Anomaly detection in scientific data using joint statistical moments

Konduri, Aditya; Kolla, Hemanth; Kegelmeyer, W. Philip; Shead, Timothy M.; Ling, Julia; Davis, Warren L.

doi:10.1016/j.jcp.2019.03.003

Title: Anomaly detection in scientific data using joint statistical moments

Journal Article · Wed Mar 13 00:00:00 EDT 2019 · Journal of Computational Physics

DOI:https://doi.org/10.1016/j.jcp.2019.03.003· OSTI ID:1502973

Konduri, Aditya ^[1]; Kolla, Hemanth ^[1]; Kegelmeyer, W. Philip ^[1]; Shead, Timothy M. ^[2]; Ling, Julia ^[3]; Davis, Warren L. ^[2]

Sandia National Lab. (SNL-CA), Livermore, CA (United States)
Sandia National Lab. (SNL-NM), Albuquerque, NM (United States)
Citrine Informatics, Redwood City, CA (United States)

We propose an anomaly detection method for multi-variate scientific data based on analysis of high-order joint moments. Using kurtosis as a reliable measure of outliers, we suggest that principal kurtosis vectors, by analogy to principal component analysis (PCA) vectors, signify the principal directions along which outliers appear. The inception of an anomaly, then, manifests as a change in the principal values and vectors of kurtosis. Obtaining the principal kurtosis vectors requires decomposing a fourth order joint cumulant tensor for which we use a simple, computationally less expensive approach that involves performing a singular value decomposition (SVD) over the matricized tensor. We demonstrate the efficacy of this approach on synthetic data, and develop an algorithm to identify the occurrence of a spatial and/or temporal anomalous event in scientific phenomena. The algorithm decomposes the data into several spatial sub-domains and time steps to identify regions with such events. Feature moment metrics, based on the alignments of the principal kurtosis vectors, are computed at each sub-domain and time step for all features to quantify their relative importance towards the overall kurtosis in the data. Accordingly, spatial and temporal anomaly metrics for each sub-domain are proposed using the Hellinger distance of the feature moment metric distribution from a suitable nominal distribution. Finally, we apply the algorithm to two turbulent auto-ignition combustion cases and demonstrate that the anomaly metrics reliably capture the occurrence of auto-ignition in relevant spatial sub-domains at the right time steps.

View Accepted Manuscript (DOE)

View Accepted Manuscript (Publisher)

Cite

Export

Save

Research Organization:: Sandia National Lab. (SNL-CA), Livermore, CA (United States); Sandia National Lab. (SNL-NM), Albuquerque, NM (United States)

Sponsoring Organization:: USDOE Office of Science (SC), Advanced Scientific Computing Research (ASCR); USDOE National Nuclear Security Administration (NNSA)

Grant/Contract Number:: NA0003525; AC04-94AL85000; FWP16-019471

OSTI ID:: 1502973

Alternate ID(s):: OSTI ID: 1502456; OSTI ID: 1636004

Report Number(s):: SAND-2019-2948J; SAND-2018-8923J; 673503

Journal Information:: Journal of Computational Physics, Vol. 387; ISSN 0021-9991

Publisher:: ElsevierCopyright Statement

Country of Publication:: United States

Language:: English

Citation Metrics:

Cited by: 5 works

Citation information provided by
Web of Science

References (19)

