A method is proposed for estimating both the weights and the state of a multiple model system with one common state vector. In this system, the weights are related to the activation of each individual model. For the r...
详细信息
A method is proposed for estimating both the weights and the state of a multiple model system with one common state vector. In this system, the weights are related to the activation of each individual model. For the resulting nonlinear estimation problem a method is developed that efficiently decomposes the total problem into two separate parts, one for estimating the model weights and one for estimating the state. The method has been validated on a component, actuator and sensor fault detection and identification problem for a linearized model of an aircraft.
In this paper we discuss algorithms for the computation of the ℒ ∞ -norm of transfer functions related to descriptor systems, both in the continuous- and discrete-time context. We show how one can achieve this goal b...
详细信息
In this paper we discuss algorithms for the computation of the ℒ ∞ -norm of transfer functions related to descriptor systems, both in the continuous- and discrete-time context. We show how one can achieve this goal by computing the eigenvalues of certain structured matrix pencils. These pencils can be transformed to skew-Hamiltonian/Hamiltonian matrix pencils which are constructed by only using the original data. Furthermore, we apply a structure-preserving algorithm to compute the desired eigenvalues. In this way we increase robustness and efficiency of the method. Finally, we present numerical results in order to illustrate the advantages of our approach.
Iterative feedback tuning (IFT) is a model-free tuning method that has been proven to work well in various applications since its introduction in 1994. Several factors affect the performance of the optimization proces...
详细信息
Iterative feedback tuning (IFT) is a model-free tuning method that has been proven to work well in various applications since its introduction in 1994. Several factors affect the performance of the optimization process, and one of which is the design criterion. Historically, the weighting factor for each element in the design criterion is chosen by trial and error and results in a different value for each system tested. In this paper, a normalized design criterion is presented with a weighting factor that allows the tuning performance to be assessed across different systems. This new design criterion is then applied to various test systems using the Monte Carlo method to determine the optimal range of values of this normalized weighting factor in tuning for step input responses.
The paper deals with a problem of numerical conditioning of basic J – lossless factori-sations associated with suboptimal H ∞ – norm estimation of discrete-time processes described in the so-called delta-domain. St...
详细信息
The paper deals with a problem of numerical conditioning of basic J – lossless factori-sations associated with suboptimal H ∞ – norm estimation of discrete-time processes described in the so-called delta-domain. State space formulae for dual J – lossless factorisations of a chain scattering representation of the estimated process are given. Solutions are obtained via solving two coupled algebraic Riccati equations. A relative condition number of the delta-domain algebraic Riccati equation is employed as a measure of numerical conditioning of these solutions. A numerical example is given to show that solutions obtained for the delta operator are much better-conditioned than its counterpart versions based on the common forward shift operator.
In this paper, a robust Stackelberg game for a class of uncertain stochastic systems with state delay is investigated. After introducing some definitions and preliminaries, we derive the conditions for the existence o...
详细信息
In this paper, a robust Stackelberg game for a class of uncertain stochastic systems with state delay is investigated. After introducing some definitions and preliminaries, we derive the conditions for the existence of the robust static output feedback (SOF) Stackelberg strategy set such that the upper bounds of leader’s cost function and the weighted cost function of the followers are minimized respectively. In order to obtain the robust SOF Stackelberg strategy set, a heuristic algorithm is proposed based on the stochastic Lyapunov type matrix equations (SLMEs) and the linear matrix inequalities (LMIs). In particular, it is shown that robust convergence is guaranteed by applying the Krasnoselskii-Mann (KM) iterative algorithm. An academic numerical example is presented to demonstrate the effectiveness of the proposed method.
Mixed structured singular value analysis tests that allow the structure of the multipliers to a priori be specified have been developed. These tests can be characterized in terms of the existence of positive definite ...
详细信息
Mixed structured singular value analysis tests that allow the structure of the multipliers to a priori be specified have been developed. These tests can be characterized in terms of the existence of positive definite solutions to certain Riccati equations. Probability-one homotopy algorithms are developed to find a solution. These results easily extend to provide computationally tractable algorithms for fixed-architecture, robust control design.
The knowledge of microstructure is essential to compute mechanical properties of hot rolled steel. Depending on process data and the chemical composition of the steel, a physical model is used to calculate microstruct...
详细信息
The knowledge of microstructure is essential to compute mechanical properties of hot rolled steel. Depending on process data and the chemical composition of the steel, a physical model is used to calculate microstructure parameters such as grain size, recrystallization and the γ-α-transformation. In addition to the physical model an artificial neural network is employed to calculate mechanical properties. New results from hot rolling mills are presented. The on-line application of the monitor system allows a quick release of strips after rolling and the saving of costs for quality control measurements. Process parameters could be adjusted to achieve desired mechanical properties of the product. Process data can either be obtained on-line from the process itself or can be simulated off-line, e.g. by Siemens’ HYBRid EXpert System HYBREX. Its central component is a high-performance flowsheet simulator, which enables complex plants to be modeled, simulated and evaluated. The architecture and capability of HYBREX will be presented as well as the newly achieved possibilities by integrating the microstructure model to the simulation tool.
A Newton algorithm is discussed. This algorithm allows for studying nonlinear material behavior, such as plasticity and viscoplasticity. The material behavior is described by so-called internal variables. The model ap...
详细信息
A Newton algorithm is discussed. This algorithm allows for studying nonlinear material behavior, such as plasticity and viscoplasticity. The material behavior is described by so-called internal variables. The model applied in this paper was developed by Lemaitre and Chaboche. The Newton algorithm is implemented as a user-subroutine in the commercial finite-element method package ABAQUS. The results of a numerical analysis of the rotation of an aircraft turbine disk are presented. Cyclic loading conditions are studied, and the results are compared with numerical results in the literature.
Lyapunov equations are key mathematical objects in systems theory, analysis and design of control systems, and in many applications, including balanced realization algorithms, procedures for reduced order models, Newt...
详细信息
Lyapunov equations are key mathematical objects in systems theory, analysis and design of control systems, and in many applications, including balanced realization algorithms, procedures for reduced order models, Newton methods for algebraic Riccati equations, or stabilization algorithms. A new iterative accuracy-enhancing solver for both standard and generalized continuous- and discrete-time Lyapunov equations is proposed and investigated in this paper. The underlying algorithm and some technical details are summarized. At each iteration, the computed solution of a reduced Lyapunov equation serves as a correction term to refine the current solution of the initial equation. The best available algorithms for solving Lyapunov equations with dense matrices, employing the real Schur(-triangular) form of the coefficient matrices, are used. The reduction to Schur(-triangular) form has to be done only once, before starting the iterative process. The algorithm converges in very few iterations. The results obtained by solving series of numerically difficult examples derived from the SLICOT benchmark collections for Lyapunov equations are compared to the solutions returned by the MATLAB and SLICOT solvers. The new solver can be more accurate than these state-of-the-art solvers and requires little additional computational effort.
暂无评论