Hostname: page-component-78c5997874-j824f Total loading time: 0 Render date: 2024-11-10T09:07:20.504Z Has data issue: false hasContentIssue false
  • English
  • Français

Margin-based approach for outlier detection of industrial design data using a modified general regression neural network

Published online by Cambridge University Press:  09 February 2022

Jayaram Sivaramakrishnan Open the ORCID record for Jayaram Sivaramakrishnan [Opens in a new window] ,
Gareth Lee ,
David Parlevliet  and
Kok Wai Wong
Jayaram Sivaramakrishnan*
Affiliation:
College of Science, Health, Engineering and Education, Murdoch University, 90 South Street, Murdoch, WA6150, Australia
Gareth Lee
Affiliation:
College of Science, Health, Engineering and Education, Murdoch University, 90 South Street, Murdoch, WA6150, Australia
David Parlevliet
Affiliation:
College of Science, Health, Engineering and Education, Murdoch University, 90 South Street, Murdoch, WA6150, Australia
Kok Wai Wong
Affiliation:
College of Science, Health, Engineering and Education, Murdoch University, 90 South Street, Murdoch, WA6150, Australia
*
Author for correspondence: Jayaram Sivaramakrishnan, E-mail: j.sivaramakrishnan@murdoch.edu.au, jaya.sivaraman@hotmail.com
Get access Rights & Permissions [Opens in a new window]

Abstract

The choice of components in industrial design involves setting design parameters that typically must reside inside permissible ranges called “design margins”. This paper proposes a novel automated method called the Margin-Based General Regression Neural Network (MB-GRNN) that classifies design errors for design parameters that are outside of permissible ranges as outliers, directly from industrial design data, using an unsupervised machine learning approach. The method is based on a modified GRNN that estimates extremal margin boundaries of design parameters by self-learning the features from datasets. These extremal permissible margin boundaries are determined by “stretching out” the upper and lower GRNN surfaces using an iterative application of stretch factors (a second kernel weighting factor). The method creates a variable insensitive band surrounding the data cloud, interlinked with the normal regression function, providing upper and lower margin boundaries. These boundaries can then be used to determine outliers and to predict a range of permissible values of design parameters during design. Pushing out extremal margin boundaries reduce the false identification of outliers. This classification technique could be used by industrial engineers to detect likely outliers and to predict a range of permissible output limits for chosen design parameters. The efficacy of this method has been validated against the widespread Parzen window method by comparing experimental results from three multivariate datasets. It was found that the two methods have different but complementary capabilities. The MB-GRNN also uses a modified algorithm for estimating the smoothing parameter using a combination of clustering, k-nearest neighbor, and localized covariance matrix.

Keywords

Decision supportdesign marginGRNNindustrial designmachine learningoutlier detectionParzen window
Type
Research Article
Information
AI EDAM , Volume 36 , 2022 , e7
Check for updates
Copyright
Copyright © The Author(s), 2022. Published by Cambridge University Press

