Abstract
Proportional-Integral-Derivative (PID) controller is broadly used to control industrial plants. However, intelligent control strategies are able to improve the performance of a PID controller. Some of them can do so without removing the current controller. One of these techniques is called Feedback-Error-Learning (FEL), which adds an artificial neural network (ANN) to the closed-loop system, alongside a PID controller, to improve the control. FEL strategy is inspired on a neurocomputacional model for control and learning of voluntary movement. Nevertheless, the introduced ANN may reach a local minimum and the performance of the control system may not improve anymore. Therefore, another ANN can be inserted to improve the control even further. Such a strategy is called Multi-Network-Feedback-Error-Learning (MNFEL) and is a FEL improvement that uses multiple ANNs instead of a single one. An automatic approach to insert new ANNs based on the standard deviation of the plant’s error was proposed to substitute the manual insertion driven by a specialist. Even though interesting results with the automatic approach have been found, this method was applied to only one system that was the burner group of a Brazilian mining ore company. To validate the method on a different system, this work applies the MNFEL with automatic insertion to a cooling coil nonlinear system simulated by a Hammerstein-Wiener model. Results with this approach indicate improvement in both systems. The automatic approach reached smaller overshoot when compared with manual approach. Graphical analysis points out better performance achieved by FEL over PID-only and MNFEL over FEL.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Aafaque, M., Kadri, M.B.: Dynamic fuzzy modeling of cooling coil system. In: 17th International Multi-Topic Conference, INMIC, Pakistan, vol. 17, pp. 349–353. IEEE (2014)
Haykin, S.S.: Neural Networks: A Comprehensive Foundation, 2nd edn., USA (1998)
Singh, A.K., Gaur, P.: Adaptive control for non-linear systems using artificial neural network and its applications applied on inverted pendulum. In: India International Conference on Power Electronics 2010, IICPE 2010, India, vol. 1, pp. 1–8. IEEE (2010)
Qian, K., Chen, Z.: Dynamic inversion base on neural network applied to nonlinear flight control system. In: 2nd International Conference on Future Computer and Communication 2010, ICFCC, China, vol. 1, pp. 699–703. IEEE (2010)
Fu, M., Liu, T., Liu, J., Gao, S.: Neural network-based adaptive fast terminal sliding mode control for a class of SISO uncertain nonlinear systems. In: IEEE International Conference on Mechatronics and Automation, ICMA 2016, China, vol. 1, pp. 1456–1560. IEEE (2016)
Kawato, M., Furukawa, K., Suzuki, R.: A hierarquical neural-network model for control and learning of voluntary movement. Biol. Cybern. 57(3), 169–185 (1987)
Miyamoto, H., Kawato, M., Setoyama, T., Suzuki, R.: Feedback-error-learning neural networks for trajectory control of a robotic manipulator. Neural Netw. 1(3), 251–265 (1998)
Gomi, H., Kawato, M.: Learning control for a closed loop system using feedback-error-learning. In: 29th IEEE Conference on Decision and Control on Proceedings, United States, pp. 3289–2394. IEEE (1990)
Jia, C., Li, X., Liu, D., Ding, D.: Adaptive control using multiple parallel dynamic neural networks. In: 9th Asian Control Conference 2013, ASCC, Turkey, vol. 9, pp. 1–6. IEEE (2013)
Chen, F.: Predictive control based on multi-network for a deep seabed mining robot vehicle. In: 30th Chinese Control Conference 2011, CCC, China, vol. 30, pp. 2632–2635. IEEE (2011)
Almeida Neto, A.: Applications of multiple neural networks in mechatronics systems. Ph.D. thesis, Technological Institute of Aeronautics, Brazil (2003). (in Portuguese)
Almeida Neto, A., Goes, L.C.S., Nascimento, C.L.: Accumulative learning using multiple ANN for flexible link control. IEEE Trans. Aerosp. Eletronic Syst. 46(2), 508–524 (2010)
Ribeiro, P.R.A., Almeida Neto, A., Oliveira, A.C.M.: Multi-network-feedback-error-leaning in pelletizing plant control. In: 2nd International Conference on Advanced Computer Control, ICACC, China, vol. 2, pp. 340–344. IEEE (2010)
de Almeida Ribeiro, P.R., de Almeida Neto, A., de Oliveira, A.C.M.: Multi-network-feedback-error-learning with automatic insertion. In: Corchado, E., Novais, P., Analide, C., Sedano, J. (eds.) Soft Computing Models in Industrial and Environmental Applications, pp. 171–178. Springer, Heidelberg (2010). https://doi.org/10.1007/978-3-642-13161-5_22
Nascimento, C.L.: Artificial neural networks in control and optimization. Ph.D. thesis, University of Manchester, United Kingdom (1994)
Wang, S.: Dynamic simulation of building VAV air-conditioning system and evaluation of EMCS on-line control strategies. Build. Environ. 34(6), 681–705 (1999)
Wu, Y., Dexter, A.L.: Modeling capabilities of fuzzy relational models. In: 12th IEEE International Conference on Fuzzy Systems, FUZZ 2003, USA, vol. 1, pp. 430–435. IEEE (2003)
Kadri, B.M.: Robust model free fuzzy adaptive controller with fuzzy and crisp feedback error learning schemes. In: 12th International Conference on Control, Automation and Systems, ICC, vol. 12, pp. 384–388. ICROS, Korea (2012)
Kadri, M.B., Dexter, A.L.: Disturbance rejection in information-poor systems using an adaptive model-free fuzzy controller. In: 28th North American Fuzzy Information Processing Society Annual Conference, NAFIPS 2009, USA, vol. 28, pp. 1–6. IEEE (2009)
Ribeiro, P.R.A., Costa, T.S., Barros, V.H., Almeida Neto, A., Oliveira, A.C.M.: Feedback-error-learning in pelletizing plant control. In: 7th Brazilian Meeting on Artificial Intelligence, ENIA, Brazil (2009)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2017 Springer International Publishing AG
About this paper
Cite this paper
Santos, A.N.V., Ribeiro, P.R.A., de Almeida Neto, A., Oliveira, A.C.M. (2017). Multi-Network-Feedback-Error-Learning with Automatic Insertion: Validation to a Nonlinear System. In: Barone, D., Teles, E., Brackmann, C. (eds) Computational Neuroscience. LAWCN 2017. Communications in Computer and Information Science, vol 720. Springer, Cham. https://doi.org/10.1007/978-3-319-71011-2_14
Download citation
DOI: https://doi.org/10.1007/978-3-319-71011-2_14
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-71010-5
Online ISBN: 978-3-319-71011-2
eBook Packages: Computer ScienceComputer Science (R0)