ANNALS OF THE UNIVERSITY OF CRAIOVA

Stephen P. Banks
Department of Automatic Control and Systems Engineering University of Sheffield

Abstract: In this paper, we consider the theory of linear, non-autonomous systems with a view to applications to nonlinear systems via a global approximation scheme. Thus we consider stability theory and explicit solutions, together with Oseledec’s theorem and exponential dichotomies.
Keywords: Linear Time-Varying Systems, Nonlinear Systems, Stability, Exponential Splitting.

Andrzej Bartoszewicz
Institute of Automatic Control, Technical University of Lódz

Abstract: The theory of variable structure systems (VSS) with sliding modes is currently one of the most significant research topics within the control engineering domain. Moreover, recently a number of important applications of the theory in the field of power electronics, motion control, robotics, bioprocess etc. have also been reported. Therefore, this paper presents a brief introduction to the theory of sliding mode control, then enumerates the most important novel trends in fundamental research in this field and finally gives some examples of successful engineering applications.
Keywords: Variable structure systems, Sliding mode control, Discrete time sliding mode, Chattering.

Andrzej Bartoszewicz, Justyna Zuk
Institute of Automatic Control, Technical University of Lódz

Abstract: In this paper a new sliding mode flow control strategy for the connectionoriented communication networks is proposed. The networks are modelled as discrete time systems with the available bandwidth acting as disturbance. The strategy employs a time-varying sliding surface, which helps reduce the initial flow rate magnitude. The proposed controller is designed in such a way that the closed-loop system stability and finite time error convergence are ensured. We demonstrate that the designed controller guarantees no bottleneck link buffer overflow and full utilization of its available bandwidth. Furthermore, transmission rates generated by the controller are always upper bounded and non-negative.
Keywords: Discrete time sliding mode control, Flow control, Connection-oriented communication networks, Time-delay systems, Time-varying sliding surface design.

Livia Buzdugan, Tiberiu Colosi
Technical Univerity of Cluj-Napoca, Automation Department

Abstract: The paper presents a method based on „the matrix with partial derivatives of the state vector” (Mpdx) for numerical modelling and simulation of processes with distributed parameters, using the Taylor series. This method can also be used for modelling and numerical simulation of pollutants dispersion.
Keywords: Partial derivative equation, State variables, Taylor series, Numerical integration, Pollutants.

Sergiu Caraman, Marian Barbu and Eugen Arinton
Department of Electrical Energy Conversion Systems, "Dunarea de Jos" University

Abstract: This paper deals with the identification of a biological wastewater treatment process with the removal of two substrates. The identification has been done on the basis of experimental data supplied by Arcelor-Mittal Steel Company. For the identification of the wastewater treatment process with two substrates a criterion of distance type among the experimental data and the model outputs has been used. For the process control a linearizing structure based on the linearizing command principle has been adopted. The linearizing control algorithm has been check in the case of parametric uncertainties (the variation of the maximum specific growth rate parameter).
Keywords: Wastewater treatment, Process identification, Linearizing control

Ioan Cautil
University of Craiova, Faculty of Electrical Engineering

Abstract: The paper presents the design of turbo decoder used in IEEE 802.16 and Hyperlan standards. The paper derives the equations of BCJR algorithm for the decoder and studies its performance. The effect of mismatches that affect the algorithm due to fixed point representation of numbers and approximate calculation of exponentials is shown. Due to iterative nature of algorithm, a scaling factor must be applied to extrinsic information exchanged between decoders. The range of the scaling factor is derived by simulation.
Keywords: Turbo codes, Iterative decoding, Scaling extrinsic information

Petr Chalupa, Vladimír Bobál
Tomas Bata University in Zlin, Faculty of Applied Informatics, Department of Process Control

Abstract: This paper is focused on usability of model predictive control (MPC) approach in the area of control of nonlinear systems. A self-tuning predictive controller is introduced and subsequently used for the control of real-time nonlinear system. The controller integrates on-line identification of an ARX model of a controlled system and a predictive control synthesis on base of the identified model. Various control requirements can be fulfilled by tuning controller’s parameters. The real-time testing has been carried out by the control of nonlinear laboratory model of interconnected tanks (DTS200 by Amira company).
Keywords: Predictive control, Adaptive control, Autoregressive models, Recursive least squares method, Nonlinear systems, Real-time systems.

