

FUZZY AND NEURAL MODELLING OF THE BIOTECHNOLOGICAL PROCESSES
Sergiu Caraman, Marian Barbu
Abstract:
This paper deals with the possibility of modelling the biotechnological processes by fuzzy and neural techniques. Two casestudies are taken into consideration: a biomass producing process (the accumulation process of the proteic mass from superior mushroom mycelium of Polyporus type) and an enzyme biosynthesis process (the alphaamylase and bacterial protease biosynthesis process with microorganism Bacillus Subtilis). The modelling technique combines two methods: mass balanced equations and artificial intelligence techniques. In both cases, the main parameters of the process (ex.: the biomass specific growth rate)are calculated based on fuzzy and neural techniques. The simulation results are compared to the real data given by the Food Industry Institute from Bucharest.
Keywords:
Biotechnological Process, Fuzzy rules, neural network, biomass specific growth rate

A PRACTICAL CONTROL METHOD FOR MULTIVARIABLE SYSTEMS
Vasile Cīrtoaje, Sanda Frāncu, Alina Baiesu
Abstract:
The main purpose of the paper is to present a robust practical method for experimentally decoupling, compensating and control of two inputtwo output process. The decoupler channels are first order lag plus dead time elements, which satisfy the following requirements: the direct channels have unit gain and two channels have dead time equal to zero. The decoupler can be simplified in addition taking two appropriate channels with the lag time constant equal to zero. After decoupling, each output of the decoupled process is controlled by a special method, which consists in monotonic compensating and standard IMC control of the both direct channels of the decoupled process. The results obtained by simulation validate the proposed control procedure.
Keywords:
decoupling controller, decoupler, monotonic compensation, standard IMC algorithm.

AN ALTERNATIVE CONTROL FOR PROCESS IN FAULT CONDITIONS
Eugen Iancu, Matei Vīnatoru
Abstract:
In this paper, the authors propose some practical behaviour in control and supervising of complex process (MIMO), in presence of actuators faults. The fault detection and isolation (FDI) problem is an inherently complex one. Because of this reason, we have considered the case when one or more actuators are blocking in a fixed position or are not supplied (in this case the servomechanism are in the total closed or total open position). The immediate goals is to preserve the stability of process and, if is possible, to control the process in a slightly degraded manner.
Keywords:
Control systems, fault detection and isolation, superheater control.

LQG/LTR CONTROLLER DESIGN
FOR ROTARY INVERTED PENDULUM QUANSER REALTIME EXPERIMENT
Cosmin Ionete
Abstract:
This experiment consists of a rigid link (pendulum) rotating in a vertical plane. The rigid link is attached to a pivot arm, which is mounted on the load shaft of a DCmotor. The pivot arm can be rotated in the horizontal plane by the DCmotor. The DCmotor is instrumented with a potentiometer. In addition, a potentiometer is mounted on the pivot arm to measure the pendulum angle. The principal objective of this experiment is to balance the pendulum in the verticalupright position and to position the pivot arm. Since the plant has two degrees of freedom but only one actuator, the system is underactuated and exhibits significant nonlinear behavior for large pendulum excursion. Our purpose is to design a robust controller in order to realize a realtime control of the pendulum position using a Quanser PC board and power module and the appropriate WinCon realtime software. For the controller design is used a wellknown robust method, called LQG/LTR (Linear Quadratic Gauss Ian/Loop Transfer Recovery) which implements an optimal statefeedback. The realtime experiment is realized in the Automatic Control laboratory.
Keywords:
LQR design, WinCon software, robust control, realtime experiment.

VARIABLE CAUSALITY DYNAMICAL SYSTEMS
Constantin Marin
Abstract:
This paper deals with dynamical systems which models physical objects whose causal inputoutput ordering is changing during their evolution. Such a system is named Variable Causality Dynamical System (VCDS). VCDSs are controlled from outside by a new input called causal ordering signal sharing the same set of state variables. In VCDS all the variables, except the causal ordering signal, are gathered in two forms of so called global variables as current global variable and desired global variable. In this paper different approaches of the causality concept are analysed and there are proposed formal definitions for covariance and causality properties of variables and relations irrespective of the time domain. An example of nonlinear VCDS is presented here but many applications in walking robots of the VCDS approach are developed including simulations in Matlab environment.
Keywords:
dynamical systems, covariance, causality, inference.

