2005年 第31卷 第1期
2005, 31(1): 1-12.
摘要:
The problem of state estimation for uncertain systems has attracted a recurring interest in the past decade. In this paper, we shall give an overview on some of the recent development in the area by focusing on the robust H2 (Kalman) filtering of uncertain discrete-time systems. The robust H2 estimation is concerned with the design of a fixed estimator for a family of plants under consideration such that the estimation error covariance is of a minimal upper bound. The uncertainty under consideration includes norm-bounded uncertainty and polytopic uncertainty. In the finite horizon case, we shall discuss a parameterized difference Riccati equation approach for systems with norm-bounded uncertainty and pinpoint the difference of state estimation between systems without uncertainty and those with uncertainty. In the infinite horizon case, we shall deal with both the norm-bounded and polytopic uncertainties using a linear matrix inequality (LMI) approach. In particular, we shall demonstrate how the conservatism of design can be improved using a slack variable technique. We also propose an iterative algorithm to refine a designed estimator. An example will be given to compare estimators designed using various techniques.
The problem of state estimation for uncertain systems has attracted a recurring interest in the past decade. In this paper, we shall give an overview on some of the recent development in the area by focusing on the robust H2 (Kalman) filtering of uncertain discrete-time systems. The robust H2 estimation is concerned with the design of a fixed estimator for a family of plants under consideration such that the estimation error covariance is of a minimal upper bound. The uncertainty under consideration includes norm-bounded uncertainty and polytopic uncertainty. In the finite horizon case, we shall discuss a parameterized difference Riccati equation approach for systems with norm-bounded uncertainty and pinpoint the difference of state estimation between systems without uncertainty and those with uncertainty. In the infinite horizon case, we shall deal with both the norm-bounded and polytopic uncertainties using a linear matrix inequality (LMI) approach. In particular, we shall demonstrate how the conservatism of design can be improved using a slack variable technique. We also propose an iterative algorithm to refine a designed estimator. An example will be given to compare estimators designed using various techniques.
2005, 31(1): 13-18.
摘要:
This paper introduces a self-learning control approach based on approximate dynamic programming. Dynamic programming was introduced by Bellman in the 1950’s for solving optimal control problems of nonlinear dynamical systems. Due to its high computational complexity, the applications of dynamic programming have been limited to simple and small problems. The key step in finding approximate solutions to dynamic programming is to estimate the performance index in dynamic programming. The optimal control signal can then be determined by minimizing (or maximizing) the performance index. Artificial neural networks are very efficient tools in representing the performance index in dynamic programming. This paper assumes the use of neural networks for estimating the performance index in dynamic programming and for generating optimal control signals, thus to achieve optimal control through self-learning.
This paper introduces a self-learning control approach based on approximate dynamic programming. Dynamic programming was introduced by Bellman in the 1950’s for solving optimal control problems of nonlinear dynamical systems. Due to its high computational complexity, the applications of dynamic programming have been limited to simple and small problems. The key step in finding approximate solutions to dynamic programming is to estimate the performance index in dynamic programming. The optimal control signal can then be determined by minimizing (or maximizing) the performance index. Artificial neural networks are very efficient tools in representing the performance index in dynamic programming. This paper assumes the use of neural networks for estimating the performance index in dynamic programming and for generating optimal control signals, thus to achieve optimal control through self-learning.
2005, 31(1): 19-31.
摘要:
This paper presents several neural network based modelling, reliable optimal control, and iterative learning control methods for batch processes. In order to overcome the lack of robustness of a single neural network, bootstrap aggregated neural networks are used to build reliable data based empirical models. Apart from improving the model generalisation capability, a bootstrap aggregated neural network can also provide model prediction confidence bounds. A reliable optimal control method by incorporating model prediction confidence bounds into the optimisation objective function is presented. A neural network based iterative learning control strategy is presented to overcome the problem due to unknown disturbances and model-plant mismatches. The proposed methods are demonstrated on a simulated batch polymerisation process.
This paper presents several neural network based modelling, reliable optimal control, and iterative learning control methods for batch processes. In order to overcome the lack of robustness of a single neural network, bootstrap aggregated neural networks are used to build reliable data based empirical models. Apart from improving the model generalisation capability, a bootstrap aggregated neural network can also provide model prediction confidence bounds. A reliable optimal control method by incorporating model prediction confidence bounds into the optimisation objective function is presented. A neural network based iterative learning control strategy is presented to overcome the problem due to unknown disturbances and model-plant mismatches. The proposed methods are demonstrated on a simulated batch polymerisation process.
2005, 31(1): 32-42.
