The models that describe the dynamics of reaction systems are usually complex and uncertain. To simplify their description it is useful to identify the states which are dependent/independent of the reactions and flows...
详细信息
The models that describe the dynamics of reaction systems are usually complex and uncertain. To simplify their description it is useful to identify the states which are dependent/independent of the reactions and flows. This allows us, for instance, to design observers which are independent of the (highly uncertain) reaction functions and, in principle, simplify the controller design. The main contribution of this paper is to show the utility of nonlinear state-dependent time-scaling to carry out these objectives. First, we prove the existence of attractive invariant manifolds, which allows us to reduce the dimension of the system under study. Second, we design reaction-independent observers, which are simpler than the existing ones, and have a guaranteed convergence rate in all operating regimes. As an application we study a fifth order baker's yeast fed-batch fermentation process model, for which we reduce the essential dynamics to the plane. A simple input-output feedback linearizing controller is designed and its asymptotic stability established.
We describe an innovative hybrid controller that uses neural networks and Multivariable Predictive control (MPC) to handle abnormal events in process applications. The controller detects abnormal situations, such as g...
详细信息
We describe an innovative hybrid controller that uses neural networks and Multivariable Predictive control (MPC) to handle abnormal events in process applications. The controller detects abnormal situations, such as grinding mill spills or mill power excursions in mineral processing, or incipient flooding in separation columns and then reconfigures the multivariable controller to stabilize the operations. Neural networks are typically used to detect and classify the abnormal situation and knowledge of processdynamics and interactions is used to reconfigure the multivariable predictive controller parameters to stabilize the operations. Thus the MPC can be configured and tuned to provide good control around the `normal' operating range, and when an upset occurs and is detected a new set of tuning parameters are used.
The cement industry can utilize a variety of wastes as fuels;e.g. waste tires, used solvents, plastic or oil wastes. Recently RDF(refuse derived fuel) has also been introduced to the kiln. The rotary kiln itself has b...
ISBN:
(纸本)8492344563
The cement industry can utilize a variety of wastes as fuels;e.g. waste tires, used solvents, plastic or oil wastes. Recently RDF(refuse derived fuel) has also been introduced to the kiln. The rotary kiln itself has been modified to adapt these conditions. Its process characteristics. therefore, have become complex with more delicate sintering conditions. The purpose of this study is to establish an adequate process model to be used in designing controlsystems of such a cement kiln. Different combustion simulations have recently been carried out using commercially available softwares for fluid dynamics. The mathematical model used there, however, is too complicated to use for controls. On the other hand the models developed exclusively for control purpose are too much simplified to reproduce the practical process. The authors have proposed a transfer function model of the control object, i.e. the kiln, based on mass balance, heat transfer and combustion reaction laws.
A new laboratory course has been developed with a set of computer-controlled experiments. The objective is to offer the students 5 to 10 different experiments that build on a generic infrastructure, namely a standard ...
详细信息
A new laboratory course has been developed with a set of computer-controlled experiments. The objective is to offer the students 5 to 10 different experiments that build on a generic infrastructure, namely a standard personal computer, an intelligent process interface and an experiment. The processor board communicates through a back plane bus with the instrumentation cards and on the other side provides standard communication interfaces such as RS232 for the communication with the personal computer. This allows the students to use their own notebook computer, which is made available to every student in the university since last year. The user interface is presently built in LabView and the combination of Matlab and Simulink is used for simulations and other analytical off-line tasks. The experimental setups are designed such that they are mobile, flexible and easy to maintain. Furthermore,they can be used for both educational and research activities in our group.
The state-dependent Riccati equation (SDRE) method is used to control the nonlinear nonminimum-phase dynamics of a continuously stirred tank reactor (CSTR). The benefits of using the SDRE method are that it can be dir...
详细信息
The state-dependent Riccati equation (SDRE) method is used to control the nonlinear nonminimum-phase dynamics of a continuously stirred tank reactor (CSTR). The benefits of using the SDRE method are that it can be directly applied to the nonminimum-phase system and hard bounds can be imposed on the control activity. From simulation analysis, it is determined that a complicated state-dependent state weighting matrix Q(x) would be required in order to produce satisfactory responses to changes in the set point of the reactor. To avoid the problem of finding such a state weighting matrix, an asymmetric solution of the SDRE is used instead of the symmetric positive-definite solution. Since there are an infinite number of asymmetric solutions, the paper discusses the process that was used in obtaining the selected asymmetric solution. Simulations are performed which produce a level of confidence that the closed loop system is asymptotically stable and is robust to parameter variations in the chemical concentration level in the educt flow.
The problem addressed in this paper is as follows. Consider a non-linear discrete time system with parameters, which can be a priori classified into two groups: constant parameters and time varying ones. It is also as...