SOME WAYS TO IMPLEMENT NTERPOLATIVE TYPE CONTROLLERS ON DSP
Lucian Peana, Adrian Korodi
Abstract:
This paper presents two similar methods to implement an interpolative type controller by using a Digital Signal Processor Starter Kit. The implemented interpolative type controller is used in a practical application. First structure is a 2dimensional interpolative controller with static behavior and the second one is a 1, 2, or 3dimensional interpolative controller with dynamic behavior. At the end of the implementation was conceived a graphical user interface for each interpolative type controller. The graphical user interface helps the programmer and also the user to reconfigure the controller. The results are satisfactory, and in most cases meet and also improve the control system requirement.
Keywords:
automation, interpolative type controllers, digital signal processors, implementation, graphical interface

NEURAL NETWORK BASED ADAPTIVE CONTROL FOR A CLASS
OF NONLINEAR PROCESSES
Emil Petre
Abstract:
In this paper an adaptive neural network controller is developed for the tracking control of a class of uncertain nonlinear processes for which both the dynamics and the process dimension may be unknown. The only informations required about the process are the measurements of the output and its relative degree. Under the assumption that the process is feedback linearizable, a neural network compensator is introduced to cancel the inverse tracking error. More precisely, we introduce a linear observer to estimate the derivatives of tracking error, the output of which is used as a teaching signal for the neural network. The linear portion of the error dynamics is stabilized by using a linear output feedback dynamic compensator. Ultimate boundedness of the tracking error and observation error are shown using Lyapunov method. Computer simulations are included to demonstrate the effectiveness of this controller.
Keywords:
Nonlinear systems, Nonlinear adaptive control, Output feedback, Neural networks.

ON THE SAFE IMPLEMENTATION OF DISTRIBUTED DELAY CONTROL LAWS
Dan Popescu
Abstract:
The stabilization by feedback control of systems with input delays may be considered in various frameworks. An approach to stabilize a such system, based on the Artstein transform, is the socalled finite spectrum assignment. In this case the control law that stabilize the system is a distributed delay control law. A difficulty in applying a control law of this form consists in the practical implementation of the integral term, which needs to be calculated online. Some recent papers outlined an instability mechanism when the distributed delay in the control law is approximated with a sum of pointwise delays, despite the asymptotic stability of the ideal closedloop system. In this paper, we use a suitable discretization rule based on piecewise constant control signals. Then, we analyze and illustrate by some numerical examples the robustness/fragility of this control law
Keywords:
time delay systems, stabilization, piecewise constant control, robustness.

DISCRETETIME PARAMETRIC RESONANCE:
BASIC NOTIONS AND CRITICAL FREQUENCIES
Vladimir Rasvan
Abstract:
The parametrically excited systems are systems with periodically varying coeficients. In these systems the phenomenon of parametric resonance is to be met, being characterized by a continuous spectrum composed of several small intervals which tend each to a critical frequency when the amplitude of the parametric excitation tends to 0. Also the amplitude of the oscillation around a critical frequency grows exponentially (instead of polynomially  the case of standard resonance). In the discrete time case the results do not migrate mutatismutandis from the continuous time one; this paper shows that there are at least two ways of defining critical frequencies in discretetime parametric resonance.
Keywords:
parametric resonance; discretetime; critical frequencies; Hamiltonian systems.

ACTIVE CONTROL LAWS FOR FLUTTER SUPPRESSION
Ioan Ursu, Marius StoiaDjeska, Felicia Ursu
Abstract:
The present paper addresses the problem of flutter prevention for weakly damped aeroelastic wing structures by means of the primary flight controls servos. In fact, an active control for a typical section in unsteady incompressible flow is thought. As mathematical model, a linear system with structured uncertainty is obtained. Firstly, the optimal linear quadratic regulator with state observer is employed to get the active control of even unstable in open loop speeds. Secondly, for control law synthesis is used the paradigm of robust servomechanism problem involving two compensators. Numerical examples that illustrate the design are presented, using the data of an experimental model in aerodynamic tunnel.
Keywords:
flutter, active control, aeroservoelasticity, linear quadratic regulator, observer, robust servomechanism problem, uncertainties, stability robustness, performance robustness.

MODELING AND CONTROL OF THE AMMONIA SYNTHESIS COLUMN
Matei Vinatoru, Eugen Iancu
Abstract:
The paper presents the possibilities of simulation of steady and dynamic regimes for ammonia synthesis columns. We present the possibilities of study of the dynamic regimes for automatic control of the temperature in catalyst layers and propose an advanced control structure using the fresh gas flows between catalyst layers as commands. We also present the possibilities of optimization of the column steady states to obtain a higher conversion factor using the control of gas flows between the catalyst layers. This optimization algorithm is part of the main advanced control program for the synthesis column.
Keywords:
Advanced control system, modeling, simulation, ammonia plant synthesis.