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

Aglodiya, A (2017) Application of artificial neural network (ANN) in chemical engineering: a review. IJARIIE 3, 5322–5328. ISSN 2395-4396 (Online).Google Scholar
Al-Mahasneh, AJ, Anavatti, SG and Garratt, MA (2018) Review of applications of generalized regression neural networks in identification and control of dynamic systems, pp. 1–5. Available at http://arxiv.org/abs/1805.11236.Google Scholar
Austin-Breneman, J, Yu, BY and Yang, MC (2015) Biased information passing between subsystems over time in complex system design. Journal of Mechanical Design 138, 1–9. doi:10.1115/1.4031745.Google Scholar
Baqqar, M, Wang, T, Ahmed, M, Gu, F, Lu, J and Ball, A (2012) A general regression neural network model for gearbox fault detection using motor operating parameters. UKACC International Conference on Control 2012, Cardiff, UK.CrossRefGoogle Scholar
Berkhin, P (2002) Clustering survey Bherkin (Accrue Software). Available at https://www.cc.gatech.edu/~isbell/reading/papers/berkhin02survey.pdf.Google Scholar
Blazquez-Garcia, A, Conde, A, Mori, U and Lozano, JA (2020) A review on outlier/anomaly detection in time series data. ACM Computing Surveys, 1–32. https://arxiv.org/abs/2002.04236v1.Google Scholar
Buchan, AB and Bolton, CJ (2009) Process and structural safety: do we need to have hazards? IET Conference Publications 2009, 2–4. doi:10.1049/cp.2009.1539.Google Scholar
Cai, Z (2001) Weighted Nadaraya-Watson regression estimation. Statistics and Probability Letters 51, 307–318. doi:10.1016/S0167-7152(00)00172-3.CrossRefGoogle Scholar
Celikoglu, HB and Cigizoglu, HK (2007) Public transportation trip flow modeling with generalized regression neural networks. Advances in Engineering Software 38, 71–79. doi:10.1016/j.advengsoft.2006.08.003.CrossRefGoogle Scholar
Cigizoglu, HK and Alp, M (2006) Generalized regression neural network in modelling river sediment yield. Advances in Engineering Software 37, 63–68. doi:10.1016/j.advengsoft.2005.05.002.CrossRefGoogle Scholar
Cover, TM and Hart, PE (1967) Nearest neighbor pattern classification. IEEE Transactions on Information Theory 13, 21–27.CrossRefGoogle Scholar
Cristianini, N and Shawe-Taylor, J (2000) An Introduction to Support Vector Machines and Other Kernel Based Learning Methods. New York: Cambridge University Press.CrossRefGoogle Scholar
De Maesschalck, R, Jouan-Rimbaud, D and Massart, DL (2000) The Mahalanobis distance. Chemometrics and Intelligent Laboratory Systems 50, 1–18. doi:10.1016/S0169-7439(99)00047-7.CrossRefGoogle Scholar
Dempster, AP (1972) Covariance selection. Biometrics 28, 157–175.CrossRefGoogle Scholar
Dowell, AM (2001) Critical safe operating parameters: “never exceed” limit and “never deviate” action. Process Safety Progress 20, 208–214. doi:10.1002/prs.680200310.CrossRefGoogle Scholar
Eckert, C and Isaksson, O (2017) Safety margins and design margins: a differentiation between interconnected concepts. 27th CIRP Design 2017.CrossRefGoogle Scholar
Eckert, C, Isaksson, O and Earl, C (2019) Design margins: a hidden issue in industry. Design Science 5, 1–24. doi:10.1017/dsj.2019.7.CrossRefGoogle Scholar
Forest, J (2018) Know your limits. Process Safety Progress 37, 498–501. doi:10.1002/prs.12000.CrossRefGoogle Scholar
Ge, Z, Song, Z, Ding, SX and Huang, B (2017) Data mining and analytics in the process industry: the role of machine learning. IEEE Access 5, 20590–20616. doi:10.1109/ACCESS.2017.2756872.CrossRefGoogle Scholar
He, X and He, S (2011) Fault detection of excavator's hydraulic system using dynamic general regression neural network. Applied Mechanics and Materials 48–49, 511–514. doi:10.4028/www.scientific.net/AMM.48-49.511.CrossRefGoogle Scholar
Hodge, VJ and Austin, J (2004) A survey of outlier detection methodologies. Artificial Intelligence Review 22, 85–126. doi:10.4324/9781315744988-22.CrossRefGoogle Scholar
