Authors: Cengic, Goran; Akesson, Knut

Abstract: A new control software development method is presented. It uses IEC 61499 function blocks for control software programming and provides tools for simulation, execution, automatic model generation and formal verification of the control code during the development. Simulation and execution are supported by the same tool, the Fuber runtime environment. Formal modeling is done using extended finite automata (EFA) and an automatic model generation tool. Formal verification shows the behavior of the closed-loop system. i.e. when control code is executed against the model of the process. The model can use a non-deterministic execution control chart (ECC) in the process model block. The control code and the process model are expressed using the IEC 61499 language in order to avoid maintenance of the process model and control code in different languages, thus making it easier to use the formal verification in the control software development.

Keywords: Dependable manufacturing systems control; Discrete event systems in manufacturing; Supervisory control

Identifier: 10.3182/20080706-5-KR-1001.00003

Conference: Proceedings of the 17th IFAC World Congress, 2008

Location: COEX, Korea, South

Start Date: Sun Jul 06 2008 - End Date: Fri Jul 11 2008

Authors: McDonald, Mary K.; Schwaber, James S.; Ogunnaike, Babatunde A.

Abstract: Alcoholism and alcohol withdrawal, complicated physiological conditions with significant consequences for physical and emotional health, have been the focus of extensive studies in various scientific disciplines for centuries. In America alone, alcoholism is known to affect 14 million people; it accounts for an estimated annual cost of $100 billion in healthcare and related productivity losses, and according to the Centers for Disease Control, there were 34,833 chronic alcohol-related deaths in 2001. Various experimental studies have strongly suggested that many of the physiological consequences of alcoholism and alcohol withdrawal may be associated with alcohol-induced inhibition of N-methyl-Daspartate (NMDA) glutamate receptors, but the mechanisms are not completely known. (Glutamate is an important excitatory neurotransmitter and its NMDA receptors are expressed widely throughout the brain.) In this paper, we postulate the following hypothesis: that the level of unblocked NMDA receptors is controlled in alcoholism to maintain at a constant level, the amount of glutamate release and extent of glutamatergic excitation; that ethanol blocks these receptors, and additional receptors are generated to compensate for the loss; we then develop and analyze a control system model predicated on this hypothesis. Our model shows, among other things, that upon cessation of alcohol consumption, no additional generation of NMDA receptors occurs, but the number of unblocked NMDA receptors increases dramatically, leading to the excitotoxicity observed clinically. This first attempt at a control system representation of the NMDA receptor’s role is thus able to capture the essence of some key organism-wide responses observed in alcoholism and alcohol withdrawal. Future work will involve incorporating additional molecular details to obtain a more accurate model which can then be potentially useful in postulating treatment regimens for alcohol withdrawal.

Keywords: Estimation and control in systems biology

Identifier: 10.3182/20080706-5-KR-1001.02685

Conference: Proceedings of the 17th IFAC World Congress, 2008

Location: COEX, Korea, South

Start Date: Sun Jul 06 2008 - End Date: Fri Jul 11 2008

Authors: Kimura, Hidenori

Abstract: Advances in medical science and technology inevitably have generated the increasing ramification of disciplines of clinical medicine. Each medical professional tends to be restricted in his/her own specialty, in terms of knowledge, methodology and viewpoint, and often leads to the destruction of the harmonious balance of human body through excessive concentration of a local situation. Human body must be in balance as a complicated system whose behaviors are regulated though huge network of control mechanisms. The medical treatments must be re-considered from this point of view of the totality of human body in face of coming aged society. The integration must be emphasized instead of ramification. The so-called systems medicine is regarded as a methodology of clinical science of the next generation to overcome various bottlenecks of the modern medicine, as well as medical care systems. A core of the system medicine lies in digital technology, particularly modeling and control of human viscera and their protection systems. We propose the notion of model-based medicine as a concrete discipline of system medicine which is based on the well-known internal model control. The methodology of constructing integrative model of human body for model-based medicine is proposed. A preliminary example of model-based medicine for use in ICU(intensive care unit) is shown. As a more realistic example, a new model of glucose-insulin-glucagon dynamics which involves the allostatic effect of neuronal-hormonal regulation is derived, based on which a treatment strategy of model-base medicine is proposed. Some theoretical issues of systems medicine will be discussed.

