LSTM-VAE를 활용한 기계시설물 장치의 이상 탐지 시스템
Anomaly Detection System in Mechanical Facility Equipment: Using Long Short-Term Memory Variational Autoencoder
0
조회
0
다운로드
9
인용
0
참고문헌
키워드
초록
Purpose: The purpose of this study is to compare machine learning models for anomaly detection of mechanical facility equipment and suggest an anomaly detection system for mechanical facility equipment in subway stations. It helps to predict failures and plan the maintenance of facility. Ultimately it aims to improve the quality of facility equipment. Methods: The data collected from Daejeon Metropolitan Rapid Transit Corporation was used in this experiment. The experiment was performed using Python, Scikit-learn, tensorflow 2.0 for preprocessing and machine learning. Also it was conducted in two failure states of the equipment. We compared and analyzed five unsupervised machine learning models focused on model Long Short-Term Memory Variational Autoencoder(LSTM-VAE). Results: In both experiments, change in vibration and current data was observed when there is a defect. When the rotating body failure was happened, the magnitude of vibration has increased but current has decreased. In situation of axis alignment failure, both of vibration and current have increased. In addition, model LSTM-VAE showed superior accuracy than the other four base-line models. Conclusion: According to the results, model LSTM-VAE showed outstanding performance with more than 97% of accuracy in the experiments. Thus, the quality of mechanical facility equipment will be improved if the proposed anomaly detection system is established with this model used.
Abstract
Purpose: The purpose of this study is to compare machine learning models for anomaly detection of mechanical facility equipment and suggest an anomaly detection system for mechanical facility equipment in subway stations. It helps to predict failures and plan the maintenance of facility. Ultimately it aims to improve the quality of facility equipment. Methods: The data collected from Daejeon Metropolitan Rapid Transit Corporation was used in this experiment. The experiment was performed using Python, Scikit-learn, tensorflow 2.0 for preprocessing and machine learning. Also it was conducted in two failure states of the equipment. We compared and analyzed five unsupervised machine learning models focused on model Long Short-Term Memory Variational Autoencoder(LSTM-VAE). Results: In both experiments, change in vibration and current data was observed when there is a defect. When the rotating body failure was happened, the magnitude of vibration has increased but current has decreased. In situation of axis alignment failure, both of vibration and current have increased. In addition, model LSTM-VAE showed superior accuracy than the other four base-line models. Conclusion: According to the results, model LSTM-VAE showed outstanding performance with more than 97% of accuracy in the experiments. Thus, the quality of mechanical facility equipment will be improved if the proposed anomaly detection system is established with this model used.
참고문헌 (29)
- Akcay S, "Skip-ganomaly: Skip connected and adversarially trainedencoder-decoder anomaly detection", 2019 International Joint Conference on Neural Networks (IJCNN). IEEE (2019)
- Variational Autoencoder based AnomalyDetection using Reconstruction Probability, "An J. and Cho S", SNU Data MiningCenter, 1~18 (2015)
- Bank D., "Autoencoders", arXiv eprints., ~ (2020)
- Baur C., "Deep autoencoding models for unsupervised anomaly segmentation in brain MR images", Proceeding of International MICCAI Brainlesion Workshop, 11383, pp.161~169 (2018)
- Chalapathy R, "Deep Learning for Anomaly Detection: A Survey", CoRR, abs, pp.1901.03407 (2019), ~
- Chandola V., "Anomaly detection : A survey", ACM computing surveys(CSUR), 41(3), pp.15~ (2009)
- Anomaly Detection & Prediction Decoded: 6 Industries. Copious Challenges. Extraordinary Impact, "DataRPM, P.", ~ (2017)
- E. S. Lee, "Trends in AI Technology for Smart Manufacturing in the Future", Electronics and Telecommunications Trends, 35(1), pp.60~70 (2020)
- Guo Y., "Multidimensional Time Series Anomaly Detection: A GRU-based Gaussian Mixture Variational Autoencoder Approach", Proceedings of the 10th Asian Conference on Machine Learning, 95, pp.97~112 (2018)
- Harmeling S., "From outliers to prototypes : ordering data", Neurocomputing, 69(13-15), pp.1608~1618 (2006)
- Hasan M., "Attack and anomaly detection in IoT sensors in IoT sites using machine learning approaches", Internet of Things, 7, pp.100059~ (2019)
- Hodge V., "A survey of outlier detection methodologies", Artificial Intelligence Review, 22(2), pp.85~126 (2004)
- 김준세, "타겟 샘플링 검사를 통한 출하품질 향상에 관한 사례 연구", 품질경영학회지, 49(3), pp.421~431 (2021), 한국품질경영학회[DOI]
- K. H. Sun, "Vision-Based Fault Diagnostics Using Explainable Deep Learning with Class Activation Map", IEEE Access, 8, pp.129169~129179 (2020)
- Li D, "Anomaly detection with generative adversarial networks for multivariate time series", arXiv preprint arXiv:1809.04758, ~ (2018)
- 김명준, "오차항과 러닝 기법을 활용한 예측진단 시스템 개선 방안 연구", 품질경영학회지, 47(4), pp.783~793 (2019), 한국품질경영학회[DOI]
- 서명교, "오토인코더를 이용한 열간 조압연설비 상태모니터링과 진단", 품질경영학회지, 47(1), pp.75~86 (2019), 한국품질경영학회[DOI]
- Maesschalck R D, "The Mahalanobis Distance", Chemometrics and Intelligent Laboratory Systems, 50(1), pp.1~18 (2000)
- Malhotra P, "Long short term memory networks for anomaly detection in time series", Proceedings of Presses Universitaires de Louvain, 89, pp.89~94 (2015)
- Munawar A, "Limiting the reconstruction capability of generative neural network using negative learning", 2017 IEEE 27th International Workshop on Machine Learning for Signal Processing (MLSP) (2017), IEEE
- N. Y. Choi, "Detecting user behavior anomalies using Generative Adversarial Networks", Intelligence Information Research, 25(3), pp.43~62 (2019)
- Park D, "A Multimodal Anomaly Detector for Robot-Assisted Feeding Using an LSTM-Based Variational Autoencoder", IEEE Robotics and Automation Letters, 3(3), pp.1544~1551 (2018)
- Perera P, "Ocgan: One-class novelty detection using gans with constrained latent representations" (2019), Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition
- Pokrajac D, "Incremental Local Outlier Detection for Data Streams", CIDM Conference (2007)
- Rousseeuw P., "Robust Regression and Outlier Detection" (2003), Wiley
- Son V H, "Anomaly detection with adversarial dual autoencoders", arXiv eprint, ~ (2019)
- Wei Q, "Anomaly detection for medical images based on a one-class classification" (2018), Proceeding of Medical Imaging 2018: Computer-Aided Diagnosis
- Yamanaka Y, "Autoencoding binary classifiers for supervised anomaly detection", Pacific Rim International Conference on Artificial Intelligence (2019)
- Zong B, "Deep autoencoding gaussian mixture model for unsupervised anomaly detection", Proceeding of International Conference on Learning Representations (2018)