Islam, MM, Lee, G and Hettiwatte, SN (2017) Application of a general regression neural network for health index calculation of power transformers. International Journal of Electrical Power and Energy Systems 93, 308–315. doi:10.1016/j.ijepes.2017.06.008.CrossRefGoogle Scholar
Jagan, J, Samui, P and Kim, D (2019) Reliability analysis of simply supported beam using GRNN, ELM and GPR. Structural Engineering and Mechanics 71, 739–749. doi:10.12989/sem.2019.71.6.739.Google Scholar
Jiang, P and Chen, J (2016) Displacement prediction of landslide based on generalized regression neural networks with K-fold cross-validation. Neurocomputing 198, 40–47. doi:10.1016/j.neucom.2015.08.118.CrossRefGoogle Scholar
Kartal, S, Oral, M and Ozyildirim, BM (2018) Pattern layer reduction for a generalized regression neural network by using a self-organizing map. International Journal of Applied Mathematics and Computer Science 28, 411–424. doi:10.2478/amcs-2018-0031.CrossRefGoogle Scholar
Kim, W and Katipamula, S (2018) A review of fault detection and diagnostics methods for building systems. Science and Technology for the Built Environment 24, 3–21. doi:10.1080/23744731.2017.1318008.CrossRefGoogle Scholar
Kulkarni, SG, Chaudhary, AK, Nandi, S, Tambe, SS and Kulkarni, BD (2004) Modeling and monitoring of batch processes using principal component analysis (PCA) assisted generalized regression neural networks (GRNN). Biochemical Engineering Journal 18, 193–210. doi:10.1016/j.bej.2003.08.009.CrossRefGoogle Scholar
Kumar, G and Malik, H (2016) Generalized regression neural network based wind speed prediction model for western region of India. International Conference on Advances in Computing & Communications, Cochin, India.CrossRefGoogle Scholar
Lee, GE and Zaknich, A (2015) A mixed-integer programming approach to GRNN parameter estimation. Information Sciences 320, 1–11. doi:10.1016/j.ins.2015.04.052.CrossRefGoogle Scholar
Lee, H, Kim, S and Jun, K (2018) The study for storm surge prediction using generalized regression neural networks. Journal of Coastal Research 85, 781–785. doi:10.2112/si85-157.1.CrossRefGoogle Scholar
Li, H, Zhao, J, Ni, X and Zhang, X (2018) Fault diagnosis for machinery based on feature extraction and general regression neural network. International Journal of Systems Assurance Engineering and Management 9, 1034–1046. doi:10.1007/s13198-018-0726-9.CrossRefGoogle Scholar
Linde, Y, Buzo, A and Gray, RM (1980) An algorithm for vector quantizer design. IEEE Transactions on Communications 28, 84–95. doi:10.1109/TCOM.1980.1094577.CrossRefGoogle Scholar
May, RJ, Maier, HR, Dandy, GC and Nixon, JB (2004) General regression neural networks for modeling disinfection residual in water distribution systems. World Water Congress, Marrakech, Morocco.CrossRefGoogle Scholar
Modarres, M (2009) Advanced nuclear power plant regulation using risk-informed and performance-based methods. Reliability Engineering and System Safety 94, 211–217. doi:10.1016/j.ress.2008.02.019.CrossRefGoogle Scholar
Mohandes, SR, Zhang, X and Mahdiyar, A (2019) A comprehensive review on the application of artificial neural networks in building energy analysis. Neurocomputing 340, 55–75. doi:10.1016/j.neucom.2019.02.040.CrossRefGoogle Scholar
Mussa, HY, Mitchell, JBO and Afzal, AM (2015) The Parzen window method: in terms of two vectors and one matrix. Pattern Recognition Letters 63, 30–35. doi:10.1016/j.patrec.2015.06.002.CrossRefGoogle ScholarPubMed
Nadaraya, EA (1964) On estimating regression. Theory of Probability & Its Applications 9, 141–142. doi:10.1137/1109020.CrossRefGoogle Scholar
Niu, D, Liang, Y and Hong, W-C (2017) Wind speed forecasting based on EMD and GRNN optimized by FOA. Energies 10, 12. doi:10.3390/en10122001.CrossRefGoogle Scholar
Niu, Q, Tong, Q, Cao, J, Zhang, Y and Liu, F (2019) On-line prediction remaining useful life for ball bearings via grey NARX. Journal of Vibroengineering. doi:10.21595/jve.2018.20120.Google Scholar
Ou, TC and Hong, CM (2014) Dynamic operation and control of microgrid hybrid power systems. Energy 66, 314–323. doi:10.1016/j.energy.2014.01.042.CrossRefGoogle Scholar
