machinery based on the statistical parameters of wavelet packet paving and a generic support vector regressive classifier[J]. Measurement, 2013, 46:1551-1564.
[11] HE K, YANG Z, BAI Y, et al. Intelligent fault diagnosis of delta 3D printers using attitude sensors based on support vector machines[J]. Sensors, 2018, 18:1298.
[12] ZHAO R, YAN R, CHEN Z, et al. Deep learning and its applications to machine health monitoring[J]. Mechanical Systems and Signal Processing, 2019, 115:213-237.
[13] HINTON G E, SALAKHUDINOV R. Reducing the dimensionality of data with neural networks[J]. Science, 2006, 313:504-507.
[14] AHMED H, TAHIR M. Accurate attitude estimation of a moving land vehicle using low-cost MEMS IMU sensors[J]. IEEE Transactions on Intelligent Transportation Systems, 2017, 18, 1723-1739.
[15] SALAKHUTDINOV R, HINTON G. An efficient learning procedure for deep Boltzmann machines[J]. Neural Computation, 2012, 24(8):1967-2006.
[16] LI C, SANCHEZ R V, ZURITA G, et al. Multimodal deep support vector classification with homologous features and its application to gearbox fault diagnosis[J]. Neurocomputing, 2015, 168:119-127.
[17] 张绍辉. 集成参数自适应调整及隐含层降噪的深层RBM算法[J]. 自动化学报, 2017, 43(5):855-865. ZHANG Shaohui. Deep RBM algorithm with adaptive adjustment parameters and de-noising in hidden layer[J]. Acta Automatica Sinica, 2017, 43(5):855-865.
[18] TANG Y. Deep learning using linear support vector machines, in:Workshop on representational learning[C]//The 30th International Conference on Machine Learning, Atlanta, USA, June 16-21, 2013:1-6.

基于次优网络深度学习的3D打印机故障诊断

Fault Diagnosis for 3D Printers Using Suboptimal Networked Deep Learning

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	裴洪, 胡昌华, 司小胜, 张建勋, 庞哲楠, 张鹏. 基于机器学习的设备剩余寿命预测方法综述[J]. 机械工程学报, 2019, 55(8): 1-13.
[2]	廖中源, 王英俊, 王书亭. 基于拓扑优化的变密度点阵结构体优化设计方法[J]. 机械工程学报, 2019, 55(8): 65-72.
[3]	沈长青, 汤盛浩, 江星星, 石娟娟, 王俊, 朱忠奎. 独立自适应学习率优化深度信念网络在轴承故障诊断中的应用研究[J]. 机械工程学报, 2019, 55(7): 81-88.
[4]	熊鹏, 汤宝平, 邓蕾, 赵明航. 基于动态加权密集连接卷积网络的变转速行星齿轮箱故障诊断[J]. 机械工程学报, 2019, 55(7): 52-57.
[5]	王奉涛, 刘晓飞, 敦泊森, 邓刚, 韩清凯, 李宏坤. 基于萤火虫优化的核自动编码器在中介轴承故障诊断中的应用[J]. 机械工程学报, 2019, 55(7): 58-64.
[6]	胡茑庆, 陈徽鹏, 程哲, 张伦, 张宇. 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.
[7]	刘腾飞, 田小永, 朱伟军, 李涤尘. 连续碳纤维增强聚乳酸复合材料3D打印及回收再利用机理与性能[J]. 机械工程学报, 2019, 55(7): 128-134.
[8]	雷亚国, 杨彬, 杜兆钧, 吕娜. 大数据下机械装备故障的深度迁移诊断方法[J]. 机械工程学报, 2019, 55(7): 1-8.
[9]	姜洪开, 邵海东, 李兴球. 基于深度学习的飞行器智能故障诊断方法[J]. 机械工程学报, 2019, 55(7): 27-34.
[10]	訾斌, 尹泽强, 李永昌, 赵涛. 基于YOLO模型的柔索并联机器人移动构件快速定位方法[J]. 机械工程学报, 2019, 55(3): 64-72.
[11]	翁军华,张剑锐,李达,郑海兴. 光伏系统直流电弧故障检测的抗干扰技术综述[J]. 电气工程学报, 2019, 14(3): 9-15.
[12]	汪久根, 柯梁亮. 基于残差网络的RV减速器故障诊断[J]. 机械工程学报, 2019, 55(3): 73-80.
[13]	朱伟军, 陈言坤, 张志坤, 田小永, 李涤尘. 可拉伸柔性电路的原位封装3D打印工艺[J]. 机械工程学报, 2019, 55(15): 64-70.
[14]	姜世杰, 史银芳, SIYAJEU Yannick, 赵春雨. 路径宽度对FDM薄板固有特性的影响机理研究[J]. 机械工程学报, 2019, 55(15): 226-232.
[15]	周贺飞, 兰红波, 李红珂, 许权, 赵佳伟, 张广明. 基于电场驱动喷射沉积微尺度3D打印制造金属网栅透明电磁屏蔽玻璃的研究[J]. 机械工程学报, 2019, 55(15): 56-63.