Intelligent Anomaly Detection for Industrial Product Quality Inspection Based on Gaussian Restricted Boltzmann Machine

doi:10.3969/j.issn.1007-7375.2022.06.018

Abstract

Abstract: In order to improve the quality anomaly detection performance of industrial products based on complex data features, an intelligent anomaly detection method FE-GRBM for industrial product quality inspection is proposed based on the comprehensive advantages of Gaussian Restricted Boltzmann machine in nonlinear modeling and computational complexity. Specifically, the free energy based model training and detection strategies are studied, using the mathematical relationship between free energy and natural logarithm of edge probability of GRBM. The method are verified by the quality inspection of cigarette products in real case. The results show that the average performance of FE-GRBM is 0.29 higher than the highest score among the three traditional methods, and 0.04 higher than that of RE-GRBM, which shows the superiority of our method.

Key words: anomaly detection, restricted Boltzmann machine (RBM), free-energy function, quality inspection, cigarette products

CLC Number:

TS452

HUANG Cong, NONG Yingxiong, ZHANG Yi. Intelligent Anomaly Detection for Industrial Product Quality Inspection Based on Gaussian Restricted Boltzmann Machine[J]. Industrial Engineering Journal, 2022, 25(6): 152-159.

References

[1] FRANCESCHINI F, GALETTO M, GENTA G, et al. Selection of quality-inspection procedures for short-run productions[J]. The International Journal of Advanced Manufacturing Technology, 2018, 99(9): 2537-2547
[2] 曾婷, 黄东军. 智能视频监控系统异常行为检测算法研究综述[J]. 计算机测量与控制, 2021, 29(7): 1-6
ZENG Ting, HUANG Dongjun. A survey of detection algorithms for abnormal behaviors in intelligent video surveillance system[J]. Computer Measurement & Control, 2021, 29(7): 1-6
[3] 刘晓华, 高俊成, 王少林, 等. 智能电网异常用电检测框架[J]. 自动化技术与应用, 2020, 39(12): 124-128
LIU Xiaohua, GAO Juncheng, WANG Shaolin, et al. Smart grid abnormal power detection framework[J]. Techniques of Automation and Applications, 2020, 39(12): 124-128
[4] 丁小欧, 于晟健, 王沐贤, 等. 基于相关性分析的工业时序数据异常检测[J]. 软件学报, 2020, 31(3): 726-747
DING Xiaoou, YU Shengjian, WANG Muxian, et al. Anomaly detection on industrial time series based on correlation analysis[J]. Journal of Software, 2020, 31(3): 726-747
[5] 盛铭, 陈凌珊, 汪俊杰, 等. 基于单分类支持向量机的CAN总线异常检测方法[J]. 汽车技术, 2020(5): 21-25
SHENG Ming, CHEN Lingshan, WANG Junjie, et al. Abnormity detection method for in-vehicle CAN bus based on one-class SVM[J]. Automobile Technology, 2020(5): 21-25
[6] 张丹丹, 游子毅, 郑建, 等. 基于改进的局部异常因子检测的优化聚类算法[J]. 微电子学与计算机, 2019, 36(11): 43-48
ZHANG Dandan, YOU Ziyi, ZHENG Jian, et al. Optimal clustering algorithm based on modified local outlier factor detection[J]. Microelectronics & Computer, 2019, 36(11): 43-48
[7] 张怀峰, 皮德常, 董玉兰. iBTC: 一种基于独立森林的移动对象轨迹聚类算法[J]. 计算机科学, 2019, 46(1): 251-259
ZHANG Huaifeng, PI Dechang, DONG Yulan. iBTC: a trajectory clustering algorithm based on isolation forest[J]. Computer Science, 2019, 46(1): 251-259
[8] STOJANOVIC L , DINIC M , STOJANOVIC N, et al. Big-data-driven anomaly detection in industry (4.0) : an approach and a case study[C]//2016 IEEE International Conference on Big Data (Big Data). Washington DC, USA: IEEE, 2016: 1647-1652.
[9] HINTON G, OSINDERO S, TEH Y W. A fast learning algorithm for deep belief nets[J]. Neural Computation, 2014, 18(7): 1527-1554
[10] CHERLA S, TRAN S N, GARCEZ A D, et al. Generalising the discriminative restricted boltzmann machines[C]//Artificial Neural Networks and Machine Learning-ICANN 2017. Alghero, Italy: Springer, 2017: 10614.
[11] ZHAI S, CHENG Yu, LU W, et al. Deep structured energy based models for anomaly detection[C]//The 33^rd International Conference on Machine Learning. New York, USA: JMLR: W&CP, 2016: 48.
[12] DO K, TRAN T, PHUNG D, et al. Outlier detection on mixed-type data: an energy-based approach[C]//Advanced data mining and applications 2016. Australia: Springer, 2016: 10086.
[13] FISCHER A, IGEL C. An introduction to restricted Boltzmann machines[C]//Pattern Recognition, Image Analysis, Computer Vision, and Applications 2012. Berlin Heidelberg: Springer, 2012: 7441.
[14] HINTON G. Training products of experts by minimizing contrastive divergence[J]. Neural Computation, 2002, 14(8): 1771-1800
[15] KRIZHEVSKY A. Learning multiple layers of features from tiny images[R/OL]. (2012-05) . http://www.cs.toronto.edu/~kriz/learning-features-2009-TR.pdf.
[16] HINTON G. A practical guide to training restricted Boltzmann machines[J]. Momentum, 2010, 9(1): 926-947
[17] HE R, LI Q, BO A, et al. A kernel-power-density-based algorithm for channel multipath components clustering[J]. IEEE Transactions on Wireless Communications, 2017, 16(11): 7138-7151
[18] 瞿先中, 张劲, 蒋士盛, 等. 细支卷烟烟支物理指标相关性分析[J]. 湖南文理学院学报 (自然科学版), 2020, 32(4): 64-68
QU Xianzhong, ZHANG Jing, JIANG Shisheng, et al. Correlation analysis of physical indicators in slim cigarettes[J]. Journal of Hunan University of Arts and Science (Science and Technology), 2020, 32(4): 64-68
[19] PEDREGOSA F, VAROQUAUX G, GRAMFORT A, et al. Scikit-learn: machine learning in python[J]. Journal of Machine Learning Research, 2011, 12(1): 2825-2830