Pal, M (2011) Modelling pile capacity using generalised regression neural network. Proceedings of Indian Geotechnical Conference, pp. 811–814.Google Scholar
Pal, M and Deswal, S (2008) Modeling pile capacity using support vector machines and generalized regression neural network. Journal of Geotechnical and Geoenvironmental Engineering 134, 7.CrossRefGoogle Scholar
Pandian, S, Hassim, MH, Ng, RTL and Hurme, M (2015) Designing an inherently healthier process based on inherently safer design (ISD) concept: research and development stage. Clean Technologies and Environmental Policy 17, 1247–1259. doi:10.1007/s10098-015-0951-8.CrossRefGoogle Scholar
Parzen, E (1962) On the estimation of probability density functions and mode. Annals of Mathematical Statistics 33, 1065–1076.CrossRefGoogle Scholar
Patil, S and Chouksey, PD (2016) A survey on: distance based outlier detection. International Journal of Science and Research (IJSR) 5, 280–282. doi:10.21275/v5i1.nov152729.Google Scholar
Patton, JB (1995) Brushless DC motor control using a general regression neural network. Proceedings of IECON ‘95 - 21st Annual Conference on IEEE Industrial Electronics, Orlando, USA.Google Scholar
Repository, UCI Machine Learning (1993) AutoMPG Data Set. Available at http://archive.ics.uci.edu/ml/datasets/Auto+MPG.Google Scholar
Repository, UCI Machine Learning (2007) Concrete Compressive Strength Data Set. Available at http://archive.ics.uci.edu/ml/datasets/Concrete+Compressive+Strength.Google Scholar
Repository, UCI Machine Learning (2012) Energy Efficiency Data Set. Available at https://archive.ics.uci.edu/ml/datasets/Energy+efficiency.Google Scholar
Richardson, M (2012) Standardizing Safe Operating Limit Information. 15th Annual International Symposium, Houston, Texas, October 23–25.Google Scholar
Singh, K and Cantt, M (2012) Outlier detection: applications and techniques. International Journal of Computer Science Issues 9, 307–323.Google Scholar
Specht, DF (1991) A general regression neural network. IEEE Transactions on Neural Networks 2, 568–576. doi:10.1006/brcg.1996.0066.CrossRefGoogle ScholarPubMed
Stauffer, T and Chastain-Knight, D (2019) Do not let your safe operating limits leave you S-O-L (out of luck). 15th Global Congress on Process Safety, New Orleans, LA.CrossRefGoogle Scholar
Sutherland, PE (2011) Safe operating limits: generator capability study for offshore oil platforms. IEEE Industry Applications Magazine 17, 14–19. doi:10.1109/MIAS.2010.939648.CrossRefGoogle Scholar
Wang, H, Bah, MJ and Hammad, M (2019) Progress in outlier detection techniques: a survey. IEEE Access 7, 107964–108000. doi:10.1109/ACCESS.2019.2932769.CrossRefGoogle Scholar
Watson, G (1964) Smooth regression analysis. Sankhyā: the Indian Journal of Statistics; Series A 26, 359–372.Google Scholar
Willey, RJ (2014) Layer of protection analysis. Procedia Engineering 84, 12–22. doi:10.1016/j.proeng.2014.10.405.CrossRefGoogle Scholar
Wynn, DC and Eckert, CM (2017) Perspectives on iteration in design and development. Research in Engineering Design 28, 153–184. doi:10.1007/s00163-016-0226-3.CrossRefGoogle Scholar
Xu, L-Y, Zhang, M, Zhu, W and He, Y-L (2013) Comparison of geometric and arithmetic means for bandwidth selection in NWKE. Proceedings of the 2013 International Conference on Machine Learning and Cybernetics, Tianjin, China.Google Scholar
Xu, X, Liu, H, Li, L and Yao, M (2018) A comparison of outlier detection techniques for high-dimensional data. International Journal of Computational Intelligence Systems 11, 652–662. doi:10.2991/ijcis.11.1.50.CrossRefGoogle Scholar
Zhang, XH, Wang, QJ, Zhu, JJ and Zhang, H (2012) Application of general regression neural network to the prediction of LOD change. Chinese Astronomy and Astrophysics 36, 86–96. doi:10.1016/j.chinastron.2011.12.010.CrossRefGoogle Scholar
Zimek, A, Schubert, E and Kriegel, HP (2012) A survey on unsupervised outlier detection in high-dimensional numerical data. Statistical Analysis and Data Mining 5, 363–387. doi:10.1002/sam.11161.CrossRefGoogle Scholar