Radu Dobrescu, Virginia Ecaterina Oltean, Dan Popescu
Faculty of Control and Computers Politehnica University of Bucharest

Abstract: There is an increasing interest in modelling biochemical pathway as dynamic systems. The main challenge is the necessity to drastically decrease the huge number of parameters involved in the biological process, thus operating a selection, which renders analytical dynamic models tractable. This paper comments a three order non-linear differential model of a sepsis shock and proposes a MATLAB-Simulink simulation model which captures both the free evolution and the treatment of the infection. Additionally, the opportunities offered by using hybrid modelling are underlined and two applications based on models that include resetting hybrid systems are presented.
Keywords: Sepsis, Complex system, Cellular pathways, Hybrid modelling, Adaptive simulation model.

Vasile Dragan^a and Adrian-Mihail Stoica^b
aInstitute of Mathematics of the Romanian Academy
^bUniversity „Politehnica” of Bucharest, Faculty of Aerospace Engineering

Abstract: The purpose of the paper is to present an estimation method of the state vector of linear stochastic systems corrupted both with additive and multiplicative white noise. The observer design is based on an H2-type optimization problem for the considered class of stochastic systems. The solution is given in terms of the solution of a specific system of Riccati and Lyapunov equations which in the absence of the multiplicative noise, reduces at the well-known filtering equation in Kalman-Bucy filtering. A numerical example illustrates the theoretical developments.
Keywords: Stochastic systems, Multiplicative noise, Stability, Estimation theory, Kalman filters, Riccati equations, Iterative methods.

Eva-Henrietta Dulf, Andreea Stegaru, Tudor Buzdugan, Clement Festila
Technical University of Cluj-Napoca, Department of Automatic Control

Abstract: The chemical processes developed in a separation plant are very complex and many details are not yet known in totality. The Labview project is an important step forward for the correlation of the theoretical scientific research results with the concrete information acquired from the experimental plant, and also in the analysis and better understanding of the process operation. The paper presents a short description of the column monitoring and control system for main variables that ensure an efficient and safe column operation.
Keywords: Cryogenic isotope separation column, Monitoring, control.

Jozef Futó¹, František Krepelka², Lucia Ivanicová^2
1Institute of Control and Informatization of Production Processes Technical University of Kosice, Slovakia
²Institute of Geotechnics, Slovak Academy of Sciences, Kosice

Abstract: The paper describes a possibility for identification of indentor-rock system at rock cutting using acoustic models. Based on parallel measurements by sound level meter Mediator 2238 and electret microphone with scanning and archiving the acoustic signal to PC, several models were designed, providing the description of observed system based on acoustic behaviour of environment. The paper contains characterization of three such models. First two were based on measurements with Mediator 2238 and the third model was based on the values obtained by electret microphone.
Keywords: Identification, Rock cutting process, Acoustic signal, Processing of acoustic signal, Frequency analysis, Sound level meter Mediator 2238.

Ionela Iancu^* and Eugen Iancu^**
* Department of Physiology and Medical Informatics, University of Medicine and Pharmacy of Craiova
^** Department of Automation and Mechatronics, University of Craiova

Abstract: Cardiovasculary diseases are affecting a constantly growing number of people. Due to this reason we are continously looking for new ways of investigations. The processing of the audio signals produced by the heart also belong to this category. Along this paper the authors present the results of the combined time-requency analysis for short-time periods applied to cardiac signals useful for diagnosis in medical practice.
Keywords: heart sound, spectral analyse, fault detection, fault isolation, automated diagnosis.

Ciprian Lupu, Catalin Petrescu, Puiu Craciunescu, George Condur
University “Politehnica” of Bucharest, Department of Automatic Control and Computer Science

Abstract: Systems with inverse model represent one of the successful solutions for the real-time control of the nonlinear processes. The use of these structures imposes solving some specific problems, like determination of static characteristic of the process, construction of inverse model or robust control law design. The paper proposes a structure and the correspondent methods of designing and implementation of inverse model command. The applicability of the method is proved using a real-time structure with an RST control algorithm. In the end, its software implementation and the obtained results are also shown.
Keywords: Control systems, Inverse model, Robustness, Real-time systems.

Camelia Maican, Matei Vinatoru, Eugen Iancu, Gabriela Canureci
Automation and Mechatronics Department, University of Craiova

Abstract: This paper present two methods for the control of the temperature in the steam overheater. In the first case it is use a control structure that use only one translator and one PI controller. In the second case it study a complex structure of control which makes the rise of the robust structure of control which deals with false information from the process, using three temperature sensors settled in equivalent positions.
Keywords: Overheat, Steam, Injection flow, Boiler, Translator.

