面向传感器突发故障的监测模型自适应快速调整技术研究

钱政; 程淑玉; 夏红霞

doi:10.13433/j.cnki.1003-8728.20230102

面向传感器突发故障的监测模型自适应快速调整技术研究

doi: 10.13433/j.cnki.1003-8728.20230102

1.
安徽电子信息职业技术学院信息工程学院, 安徽蚌埠 233000
2.
安徽电子信息职业技术学院软件工程学院, 安徽蚌埠 233000

基金项目:

安徽高校自然科学研究项目 KJ2021A1488

安徽省职业教育创新发展试验区建设项目 WJ-ZYPX-011

详细信息

作者简介:
钱政(1983-), 副教授, 硕士, 研究方向为人工智能与数据库, threean@163.com

中图分类号: TP212;TH17
计量
- 文章访问数: 78
- HTML全文浏览量: 11
- PDF下载量: 16
- 被引次数: 0
出版历程
- 收稿日期: 2022-04-29
- 刊出日期: 2023-04-25

Study on Adaptive Fast Adjustment of Monitoring Model for Sensor Sudden Fault

1.
College of Information Engineering, Anhui Vocational College of Electronics and Information Technology, Bengbu 233000, Anhui, China
2.
College of Software Engineering, Anhui Vocational College of Electronics and Information Technology, Bengbu 233000, Anhui, China

摘要

摘要: 针对多传感器监测场景中传感器突发故障导致监测模型失效的问题, 本文提出了一种堆叠自编码器深度学习模型的自适应快速调整方法。根据传感器故障时采集数据的分布特点自适应调整原始数据集, 采用调整后的数据集正向传播和监测误差反向传播微调模型更新模型的权重和偏置, 实现监测模型自适应快速调整。以机械加工过程中刀具磨损状态监测为例, 采用加州大学伯克利分校的BEST实验室提供的刀具数据集验证了自适应调整方法的有效性。结果表明, 该方法可解决当传感器突发故障时, 在实时监测不中断的情况下, 自适应调整后的监测模型可以准确地对刀具状态进行监测。
- 传感器突发故障 /
- 堆叠自编码器 /
- 模型自适应调整 /
- 刀具状态监测
Abstract: To solve the sensor fault leading to failure of monitoring model for multi-sensor monitoring scenarios, an adaptive fast adjustment method of stacked autoencoder (SAE) deep learning model is proposed in this paper. The original data set was adjusted adaptively according to the distribution characteris-tics of the data collected during sensor fault, the weight and bias of the model are updated by using the adjusted data set forward propagation and the monitoring error back propagation to fine-tune model, and the adaptive fast adjustment of the monitoring model was realized. Taking tool wear monitoring in machining as an example, the effectiveness of the proposed method was verified with the data set provided by the BEST lab at UC Berkeley. The results show that the adaptively adjusted monitoring model can accurately monitor the tool state under the condition that the real-time monitoring is not interrupted when the sensor breaks down suddenly.
- sensor sudden fault /
- stack autoencoder /
- model adaptive adjustment /
- tool condition monitoring

HTML全文

图 1 基于模型自适应调整方法的监测方案

下载: 全尺寸图片幻灯片

图 2 自编码器和堆叠自编码器网络

下载: 全尺寸图片幻灯片

图 3 模型自适应调整方案

下载: 全尺寸图片幻灯片

图 4 自编码器训练集和测试集的混淆矩阵

下载: 全尺寸图片幻灯片

图 5 模型自适应调整时间与调整后模型准确率

下载: 全尺寸图片幻灯片

图 6 模型自适应调整时间与调整后模型准确率

下载: 全尺寸图片幻灯片

表 1 模型自适应调整时间与调整后模型准确率 %

调整时间/s	故障传感器序号
调整时间/s	1	2	3	4	5	6
0	71.58	65.75	97.43	63.70	71.23	75.17
1	95.38	82.19	96.92	85.96	89.04	93.32
2	96.23	85.27	97.95	88.87	92.64	92.81
3	96.92	87.67	97.43	92.12	94.69	95.38
4	97.43	89.38	97.60	91.44	95.72	95.89
5	97.26	92.29	97.95	93.15	94.86	95.55
6	97.09	93.32	98.12	94.01	95.21	95.21
7	97.43	93.32	97.95	93.49	94.86	95.21
8	97.26	90.58	95.89	93.66	95.21	95.72
9	97.43	93.15	97.26	94.86	95.03	95.38
10	97.60	92.64	97.60	94.69	95.03	95.55