详细信息
The problem addressed in this paper is as follows. Consider a non-linear discrete time system with parameters, which can be a priori classified into two groups: constant parameters and time varying ones. It is also assumed that loose attributes like 'slow' or 'fast' express the only prior information concerning the dynamics of varying parameters. The question of interest is how to estimate the constant parameters of the process (on the basis of a recorded batch of input and output data) so that possible alterations of the parameters that are expected to vary with time are taken into account. To deal with this kind of non-linear fixed-interval smoothing problem a two-level procedure based on rectangular window technique and Gauss-Newton method is presented. All the computations are performed off-line. The procedure is applicable in a pre-processing stage of adaptive parameter estimation where the parameters that are evidently time invariant are identified from past records. Thus the number of parameters that have to be tracked (i.e. estimated) on-line is reduced. The efficiency of the scheme is illustrated on a simulated industrial process.
processcontrol has been recognized as an important means of improving the performance and consistency of thermoplastic parts. However, no single control strategy or system design has been universally accepted, and mo...
详细信息
processcontrol has been recognized as an important means of improving the performance and consistency of thermoplastic parts. However, no single control strategy or system design has been universally accepted, and molding systems continue to produce defective components during production. The capability of the injection molding process is limited by the thermal and flow dynamics of the heated polymer melt. This paper discusses some of the difficulties posed by the complex and distributed nature of the injection molding process. The flow and thermal dynamics of the process are analyzed with respect to transport and rheology. Then, two novel processing methods are described to enable in-cycle flow, pressure, and thermal control. Simulation and experimental results demonstrate effectiveness of these innovations to increase the consistency and flexibility in polymer processing. Such system design changes simplify the requisite control structures while improving the process robustness and productivity.
The high computation and visualization power now available at the control engineer's desk opens the perspective of embedding CACSD Technology into a virtual engineering environment. This enables concurrent interac...
详细信息
The high computation and visualization power now available at the control engineer's desk opens the perspective of embedding CACSD Technology into a virtual engineering environment. This enables concurrent interaction of control engineering with multidisciplinary high-fidelity system dynamics modeling, and supports a fundamental change of the conventional control design tuning process in a computation framework of fast nonlinear simulation, automatic search for feasible solutions, and interactive multiobjective optimization to trade-off conflicting design requirements. Such an environment promotes active decision support to foster the control engineer's role as the prime decision maker in system dynamics design.
The proceedings contain 41 papers. The special focus in this conference is on System Architecture and System Software. The topics include: An introduction to DRAMA project;an introduction to OCEANS project;an introduc...
ISBN:
(纸本)3540659692
The proceedings contain 41 papers. The special focus in this conference is on System Architecture and System Software. The topics include: An introduction to DRAMA project;an introduction to OCEANS project;an introduction to NANOS project;synchronous communication of molecular and electronic structure information in a distributed computing environment;profile-based selection of load value and address predictors;instruction-level microprocessor modeling of scientific applications;neural network classifiers execution on superscalar microprocessors;coping with very high latencies in petaflop computer systems;parallelization via constrained storage mapping optimization;a comparison of automatic parallelizing compiler and improvements by compiler directives;dynamically adaptive parallel programs;a feasibility study in iterative compilation;multithreading runtime support for loop and functional parallelism;parallelization and vectorization effects on a code simulating a vitreous lattice model with constrained dynamics;multi-phase hydrodynamic simulations on parallel computer;the case for a practical approach to performance prediction;evaluation of parallel form factor calculations and a static load balancing algorithm;parallel distributed optimization by resource addition and reduction;application of parallel sparse direct methods in semiconductor device and process simulation;integrity and performance in network attached storage;basic design of the earth simulator;high-performance parallel fem for geophysical applications;a graphics tool for the synthesis of automatic and user-driven program parallelization in the compilation process;parallel map recognition by pipeline control;a parallel programming environment for networks and emulation facility for data-driven real-time multi-processing.
Repetitive, or multipass, processes are a class of 2D systems of both practical and algorithmic/theoretical interest whose dynamics cannot be analysed or controlled using standard (1D) systems theory. Recently it has ...
详细信息
Repetitive, or multipass, processes are a class of 2D systems of both practical and algorithmic/theoretical interest whose dynamics cannot be analysed or controlled using standard (1D) systems theory. Recently it has been shown that the modelling of the boundary conditions, also known as the process initial conditions, is a crucial feature in the analysis and control of these processes. This paper presents some further results on the effects of so-called 'dynamic' process initial conditions on the controllability and stability properties of discrete linear repetitive processes. Previous work has shown that these dynamic process initial conditions alone can destroy the stability properties of these processes. Hence their effects must be 'adequately' accounted for the process modelling stage in order to ensure that subsequent analysis does not lead to incorrect results/conclusions. The main results developed in this paper can be summarised as follows. (i) Computationally efficient stability tests which can, in effect, be applied using standard, or 1D, linear systems tests. (ii) Characterisation of so-called pass controllability in the form of matrix rank based conditions. (iii) Conditions under which the dynamic process initial conditions can be selected to ensure stability and pass controllability.
暂无评论