In this paper, we derive a systemidentification framework for continuous-time nonlinear systems, for the first time using a simulation-focused computational Bayesian approach. Simulation approaches to nonlinear syste...
详细信息
In this paper, we derive a systemidentification framework for continuous-time nonlinear systems, for the first time using a simulation-focused computational Bayesian approach. Simulation approaches to nonlinear systemidentification have been shown to outperform regression methods under certain conditions, such as non-persistently exciting inputs and fast-sampling. We use the approximate Bayesian computation (ABC) algorithm to perform simulation-based inference of model parameters. The framework has the following main advantages: (1) parameter distributions are intrinsically generated, giving the user a clear description of uncertainty, (2) the simulation approach avoids the difficult problem of estimating signal derivatives as is common with other continuous-time methods, and (3) as noted above, the simulation approach improves identification under conditions of non-persistently exciting inputs and fast-sampling. Term selection is performed by judging parameter significance using parameter distributions that are intrinsically generated as part of the ABC procedure. The results from a numerical example demonstrate that the method performs well in noisy scenarios, especially in comparison to competing techniques that rely on signal derivative estimation.
A new methodology for model structure identification is presented. This is implemented as a new toolbox of MATLAB. The methodology developed has two parts, the first one has the objective to find a pure delay using th...
详细信息
ISBN:
(纸本)085296708X
A new methodology for model structure identification is presented. This is implemented as a new toolbox of MATLAB. The methodology developed has two parts, the first one has the objective to find a pure delay using the estimated impulse response, and the second one, uses at same time the minimal value of the identification criteria and the representative value of numerator coefficients. The graphical interface and the results obtained using some real applications are also presented.
identification results for the shaft-speed dynamics of a twin-spool aircraft gas turbine engine are presented. From responses to a range of input perturbation signals are found time-invariant small-signal, time-varyin...
详细信息
ISBN:
(纸本)085296708X
identification results for the shaft-speed dynamics of a twin-spool aircraft gas turbine engine are presented. From responses to a range of input perturbation signals are found time-invariant small-signal, time-varying small-signal and time-varying large-signal models. The presence of non-linearity and slow dynamics is inferred from these results.
In this paper the implementation of dynamic data reconciliation techniques for sequential modular models is described. The paper is organised as follows. First, an introduction to dynamic data reconciliation is given....
详细信息
ISBN:
(纸本)085296708X
In this paper the implementation of dynamic data reconciliation techniques for sequential modular models is described. The paper is organised as follows. First, an introduction to dynamic data reconciliation is given. Then, the on-line use of rigorous process models is introduced. The sequential modular approach to dynamic simulation is briefly discussed followed by a short review of the extended Kalman filter. The second section describes how the modules are implemented. A simulation case study and its results are also presented.
Support Vector Machines (SVMs) are used for systemidentification of both linear and non-linear dynamic systems, Discrete time linear models are used to illustrate parameter estimation and non-linear models demonstrat...
详细信息
ISBN:
(纸本)085296708X
Support Vector Machines (SVMs) are used for systemidentification of both linear and non-linear dynamic systems, Discrete time linear models are used to illustrate parameter estimation and non-linear models demonstrate model structure identification. The VC-dimension of a trained SVM indicates the model accuracy without using separate validation data. We conclude that SVMs have potential in the field of dynamic systemidentification, but that there are a number of significant issues to be addressed.
暂无评论