Keywords: Kinetic modelling and control of biological systems

Identifier: 10.3182/20080706-5-KR-1001.01832

Conference: Proceedings of the 17th IFAC World Congress, 2008

Location: COEX, Korea, South

Start Date: Sun Jul 06 2008 - End Date: Fri Jul 11 2008

Authors: Aracil, Javier; Acosta, José Ángel; Gordillo, Francisco

Abstract: The hybrid solution to the pendulum swinging-up and stabilizing problem introduced by Åström and Furuta is based in two steps: an energy injection and a linear stabilization around the desired inverted position. However the energy injection stage only considers the pendulum, and not the motion of the pivot. Furthermore, for the stabilization stage linear law, only a very small basin of attraction can be guaranteed. In this paper the energy controller is enlarged to cope with the pivot dynamics and a nonlinear controller is introduced for the stabilization stage with a larger basin of attraction. The approach proposed allows to cope both with the pendulum on a cart and the Furuta one. Experiments with a laboratory Furuta pendulum are included.

Keywords: Application of nonlinear analysis and design; Nonlinear system control

Identifier: 10.3182/20080706-5-KR-1001.01301

Conference: Proceedings of the 17th IFAC World Congress, 2008

Location: COEX, Korea, South

Start Date: Sun Jul 06 2008 - End Date: Fri Jul 11 2008

Authors: Belkoura, Lotfi; Richard, Jean-Pierre; Fliess, Michel

Abstract: This paper deals with on-line identification of delay systems. Based on non-asymptotic techniques, the estimation approach reduces to solving polynomials or eigenvalue problems. Numerical simulations with noisy data but also with slowly time varying parameters and delay are provided.

Keywords: Systems with time-delays

Identifier: 10.3182/20080706-5-KR-1001.01067

Conference: Proceedings of the 17th IFAC World Congress, 2008

Location: COEX, Korea, South

Start Date: Sun Jul 06 2008 - End Date: Fri Jul 11 2008

Authors: Yu, Aiqing; Gu, Xingsheng; Jiao, Bin

Abstract: Identical parallel machine scheduling problems with sequence dependent setup times, to minimize the total completion time are studied in this paper. A mixed-integer programming formulation of this problem is presented. And a neural computation architecture based on a Coupled Transiently Chaotic Neural Network is introduced to construct the model. The transiently chaotic dynamics are defined after the energy function is constructed by a penalty function approach. Tradeoff problem existing among the penalty terms included in the energy function is overcome by using time-varying penalty parameters. Simulation results tested on different problems with 100 random initial conditions show that this approach converges to near-optimal or optimal solutions and outperforms the Hopfield neural networks.

Keywords: Job and activity scheduling; Intelligent manufacturing systems; Discrete event systems in manufacturing

Identifier: 10.3182/20080706-5-KR-1001.02519

Conference: Proceedings of the 17th IFAC World Congress, 2008

Location: COEX, Korea, South

Start Date: Sun Jul 06 2008 - End Date: Fri Jul 11 2008

Authors: Fedele, Giuseppe; Coluccio, Loredana

Abstract: This paper proposes a non iterative algorithm for the identification of Hammerstein model, using the sampled output data obtained from the step response, giving a continuoustime model for the linear part and a point-wise estimation of the nonlinear one. Key in the derivation of the results is the algebraic derivative method in the frequency domain yielding exact formula in terms of multiple integrals of the output signal, when placed in the time domain. By investigating the connection between such integrals and parameters to be estimated, a set of three linear regression equations is proposed. The first equation is used to estimate the structure of poles in the linear part, the second to estimate a point of the nonlinearity, the third to estimate the structure of zeros in the linear part. No a priori knowledge of the structure of the nonlinearity is required. The proposed algorithm is numerically robust, since it is based only on least squares estimation. Simulation results validate the proposed algorithm.

