一种基于LSTM自动编码机的工业系统异常检测方法

doi:10.11959/j.issn.1000-0801.2020188

Abstract

Abstract:

In the context of the industrial internet,automatic and effective anomaly detection methods are of great significance to the safe and stable production of industrial systems.Traditional anomaly detection methods have the disadvantages of requiring a large number of labeled samples,and not adapting to high-dimensional time series data.Aiming at these limitations,an industrial system anomaly detection method based on LSTM (long short-term memory)auto-encoder was proposed.Firstly,to address the limitation of relying on labeled samples,an encoder used to learn the features and patterns of a large number of normal samples in an unsupervised manner.Then,anomaly detection was performed via reconstructing samples and calculating the reconstruction error.Secondly,to adapt to high-dimensional time series data,a BiLSTM (bidirectional LSTM) was used as an encoder,and then the potential characteristics of multi-dimensional time series data were mined.Experiments based on a real paper industry data set which demonstrate this method has improved the existing unsupervised anomaly detection methods in various indicators,and the overall accuracy of the detection has reached 93.4%.

Key words: abnormal detection, industrial internet, auto-encoder, LSTM

CLC Number:

TN929.5

Xiaojun SHEN,Yanan GE,Zhihao SHEN,Yangdan NI,Mingqi LV,Zhengqiu WENG. An LSTM auto-encoder based anomaly detection for industrial system[J]. Telecommunications Science, 2020, 36(7): 136-145.

Figures/Tables 10

References 24

[1]	ZHOU L Y , GUO H Q . Anomaly detection methods for IoT Networks[C]// Proceedings of 2018 IEEE International Conference on Service Operations and Logistics,and Informatics (SOLI). Piscataway:IEEE Press, 2018: 214-219.
[2]	GENGE B , PIROSKA H , CALIN E . Anomaly detection in aging industrial internet of things[J]. IEEE Access, 2019(99):1.
[3]	WONG T , LUO Z . Recurrent auto-encoder model for large-scale industrial sensor signal analysis[C]// Proceedings of International Conference on Engineering Applications of Neural Networks. Berlin:Springer, 2018: 203-216
[4]	HUANG H K . Real-time fault-detection for IoT facilities using GBRBM-based DNN[J]. IEEE Internet of Things Journal, 2019(10): 1-1.
[5]	RIEGER T , . Fast predictive maintenance in industrial internet of things (IIoT) with deep learning (DL):a review[C]// Proceedings of 2019 Collaborative European Research Conference (CERC2019).[S.l.:s.n]. 2019: 69-79.
[6]	KANAWADAY A , SANE A . Machine learning for predictive maintenance of industrial machines using IoT sensor data[C]// Proceedings of 2017 8th IEEE International Conference on Software Engineering and Service Science (ICSESS). Piscataway:IEEE Press, 2017: 107-110.
[7]	ABU A , HANEEN A . Effects of distance measure choice on K-nearest neighbor classifier performance:a review[J]. Big Data, 2019(7): 221-248.
[8]	BREUNING M , . LOF:identifying density-based local outliers[C]// Proceedings of the 2000 ACM SIGMOD International Conference on Management of Data. New York:ACM Press, 2000,29(2): 93-104.
[9]	TANG J , . Enhancing effectiveness of outlier detections for low density patterns[C]// Proceedings of Pacific-Asia Conference on Knowledge Discovery and Data Mining. Berlin:Springer, 2002.
[10]	GRIFFIN J M , DOBERTI A J , VALBORT H ,et al. Multiple classification of the force and acceleration signals extracted during multiple machine processes:part 1 intelligent classification from an anomaly perspective[J]. International Journal of Advanced Manufacturing Technology, 2017(93): 1-13.
[11]	NANDURI A , SHERRY L . Anomaly detection in aircraft data using recurrent neural networks (RNN)[C]// Proceedings of 2016 Integrated Communications Navigation and Surveillance (ICNS). Piscataway:IEEE Press, 2016.
[12]	JANSSENS O . Convolutional neural network based fault detection for rotating machinery[J]. Journal of Sound and Vibration, 2016(377): 331-345.
[13]	AMRUTHNATH N , TARUN G . A research study on unsupervised machine learning algorithms for early fault detection in predictive maintenance[C]// Proceedings of 2018 5th International Conference on Industrial Engineering and Applications (ICIEA). Piscataway:IEEE Press, 2018: 355-361.
[14]	DIEZ-OLIVAN A , RUBIO J A P . Kernel-based support vector machines for automated health status assessment in monitoring sensor data[J]. The International Journal of Advanced Manufacturing Technology, 2018,95(1-4): 327-340.
[15]	JOSHI S , PHOHA V . Investigating hidden Markov models capabilities in anomaly detection[C]// Proceedings of the 43rd Annual Southeast Regional Conference. New York:ACM Press, 2005: 98-103.
[16]	FENG T Z , DU Z H , SUN Y K ,et al. Real-time anomaly detection of short-time-scale gwac survey light curves[C]// Proceedings of 2017 IEEE International Congress on Big Data (BigData Congress). Piscataway:IEEE Press, 2017: 224-231.
[17]	SERDIO F , LUGHOFER E , PICHLER K ,et al. Residual-based fault detection using soft computing techniques for condition monitoring at rolling mills[J]. Information Sciences, 2014(259): 304-320.
[18]	KINGMA D P . Auto-encoding variational bayes[J]. arXiv:1312.6114. 2013
[19]	LU C , WANG Z Y , QIN W L ,et al. Fault diagnosis of rotary machinery components using a stacked denoising autoencoder-based health state identification[J]. Signal Processing, 2016(130): 377-388.
[20]	HOCHREITER S , SCHMIDHUBER J . Long short-term memory[J]. Neural computation, 1997(9): 1735-1780.
[21]	RANJAN C . Dataset:rare event classification in multivariate time series[J]. arXiv:1809.10717. 2018
[22]	SCHOLKOPF B , WILLIAMSONET R , et al . Support vector method for novelty detection[C]// Proceedings of the 12th International Conference on Neural Information Processing Systems. Cambridge:MIT Press, 1999: 582-588.
[23]	LIU F T , TING K M , ZHOU Z H . Isolation-based anomaly detection[J]. ACM Transactions on Knowledge Discovery from Data (TKDD), 2012,6(1): 1-39.
[24]	MALHOTRA P . LSTM-based encoder-decoder for multi-sensor anomaly detection[J]. arXiv:1607.00148. 2016