Mihaela-Hanako Matcovschi, Octavian Pastravanu
Department of Automatic Control and Applied Informatics Technical University “Gh. Asachi” of Iasi

Abstract: The paper considers a class of continuous-time linear systems with uncertain parameters and develops a methodology for the design of linear feedback laws which, besides stabilization, ensure the constraining of the state-space trajectories inside contractive boxes exponentially decreasing to the equilibrium point. These constraints allow a continuous and individual monitoring of each state variable, despite the parametric uncertainties. The design methodology is based on the positive invariance of the contractive boxes with respect to the dynamics of interval matrix systems. It is shown that the class of matrices that achieve the box-constrained stabilization can be characterized as the solution set of two equivalent linear inequalities. This characterization is further used to derive the design technique as a linear programming problem.
Keywords: Constrained stabilization, Parametric uncertain systems, Interval dynamic systems, Matrix inequalities, Linear programming.

Emil Petre, Dan Selisteanu
Department of Automatic Control, University of Craiova

Abstract: This paper presents some aspects concerning the estimation of the parameters in the input-output chlorine residuals models of drinking water distribution networks (DWDNs) used for transportation of treated drinkable water from the water treating plants to consumers. To kill the microorganisms that can cause the waterborne ills, disinfection is usually the final treatment stage in the drinking water plants. Usually, the disinfectant used in DWDNs is chlorine because it is inexpensive and effectively annihilates a variety of disease-causing organisms. To maintain high-quality drinkable water in such distribution networks it is necessary to regulate the chlorine residual concentration within a prescribed set of bounds. Since the DWDNs are large scale systems with high uncertainties and time varying delays, in order to obtain useful models for control, in this paper, some aspects regarding the estimation of parameters of the input-output chlorine residuals models are analysed. The model parameters are time varying and heavily dependent on the hydraulics which is a main source of uncertainties in chlorine concentration models. Using the input/output data and a priori knowledge of the system, in this paper will be presented a method for the estimation of bounding values corresponding to model parameters so that the obtained model could be used in a robust predictive control method.
Keywords: Nonlinear systems, Time delay systems, Modelling, Parameter estimation.

Vladimir Rasvan, Dan Popescu, Daniela Danciu
Dept. of Automatic Control, University of Craiova

Abstract: This paper presents some network models in connection to basic qualitative dichotomies: stability/ instability and synchronization/chaos. Another dichotomy deals with interconnections which may be delayed or diffusive. A special model leading to time delay partial differential equations is presented.
Keywords: Oscillator networks, Synchronization, Diffusive interconnections

Monica Roman, Eugen Bobasu, Dorin Sendrescu
University of Craiova, Faculty of Automation, Computers and Electronics, Department of Automation

Abstract: In this paper is presented an approach on the bond graph modeling applied to mechanical oscillating systems, in this case a Quanser flexible link experiment, followed by the simulation of the resulted mathematical model. The system was decomposed into subsystems that were modeled separately. The obtained subsystems generated submodels and the overall model was then built up by combining these separate structures. The model was built up using BondSim software and we compared the results of linear and nonlinear models simulation to identify the nonlinear factor influence upon the output parameters evolution.
Keywords: Bond Graphs, Flexible Arm, Modeling, Simulation.

David Samek^*, Lubomir Macku^**
*Tomas Bata University in Zlin, Faculty of Technology, Department of Production Engineering
^**Tomas Bata University in Zlin, Faculty of Applied Informatics, Department of Electrotechnics and Measurements

Abstract: The aim of this paper is to present simulation of model predictive control of chemical exothermic semi-batch reactor model, while the MPC controller uses an artificial neural network as a predictor. A first order chemical reaction is considered to be running in the reactor. The reaction is strongly exothermic so the in-reactor temperature is rising very fast due to reaction component dosing. Thus, the temperature control is necessary. The system is nonlinear because of chemical reaction kinetics, so its control is difficult by classical methods.
Keywords: Modelling, Simulation, Artificial neural networks, Model predictive control, Batch reactor.

Adrian Toader, Ioan Ursu
Elie Carafoli National Institute for Aerospace Research

Abstract: Based on recent works of the authors, the present paper considers the problem of enhancing the dynamic response of a new type of actuation system, the hydrostatic servoactuator, by developing a backstepping position control law for these actuators, as actuating a primary flight control surface. Indeed, the need for fast time constants is common in aerospace systems. Thus, our approach concerns the means, often avoided, of applied control to perform a controlled system, otherwise limited by hardware constraints.
Keywords: Electrohydrostatic actuators, Backstepping control synthesis, Nonlinear systems, Numerical simulation, Control Lyapunov functions