Keywords: Identification for control; Nonlinear system identification

Identifier: 10.3182/20080706-5-KR-1001.01085

Conference: Proceedings of the 17th IFAC World Congress, 2008

Location: COEX, Korea, South

Start Date: Sun Jul 06 2008 - End Date: Fri Jul 11 2008

Authors: Han, Qing-Long

Abstract: This paper is concerned with stability of linear neutral systems. Firstly, a new approach, a delay decomposition approach, is proposed to deal with the stability issue. The idea of the approach is that the delay interval is uniformly divided into N segments with N a positive integer, and a proper Lyapunov-Krasovskii functional is chosen with different weighted matrices corresponding to different segments in the Lyapunov-Krasovskii functional. Secondly, based on the delay decomposition approach, some new delay-dependent stability criteria for linear neutral systems are derived. These criteria are much less conservative and include some existing results as their special cases. Numerical examples show that significant improvement using the delay decomposition approach is achieved over some existing method even for coarse delay decomposition.

Keywords: Systems with time-delays; Lyapunov methods; LMIs

Identifier: 10.3182/20080706-5-KR-1001.00439

Conference: Proceedings of the 17th IFAC World Congress, 2008

Location: COEX, Korea, South

Start Date: Sun Jul 06 2008 - End Date: Fri Jul 11 2008

Authors: Du, Baozhu; Lam, James; Shu, Zhan

Abstract: This paper introduces a new effective approach to study the stability of neutral systems. By employing a special Lyapunov-Krasovskii functional form based on delay partitioning, delay-dependent stability criteria are established for the nominal and the uncertain case (polytopic type) in terms of linear matrix inequalities (LMI). Numerical examples are employed to illustrate that the delay-partitioning projection approach can be applied to estimate the upper bounds for the delays for the system to maintain stability. Judging from these numerical results, the stability criteria obtained are less conservative than those of existing methods.

Keywords: Robust time-delay systems; Convex optimization; Relaxations

Identifier: 10.3182/20080706-5-KR-1001.02090

Conference: Proceedings of the 17th IFAC World Congress, 2008

Location: COEX, Korea, South

Start Date: Sun Jul 06 2008 - End Date: Fri Jul 11 2008

Authors: Shen, Chao; Jing, Yuanwei; Wang, Qingli; Ban, Ying

Abstract: This paper investigates the problem of stability analysis for a polytopic system with time-varying delay via parameter-dependent Lyapunov functions. By a relaxation approach with slack matrices and a descriptor model transformation, a new robust delay-dependent stability criterion is expressed as a set of linear matrix inequalities (LMIs) with less computational burden. This criterion combined with fault tolerant techniques can be employed for robust reliable controller synthesis for an aircraft dynamic system with multiple operating points. The resulting flight control system remains stable when actuator faults occur. The simulation results illustrate the effectiveness of the proposed approach.

Keywords: Robust time-delay systems; Robust linear matrix inequalities; Robust control applications

Identifier: 10.3182/20080706-5-KR-1001.01931

Conference: Proceedings of the 17th IFAC World Congress, 2008

Location: COEX, Korea, South

Start Date: Sun Jul 06 2008 - End Date: Fri Jul 11 2008

Authors: Zhang, Qingling; Zheng, Meng; Song, Min; Li, Chunji; Duan, Xiao-dong; An, Yichun

Abstract: In this paper, we consider a class of networked control systems (NCSs) with normbounded uncertainties. Using the continuous modelling method, the NCSs can be described as delayed differential equations (DDEs), which can be viewed as a general form of the NCSs model, where the effect of the network-introduced delay and data packet dropout are simultaneously considered. Robust exponentially asymptotical stability criterion are derived based on a delaydependent method. Robust output feedback controller can also be determined by solving a set of linear matrix inequalities.