摘要:
This paper summarizes recent progress by the authors in developing two solution frame-works for dual control. The first solution framework considers a class of dual control problems where there exists a parameter uncertainty in the observation equation of the LQG problem. An analytical active dual control law is derived by a variance minimization approach. The issue of how to determine an optimal degree of active learning is then addressed, thus achieving an optimality for this class of dual control problems. The second solution framework considers a general class of discrete-time LQG problems with unknown parameters in both state and observation equations. The best possible (partial) closed-loop feedback control law is derived by exploring the future nominal posterior probabilities, thus taking into account the effect of future learning when constructing the optimal nominal dual control.
This paper summarizes recent progress by the authors in developing two solution frame-works for dual control. The first solution framework considers a class of dual control problems where there exists a parameter uncertainty in the observation equation of the LQG problem. An analytical active dual control law is derived by a variance minimization approach. The issue of how to determine an optimal degree of active learning is then addressed, thus achieving an optimality for this class of dual control problems. The second solution framework considers a general class of discrete-time LQG problems with unknown parameters in both state and observation equations. The best possible (partial) closed-loop feedback control law is derived by exploring the future nominal posterior probabilities, thus taking into account the effect of future learning when constructing the optimal nominal dual control.
2005, 31(1): 43-55.
摘要:
In this paper, we shall summarize a new approach to robust and fault tolerant control proposed recently by the author. This approach is based on a variation of all controller parametrization. This robust and fault-tolerant control design consists of two parts: a nominal performance controller and a robustness controller, and works in such a way that when a component (sensor, actuator, etc.) failure is detected, the controller structure is reconfigured by adding a robustness loop to compensate the fault. We shall illustrate how this strategy works under various situations.
In this paper, we shall summarize a new approach to robust and fault tolerant control proposed recently by the author. This approach is based on a variation of all controller parametrization. This robust and fault-tolerant control design consists of two parts: a nominal performance controller and a robustness controller, and works in such a way that when a component (sensor, actuator, etc.) failure is detected, the controller structure is reconfigured by adding a robustness loop to compensate the fault. We shall illustrate how this strategy works under various situations.
2005, 31(1): 56-63.
摘要:
Internet-based process control is becoming new generations of control systems, in which the Internet is used as a platform for global remote monitoring and control. The obvious benefit is to enable global collaboration between operators from geographically dispersed locations, data sharing and data provision for remote monitoring and control. However, connection to an open network and the use of universal technology present new problems that did not exist with the conventional design and construction of control systems, such as time delay and data loss in Internet transmission and security. This paper reviews the latest research results and presents design guidance of Internet based monitoring and control systems.
Internet-based process control is becoming new generations of control systems, in which the Internet is used as a platform for global remote monitoring and control. The obvious benefit is to enable global collaboration between operators from geographically dispersed locations, data sharing and data provision for remote monitoring and control. However, connection to an open network and the use of universal technology present new problems that did not exist with the conventional design and construction of control systems, such as time delay and data loss in Internet transmission and security. This paper reviews the latest research results and presents design guidance of Internet based monitoring and control systems.
2005, 31(1): 64-74.
摘要:
We present a nonlinear flow control scheme based on a buffer management model with physical constraints. It extends previous result of Pitsillides et al. in [6] by improving the queue length regulation for better service of network traffics. Besides a single node system, we also address the decentralized control of many cascaded nodes. The proposed discontinuous controller asymptot- ically regulates the buffer queue length at the output port of a router/switch to a constant reference value, under unknown time varying interfering traffics and saturation constraints on control input and states. Its continuous approximation achieves practical regulation with an ultimate bound on the regulation error tunable by a design parameter.
We present a nonlinear flow control scheme based on a buffer management model with physical constraints. It extends previous result of Pitsillides et al. in [6] by improving the queue length regulation for better service of network traffics. Besides a single node system, we also address the decentralized control of many cascaded nodes. The proposed discontinuous controller asymptot- ically regulates the buffer queue length at the output port of a router/switch to a constant reference value, under unknown time varying interfering traffics and saturation constraints on control input and states. Its continuous approximation achieves practical regulation with an ultimate bound on the regulation error tunable by a design parameter.
2005, 31(1): 75-83.
摘要:
Research in control systems, sensor fusion and haptic interfaces is reviewed.
Research in control systems, sensor fusion and haptic interfaces is reviewed.
2005, 31(1): 84-91.
摘要:
The paper consists of three topics on control theory and engineering applications, namely bifurcation control, manufacturing planning, and formation control. For each topic, we summarize the control problem to be addressed and some key ideas used in our recent research. Interested readers are referred to related publications for more details. Each of the three topics in this paper is technically independent from the other ones. However, all three parts together reflect the recent research activities of the first author, jointly with other researchers in different fields.