下载: 导出CSV

表 2 模型自适应调整时间与调整后模型准确率 %

故障传感器序号	模型自适应调整时间/s
故障传感器序号	0	1	2	3	4	5	6	7	8	9	10
1和2	56.51	78.08	83.73	85.62	89.04	89.73	89.90	91.10	90.92	91.44	91.10
1和3	72.95	95.72	97.26	97.09	97.26	97.77	97.26	97.77	97.60	97.77	97.43
1和4	47.26	80.14	86.47	88.87	89.90	91.95	91.95	91.95	93.66	93.66	93.49
1和5	57.71	80.48	86.64	91.78	93.49	94.01	93.66	94.01	94.35	94.35	94.52
1和6	74.49	88.87	90.92	93.15	94.52	94.52	94.52	94.69	95.38	95.55	95.72
2和3	67.47	82.02	84.42	86.47	87.84	88.01	90.24	90.92	90.92	90.92	91.95
2和4	53.42	79.97	79.79	80.48	78.25	83.90	86.13	85.96	85.96	86.82	89.04
2和5	55.99	61.99	73.12	79.62	84.25	86.30	86.30	88.18	89.04	87.84	89.73
2和6	56.16	70.38	78.42	78.94	84.76	85.10	84.25	86.47	87.16	88.53	88.87
3和4	67.12	87.33	88.36	91.95	92.47	94.18	93.66	94.35	94.86	94.86	94.86
3和5	70.72	87.84	92.64	94.86	96.06	95.55	95.03	96.06	96.23	96.58	96.06
3和6	74.49	92.64	91.78	94.18	92.81	94.18	95.38	95.72	95.38	95.55	95.72
4和5	67.47	85.45	86.30	91.10	91.61	90.41	91.44	91.10	92.64	92.29	92.12
4和6	74.32	84.93	86.99	88.18	89.38	90.07	90.75	90.92	92.47	92.29	91.61
5和6	63.01	74.66	80.31	82.71	84.93	86.13	85.27	87.50	87.50	88.18	88.01

下载: 导出CSV

参考文献(15)

[1]	CHAMOLA V, VINEET A, NAYYAR A, et al. Brain-computer interface-based humanoid control: A review[J]. Sensors, 2020, 20(13): 3620. doi: 10.3390/s20133620
[2]	陈唯实, 黄毅峰, 卢贤锋. 多传感器融合的无人机探测技术应用综述[J]. 现代雷达, 2020, 42(6): 15-29. CHEN W S, HUANG Y F, LU X F. Survey on application of multi-sensor fusion in UAV detection technology[J]. Modern Radar, 2020, 42(6): 15-29. (in Chinese)
[3]	LI N P, GEBRAEEL N, LEI Y G, et al. Remaining useful life prediction based on a multi-sensor data fusion model[J]. Reliability Engineering & System Safety, 2021, 208: 107249.
[4]	CHEN G, LIU Z G, YU G, et al. A new view of multisensor data fusion: research on generalized fusion[J]. Mathematical Problems in Engineering, 2021, 2021: 5471242.
[5]	JIAO W Z, TIAN C, CHANG Q, et al. A new multi-sensor integrated index for drought monitoring[J]. Agricultural and Forest Meteorology, 2019, 268: 74-85. doi: 10.1016/j.agrformet.2019.01.008
[6]	QU Y H, YANG M H, ZHANG J Q, et al. An outline of multi-sensor fusion methods for mobile agents indoor navigation[J]. Sensors, 2021, 21(5): 1605. doi: 10.3390/s21051605
[7]	马伟宁. 基于多传感器信息技术的变电站热故障监测方法[D]. 唐山: 华北理工大学, 2017. MA W N. Thermal fault monitoring method for substation based on multi-sensors information technology[D]. Tangshan: North China University of Science and Technology, 2017. (in Chinese)
[8]	KUMAR M, GARG D P, ZACHERY R A. A generalized approach for inconsistency detection in data fusion from multiple sensors[C]//2006 American Control Conference. Minneapolis, MN, USA: IEEE, 2006: 6.
[9]	乔宁国. 基于多传感器数据融合的高速列车传动系统故障诊断与健康状态预测[D]. 长春: 吉林大学, 2019. QIAO N G. Fault diagnosis and health prediction of high-speed train transmission system based on multi-sensor fusion[D]. Changchun: Jilin University, 2019. (in Chinese)
[10]	刘思辰, 杨飞然, 杨军. 基于多传感器融合的刀具剩余寿命预测[J]. 振动与冲击, 2021, 40(17): 47-54. https://www.cnki.com.cn/Article/CJFDTOTAL-ZDCJ202117007.htm LIU S C, YANG F R, YANG J. Tool residual life prediction based on multi-sensor fusion[J]. Journal of Vibration and Shock, 2021, 40(17): 47-54. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-ZDCJ202117007.htm
[11]	QIAO H H, WANG T Y, WANG P, et al. A time- distributed spatiotemporal feature learning method for machine health monitoring with multi-sensor time series[J]. Sensors, 2018, 18(9): 2932. doi: 10.3390/s18092932
[12]	YAO X M, LI S B, YAO Y, et al. Health monitoring and diagnosis of equipment based on multi-sensor fusion[J]. International Journal of Online and Biomedical Engineering, 2018, 14(4): 4-19.
[13]	SAMARAS S, DIAMANTIDOU E, ATALOGLOU D, et al. Deep learning on multi sensor data for counter UAV applications-A systematic review[J]. Sensors, 2019, 19(22): 4837. doi: 10.3390/s19224837
[14]	李宏坤, 郝佰田, 代月帮, 等. 基于压缩感知和加噪堆栈稀疏自编码器的铣刀磨损程度识别方法研究[J]. 机械工程学报, 2019, 55(14): 1-10. https://www.cnki.com.cn/Article/CJFDTOTAL-JXXB201914001.htm LI H K, HAO B T, DAI Y B, et al. Wear status recognition for milling cutter based on compressed sensing and noise stacking sparse auto-encoder[J]. Journal of Mechanical Engineering, 2019, 55(14): 1-10. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-JXXB201914001.htm
[15]	来杰, 王晓丹, 向前, 等. 自编码器及其应用综述[J]. 通信学报, 2021, 42(9): 218-230. LAI J, WANG X D, XIANG Q, et al. Review on autoencoder and its application[J]. Journal on Communications, 2021, 42(9): 218-230. (in Chinese)