Keywords: Control under communication constraints; Control over networks; Control and estimation with data loss

Identifier: 10.3182/20080706-5-KR-1001.01115

Conference: Proceedings of the 17th IFAC World Congress, 2008

Location: COEX, Korea, South

Start Date: Sun Jul 06 2008 - End Date: Fri Jul 11 2008

Authors: Hagiwara, Takaaki; Yamada, Kou

Abstract: In this paper, we examine a design method of modified PID (Proportional-Integral-Derivative) controllers for multiple-input/multiple-output plants. PID controller structure is the most widely used one in industrial applications. Recently the parametrization of all stabilizing PID controller has been considered. Yamada and Hagiwara proposed a design method of modified PID controllers such that modified PID controllers make the closed-loop system for unstable plants stable and the admissible sets of P-parameter, I-parameter and D-parameter are independent from each other. However no method has been published to guarantee the stability of PID control system for multiple-input/multiple-output plants and the admissible sets of P-parameter, I-parameter and D-parameter to guarantee the stability of PID control system are independent from each other. In this paper, we propose a design method of modified PID controllers such that the modified PID controller make the closed-loop system for multiple-input/multiple-output plants stable and the admissible sets of P-parameter, I-parameter and D-parameter are independent from each other.

Keywords: Analytic design; Time-invariant systems

Identifier: 10.3182/20080706-5-KR-1001.00982

Conference: Proceedings of the 17th IFAC World Congress, 2008

Location: COEX, Korea, South

Start Date: Sun Jul 06 2008 - End Date: Fri Jul 11 2008

Authors: Yamada, Kou; Hagiwara, Takaaki; Takenaga, Hiroshi; Kobayashi, Masahiko

Abstract: The multi-period repetitive control system is a type of servomechanism for the periodic reference input. Even if the plant does not include time-delay, the transfer function from the periodic reference input to the output and that from the disturbance to the output of the multi-period repetitive control system generally have an infinite number of poles. In order to specify the input-output characteristic and the disturbance attenuation characteristic easily, Yamada et al. propose the concept of simple multi-period repetitive control systems such that the controller works as a stabilizing multi-period repetitive controller and the transfer function from the periodic reference input to the output and that from the disturbance to the output have a finite number of poles. However, the method by Yamada et al. cannot apply for the plant with uncertainty. The purpose of this paper is to propose the concept of robust stabilizing simple multi-period repetitive controllers for the plant with uncertainty and to clarify the parametrization of all robust stabilizing simple multi-period repetitive controllers.

Keywords: Iterative learning control

Identifier: 10.3182/20080706-5-KR-1001.00254

Conference: Proceedings of the 17th IFAC World Congress, 2008

Location: COEX, Korea, South

Start Date: Sun Jul 06 2008 - End Date: Fri Jul 11 2008

Authors: Ben Gaid, Mongi; Simon, Daniel; Sename, Olivier

Abstract: The problem of the integrated control and weakly-hard real-time scheduling is addressed. First, an abstract model of control tasks execution is introduced, allowing the establishment of a formal relationship linking control performance to deadline misses. Then, the notion of accelerable control task is introduced. An accelerable control task has the property that more executions are performed, better is the control performance. Thanks to this latter property, it becomes straightforward to design the control laws according to the average execution times of control tasks, and guaranteeing that in the worst-case scenario, the minimal allowable performance will be achieved. Based on Bellman optimality principle, sufficient conditions for a given control task to be accelerable are stated. A design method of optimal control laws for the weakly-hard execution model is then proposed.

Keywords: Control under computation constraints; Networked embedded control systems

Identifier: 10.3182/20080706-5-KR-1001.01736

Conference: Proceedings of the 17th IFAC World Congress, 2008

Location: COEX, Korea, South

Start Date: Sun Jul 06 2008 - End Date: Fri Jul 11 2008

Authors: Palacios, Francisco; Rodellar, Jose; Rossell, Josep M.

