Latest issue

International Review of Automatic Control (Theory and Applications) - September 2010 - Papers

Fuzzy Model-Based Black Box System Control
A Combined Adaptive Backstepping-Sliding Mode Control of a Class of Uncertain Nonlinear Systems
Robust State Observers for Nonlinear Systems
Sequential Bayesian state and parameter Estimation of Non linear Stochastic Systems
An output controllability of bilinear distributed system
On-line Identification Method Based on Dynamic Neural Network
Comparison between Self Tuning PI Voltage Control of DFIG and a Combinational Control for Improved Wind Turbines
Computation of Time Delay Margin for DC Motor Speed Control System with Time Delays
Synchronous Generator’s Dynamic Parameters Estimation by a Fuzzy Index and Neural-based Observer
Two Novel Proposed Controllers for a Wind Energy Conversion System
Modeling the Synchronous Generator Excitation Systems Using Artificial Neural Network
Damping Power Systems Oscillations Using Thyristor Controlled Series Capacitor(TCSC)
Fuzzy Control of a Thruster Driven AUV
Identification of nonlinear complex systems using uncoupled state fuzzy model for liquid level process
Monocular Mapless Navigation in Streets and Highways
A New Contribution of an Uncoupled State Multimodel Predictive Control: Experimental Validation on a Chemical Reactor
Simulated Annealing Approach Based HVDC FACTS Controller for Power Systems Stabilization

International Review of Automatic Control - Papers

Fuzzy Model-Based Black Box System Control
by P. Fedor, D. Perduková

Vol. 3. n. 5, pp. 438-442

Abstract - The paper deals with the design and simulation of properties of an unknown system fuzzy controller. In the first step of the design process, a black-box system fuzzy model is formed on basis measured input-output data. The model is then exploited for identifying sets of control input signals that will transfer the system from any point of its state space to the selected final point. In the last phase of the design procedure, a fuzzy controller is generated on basis of the trajectories found, enabling transition of the system to the desired final point. The design results are validated by simulation in the MATLAB programme package on a system represented by a drive with a DC motor.

Keywords: black-box system control, fuzzy modelling, fuzzy control, DC drive.

A Combined Adaptive Backstepping-Sliding Mode Control of a Class of Uncertain Nonlinear Systems
by R. Ben Khaled, C. Mnasri, M. Gasmi

Vol. 3. n. 5, pp. 443-451

Abstract - In this paper, an adaptive control using a combined backstepping sliding mode control algorithm is proposed for a class of uncertain nonlinear system in strict feedback form. It is proved that the proposed robust adaptive scheme can guarantee that all the signals in the closed loop systems are globally uniformly ultimately bounded and the tracking error converges asymptotically to zero. The efficacy of the proposed algorithm is investigated by an illustrative simulation example of one link robot.

Keywords: Adaptive Control, Nonlinear System, Backstepping technique, Sliding Mode Control.

Robust State Observers for Nonlinear Systems

by Ramzi Ben Messaoud, Nadia Zanzouri, Mekki Ksouri

Vol. 3. n. , pp. 452-456

Abstract - This paper proposes a new design of an unknown inputs observer for nonlinear systems. It represents an extension of the Luenberger observer with unknown inputs using the Taylor expansion for first order. The aim of this observer is state system estimation. The necessary and sufficient conditions for the existence of the observer are given. Numerical examples are given to illustrate the attractiveness and the simplicity of the new design procedure.

Keywords: Nonlinear system, Unknown input observer, State estimation.

Sequential Bayesian state and parameter Estimation of Non linear Stochastic Systems
by F. Souibgui, F. Ben Hmida, A. Chaari, M. Gossa

Vol. 3. n. 5, pp. 457-465

Abstract - This paper presents the problem of Bayesian state and parameter estimation in stochastic non linear systems. However, the presence of parameters in the system equations makes their use more complex and can even impair their convergence properties. In this paper, a new formulation of Particle Filter to parameter estimation for non linear systems is proposed from the sequential Bayesian approach theory. Based on this new technique, a consistent estimation of these parameters can be obtained to that of the state variables is presented. We compare the performance with that using the Extended Kalman Filtering. In simulation tests it was observed that for a large number of particles and residual resampling, the new strategy of Particle Filter could succeed in accurately estimating the joint state and parameter for strongly non linear system Gaussian or non- Gaussian, while the Extended Kalman Filter does not.

