DOI QR코드

DOI QR Code

Anomaly detection in particulate matter sensor using hypothesis pruning generative adversarial network

  • Received : 2020.03.09
  • Accepted : 2020.07.10
  • Published : 2021.06.01

Abstract

The World Health Organization provides guidelines for managing the particulate matter (PM) level because a higher PM level represents a threat to human health. To manage the PM level, a procedure for measuring the PM value is first needed. We use a PM sensor that collects the PM level by laser-based light scattering (LLS) method because it is more cost effective than a beta attenuation monitor-based sensor or tapered element oscillating microbalance-based sensor. However, an LLS-based sensor has a higher probability of malfunctioning than the higher cost sensors. In this paper, we regard the overall malfunctioning, including strange value collection or missing collection data as anomalies, and we aim to detect anomalies for the maintenance of PM measuring sensors. We propose a novel architecture for solving the above aim that we call the hypothesis pruning generative adversarial network (HP-GAN). Through comparative experiments, we achieve AUROC and AUPRC values of 0.948 and 0.967, respectively, in the detection of anomalies in LLS-based PM measuring sensors. We conclude that our HP-GAN is a cutting-edge model for anomaly detection.

Keywords

Acknowledgement

We are grateful to all the members of our team and to SK Planet Co., Ltd., who have supported this research, not only via data collection, but also by providing equipment for the experiment.