Abstract: In this paper a quadratic guaranteed cost control problem for a class of linear continuous-time state-delay systems with norm-bounded uncertainties is considered. We will suppose that the systems are composed by two overlapped subsystems but the results can be easily extended to any number of subsystems. The main objective is to design overlapping guaranteed cost controllers with tridiagonal gain matrices for these kind of systems by using a linear matrix inequality (LMI) approach. With this idea in mind, we present a design strategy to reduce the computational burden and to increase the feasibility in the LMI problem. In this context, the use of so-called complementary matrices play an important role. A simple example to illustrate the advantages achieved by using the proposed method is supplied.

Keywords: Controller constraints and structure; Decentralization; LMIs

Identifier: 10.3182/20080706-5-KR-1001.01471

Conference: Proceedings of the 17th IFAC World Congress, 2008

Location: COEX, Korea, South

Start Date: Sun Jul 06 2008 - End Date: Fri Jul 11 2008

Authors: Wattamwar, Satyajit; Weiland, Siep; Backx, Ton

Abstract: Finite element or finite volume discretizations of distributed parameter systems (DPS) typically lead to high order finite dimensional systems. Model approximation is then an important first step towards the construction of optimal controllers. However, model reduction methods hardly take model uncertainties and parameter variations into account. As such, reduced order models are not well equipped when uncertain system parameters vary in time. This is particularly true when system behavior does not depend continuously on the parameters. It is shown in this paper that the performance of reduced order models inferred from Galerkin projections and proper orthogonal decompositions can deteriorate considerable when system parameters vary over bifurcation points. Motivated by these observations, we propose a detection mechanism based on reduced order models and proper orthogonal decompositions that allows to characterize the influence of parameter variations around a bifurcation value. for this, a hybrud model structure is proposed. The ideas are applied on the example of a tubular reactor. In particular, this paper discusses the difficulties in approximating the transition from extinction to ignited state in a tubular reactor.

Keywords: Nonlinear model reduction; Process observation and parameter estimation; Process modeling and identification

Identifier: 10.3182/20080706-5-KR-1001.01856

Conference: Proceedings of the 17th IFAC World Congress, 2008

Location: COEX, Korea, South

Start Date: Sun Jul 06 2008 - End Date: Fri Jul 11 2008

Authors: Doraiswami, Rajamani; Diduch, C.P.; Tang, Jiong

Abstract: This paper considers a new approach to fault detection and isolation (FDI) for systems modeled as an interconnection of subsystems that are each subject to parametric faults. The paper develops i) the concept of a diagnostic model that parameterizes all possible subsystem faults, ii) an off-line scheme for identification of the diagnostic model, iii) a parity equation that results in a residual that is a linear function of the change in the diagnostic parameters and iv) a fault isolation scheme that does not require a recursive least squares type identifier.

Keywords: Fault detection and diagnosis; Nonlinear system identification; Identification for control

Identifier: 10.3182/20080706-5-KR-1001.01246

Conference: Proceedings of the 17th IFAC World Congress, 2008

Location: COEX, Korea, South

Start Date: Sun Jul 06 2008 - End Date: Fri Jul 11 2008

Authors: Cao, Erbao

Abstract: The vehicle routing problem with simultaneous delivery and pick-up and time windows (VRPSDPTW) is presented from the point of strategic view that combines the logistics and reverse logistics (bidirectional logistics). A general mixed integer programming mathematic model of VRP-SDPTW is constructed, it can transform into other classical vehicle routing problems by setting different parameters. A hybrid optimization algorithm (HOA) is proposed, which is based principally on the combination of the differential evolution (DE) theory and genetic algorithm (GA). In operation process, we firstly adopted the novel decimal coding to construct the chromosome, and then the differential evolution operator is adopted as the main optimizing scheme, while such techniques of the genetic algorithm, as the novel crossoveroperator mutation-operator are designed to improve the result. Therefore it is expected that advantages of both the differential evolution and the genetic algorithm are kept so that the hybrid optimum algorithm can optimize large scale highly nonlinear problems well and improve the efficiency of the optimization process, and can avoid effectively the common defects of early convergence and the diversity of population in traditional genetic algorithm. The computer simulations are used to compare the performance of the proposed method with GA and DE, and numerical results show that the performance of the proposed method is better than the other method.

