There have been numerous methods for learning and predicting time series ranging from the traditional time-series analyses to recent approaches using neural networks. A central issue common to all of them is the deter...
详细信息
There have been numerous methods for learning and predicting time series ranging from the traditional time-series analyses to recent approaches using neural networks. A central issue common to all of them is the determination of model structure. Both mean prediction error and An Information Criterion (AIC) are useful in model selection;the model with the smallest mean prediction error or AIC is selected from among a set of models as the best one. In this way they give a solution to the problem of model selection. Due to huge search space, however, the mean prediction error or AIC alone is not powerful enough to find the best model structure from among all the candidates. In the present paper the authors propose to use both a structural learning with forgetting and the mean prediction error or AIC to find a model with better generalization ability. Jordan networks and buffer networks, popular in the modeling of time series, are examined in this paper. The structural learning with forgetting and backpropagation (BP) learning are applied to compare the learning and prediction performance of these two types of models. Simulation results demonstrate that the structural learning with forgetting has better generalization ability than BP learning both in Jordan networks and buffer networks.
An adaptive controller with improved performance characteristics is introduced. The proposed controller extends recent results in this area since it achieves performance improvement of the zero-state output error in t...
详细信息
An adaptive controller with improved performance characteristics is introduced. The proposed controller extends recent results in this area since it achieves performance improvement of the zero-state output error in the presence of some uncertainty on the high-frequency gain of the plant.
This paper presents the design and implementation of an active control mechanism for noise cancellation and vibration suppression within an adaptive control framework. A control mechanism is designed within a feedforw...
详细信息
This paper presents the design and implementation of an active control mechanism for noise cancellation and vibration suppression within an adaptive control framework. A control mechanism is designed within a feedforward control structure on the basis of optimum cancellation at an observation point. The design relations are formulated such that to allow on-line design and implementation and thus result in a self-tuning control algorithm. The algorithm is implemented on an integrated digital signal processing (DSP) and transputer system. Simulation results verifying the performance of the algorithm are presented and discussed. (C) 1996 Academic Press Limited
This paper presents an investigation into the utilisation of digital signal processing and parallel processing techniques for the real-time simulation of a flexible manipulator system. A finite dimensional simulation ...
详细信息
This paper presents an investigation into the utilisation of digital signal processing and parallel processing techniques for the real-time simulation of a flexible manipulator system. A finite dimensional simulation of the system is developed using a finite difference approximation to the governing dynamic equation of the manipulator. The proposed algorithm allows dynamic modification of the boundary conditions and the inclusion of a distributed actuator and sensor term in the system dynamic equation. The algorithm developed is implemented on a number of uni-processor and multi-processor, homogeneous and heterogeneous parallel architectures. The partitioning and mapping of the algorithm on the homogeneous and heterogeneous architectures is also explored. A comparison of the results of these implementations is made and discussed to establish merits of design and real-time processing requirements in the control of flexible manipulator systems. (C) 1996 Academic Press Limited
This paper formulates and analyzes a robust tracking problem for sampled data repetitive controlsystems in the presence of structured linear periodically time varying perturbations. Two pertinent issues are addressed...
详细信息
This paper formulates and analyzes a robust tracking problem for sampled data repetitive controlsystems in the presence of structured linear periodically time varying perturbations. Two pertinent issues are addressed. One is stability robustness, that is, if stability will be retained under such perturbations. The other is robust tracking, that is, if the tracking criterion will be met for all variations of the plant under the given class of perturbations. The tracking measure is taken to be the induced power-norm, which is defined as the power of the steady-state tracking error for the worst periodic input of unit power, and the tracking criterion is that only induced power-norms which are less than a prespecified bound are acceptable. A generalized notion of structured singular value, previously introduced in the literature. is used to answer these questions in the form of some necessary and sufficient conditions.
作者：
Wei-Ming LingDaniel E. RiveraDepartment of Chemical
Bio and Materials Engineering and Control Systems Engineering Laboratory Computer-Integrated Manufacturing Systems Research Center Arizona State University Tempe AZ 85287-6006
A two-step nonlinear system identification method using restricted complexity models (RCM) is proposed. In the first step, a parsimonious yet full order Volterra model is identified using the orthogonal least squares ...
详细信息
A two-step nonlinear system identification method using restricted complexity models (RCM) is proposed. In the first step, a parsimonious yet full order Volterra model is identified using the orthogonal least squares method. In the second step, using a control relevant approach, the full order model is further reduced to a restricted complexity model which is more amenable to control design and analysis. The minimization problem in the model reduction step is posed such that it can be solved using general optimization routines. A corresponding two-step model validation procedure is implemented to ensure the closed-loop performance of the resulting model. Effectiveness of the proposed method is illustrated by a polymerization reactor example.
This paper proposes a hybrid quasi-ARMAX modeling and identification scheme for nonlinear systems. The idea is to incorporate a group of certain nonlinear nonparametric models (NNMs) into a linear ARMAX structure. Par...
详细信息
ISBN:
(纸本)0780335902
This paper proposes a hybrid quasi-ARMAX modeling and identification scheme for nonlinear systems. The idea is to incorporate a group of certain nonlinear nonparametric models (NNMs) into a linear ARMAX structure. Particular effort is made to find a better compromise to the trade-off between the model flexibility and the simplicity for estimation by using knowledge information efficiently. As the result, we obtain a model equipped with a linear ARMAX structure, flexibility and simplicity. The effectiveness and usefulness of the proposed hybrid model are examined by applying it to identification and control of nonlinear systems.
The regulation problem of linear discrete-time systems under state and control constraints is investigated. In the first part of the paper necessary and sufficient conditions for the existence of a solution to the con...
详细信息
The regulation problem of linear discrete-time systems under state and control constraints is investigated. In the first part of the paper necessary and sufficient conditions for the existence of a solution to the constrained control problem are established. The constructive form of the proof of this result provides also a method for the derivation of a control law transferring to the origin any state belonging to a given set of initial states while respecting the state and control constraints. Then a design technique for the determination of a solution to the constrained control problem is developed. The proposed technique is based on the reduction of the constrained control problem to simple linear programming problems.
A fundamental limitation in achieving high performance control for multivariable plants is associated with the uncertainty in the modelling of the plant. Previous measures of robust stability uncertainty have typicall...
详细信息
A fundamental limitation in achieving high performance control for multivariable plants is associated with the uncertainty in the modelling of the plant. Previous measures of robust stability uncertainty have typically been based on complex plant perturbations, which often lead to excessively conservative controller design. It is the purpose of this paper to propose a nonconservative measure of plant model uncertainty; this is done by utilizing results obtained on the real stability radius problem. A CAD approach to synthesize controllers for a plant which has a specified degree of plant model uncertainty is then proposed. Some application studies of the procedure are included.
This paper presents a robust fault detection system (FDS) for dynamic systems with unmodeled dynamics. In the FDS, umnodeled dynamics is first qualified as soft bound, which as well as model parameters are estimated u...
详细信息
This paper presents a robust fault detection system (FDS) for dynamic systems with unmodeled dynamics. In the FDS, umnodeled dynamics is first qualified as soft bound, which as well as model parameters are estimated using a robust identification algorithm. Then as a fault detection index, Kullback discrimination information (KDI) is derived into a feasible form and an index of umnodeled dynamics is also introduced. A decision making scheme is thus developed so that fault detection is carried out based on the KDI, the index of umnodeled dynamics and other prior information about the system.
暂无评论