Abstract
Knowledge extraction from large amounts of data is an effective approach for analysis and monitoring of industrial processes. The self-organizing map (SOM) is a useful method for this purpose, because it is able to discover low-dimensional structures on high-dimensional spaces and produce a mapping on an ordered low-dimensional space that can be visualized and preserves the most important relationships. With the aim to extract knowledge about the dynamics of industrial processes, we define 2D SOM maps that represent dynamic features which are useful for usual tasks in control engineering such as the analysis of the time response, the coupling among variables, or the difficulties in control of MIMO (multiple-input and multiple-output) systems. Those new maps make it possible to discover, increase or confirm knowledge about the system, spanned through the entire operation range. A well-known quadruple-tank MIMO system was used to test the usefulness of these maps. First, we perform an analysis of the theoretical dynamic behaviors obtained from the physical equations of the system. After that, we carry out an analysis of experimental data from an industrial pilot plant.
Similar content being viewed by others
References
Abonyi J, Nemeth S, Csaba V, Vesanto PAJ (2003) Process analysis and product quality estimation by self-organizing maps with an application to polyethylene production. Comput Ind 52(3):221–234. doi:10.1016/S0166-3615(03)00128-3
Albertos P, Sala A (2004) Multivariable control systems: an engineering approach. Springer, London
Barreto GA, Araujo AFR (2002) Identification and control of dynamical systems using the self-organizing map. IEEE Trans Neural Netw 15(5):1244–1259. doi:10.1109/TNN.2004.832825
Cho J, Principe JC, Erdogmus D, Motter MA (2006) Modeling and inverse controller design for an unmaned aerial vehicle based on the self-organizing map. IEEE Trans Neural Netw 17(2):445–460. doi:10.1109/TNN.2005.863422
Corona F, Mulas M, Baratti R, Romagnoli J (2010) On the topological modeling and analysis of industrial process data using the som. Comput Chem Eng 34(12):2022–2032. doi:10.1016/j.compchemeng.2010.07.002
Dayal B, MacGregor J (1997) Multi-output process identification. J Process Cont 7(4):269–282
Díaz I, Cuadrado AA, Díez AB, Fuertes JJ, Domínguez M, Prada MA (2009) Visualization of MIMO process dynamics using local dynamic modelling with self organizing maps. In: Palmer-Brown D, Draganova C, Pimenidis E, Mouratidis H (eds) Engineering Applications of Neural Networks, Communications in Computer and Information Science, vol. 43, pp. 119–130. Springer, Berlin Heidelberg (2009). doi:10.1007/978-3-642-03969-0_12
Díaz I, Cuadrado AA, Díez AB, Ojea G (2005) Modelado visual de procesos industriales. Revista Iberoamericana de Automática e Informática Industrial 2(4):101–112
Díaz I, Domínguez M, Cuadrado AA, Fuertes JJ (2008) A new approach to exploratory analysis of system dynamics using SOM. Application to industrial processes. Expert Syst Appl 34(4):2953–2965. doi:10.1016/j.eswa.2007.05.031
Domínguez M, Reguera P, Fuertes JJ, Díaz I, Cuadrado AA (2007) Internet-based remote supervision of industrial processes using self-organizing maps. Eng Appl Artif Intell 20(6):757–765. doi:10.1016/j.engappai.2006.11.017
Eckerson W (1995) Three tier client/server architectures: achieving scalability, performance, and efficiency in client/server applications. Open Inf Syst 3(20):46–50
Fuertes JJ, Domínguez M, Reguera P, Prada MA, Díaz I, Cuadrado AA (2010) Visual dynamic model based on self-organizing maps for supervision and fault detection in industrial processes. Eng Appl Artif Intell 23(1):8–17. doi:10.1016/j.engappai.2009.06.001
Fuertes JJ, Prada MA, Domínguez M, Reguera P, Díaz I, Díez AB (2008) Visualization of dynamic parameters of a multivariable system using Self-organizing maps. In: 17th IFAC World Congress. Seoul. doi:10.3182/20080706-5-KR-1001.01863
Heikkinen M, Hiltunen T, Liukkonen M, Kettunen A, Kuivalainen R, Hiltunen Y (2009) A modelling and optimization system for fluidized bed power plants. Expert Syst Appl 36(7):10,274–10,279. doi:10.1016/j.eswa.2009.01.072
Johansson KH (2000) The quadruple–tank process: a multivariable laboratory process with an adjustable zero. IEEE Transact Control Syst Technol 8(3):456–465
Kohonen T (1990) The self-organizing map. In: Proceedings of the IEEE 78:1464–1480
Kohonen T (2001) Self-organizing maps, 3rd edn. Springer, New York Inc, Secaucus, NJ
Kohonen T, Oja E, Simula O, Visa A, Kangas J (1996) Engineering applications of the self-organizing map. Proc IEEE 84(10):1358–84
Kwak S, Lim C, Metzger R Jr, Rodriguez A (1999) Multivariable submarine control system analysis and design using an interactive visualization tool. In: Proc. of the 1999 American Control Conference 5:3427–3431. IEEE
Lee JA, Verleysen M (2007) Nonlinear dimensionality reduction. Information science and statistics. Springer, New York
Ljung L (2002) Identification for control: simple process models. Proceedings of the 41st IEEE Conference on Decision and Control, 2002 4:4652–4657. doi:10.1109/CDC.2002.1185112
Oja M, Kaski S, Kohonen T (2003) Bibliography of self-organizing map (SOM) papers: 1998–2001 addendum. Neural Comput Surv 3:1–156
Peng H, Wu J, Inoussa G, Deng Q, Nakano K (2009) Nonlinear system modeling and predictive control using the RBF nets-based quasi-linear ARX model. Control Eng Pract 17(1):59–66. doi:10.1016/j.conengprac.2008.05.005
Pölzlbauer G, Dittenbach M, Rauber A (2006) Advanced visualization of self-organizing maps with vector fields. Neural Netw (19):911–922. doi:10.1016/j.neunet.2006.05.013
Principe J, Wang L, Motter M (1998) Local dynamic modeling with self-organizing maps and applications to nonlinear system identification and control. Proc IEEE 86(11):2240–2258
Skogestad S, Postlethwaite I (1996) Multivariable Feedback Control – Analysis and Design. Wiley, Chichester
Souza LGM, Barreto GA (2010) On building local models for inverse system identification with vector quantization algorithms. Neurocomputing 73(10–12):1993–2005. doi:10.1016/j.neucom.2009.10.021
Tasdemir K, Merényi E (2009) Exploiting data topology in visualization and clustering of self-organizing maps. IEEE Transact Neural Netw 20(4):549–562. doi:10.1109/TNN.2008.2005409
Venkatasubramanian V, Rengaswamy R, Kavuri S, Yin K (2003) A review of process fault detection and diagnosis part III: Process history based methods. Comput Chem Eng 27(3):327–346. doi:10.1016/S0098-1354(02)00162-X
Vesanto J (1999) SOM-based data visualization methods. Intell Data Anal 3(2):111–126
Acknowledgments
This work was supported in part by the Spanish Ministerio de Ciencia e Innovación (MICINN) and the European FEDER funds under grants DPI2009-13398-C02-01 and DPI2009-13398-C02-02. A. Morán was supported by a grant from the Consejería de Educación de la Junta de Castilla y León and the European Social Fund.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Fuertes, J.J., Domínguez, M., Díaz, I. et al. Visualization maps based on SOM to analyze MIMO systems. Neural Comput & Applic 23, 1407–1419 (2013). https://doi.org/10.1007/s00521-012-1090-3
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00521-012-1090-3