The paper consists of three topics on control theory and engineering applications, namely bifurcation control, manufacturing planning, and formation control. For each topic, we summarize the control problem to be addressed and some key ideas used in our recent research. Interested readers are referred to related publications for more details. Each of the three topics in this paper is technically independent from the other ones. However, all three parts together reflect the recent research activities of the first author, jointly with other researchers in different fields.
2005, 31(1): 92-104.
摘要:
The development of control techniques to mitigate the effects of unknown hysteresis preceding with plants has recently re-attracted significant attention. In this paper, we first give a brief review of presently developed hysteresis models and hysteresis compensating control methods. Then, with the use of the Prandtl-Ishlinskii hysteresis model, we propose a robust adaptive control scheme. The novelty is that the model of hysteresis nonlinearities is firstly fused with the available control techniques without necessarily constructing a hysteresis inverse. The global stability of the adaptive system and tracking a desired trajectory to a certain precision are achieved. Simulations performed on a nonlinear system illustrate and clarify the approach.
The development of control techniques to mitigate the effects of unknown hysteresis preceding with plants has recently re-attracted significant attention. In this paper, we first give a brief review of presently developed hysteresis models and hysteresis compensating control methods. Then, with the use of the Prandtl-Ishlinskii hysteresis model, we propose a robust adaptive control scheme. The novelty is that the model of hysteresis nonlinearities is firstly fused with the available control techniques without necessarily constructing a hysteresis inverse. The global stability of the adaptive system and tracking a desired trajectory to a certain precision are achieved. Simulations performed on a nonlinear system illustrate and clarify the approach.
2005, 31(1): 123-131.
摘要:
Many industrial processes have compositive complexities including multivariable, strong coupling, nonlinearity, time-variant and operating condition variations. Combining multivariable adaptive decoupling control with neural networks, this paper presents a multivariable neural network-based decoupling control algorithm. This control algorithm is integrated with distributed control technique and intelligent control technique, and a three-leveled intelligent decoupling control system consisting of basic control level, coordinating control level, and management and decision level is developed. The configuration and function of the control system are discussed in detail. This system has been successfully applied in ball mill pulverizing systems of 200MW power units, and remarkable benefits have been obtained.
Many industrial processes have compositive complexities including multivariable, strong coupling, nonlinearity, time-variant and operating condition variations. Combining multivariable adaptive decoupling control with neural networks, this paper presents a multivariable neural network-based decoupling control algorithm. This control algorithm is integrated with distributed control technique and intelligent control technique, and a three-leveled intelligent decoupling control system consisting of basic control level, coordinating control level, and management and decision level is developed. The configuration and function of the control system are discussed in detail. This system has been successfully applied in ball mill pulverizing systems of 200MW power units, and remarkable benefits have been obtained.
2005, 31(1): 132-142.
摘要:
In this paper we review the recent advances in three sub-areas of iterative learning control (ILC): 1) linear ILC for linear processes, 2) linear ILC for nonlinear processes which are global Lipschitz continuous (GLC), and 3) nonlinear ILC for general nonlinear processes. For linear processes, we focus on several basic configurations of linear ILC. For nonlinear processes with linear ILC, we concentrate on the design and transient analysis which were overlooked and missing for a long period. For general classes of nonlinear processes, we demonstrate nonlinear ILC methods based on Lyapunov theory, which is evolving into a new control paradigm.
In this paper we review the recent advances in three sub-areas of iterative learning control (ILC): 1) linear ILC for linear processes, 2) linear ILC for nonlinear processes which are global Lipschitz continuous (GLC), and 3) nonlinear ILC for general nonlinear processes. For linear processes, we focus on several basic configurations of linear ILC. For nonlinear processes with linear ILC, we concentrate on the design and transient analysis which were overlooked and missing for a long period. For general classes of nonlinear processes, we demonstrate nonlinear ILC methods based on Lyapunov theory, which is evolving into a new control paradigm.
2005, 31(1): 143-150.
摘要:
A quite great progress of the supervisory control theory for discrete event systems (DES) has been made in the past nearly twenty years, and now, automata, formal language and Petri nets become the main research tools. This paper focus on the Petri nets based supervisory control theory of DES. Firstly, we review the research results in this field, and claim that there generally exists a problem in Petri nets based supervisory control theory of DES, that is, the deadlock caused by the controller introduced to enforce the given specification occurs in the closed-loop systems, especially the deadlock occurs in the closed-loop system in which the original plant is live. Finally, a possible research direction is presented for the solution of this problem.