Keywords: Job and activity scheduling; Logistics in manufacturing

Identifier: 10.3182/20080706-5-KR-1001.01791

Conference: Proceedings of the 17th IFAC World Congress, 2008

Location: COEX, Korea, South

Start Date: Sun Jul 06 2008 - End Date: Fri Jul 11 2008

Authors: Izadi, Iman; Shah, Sirish; Chen, Tongwen

Abstract: Non-uniformly sampled systems are widely found in industry. In these systems the process output is sampled and the control input is generated at non-uniformly distributed time instants. In this paper, an optimal residual generator is developed for fault detection in non-uniformly sampled systems. In the direct approach used here, the intersample behavior of fault and disturbance is captured by introducing operators that map continuous-time signals to discrete-time signals. No periodicity assumption is made for the sampling instants.

Keywords: Fault detection and diagnosis; Switched discrete and hybrid systems

Identifier: 10.3182/20080706-5-KR-1001.01717

Conference: Proceedings of the 17th IFAC World Congress, 2008

Location: COEX, Korea, South

Start Date: Sun Jul 06 2008 - End Date: Fri Jul 11 2008

Authors: Chi, Ronghu; Sui, Shulin; Yu, Lei

Abstract: A discrete-time adaptive ILC scheme is presented for systems with time-varying parametric uncertainties. Using the analogy between the discrete-time axis and the iterative learning axis, the new AILC can incorporate a recursive Least-Squares algorithm, hence the learning gain can be tuned iteratively along the learning axis and pointwisely along the time axis. When the initial states are random and the reference trajectory is iteration-varying, the new AILC can achieve the pointwise convergence over a finite time interval asymptotically along the iterative learning axis. An extension of the new AILC is also developed by using nonlinear data weighting to systems without assuming any growth conditions on the nonlinearity.

Keywords: Iterative learning control; Nonlinear adaptive control; Gain scheduling

Identifier: 10.3182/20080706-5-KR-1001.02445

Conference: Proceedings of the 17th IFAC World Congress, 2008

Location: COEX, Korea, South

Start Date: Sun Jul 06 2008 - End Date: Fri Jul 11 2008

Authors: Xu, Jian-Xin; Abidi, Khalid

Abstract: A new integral type sliding surface (ISM) is design for sampled-data systems for output tracking. ISM surface design is based on Output Feedback with a State Observer. Discrete-time control based on ISM achieves good tracking performance while allowing the pole assignment of m poles, which are otherwise zero in a deadbeat design. In particular, the new scheme can avoid overlarge control actions by avoiding the deadbeat response inherent in conventional sliding mode control designed for sampled-data systems. It will be shown in this work that, the discrete-time version of the sliding mode control based on the integral type sliding surface has O(T 2) tracking error for output tracking. An experimental example demonstrates the validity of the proposed scheme.

Keywords: Tracking; Output feedback control; Disturbance rejection

Identifier: 10.3182/20080706-5-KR-1001.02407

Conference: Proceedings of the 17th IFAC World Congress, 2008

Location: COEX, Korea, South

Start Date: Sun Jul 06 2008 - End Date: Fri Jul 11 2008

Authors: Zou, Yunfei; Pagilla, Prabhakar R.

Abstract: A new approach to coordination of multiple mobile robots is presented in this paper. The approach relies on the notion of constraint forces which are used in the development of the dynamics of a system of constrained particles with inertia. A familiar class of dynamic, nonholonomic robots are considered. The goal is to design a distributed coordination control algorithm for each robot in the group to achieve, and maintain, a particular formation while ensuring navigation of the group. The theory of constraint forces is used to generate a stable control algorithm for each mobile robot that will achieve, and maintain, a given formation. The advantage of the proposed method is that the formation keeping forces (constraint forces) cancel only those applied forces which act against the constraints. Another feature of the proposed distributed control algorithm is that it allows to add/remove other mobile robots into/from the formation gracefully with simple modifications of the control input. Further, the algorithm is scalable. To corroborate the theoretical approach, simulation results on a group of six robots are shown and discussed.

