Expert Systems with Applications 32 (2007) 911–918
Brain emotional learning centered intelligent control applied to neurofuzzy model of micro-heat exchanger
Hossein Rouhani a, *, one particular, Mahdi Jalili b, 2, Babak In. Araabi n,
Wolfgang Eppler c, Querido Lucas b
Mechanical Executive Department, School of Tehran, Tehran, Iran Control and Intelligent Control Center of Excellence, Electric powered and Computer system Engineering Section, University of Tehran, Tehran, Iran
Institute of Data Processing and Electronics, Forschungszentrum Karlsruhe, Germany
In this paper, a brilliant controller can be applied to control the mechanics of electrically heated micro-heat exchanger plant. First, the dynamics of the micro-heat exchanger, which provides a non-linear grow, is identiﬁed using a neurofuzzy network. To build the neurofuzzy model, a locally linear learning formula, namely, in your area linear mode tree (LoLiMoT) is used. Then, an intelligent control mechanism based on mind emotional learning algorithm is usually applied to the identiﬁed version. The brilliant controller is founded on a computational model of limbic system inside the mammalian mind. The brain emotional learning based intelligent control (BELBIC) based on PID control is followed for the micro-heat exchanger plant. The contribution of BELBIC in improving the control system performance is definitely shown by comparison with results obtained from vintage PID controller without BELBIC. The effects demonstrate excellent improvements of control action, without any extensive increase in control eﬀort for PID + BELBIC. Ó 2006 Elsevier Ltd. All rights appropriated.
Keywords: Clever control; Feeling based learning; Neurofuzzy models; Locally linear models; Nonlinear system identiﬁcation; Heat exchanger
1 . Advantages
Although commercial processes generally contain complicated
nonlinearities, most of the conventional control algorithms
derive from a linearized model of the procedure. Linear
E-mail address: [email protected] ac. ir, hossein. [email protected]ﬂ. ch (H. Rouhani), [email protected] lace. ac. ir (M. Jalili), [email protected] ac. ir (B. N. Araabi), wolfgang. [email protected] fzk. de (W. Eppler), [email protected] ir (C. Lucas).
Present address: Laboratory for Computer-Aided Design and Production, Commence of Development and Robotics, Swiss Government Institute of Technology Lausanne (EPFL), CH 1015 Lausanne, Switzerland.
a couple of
Present talk about: Laboratory to get Nonlinear Devices, School of Computer and Communication Sciences, Swiss National Institute of Technology Lausanne (EPFL), CH 1015 Lausanne, Switzerland.
0957-4174/$ - discover front subject Ó 2006 Elsevier Ltd. All legal rights reserved. doi: 10. 1016/j. eswa. 06\. 01. 047
models could be identiﬁed within a straightforward method from
method test data; e. g. via stage or impulse response. Yet , if the method is highly non-linear and be subject to large regular disturbances, a non-linear style will be required
to describe the behaviour of the procedure. For such systems
nonlinear identiﬁcation methods should be accustomed to describe the dynamic tendencies of the system, which can be attained
by means of nerve organs networks. An alternative solution approach should be to
design a nonlinear model consisting of many linear functions. The major outcome function is derived from a combination of thready models. Various training algorithms and buildings are suggested for the mentioned networks such
while locally linear model shrub (LoLiMoT), adaptive network
structured fuzzy inference system (ANFIS), Takagi–Sugeno
(TS) and piecewise linear sites (PLN) (Eppler & Beck,
1999; Jang, 1993; Nelles, 1997; Sugeno & Kang, 1988). In
this work we can make use of LoLiMoT for schooling
H. Rouhani et al. as well as Expert Systems with Applications 32 (2007) 911–918
formula of the neurofuzzy network for its rapid
and accurate procedure in control applications.
We uses brain emotional based learning intelligent...
References: Balkenius, C., & Moren, J. (2000). Emotional learning: a computational
model of the amygdale
Brandner, J. T., & Schubert, K. (2005). Fabrication and testing of
microstructure high temperature exchangers pertaining to thermal applications
Eppler, W., & Beck, H. D. (1999). Piecewise linear sites (PLN) intended for
Fatourechi, M., Lucas, C., & Khaki Sedigh, A. (2001a). Reducing control
eﬀort by means of mental learning
conference on electric engineering (ICEE2001), May, Tehran, Iran,
Fatourechi, M., Lucas, C., & Khaki Sedigh, A. (2001b). Lowering of
optimum overshoot through emotional learning
Fatourechi, Meters., Lucas, C., & Khaki Sedigh, A. (2003). Psychological learning
like a new device for development of agent based system
Fink, A., Fischer, M., Nelles, O., & Isermann, 3rd there�s r. (2000). Oversight of
nonlinear adaptive controllers based on fluffy models
Fink, A., Topfer, S., & Isermann, 3rd there�s r. (2003). non-linear model-based
control with regional linear neuro-fuzzy models
Hafner, M., Schukler, M., Nelles, O., & Isermann, R. (2001). Fast neural
networks for diesel-powered engine control design
Practice, 8, 1211–1221.
Henning, To., Brandner, L. J., & Schubert, E. (2004). Portrayal of
electrically powered micro-heat exchangers
Inoue, K., Kawabata, K., & Kobayashi, L. (1996). On a decision making
system with feelings
Jang, M. S. 3rd there�s r. (1993). Adaptive-network-based fuzzy inference system.
Lucas, C., Shahmirzadi, D., & Sheikholeslami, D. (2004). Launching
BELBIC: human brain emotional learning based intelligent controller
Moren, J. (2002). Emotion and learning: a computational model of the
amygdala, PhD Thesis, Lund school, Lund, Sweden.
Neese, Ur. (1998). Psychological disorders in evolutionary point of view. British
Diary of Medical Psycology, 71, 397–415.
Nelles, O. (1997). Orthonormal basis functions for nonlinear system
identiﬁcation with local thready model forest (LoLiMoT)
Nelles, O. (2001). Nonlinear program identiﬁcation: Via classical
approaches to neural sites and fuzzy models
Savinov, A. A. (1999). An algorithm for induction of options setvalued rules by ﬁnding prime disjunctions. In Process of the last
on-line world conference on soft computing in industrial applications
Sugeno, M., & Kang, G. T. (1988). Structure identiﬁcation of fluffy model.