Accurate frequency-domain system identification demands for reliable computational algorithms. The aim of this paper is to develop a new algorithm for parametric system identification with favorable convergence proper...
详细信息
Accurate frequency-domain system identification demands for reliable computational algorithms. The aim of this paper is to develop a new algorithm for parametric system identification with favorable convergence properties and optimal numerical conditioning. Recent results in frequency-domain instrumental variable identification are exploited, which lead to enhanced convergence properties compared to classical identification algorithms. In addition, bi-orthonormal polynomials with respect to a data-dependent bi-linear form are introduced for system identification. Hereby, optimal numerical conditioning of the relevant system of equations is achieved. This is shown to be particularly important for the class of instrumental variable algorithms, for which numerical conditioning is typically quadratic compared to alternative frequency-domain identification algorithms. Superiority of the proposed algorithm is demonstrated by means of both simulation and experimental results. (C) 2014 Elsevier Ltd. All rights reserved.
In contrast to the worst case approach, the average case setting provides less conservative insight into the quality of estimation algorithms. In this paper we consider two local average case error measures of algorit...
详细信息
In contrast to the worst case approach, the average case setting provides less conservative insight into the quality of estimation algorithms. In this paper we consider two local average case error measures of algorithms based on noisy information, in Hilbert norms in the problem element and information spaces. We define the optimal algorithm and provide formulas for its two local errors, which explicitly exhibit the influence of factors such as information, information (measurement) errors, norms in the considered spaces, a subset where approximations are allowed, and "unmodeled dynamics." Based on the error expression, we formulate in algebraic language the problem of selecting the optimal approximating subspace. The solution is given along with the specific formula for the error, which depends on the eigenvalues of a certain matrix defined by information and norms under consideration.
Modern robot control methods require knowledge of the form and coefficients of the joint friction. The main feature of the present work is the practical application of the nonlinear filtering approach of Detchmendy an...
详细信息
Modern robot control methods require knowledge of the form and coefficients of the joint friction. The main feature of the present work is the practical application of the nonlinear filtering approach of Detchmendy and Sridhar, on both computer simulation and experimental data, in frictional identification. The results confirm the feasibility of the proposed estimation approach. In addition, different models of friction have been examined for the estimation of Coulomb and viscous coefficients. In these models, asymmetry of the parameters, as an essential assumption, has been investigated and separate values determined for positive and negative directions of rotation. The experimental results justify the introduction of asymmetry, particularly for the Coulomb coefficient. It is shown that an asymmetric Coulomb and viscous frictional model is appropriate for the DC servo motor investigated. The quality of the fit of the coefficients was measured in terms of the RMS torque error, and low torque errors has been obtained. (C) 1997 Elsevier Science Ltd.
The problem of unbiased recursive identification of a plant model in closed-loop operation is considered. A particular form of an output error predictor for the closed loop is introduced. This allows one to derive a p...
详细信息
The problem of unbiased recursive identification of a plant model in closed-loop operation is considered. A particular form of an output error predictor for the closed loop is introduced. This allows one to derive a parameter estimation algorithm for the plant model that is globally asymptotically stable and asymptotically unbiased in the presence of noise. The paper presents a stability analysis in a deterministic environment and a convergence analysis in the stochastic environment. Both require a mild sufficient passivity condition to be satisfied. Simulations and real-time experiments on flexible transmission illustrate the performances of the proposed algorithm. (C) 1997 Elsevier Science Ltd.
This paper discusses the identification of rime-varying fuzzy systems. A square-root type recursive algorithm with variable forgetting factor is presented to estimate the parameters of systems. Numerical examples are ...
详细信息
This paper discusses the identification of rime-varying fuzzy systems. A square-root type recursive algorithm with variable forgetting factor is presented to estimate the parameters of systems. Numerical examples are provided to illustrate the performance of the identification algorithm. An interesting discovery of this paper is that, for a linear system, a fuzzy model may have a frequency response close to that of a conventional model, say, a difference equation model, although the orders of two models may be different.
This paper deals with the identification of three classes of (linear time-invariant, time-varying, and nonlinear) discrete-time systems via discrete orthogonal functions (DOFs). The important results of this study are...