References

  1. World Health Organization, Who air quality guidelines for particulate matter, ozone, nitrogen dioxide and sulfur dioxid : global update 2005: summary of risk assessment, WHO Press, Geneva, Switzerland, Rep. WHO/SDE/PHE/OEH/06.02, 2006.
  2. A. Faustini et al., The relationship between ambient particulate matter and respiratory mortality: a multi-city study in Italy, Eur. Respir. J. 38 (2011), 538-547. https://doi.org/10.1183/09031936.00093710
  3. Y. Du et al., Air particulate matter and cardiovascular disease: The epidemiological, biomedical and clinical evidence, J. Thorac. Dis. 8 (2016), E8-E19.
  4. G. B. Hamra et al., Outdoor particulate matter exposure and lung cancer: A systematic review and meta-analysis , Environ. Health Perspect. 122 (2014), 906-911. https://doi.org/10.1289/ehp.1408092
  5. M. Badura et al., Evaluation of low-cost sensors for ambient PM 2.5 monitoring, J. Sensors. 2018 (2018), 1-16.
  6. F. M. J. Bulot et al., Long-term field comparison of multiple low-cost particulate matter sensors in an outdoor urban environment, Sci. Rep. 9 (2019), no. 1, 7497. https://doi.org/10.1038/s41598-019-43716-3
  7. W. Shao, H. Zhang, and H. Zhou, Fine particle sensor based on multi-angle light scattering and data fusion, Sensors, 17 (2017), no. 5, 1033. https://doi.org/10.3390/s17051033
  8. J. Shi et al., Validation of a light-scattering PM2.5 sensor monitor based on the long-term gravimetric measurements in field tests, PLoS One 12 (2017), no. 11, 1-13.
  9. D. M. Himmelblau, R. W. Barker, and W. Suewatanakul, Fault classification with the aid of artificial neural networks, in Fault Detection, Supervision and Safety for Technical Processes, Pergamon, Oxford, England, 1992, pp. 541-545.
  10. B. Tellenbach et al., Accurate network anomaly classification with generalized entropy metrics, Comput. Netw. 55 (2011), 3485-3502. https://doi.org/10.1016/j.comnet.2011.07.008
  11. E. de la Hoz et al., Network anomaly classification by support vector classifiers ensemble and non-linear projection techniques, in Hybrid Artificial Intelligent Systems, Springer, Berlin, Germany, 2013, pp. 103-111.
  12. S. U. Jan et al., Sensor fault classification based on support vector machine and statistical time-domain features, IEEE Access 5 (2017), 8682-8690. https://doi.org/10.1109/ACCESS.2017.2705644
  13. K. B. Lee, S. Cheon, and C. O. Kim, A convolutional neural network for fault classification and diagnosis in semiconductor manufacturing processes, IEEE Trans. Semicond. Manuf. 30 (2017), 135-142. https://doi.org/10.1109/TSM.2017.2676245
  14. I. Steinwart, D. Hush, and C. Scovel, A classification framework for anomaly detection, J. Mach. Learn. Res. 6 (2005), 211-232.
  15. M. Sakurada and T. Yairi, Anomaly detection using autoencoders with nonlinear dimensionality reduction, in Proc. Mach. Learn. Sens. Data Anal. ( Gold Coast, Australia), 2014, pp. 4:4-4:11.
  16. C. Zhou and R. C. Paffenroth, Anomaly detection with robust deep autoencoders, in Proc. ACM SIGKDD Int. Conf. Knowl. Discov. Data Min. (Halifax, Canada), 2017, pp. 665-674.
  17. P. Malhotra et al., Long short termmemory networks for anomaly detection in time series, in ESANN 2015, Presses Universitaires de Louvain, Bruges, Belgium, 2015, p. 89.
  18. S. Chauhan and L. Vig, Anomaly detection in ECG time signals via deep long short-term memory networks, in Proc. IEEE Int. Conf. Data Sci. Adv. Anal. (Paris, France), Oct. 2015, pp. 1-7.
  19. H. Xu et al., Unsupervised anomaly detection via variational auto-encoder for seasonal KPIs in web applications, in Proc. World Wide Web Conf. (Republic and Canton of Geneva, Switzerland), Apr. 2018, pp. 187-196.
  20. T. Schlegl et al., Unsupervised anomaly detection with generative adversarial networks to guide marker discovery, in Information Processing in Medical Imaging, Springer, Boone, NC, USA, 2017, pp. 146-157.
  21. J. Donahue, P. Krahenbuhl, and T. Darrell, Adversarial feature learning, arXiv preprint, CoRR, 2016, arXiv: abs/1605.09782.
  22. S. Akcay, A. Atapour-Abarghouei, and T. P. Breckon, Ganomaly: Semi-supervised anomaly detection via adversarial training, in Computer Vision-ACCV 2018, Springer, Perth, Australia, 2019, pp. 622-637.
  23. X. Wang et al., Selfadversarial variational autoencoder with Gaussian anomaly prior distribution for anomaly detection, arXiv preprint, CoRR, 2019, arXiv: abs/1903.00904.
  24. C. Zhang and Y. Chen, Time series anomaly detection with variational autoencoders, arXiv preprint, CoRR, 2019, arXiv: abs/1907.01702.
  25. D.T. Nguyen et al., Anomaly detection with multiple-hypotheses predictions, in Proc. Int. Conf. Mach. Learn. (Long Beach, CA, USA), June 2019, pp. 4800-4809.
  26. D. P. Kingma and M. Welling, Auto-encoding variational Bayes, in Proc. Int. Conf. Learn. Representations (Banff, Canada), Apr. 2014, pp. 1-14.
  27. S. B. Serpico et al., Comparison of feature reduction techniques for classification of hyperspectral remote-sensing data, in Image and Signal Processing for Remote Sensing VIII, vol. 4885, SPIE, Crete, Greece, 2003, pp. 347-358.
  28. J. Yan et al., Effective and efficient dimensionality reduction for large-scale and streaming data preprocessing, IEEE Trans. Knowl. Data Eng. 18 (2006), 320-333. https://doi.org/10.1109/TKDE.2006.45
  29. K. Thangavel and A. Pethalakshmi, Dimensionality reduction based on rough set theory: a review, Appl. Soft Comput. 9 (2009), 1-12.
  30. A. Balzanella, A. Irpino, and R. Verde, Dimensionality reduction techniques for streaming time series: A new symbolic approach, in Classification as a Tool for Research, Springer, Berlin, Germany, 2010, pp. 381-389.
  31. Y. Park and I. D. Yun, Fast adaptive RNN encoder-decoder for anomaly detection in SMD assembly machine, Sensors 18 (2018), no. 10, 3573. https://doi.org/10.3390/s18103573
  32. T. Mikolov et al., Recurrent neural network based language model, in Proc. Annu. Conf. Int. Speech Commun. Assoc. (Chiba, Japan), Sept. 2010, pp. 1045-1048.
  33. S. S. Blackman, Multiple hypothesis tracking for multiple target tracking, IEEE Aerosp. Electron. Syst. Mag. 19 (2004), 5-18. https://doi.org/10.1109/maes.2004.1263228
  34. D. Streller and K. Dietmayer. Object tracking and classification using a multiple hypothesis approach, IEEE Intell. Veh. Symp. 19 (2004), no. 1, 808-812.
  35. C. Rupprecht et al., Learning in an uncertain world: Representing ambiguity through multiple hypotheses, in Proc. IEEE Int. Conf. Comput. Vis. (Venice, Italy), Oct. 2017, pp. 3591-3600.
  36. W. Maass, On the computational power of winner-take-all, Neural Comput. 12 (2000), 2519-2535. https://doi.org/10.1162/089976600300014827
  37. X. Glorot and Y. Bengio, Understanding the difficulty of training deep feedforward neural networks, in Proc. Int. Conf. Artif. Intell. Stat. (Sardinia, Italy), May 2010, pp. 249-256.
  38. D. P. Kingma and J. B. Adam. A method for stochastic optimization, arXiv arXiv preprint, 2014, arXiv:1412.6980.
  39. J. W. Tukey, Exploratory Data Analysis, vol. 2, Addison-Wesley, Reading, MA, USA, 1977.
  40. T. Fawcett, An introduction to ROC analysis, Pattern Recogni. Lett. 27 (2006), 861-874. https://doi.org/10.1016/j.patrec.2005.10.010
  41. T. Saito and M. Rehmsmeier, The precision-recall plot is more informative than the ROC plot when evaluating binary classifiers on imbalanced datasets, PLoS One 10 (2015), 1-21.
  42. D.-A. Clevert, T. Unterthiner, and S. Hochreiter, Fast and accurate deep network learning by exponential linear units (elus), arXiv arXiv preprint, 2015, arXiv:1511.07289.
  43. S. Anwar, K. Hwang, and W. Sung, Structured pruning of deep convolutional neural networks, arXiv preprint, CoRR, 2015, arXiv: abs/1512.08571.
  44. R. Meyes et al., Ablation studies in artificial neural networks, arXiv preprint, CoRR, 2019, arXiv: abs/1901.08644.
  45. D. Chicco. Ten quick tips for machine learning in computational biology, BioData Mining 10 (2017), no. 1, 35. https://doi.org/10.1186/s13040-017-0155-3
  46. P. Dirk et al., Why the Monte Carlo method is so important today, Wiley Interdiscip. Rev. Comput. Stat. 6 (2014), 386-392.