Keywords: Sequential Bayesian approach, Extended Kalman Filter, Particle Filter, State estimation, Parameter estimation.

An output controllability of bilinear distributed system
by E. Zerrik, M. Ould sidi

Vol. 3. n. 5, pp. 466-473

Abstract - In this work, we consider a regional controllability problem of a class of distributed bilinear systems evolving in a spatial domain Ω. A feedback control is used to steer the system state close to a desired profile at a final time T, only on a subregion ω which my be interior or on the boundary of the system domain. Our purpose is to prove that an optimal control exists, and characterized as a solution to an optimality system. Numerical algorithm is given and successfully illustrated by simulations.

Keywords: Distributed Systems, Bilinear Systems, Regional Controllability, Optimal control.

On-line Identification Method Based on Dynamic Neural Network
by A. Errachdi, I. Saad, M. Benrejeb

Vol. 3. n. 5, pp. 474-479

Abstract - In this paper, an on-line identification neural network method for dynamical nonlinear system is proposed. The corresponding algorithm introduces an adaptive learning rate. The convergence of the corresponding algorithm is based on the use of the conventional method. At every moment, parameters are obtained to reproduce the behavior of the system. The proposed algorithm may be very helpful to design an adaptive control strategy of nonlinear system. Different examples according to the complexity of the systems are considered to show the efficiency of the proposed algorithm.

Keywords: Nonlinear system identification, Neural networks, Gradient method, On-line identification.

Comparison between Self Tuning PI Voltage Control of DFIG and a Combinational Control for Improved Wind Turbines
by Hengameh Kojooyan Jafari, Ahmed Radan

Vol. 3. n. 5, pp. 480-484

Abstract - As wind energy is one of the low cost renewable energies and wind turbines are simple and reliable, wind power plants are very noteworthy power electricity sources. Nowadays doubly fed induction machine is vastly applied in wind power plants instead of other electric machines. In this paper, a doubly-fed induction generator with winding-rotor configuration, an external rotor resistance and a modeled voltage source inverter that has an adjustable phase and amplitude is used and controlled first time by a PI self tuning voltage controller and second time by a P self tuning voltage controller and a pitch controller as the combinational controller to control the output active and reactive powers. The performance of controllers is compared. Furthermore, the doubly-fed induction machine is simulated with an improved wind turbine maximum power tracking system for different wind speeds. Results show that PI self tuning controller is technically better with lower cost but P controller has better dynamic output responses and can be better for wind plants with pitch control technology to promote the strategy of controlling by a combinational controller.

Keywords: PI self tuning voltage controller, P self tuning combinational controller, Doubly-fed induction machine, Gearbox ratio.

Computation of Time Delay Margin for DC Motor Speed Control System with Time Delays
by Saffet Ayasun

Vol. 3. n. , pp. 485-491

Abstract - This paper investigates the effect of time delays on the stability of the time-delayed DC motor speed control system. The time delays are due to the use of measurement devices, communication links for data transfer and controller processing time. A direct and exact method based on Rekasius substitution is proposed to determine the maximum amount of time delay known as delay margin that the system can tolerate without loosing its stability. The method starts with the determination of all possible purely imaginary characteristic roots for any positive time delay. To achieve this, Rekasius substitution is first used to convert the transcendental characteristic equation into a polynomial. Then, Routh stability criterion is applied to determine the critical root, the corresponding oscillation frequency and the delay margin for stability. It is found that the speed control system becomes unstable when the time delay crosses certain critical values, delay margins for stability. Delay margins are computed for a wide range of controller gains. Finally, the time-domain simulations of Matlab/Simulink are used to validate theoretical delay margin results.

Keywords: Networked Control System, DC Motor, Stability, Time Delay, Delay Margin, Rekasius Substitution.

Synchronous Generator’s Dynamic Parameters Estimation by a Fuzzy Index and Neural-based Observer
by H. Chahkandi nejad, R. Jahani, M. Mohammad Abadi, A. Zare, J. Olamaei, H. A. Shayanfar

Vol. 3. n. 5, pp. 492-498

Abstract - This paper presents a new methodology for implementing Artificial Neural Network (ANN) observers in estimating and identifying synchronous generator dynamic parameters based on one statistic feature extraction from the operating data. Data for training the neural network observers are obtained through off-line simulations of a synchronous generator operating in a one-machine-infinite-bus environment. Nominal values of parameters are used as a deviance index in the machine model. After training, the ANN observer is tested with simulated on-line measurements to provide estimates of the dynamic parameters.

