Rimantas PupeikisReinforced methods for dynamical system identification and adaptive control | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
ISBN: | 978-3-8440-5218-3 | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Reihe: | Informatik | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Schlagwörter: | system; identification; adaptive control; methods; algorithm; observations | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Publikationsart: | Fachbuch | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Sprache: | Englisch | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Seiten: | 190 Seiten | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Abbildungen: | 83 Abbildungen | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Gewicht: | 255 g | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Format: | 21 x 14,8 cm | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Bindung: | Paperback | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Preis: | 35,80 € | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Erscheinungsdatum: | Mai 2017 | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Kaufen: | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Download: | Verfügbare Online-Dokumente zu diesem Titel: Sie benötigen den Adobe Reader, um diese Dateien ansehen zu können. Hier erhalten Sie eine kleine Hilfe und Informationen, zum Download der PDF-Dateien. Bitte beachten Sie, dass die Online-Dokumente nicht ausdruckbar und nicht editierbar sind.
Benutzereinstellungen für registrierte Online-Kunden Sie können hier Ihre Adressdaten ändern sowie bereits georderte Dokumente erneut aufrufen.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Weiterempfehlung: | Sie möchten diesen Titel weiterempfehlen? | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Rezensionsexemplar: | Hier können Sie ein Rezensionsexemplar bestellen. | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Verlinken: | Sie möchten diese Seite verlinken? Hier klicken. | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Export Zitat: |
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Zusammenfassung: | In designing adaptive control systems, one ought to determine the type of uncertainties appearing in the plant to be controlled. Some important uncertainties can occur at the output of the control system, inducing negative effects that might sometimes lead to unstable control. One of the main ones is the uncertainty arising in the output disturbance description of a plant model. Nonnormal noise, and, particularly, the presence of outliers due to occasional failures of a signal acquisition device degrade the performance and control of the stochastic dynamical system. None the less important is the uncertainty that arises due to nonlinearities of measurement devices and/or control actuators. Nonlinearities, such as friction, deadzone, saturation, preload, backlash, and hysteresis, are called hard-nonlinearities, and are common in most control systems, especially in electro-mechanical ones. Usually, such nonlinearities can occur at the output of the system, inducing negative effects. They significantly limit the performance of control systems. The influence of both uncertainties in respect of the proximity of stochastic control system output to the reference or set-up signal is analysed here theoretically and using numerical simulation by PC. The text begins with an introduction to the author's creation, where the reasons for its development are shown and explained briefly. Here the basic approaches to the problems analyzed in the book are outlined and expressed in short. Various but, of course, not all the references to well-known works are presented. Chapter 2 presents two approaches to determination of the linear time-invariant system (LTI) model order in the absence as well as in the presence of outliers in unknown additive noise. Chapter 3 considers the ordinary least-squares (LS) algorithm for parametric identification of an open-loop LTI system. As an alternative to the presence of outliers in observations, nonlinear equations are written for generating of M-estimates. A reinforced method, based on the sample medians and on the LS technique, is proposed here, too. Chapter 4 introduces various recursive parameter estimation techniques, based on the ordinary LS method, that are worked out for different additive correlated noise transfer function structures. In the case of outliers in noise, the recursive M-algorithms are presented and analysed by numerical simulation. In Chapter 5, we present a direct method for parametric estimation of stationary, slowly time-varying and suddenly jumping parameters of LTI system, acting in the closed loop. For a stationary case, we propose the approach how to check the efficiency of recursive LS (RLS) in the space of parameters. Chapter 6 considers a two-stage approach to estimate slowly time-varying parameters of the LTI system, acting in the closed loop. Here novel structures, assuring optimal identification conditions of the additive noise filter as well as that of the process noise, are worked out. In Chapter 7, the method based on data rearrangement is proposed for parametric identification, when the LTI system is followed by the hard nonlinearity of the known structure. Chapter 8 includes the self-tuning minimum variance control approach to the LTI system, followed by the saturation nonlinearity. In Chapter 9 the problem of joint time-varying parameters and time delay tracking with the estimation, using input-output observations, is analysed. The approach based on corrective operators and used to transform the multiextremal criterion into the unimodal function, in respect of a time delay, is developed. In Appendix missing data restoration algorithm is presented. It can be used to reconstruct signal samples that are deleted while censoring the outlying data. | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
» Weitere Titel von Rimantas Pupeikis. |