详细信息
This paper deals with the identification of three classes of (linear time-invariant, time-varying, and nonlinear) discrete-time systems via discrete orthogonal functions (DOFs). The important results of this study are as follows. (1) The new discrete-pulse orthogonal functions (DPOFs) approach is much simpler than that of Horng and Ho (1987). (2) The identification algorithms derived via DPOFs are computationally the simplest of all the algorithms developed via discrete Laguerre polynomials (DLPs), or discrete Legendre orthogonal polynomials (DLOPs). (3) The DPOF-based algorithms and the standard well known least squares algorithms are identically one and the same for discrete-time system identification. In view of points (2) and (3), it is concluded that the DOF approach for system identification is not an attractive approach computationally. (C) 1998 Elsevier Science Ltd.
We present an elimination theory-based method for solving equality-constrained multivariable polynomial least-squares problems in system identification. While most algorithms in elimination theory rely upon Groebner b...
详细信息
We present an elimination theory-based method for solving equality-constrained multivariable polynomial least-squares problems in system identification. While most algorithms in elimination theory rely upon Groebner bases and symbolic multivariable polynomial division algorithms, we present an algorithm which is based on computing the nullspace of a large sparse matrix and the zeros of a scalar, univariate polynomial. (C) 2014 Elsevier Ltd. All rights reserved.
Design of filters ensuring convergence of recursive estimation algorithms in the presence of uncertainty in the plant model is a key problem in the area of identification and adaptive control. This paper addresses the...
详细信息
Design of filters ensuring convergence of recursive estimation algorithms in the presence of uncertainty in the plant model is a key problem in the area of identification and adaptive control. This paper addresses the problem of designing filters ensuring strict positive realness of a family of uncertain polynomials over an assigned region of the complex plane. The uncertainty is assumed to be both structured and unstructured. When the structured uncertainty of the family is represented through regions of root location of a certain shape (for instance, circles centered on the real axis), an optimal solution of the problem is provided. For more general uncertainty regions, a simple procedure for constructing sub-optimal solutions to the filter design problem is proposed. A numerical example is fully developed to show the effectiveness of the proposed approach.
Indirect closed-loop identification assumes the knowledge of the controller. In this brief, the constrained version of the instrumental variable (IV) methods is developed. It is based on using the known controller par...
详细信息
Indirect closed-loop identification assumes the knowledge of the controller. In this brief, the constrained version of the instrumental variable (IV) methods is developed. It is based on using the known controller parameters to impose linear constraints upon the closed-loop system parameters and then solving the constrained estimation problem by the La-grange method. The developed constrained IV methods outperform the unconstrained counterparts in, such aspects as insensitivity to common factors, estimation accuracy and robustness against noise. Computer simulations are presented which not only support the theoretical analysis, but also give good insight into the properties of the developed constrained IV methods.
作者：
Oku, HKimura, HUniv Twente
Fac Appl Phys Syst & Control Engn Div NL-7500 AE Enschede Netherlands Univ Tokyo
Grad Sch Frontier Sci Dept Complex Sci & Engn Bunkyo Ku Tokyo 1138656 Japan
Sometimes we obtain some prior information about a system to be identified, e.g., the order, model structure etc. In this paper, we consider the case where the order of a MIMO system to be identified is a priori known...
详细信息
Sometimes we obtain some prior information about a system to be identified, e.g., the order, model structure etc. In this paper, we consider the case where the order of a MIMO system to be identified is a priori known. Recursive subspace state-space system identification algorithms presented here are based on the gradient type subspace tracking method used in the array signal processing. The algorithms enable us to estimate directly the subspace spanned by the column vectors of the extended observability matrix of the system to be identified without performing the singular value decomposition. Also, a new convergence proof of the gradient type subspace tracking is given in this paper. Under the condition of a step size between 0 and 1, we prove the convergence property of the recursive equation of the gradient type subspace tracking. A numerical example illustrates that our algorithm is more robust with respect to the choice of the initial values than the corresponding PAST one. (C) 2002 Elsevier Science Ltd. All rights reserved.
暂无评论