Keywords: Synchronous Generator, Dynamic Parameters, On-line Estimation, Artificial Neural Networks, Operating Data.

Two Novel Proposed Controllers for a Wind Energy Conversion System
by B. Sobhani, M. S. Toulabi, H. Ebrahimian

Vol. 3. n. 5, pp. 499-506

Abstract - Nonlinear characteristics of the wind speed and generator model in wind energy conversion systems (WECS) demands for high-quality adaptive controllers to ensure both its stability and performance. In this paper, two novel controllers for doubly fed induction machine are used in generating mode for large scale grid – connected variable speed wind turbines. To enhance the close loop performance, one of these controllers is proposed as a conventional PID controller that its coefficients are tuned using a learning algorithm named; Reinforcement Learning (RL) and another controller is a robust learning fuzzy controller which is used to fulfill the same aim. In both these two proposed controllers, maximizing power including speed and torque control is considered. The simulation results substantiate the suitable behaviors of the proposed controllers.

Keywords: Reinforcement Learning; Fuzzy Control; Doubly Fed Induction Generator; Wind Energy Conversion System.

Modeling the Synchronous Generator Excitation Systems Using Artificial Neural Network
by Ashraf Mohamed Hemeida

Vol. 3. n. 5, pp. 507-515

Abstract - Application of an artificial neural network (ANN) approach for modeling the synchronous generator excitation system is presented in this paper. The proposed approach depends up on the expected operating conditions of the synchronous machines due to the flexibility of the used function of neural network. To ensure well structure artificial neural network exciter model for the synchronous machine, the training data is taken for a wide range of operating conditions and for different types of fault. Different neural network structure is considered using try and error to select the best network structure according to the system performance. The nonlinear model of studied power system is considered. The 7th order synchronous machine model is considered in this study. The time simulation of the studied single machine infinite bus power system shows that the proposed ANN model for the synchronous machine exciter has better performance than that classical one. The system equations is modeled and solved by MATLAB software.

Keywords: Synchronous Generator – Governor control - Exciter control – Artificial Neural Network.

Damping Power Systems Oscillations Using Thyristor Controlled Series Capacitor(TCSC)
by Ashraf Mohamed Hemeida, Y. A. Mobarak, A. El-bahnasawy

Vol. 3. n. 5, pp. 516-523

Abstract - This paper presents a developed Thyristor-Controlled Series Capacitor(TCSC) to provide a moderate phase advance for a wide range of interested frequencies in order to compensate the inherent lag between the field excitation and the electrical torque induced to ensure robust damping system oscillations. The performance of the proposed TCSC is compared with Static Var Compensator (SVC) used to add damping to the rotor oscillations of the synchronous machine by controlling its excitation currents during disturbances. Digital time responses simulations of the nonlinear system performance during and after severe disturbances have proved effectiveness of the developed (TCSC). The Studied power system consists of two identical area power system, each area consists of two machine. The Matlab/Simulink toolbox is used for the digital simulation.

Keywords: Thyristor-Controlled Series Capacitor - Static Var Compensator - Power System Stability.

Fuzzy Control of a Thruster Driven AUV
by Valiollah Ghaffari, Paknosh Karimaghaee

Vol. 3. n. 5, pp. 524-534

Abstract - Underwater vehicles have many important scientific and commercial applications. Therefore, their control and performance are widely investigated. The present paper is dealing with modeling and control of an underwater vehicle with six degree of freedom. This vehicle has two differential thrusters such that with total increasing in the speed of these thrusters, heave varies while differential increase in the speed of the thrusters, yaw changes. In trim condition, the speeds of the thrusters are identical and constant. In the way to control this vehicle, at first, it is modeled as a two input-two output system, which has 12 state variables. Then appropriate control configuration is selected. Lastly, an intelligent fuzzy control is applied to the model and results are compared with other classical control methods.

Keywords: AUV; Dynamical Modeling; Fuzzy Controller; PID Controller; Differential Thrusters.