Metrics

Recommended 0

No Suggested Reading articles found!

标签类型	训练集	验证集	测试集
正常（y=0）	11 692	2 926	3 650
异常（y=1）	0	17	29

	网络层	单元数	输出维度	参数个数/个
编码器	BiLSTM	32	（5，64）	23 552
	LSTM	32	（5，32）	12 416
	LSTM	16	（1，16）	3 136
	RepeatVector	—	（5，16）	0
解码器	LSTM	16	（5，16）	2 112
	LSTM	32	（5，32）	6 272
	Dense	59	（5，59）	1 947

模型	TPR	ACC	AUC
BiLSTM-AE	55.2%	93.4%	88.9%
iForest	31%	92.7%	72.1%
OC-SVM	24.1%	92.4%	-
LSTM-AE	48%	90.4%	81%

An LSTM auto-encoder based anomaly detection for industrial system

RichHTML

PDF下载

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 10

References 24

Related Articles 15

Metrics

Recommended 0

[1]	Jianbin WANG, Shuchun WANG, Shangjin LIAO, Shuyuan SHI. Research on 5G base station energy saving system based on DCNN-LSTM load prediction algorithm [J]. Telecommunications Science, 2023, 39(4): 133-141.
[2]	Shiying ZHENG, Yuan LI, Bohan YANG, Shuai MA, Shanpeng XIAO. Technology evolution and industry development of 5G + industrial proximity network [J]. Telecommunications Science, 2022, 38(Z1): 17-27.
[3]	Yaoqi LIU, Hongguang LI, Yiqing ZHOU, Jinglin SHI, Yingjiao MA, Manli QIAN. Industrial 5G based digital twin control platform for flexible intelligent manufacturing [J]. Telecommunications Science, 2022, 38(Z1): 174-183.
[4]	Wenzhe WANG, Gang AN, Jian WU, Wendong ZHAO, Yangang WANG, Yunfeng ZHAO, Zixiang DI, Zhaojie GU, Guixiang ZOU. An industrial internet business management platform that supports the cloud-native [J]. Telecommunications Science, 2022, 38(Z1): 203-213.
[5]	Bin ZHUGE, Linchao WANG, Yang SONG, Yu SHAO, Ligang DONG, Xian JIANG. Routing planning and handoff based on LSTM traffic prediction [J]. Telecommunications Science, 2022, 38(8): 86-100.
[6]	Nan FENG, Helin LIU, Quan ZHOU, Haijun YANG, Mineng FU, Daping ZHONG. Research on 5G core network service model intelligent prediction [J]. Telecommunications Science, 2022, 38(8): 111-119.
[7]	Feng WANG, Qingmin YU, Ying HUANG, Shihui Duan. Research on key technology and development of industrial Internet network [J]. Telecommunications Science, 2022, 38(7): 106-113.
[8]	Xuesen MA, Shuyou CHEN, Xiangdong XU, Zhaokun CHU. Neural network and Markov based combination prediction algorithm of video popularity [J]. Telecommunications Science, 2021, 37(8): 18-26.
[9]	Chengyu ZHANG, Hongwu LI, Yang QU, Jinwu WEI. 5G edge computing development and application for industrial internet [J]. Telecommunications Science, 2021, 37(1): 129-136.
[10]	Jinyu ZHU,Shihui DUAN,Hengsheng ZHANG,Xuetao YU. Analysis on the application necessity of time-sensitive network technology in the field of industrial interconnection [J]. Telecommunications Science, 2020, 36(5): 115-124.
[11]	Rongrong JIANG,Zhengqiu WENG,Tieming CHEN. Development of industrial internet platform and its security challenges [J]. Telecommunications Science, 2020, 36(3): 6-10.
[12]	Yuan CHEN,Tieming CHEN,Qijie SONG,Dongjie MA. IBEXSec:a universal security framework for industrial internet terminal access [J]. Telecommunications Science, 2020, 36(3): 19-26.
[13]	Yufei CUI,Yunyong ZHANG,Di ZHANG,Xiaoxia WANG,Siyu PAN,Kun LI. Analysis and development suggestions of operators in the field of industrial internet identification analysis system [J]. Telecommunications Science, 2020, 36(2): 109-116.
[14]	Xiaoxia WANG,Hui ZHAO,Yufei CUI,Kun LI,Xiao WANG. Research on B2B business model of operators in 5G era [J]. Telecommunications Science, 2020, 36(11): 149-155.
[15]	Yingqi WEI,Weiping LIN,Ying LI,Hua TAN. Analysis of operator’s flow middle platform in 5G⁺native system [J]. Telecommunications Science, 2020, 36(10): 147-152.