Anomaly detection: A survey Chandola, Varun; Banerjee, Arindam; Kumar, Vipin ACM Computing Surveys, Vol. 41, Issue 3, p. 1-58 https://doi.org/10.1145/1541880.1541882	journal	July 2009
Using feature importance metrics to detect events of interest in scientific computing applications Ling, Julia; Kegelmeyer, W. Philip; Aditya, Konduri 2017 IEEE 7th Symposium on Large Data Analysis and Visualization (LDAV) https://doi.org/10.1109/LDAV.2017.8231851	conference	October 2017
Numerically stable, scalable formulas for parallel and online computation of higher-order multivariate central moments with arbitrary weights Pébay, Philippe; Terriberry, Timothy B.; Kolla, Hemanth Computational Statistics, Vol. 31, Issue 4 https://doi.org/10.1007/s00180-015-0637-z	journal	March 2016
Procedures for Detecting Outlying Observations in Samples Grubbs, Frank E. Technometrics, Vol. 11, Issue 1 https://doi.org/10.1080/00401706.1969.10490657	journal	February 1969
On the evaluation of unsupervised outlier detection: measures, datasets, and an empirical study Campos, Guilherme O.; Zimek, Arthur; Sander, Jörg Data Mining and Knowledge Discovery, Vol. 30, Issue 4 https://doi.org/10.1007/s10618-015-0444-8	journal	January 2016
A Comparative Evaluation of Unsupervised Anomaly Detection Algorithms for Multivariate Data Goldstein, Markus; Uchida, Seiichi PLOS ONE, Vol. 11, Issue 4 https://doi.org/10.1371/journal.pone.0152173	journal	April 2016
Kurtosis as Peakedness, 1905–2014. R.I.P. Westfall, Peter H. The American Statistician, Vol. 68, Issue 3 https://doi.org/10.1080/00031305.2014.917055	journal	July 2014
Tensor Decompositions and Applications Kolda, Tamara G.; Bader, Brett W. SIAM Review, Vol. 51, Issue 3 https://doi.org/10.1137/07070111X	journal	August 2009
Symmetric Tensors and Symmetric Tensor Rank Comon, Pierre; Golub, Gene; Lim, Lek-Heng SIAM Journal on Matrix Analysis and Applications, Vol. 30, Issue 3 https://doi.org/10.1137/060661569	journal	January 2008
Independent component analysis and (simultaneous) third-order tensor diagonalization de Lathauwer, L.; de Moor, B.; Vandewalle, J. IEEE Transactions on Signal Processing, Vol. 49, Issue 10 https://doi.org/10.1109/78.950782	journal	January 2001
Advanced compression-ignition engines—understanding the in-cylinder processes Dec, John E. Proceedings of the Combustion Institute, Vol. 32, Issue 2 https://doi.org/10.1016/j.proci.2008.08.008	journal	January 2009
The Reheat Concept: The Proven Pathway to Ultralow Emissions and High Efficiency and Flexibility Güthe, Felix; Hellat, Jaan; Flohr, Peter Journal of Engineering for Gas Turbines and Power, Vol. 131, Issue 2 https://doi.org/10.1115/1.2836613	journal	December 2008
Direct numerical simulation of flame stabilization assisted by autoignition in a reheat gas turbine combustor Aditya, Konduri; Gruber, Andrea; Xu, Chao Proceedings of the Combustion Institute, Vol. 37, Issue 2 https://doi.org/10.1016/j.proci.2018.06.084	journal	January 2019
Trigger Detection for Adaptive Scientific Workflows Using Percentile Sampling Bennett, Janine C.; Bhagatwala, Ankit; Chen, Jacqueline H. SIAM Journal on Scientific Computing, Vol. 38, Issue 5 https://doi.org/10.1137/15M1027942	journal	January 2016
Three-dimensional direct numerical simulation of a turbulent lifted hydrogen jet flame in heated coflow: a chemical explosive mode analysis Lu, T. F.; Yoo, C. S.; Chen, J. H. Journal of Fluid Mechanics, Vol. 652 https://doi.org/10.1017/S002211201000039X	journal	May 2010
Terascale direct numerical simulations of turbulent combustion using S3D Chen, J. H.; Choudhary, A.; de Supinski, B. Computational Science & Discovery, Vol. 2, Issue 1 https://doi.org/10.1088/1749-4699/2/1/015001	journal	January 2009
Scalar mixing in direct numerical simulations of temporally evolving plane jet flames with skeletal CO/H2 kinetics Hawkes, Evatt R.; Sankaran, Ramanan; Sutherland, James C. Proceedings of the Combustion Institute, Vol. 31, Issue 1 https://doi.org/10.1016/j.proci.2006.08.079	journal	January 2007
Direct numerical simulations of HCCI/SACI with ethanol Bhagatwala, Ankit; Chen, Jacqueline H.; Lu, Tianfeng Combustion and Flame, Vol. 161, Issue 7 https://doi.org/10.1016/j.combustflame.2013.12.027	journal	July 2014
A Multilinear Singular Value Decomposition De Lathauwer, Lieven; De Moor, Bart; Vandewalle, Joos SIAM Journal on Matrix Analysis and Applications, Vol. 21, Issue 4 https://doi.org/10.1137/s0895479896305696	journal	January 2000

Similar Records

Fast and efficient identification of anomalous galaxy spectra with neural density estimation

Journal Article · Tue Sep 12 00:00:00 EDT 2023 · Monthly Notices of the Royal Astronomical Society · OSTI ID:1502973

Böhm, Vanessa; Kim, Alex G.; Juneau, Stéphanie

MAGNETO-FRICTIONAL MODELING OF CORONAL NONLINEAR FORCE-FREE FIELDS. I. TESTING WITH ANALYTIC SOLUTIONS

Journal Article · Sat Sep 10 00:00:00 EDT 2016 · Astrophysical Journal · OSTI ID:1502973

Guo, Y.; Keppens, R.; Xia, C.

Improving Subsurface Stress Characterization for Carbon Dioxide Storage Projects by Incorporating Machine Learning Techniques

Technical Report · Thu Mar 30 00:00:00 EDT 2023 · OSTI ID:1502973

Ampomah, William; Huang, Lianjie; Balch, Robert; +13 more

Related Subjects

97 MATHEMATICS AND COMPUTING
anomaly detection
scientific computing
co-kurtosis
tensor decomposition
Hellinger distance
auto-ignition
Hellinger distance
Auto-ignition

Title: Anomaly detection in scientific data using joint statistical moments

Citation Formats

References (19)

Similar Records

Related Subjects