Identification of nonlinear complex systems using uncoupled state fuzzy model for liquid level process
by S. Moez, C. Lotfi, C. Abdelkader, B. Fayçal, G. Moncef

Vol. 3. n. 5, pp. 535-544

Abstract - In this work, a modified fuzzy model clustering algorithm for identification of nonlinear complex systems using multiple-model approach. This new algorithm uses uncoupled state multiple-model taking in account the nonlinear system is affected by measurement noise (unknown samples in the data). This later is based to adding a second regularization term in objective function; in order to take in account the data are noisy. The proposed method allows the identification of the premise and the consequence parameters together via iterative minimization using four criteria.
A comparative study is presented. Validation results involving simulations of Gas furnace data and the liquid level process have demonstrated the practicality of the algorithm.

Keywords: Multiple-models, uncoupled state multiple-models, nonlinear system identification, noise clustering algorithm.

Monocular Mapless Navigation in Streets and Highways
by Reza Rasoolinezhad, Mehdi Ebrahimi, Reza Ebrahimpour

Vol. 3. n. 5, pp. 545-549

Abstract - Autonomous navigation between specified lines in streets and highways is among high demanded research topics in modern cities and factories nowadays. We developed a novel algorithm based on Hough transform and K-means clustering technique to find heading direction. In this paper, we present a new procedure to navigate through streets and highways without using any a priori information to the robot. This work which focuses on outdoor cases is the continuation of prior work of authors in indoor environments, especially, in corridors. Monocular vision has been used to serve this purpose. The heading point is determined by clustering of lines with 45° orientation. The robot uses these clusters to perceive the environment and to find an appropriate path of navigation.

Keywords: Mapless Navigation, Hough Transform, K-Means Clustering.

A New Contribution of an Uncoupled State Multimodel Predictive Control: Experimental Validation on a Chemical Reactor
by Anis Messaoud, Majda Ltaief, Ridha Ben Abdennour

Vol. 3. n. 5, pp. 550-559

Abstract - The multimodel approach is recently developed in order to resolve the problems of the increasing complexity of many industrial processes. The basis of the multiple model approach is the decomposition of the operating space of the system into a finite number of operating zones. In this paper, we propose an uncoupled state multimodel predictive control based on a commutation algorithm, to the problem of a semi-batch reactor control. This commutation is controlled by partial predictors/observers associated to the local controllers. The obtained results are very satisfactory and show clearly that the developed strategy leads to a good closed loop performances in terms of precision and tracking.

Keywords: Supervisor, Uncoupled State Multimodel, Experimental Validation, Multimodel Predictive Control, Multiobserver.

Simulated Annealing Approach Based HVDC FACTS Controller for Power Systems Stabilization
by Ashraf Mohamed Hemeida

Vol. 3. n. 5, pp. 560-567

Abstract - This paper aims to apply Robust PI controller based Simulating Annealing approach to HVDC FACTS in order to damp the power oscillations and improve the power system performance. The essential feature of the HVDC FACTS is that the power flow control is easy to control and regulate. The HVDC consists of two converter circuits namely rectifier and inverter circuits . The power flow control can be achieved easily through rectifier circuit . The rectifier and inverter current control loops are developed to generate the rectifier firing angle and inverter extinction angle signals respectively. To enhance the power system operation and restore the power system stability after applying different disturbances is performed by adding supplementary signal to rectifier control loop. The supplementary signal is the output of proportional-integral PI controller based on simulating annealing approach . The input signal to the proposed PI controller is the synchronous machine speed deviation. The simulated annealing optimization approach is utilized to find the proposed PI gains over wide range of operating conditions. The simulated annealing based PI controller gains is adopted to extract the stabilizing signal . The studied power system composed synchronous generator with control systems feeding large power system via Parallel AC/ DC transmission systems. The linearized dynamic model of the above system is developed . The power system under study with the proposed simulated annealing PI stabilizer is simulated digitally. Different disturbances are applied to the studied power system such ac load change, mechanical power input, excitation voltage reference, rectifier and inverter current disturbances. The simulation results prove the effectiveness and robustness of the proposed PI controller in terms of fast damping of power system oscillations , restoring the power system stability quickly over wide range of operating condition and external disturbances.

Keywords: HVDC Power System – Robust PI Controller - Simulating Annealing Approach – power system stability.