From Data to Alarms: Data-driven Anomaly Detection Techniques in Industrial Settings
Description
This paper introduces a data-driven methodology for anomaly detection in industrial processes. Our focus is on minimizing misclassifications of normal operations and enhancing anomaly and outlier detection. This optimization is based on presumed ground truth (GT) labels associated with a dependent variable (isobutane concentration). Utilizing a moving-horizon approach on an extensive industrial dataset, we perform a comprehensive evaluation of filtering algorithms, and present a representative outlier classification. Secondly, effective anomaly detection, distinct from outlier detection, is achieved by integrating a regression model trained on measurements from independent process variables to fit the dependent variable. Trained regression models consistently achieve effective prediction, staying within an approved process tolerance.
Files
fabermanuscript.pdf
Files
(380.9 kB)
Name | Size | Download all |
---|---|---|
md5:6c4bed04f3b9edc61b8efe3a3e5e822a
|
380.9 kB | Preview Download |
Additional details
Funding
- FrontSeat – Fostering Opportunities Towards Slovak Excellence in Advanced Control for Smart Industries 101079342
- European Commission
- Efficient control of industrial plants using data VEGA 1/0691/21
- The Vega Science Trust
- Energy-efficient Safe and Secure Process Control APVV-20-0261
- Slovak Research and Development Agency
References
- D. O. Afanasyev, E. A. Fedorova, 2019. On the impact of outlier filtering on the electricity price forecasting accuracy. Applied Energy 236, 196–210
- A. Blázquez-García, A. Conde, U. Mori, J. A. Lozano, 2021. A review on outlier/anomaly detection in time series data. ACM Comput. Surv. 54 (3)
- K. M. Carter, W. W. Streilein, 2012. Probabilistic reasoning for streaming anomaly detection. In: 2012 IEEE Statistical Signal Processing Workshop (SSP). pp. 377–380
- A. V. Oppenheim, 1999. Discrete-time signal processing. Pearson Education India
- IBM Corporation, 2021. SPSS Modeler 18.1.0, Outliers. URL https://www.ibm.com/docs/en/s- pss-modeler/18.1.0?topic=series-outliers
- Z. Jin, J. Zhao, L. Ding, S. Chakrabarti, E. Gryazina, V. Terzija, 2022. Power system anomaly detection using innovation reduction properties of iterated extended Kalman filter. International Journal of Electrical Power & Energy Systems 136, 107613
- F. Santosa, W. W. Symes, 1986. Linear inversion of band-limited reflection seismograms. SIAM Journal on Scientific and Statistical Computing 7 (4), 1307–1330
- T. Tziolas, K. Papageorgiou, T. Theodosiou, E. Papageorgiou, T. Mastos, A. Papadopoulos, 2022. Autoencoders for anomaly detection in an industrial multivariate time series dataset. Engineering Proceedings 18 (1)
- Y. S. Yoon, W. Jeong, J. Kim, M. Seok, J. Park, J. Bae, K. Lee, J. H. Lee, 2022. Development of inferential sensor and real-time optimizer for a vacuum distillation unit by recurrent neural network modeling of time series data. Computers & Chemical Engineering 168, 108039
- S. W. Roberts, Aug. 1959. Control chart tests based on geometric moving averages. Technometrics 1 (3), 239–250
- K. Pearson, 1901. Liii. on lines and planes of closest fit to systems of points in space. The London, Edinburgh, and Dublin Phil. Magazine and Journal of Science 2 (11), 559–572
- P. Rousseeuw, K. Driessen, 08 1999. A fast algorithm for the minimum covariance determinant estimator. Technometrics 41, 212–223
- R. Fáber, K. Ľubušký, R. Paulen, 2023. Machine Learning-based Classification of Online Industrial Datasets. 2023 24th Int. Conf. on Process Control, 132–137