Keywords: Multi-vehicle systems; Autonomous vehicle navigation, guidance and control; Motion control

Identifier: 10.3182/20080706-5-KR-1001.02050

Conference: Proceedings of the 17th IFAC World Congress, 2008

Location: COEX, Korea, South

Start Date: Sun Jul 06 2008 - End Date: Fri Jul 11 2008

Authors: Ivanescu, Mircea; Florescu, Mihaela Cecilia; Popescu, Nirvana; Popescu, Decebal

Abstract: The grasping control problem for a hyperredundant arm is studied. Frist, dynamic model of the arm is analyzed. The control problems are divided in the subproblems: the position control in a desired reaching area, the control of the arm around the object-load and the force control of grasping. The difficulties determined by the complexity of the non-linear integral-differential equations are avoided by using a very basic energy relationship of this system. First, the dynamic control of the arm for a desired reaching area is inferred. Then, the position controland the force control for grasping are discussed. Numerical simulation are presented.

Keywords: Modelling; Robotics technology; Robots manipulators

Identifier: 10.3182/20080706-5-KR-1001.02645

Conference: Proceedings of the 17th IFAC World Congress, 2008

Location: COEX, Korea, South

Start Date: Sun Jul 06 2008 - End Date: Fri Jul 11 2008

Authors: Pan, Yan; Alpcan, Tansu; Pavel, Lacra

Abstract: This paper studies constrained optical signal-to-noise ratio (OSNR) problem via a distributed optimization approach. In multi-channel optical systems, the signal over an optical link can be regarded as an interfering noise for others, which leads to OSNR degradation. Regulating the input optical power at the Source (transmitter) aims to achieve satisfactory OSNR level at the Destination (receiver) for each channel. Moreover, because all wavelength-multiplexed channels in a link share the same optical fiber, the total input power in a link has to be below the nonlinearity threshold, which corresponds to a link capacity constraint. We formulate the OSNR optimization problem as one of utility maximization with the objectives of achieving an OSNR target level for each channel while minimizing the interference and also satisfying the link capacity constraint. We derive conditions for the existence of a unique optimal solution, leading to a basis for an admission control scheme. By using a Lagrangian relaxation approach we propose two distributed update algorithms: a primal algorithm and a dual algorithm, and study their convergence properties both theoretically and numerically.

Keywords: Control of networks; Complex system management; Cooperative systems

Identifier: 10.3182/20080706-5-KR-1001.00487

Conference: Proceedings of the 17th IFAC World Congress, 2008

Location: COEX, Korea, South

Start Date: Sun Jul 06 2008 - End Date: Fri Jul 11 2008

Authors: Gutiérrez González, Luisa Paulina; Odloak, Darci; Sotomayor, Oscar; Alvarez, Hernan

Abstract: This paper presents a new dual mode nonlinear model predictive controller (NMPC) that is based on the combination of the finite horizon NMPC with the infinite horizon predictive controller (IHMPC). The resulting nonlinear controller is shown to be stable when the IHMPC is globally stabilizing. The main advantage of the proposed controller in comparison to the IHMPC is a better performance as the model nonlinearities are taken into account in the computation of the control law. The advantage of the proposed controller compared to the existing dual mode NMPC is that constraints are also considered in the linear controller that is supposed to control the system when the state enters the terminal set. The performance of the proposed controller is compared to the stable IHMPC through simulation of an industrial styrene polymerization reactor.

Keywords: Nonlinear system control; Constrained control; Controller constraints and structure

Identifier: 10.3182/20080706-5-KR-1001.02059

Conference: Proceedings of the 17th IFAC World Congress, 2008

Location: COEX, Korea, South

Start Date: Sun Jul 06 2008 - End Date: Fri Jul 11 2008