Time filter

Source Type

El Abidine Skhiri M.Z.,University of Sfax | Chtourou M.,University of Sfax | Chtourou M.,Research Unit on Intelligent Control
International Journal of Wavelets, Multiresolution and Information Processing | Year: 2013

This paper investigates the applicability of the constructive approach proposed in Ref. 1 to wavelet neural networks (WNN). In fact, two incremental training algorithms will be presented. The first one, known as one pattern at a time (OPAT) approach, is the WNN version of the method applied in Ref. 1. The second approach however proposes a modified version of Ref. 1, known as one epoch at a time (OEAT) approach. In the OPAT approach, the input patterns are trained incrementally one by one until all patterns are presented. If the algorithm gets stuck in a local minimum and could not escape after a fixed number of successive attempts, then a new wavelet called also wavelon, will be recruited. In the OEAT approach however, all the input patterns are presented one epoch at a time. During one epoch, each pattern is trained only once until all patterns are trained. If the resulting overall error is reduced, then all the patterns will be retrained for one more epoch. Otherwise, a new wavelon will be recruited. To guarantee the convergence of the trained networks, an adaptive learning rate has been introduced using the discrete Lyapunov stability theorem. © 2013 World Scientific Publishing Company.

Ben Amor N.,Research Unit on Intelligent Control | Ben Amor N.,TU Chemnitz | Kanoun O.,TU Chemnitz | Derbel N.,Research Unit on Intelligent Control
International Multi-Conference on Systems, Signals and Devices, SSD'11 - Summary Proceedings | Year: 2011

Progress in low power Very Large Scale Integration (VLSI) design open up possibilities of powering wireless devises from scavenged ambient energy. An energy management system should be designed to manage the harvested energy. DC to DC converters represent the key element in energy management systems. Knowing the diversity of DC-DC converters generally designed for specific applications, basic converters are under scope in this paper. Operation principles and simulation patterns of basic DC to DC converters are first presented. Parameter simulation occurring the effect of duty cycle, inductance, capacity and switch polarisation is studied in depth. The gaol of this paper is to show main parameters effect on output voltage behaviour. Simulation results show that the main parameters in DC-DC converters design play a key role to avoid extra-lose of harvested energy. © 2011 IEEE.

Khiareddine A.,University of Monastir | Salah C.B.,Research Unit on Intelligent Control | Mimouni M.F.,University of Monastir
2013 International Conference on Control, Decision and Information Technologies, CoDIT 2013 | Year: 2013

The purpose of this paper is the regulation of water level in a photovoltaic pumping system. To reach this objective, we have developed an algorithm with Matlab / simulink which gives as result, the value of the reference speed required for vector control of induction motor taking into account the level of water in the tank and the variation of solar illumination. The string of photovoltaic pumping is simulated under matlab-Simulink. © 2013 IEEE.

Jallouli-Khlif R.,Research Unit on Intelligent Control | Melchior P.,University of Bordeaux 1 | Derbel N.,Research Unit on Intelligent Control | Oustaloup A.,University of Bordeaux 1
International Journal of Modelling, Identification and Control | Year: 2012

Preshaping approach is used to reduce system vibration in motion control. Desired systems inputs are altered so that the system finishes the requested move without residual vibration. This technique, developed by Singer and Seering, is used for flexible structure control, particularly in the aerospace field. In a previous work, this method was extended for explicit fractional derivative systems and applied to second generation CRONE control, the robustness was also studied. CRONE (the French acronym of 'commande robuste d'ordre non-entier') control system design is a frequency-domain-based methodology using complex fractional integration. It permits the robust control of perturbed linear plants using the common unity feedback configuration. This paper presents the extension of the preshaping approach to generalised derivative systems and its application to third generation CRONE control. Then, to ensure robust path tracking, a robust algorithm for the shaper synthesis is developed. Copyright © 2012 Inderscience Enterprises Ltd.

Drira A.,Research Unit on Intelligent Control | Derbel N.,Research Unit on Intelligent Control
International Multi-Conference on Systems, Signals and Devices, SSD'11 - Summary Proceedings | Year: 2011

In this paper, we have developed a feedforward neural networks to detect and to diagnosis rotor fault on induction motors using stator currents. In the first step, causes and effects of rotor fault have been studied, particularly, the number of broken bars has been considered. Then, in the second step, the number of broken rotor bars has been localized by Artificial Neural Networks (ANN), using the Fast Fourier Transform. Simulation results show that the Neural Network proposed approach presents a good tools for the diagnostic of induction machines. © 2011 IEEE.

Krichen Masmoudi N.,Research Unit on Intelligent Control | Rekik C.,Research Unit on Intelligent Control | Djemel M.,Research Unit on Intelligent Control | Derbel N.,Research Unit on Intelligent Control
ICMLC 2010 - The 2nd International Conference on Machine Learning and Computing | Year: 2010

This paper presents a method to compute optimal control strategies of discrete large scale nonlinear systems by using hierarchical fuzzy systems. The method is based on the decomposition principle of the global system into interconnected subsystems becoming easier to study. Then, the differential dynamic programming procedure is applied in order to obtain the rule basis. After that, we construct limpid-hierarchicalMamdani fuzzy system in order to compute optimal control laws, for each subsystem. Simulation results of a rotary crane show that the proposed method yields to satisfactory performances. The robustness of the proposed approach is verified. © 2010 IEEE.

Nasr M.B.,Research Unit on Intelligent Control | Chtourou M.,Research Unit on Intelligent Control
Applied Soft Computing Journal | Year: 2011

This paper presents a novel hybrid algorithm for feedforward neural networks, called a self organizing map-based initialization for hybrid training based on a two stage learning approach. First stage, a structure learning scheme which includes adding hidden neurons is used to determine the network size. Second stage, a FN (fuzzy neighborhood)-based hybrid learning scheme which we have recently proposed is used to adjust the network parameters. In this approach the weights between input and hidden layers are firstly adjusted by Kohonen algorithm with fuzzy neighborhood, whereas the weights connecting hidden and output layers are adjusted using gradient descent method. Four simulation examples are provided to demonstrate the efficiency of the approach compared with other well-known and recently proposed learning methods. © 2011 Elsevier B.V. All rights reserved.

Zribi A.,Research Unit on Intelligent Control | Chtourou M.,Research Unit on Intelligent Control | Djemel M.,Research Unit on Intelligent Control
International Multi-Conference on Systems, Signals and Devices, SSD'11 - Summary Proceedings | Year: 2011

For linear plants, IMC have been shown good robustness properties against disturbances and model mismatches. However, when uncertain processes are concerned, the original IMC structure cannot be directly used for control system implementation. In this paper, an internal multiple model control (IMMC) based on linear model's library is introduced. This approach supposes the definition of a set of local models developed within each cluster. Such decomposition is obtained using the fuzzy c-means (FCM) clustering algorithm. Then the gap metric concept has been designed as a criterion to analyze the relationships among candidate models to get a reduced model bank. © 2011 IEEE.

Loading Research Unit on Intelligent Control collaborators
Loading Research Unit on Intelligent Control collaborators