A quite great progress of the supervisory control theory for discrete event systems (DES) has been made in the past nearly twenty years, and now, automata, formal language and Petri nets become the main research tools. This paper focus on the Petri nets based supervisory control theory of DES. Firstly, we review the research results in this field, and claim that there generally exists a problem in Petri nets based supervisory control theory of DES, that is, the deadlock caused by the controller introduced to enforce the given specification occurs in the closed-loop systems, especially the deadlock occurs in the closed-loop system in which the original plant is live. Finally, a possible research direction is presented for the solution of this problem.
2005, 31(1): 151-160.
摘要:
This paper presents a new algorithm designed to control the shape of the output probability density function (PDF) of singular systems subjected to non-Gaussian input. The aim is to select a control input uk such that the output PDF is made as close as possible to a given PDF. Based on the B-spline neural network approximation of the output PDF, the control algorithm is formulated by extending the developed PDF control strategies of non-singular systems to singular systems. It has been shown that under certain conditions the stability of the closed-loop system can be guaranteed. Simulation examples are given to show the effectiveness of the proposed control algorithm.
This paper presents a new algorithm designed to control the shape of the output probability density function (PDF) of singular systems subjected to non-Gaussian input. The aim is to select a control input uk such that the output PDF is made as close as possible to a given PDF. Based on the B-spline neural network approximation of the output PDF, the control algorithm is formulated by extending the developed PDF control strategies of non-singular systems to singular systems. It has been shown that under certain conditions the stability of the closed-loop system can be guaranteed. Simulation examples are given to show the effectiveness of the proposed control algorithm.
2005, 31(1): 161-174.
摘要:
This paper presents an introductory overview on the development of fault-tolerant control systems. For this reason, the paper is written in a tutorial fashion to summarize some of the important results in this subject area deliberately without going into details in any of them. However, key references are provided from which interested readers can obtain more detailed information on a particular subject. It is necessary to mention that, throughout this paper, no efforts were made to provide an exhaustive coverage on the subject matter. In fact, it is far from it. The paper merely represents the view and experience of its author. It can very well be that some important issues or topics were left out unintentionally. If that is the case, the author sincerely apologizes in advance.After a brief account of fault-tolerant control systems, particularly on the original motivations, and the concept of redundancies, the paper reviews the development of fault-tolerant control systems with highlights to several important issues from a historical perspective. The general approaches to fault-tolerant control has been divided into passive, active, and hybrid approaches. The analysis techniques for active fault-tolerant control systems are also discussed. Practical applications of fault- tolerant control are highlighted from a practical and industrial perspective. Finally, some critical issues in this area are discussed as open problems for future research/development in this emerging field.
This paper presents an introductory overview on the development of fault-tolerant control systems. For this reason, the paper is written in a tutorial fashion to summarize some of the important results in this subject area deliberately without going into details in any of them. However, key references are provided from which interested readers can obtain more detailed information on a particular subject. It is necessary to mention that, throughout this paper, no efforts were made to provide an exhaustive coverage on the subject matter. In fact, it is far from it. The paper merely represents the view and experience of its author. It can very well be that some important issues or topics were left out unintentionally. If that is the case, the author sincerely apologizes in advance.After a brief account of fault-tolerant control systems, particularly on the original motivations, and the concept of redundancies, the paper reviews the development of fault-tolerant control systems with highlights to several important issues from a historical perspective. The general approaches to fault-tolerant control has been divided into passive, active, and hybrid approaches. The analysis techniques for active fault-tolerant control systems are also discussed. Practical applications of fault- tolerant control are highlighted from a practical and industrial perspective. Finally, some critical issues in this area are discussed as open problems for future research/development in this emerging field.
2005, 31(1): 105-122.
摘要:
Multirate systems are abundant in industry; for example, many soft-sensor design problems are related to modeling, parameter identification, or state estimation involving multirate systems. The study of multirate systems goes back to the early 1950s, and has become an active research area in systems and control. This paper briefly surveys the history of development in the area of multirate systems, and introduces some basic concepts and latest results on multirate systems, including a polynomial transformation technique and the lifting technique as tools for handling multirate systems, lifted state space models, parameter identification of dual-rate systems, how to determine fast single-rate models from dual-rate models and directly from dual-rate data, and a hierarchical identification method for general multirate systems. Finally, some further research topics for multirate systems are given.
Multirate systems are abundant in industry; for example, many soft-sensor design problems are related to modeling, parameter identification, or state estimation involving multirate systems. The study of multirate systems goes back to the early 1950s, and has become an active research area in systems and control. This paper briefly surveys the history of development in the area of multirate systems, and introduces some basic concepts and latest results on multirate systems, including a polynomial transformation technique and the lifting technique as tools for handling multirate systems, lifted state space models, parameter identification of dual-rate systems, how to determine fast single-rate models from dual-rate models and directly from dual-rate data, and a hierarchical identification method for general multirate systems. Finally, some further research topics for multirate systems are given.
