This is the first book dedicated to direct continuous-time model identification for 15 years. It cuts down on time spent hunting through journals by providing an overview of much recent research in an increasingly busy field. The CONTSID toolbox discussed in the final chapter gives an overview of developments and practical examples in which MATLAB® can be used for direct time-domain identification of continuous-time systems. This is a valuable reference for a broad audience.
In view of the importance of system identification, the International Federation of Automatic Control (IFAC) and the International Federation of Operational Research Societies (IFORS) hold symposia on this topic every three years. Interest in continuous time approaches to system identification has been growing in recent years. This is evident from the fact that the of invited sessions on continuous time systems has increased from one in the 8th number Symposium that was held in Beijing in 1988 to three in the 9th Symposium in Budapest in 1991. It was during the 8th Symposium in August 1988 that the idea of bringing together important results on the topic of Identification of continuous time systems was conceived. Several distinguished colleagues, who were with us in Beijing at that time, encouraged us by promising on the spot to contribute to a comprehensive volume of collective work. Subsequently, we contacted colleagues all over the world, known for their work in this area, with a formal request to contribute to the proposed volume. The response was prompt and overwhelmingly encouraging. We sincerely thank all the authors for their valuable contributions covering various aspects of identification of continuous time systems.
Among the many techniques for designing linear multivariable analogue controllers, the two most popular optimal ones are H2 and H-infinity optimization. The fact that most new industrial controllers are digital provides strong motivation for adapting or extending these techniques to digital control systems. This book, now available as a corrected reprint, attempts to do so. Part I presents two indirect methods of sampled-data controller design: These approaches include approximations to a real problem, which involves an analogue plant, continuous-time performance specifications, and a sampled-data controller. Part II proposes a direct attack in the continuous-time domain, where sampled-data systems are time-varying. The findings are presented in forms that can readily be programmed in, e.g., MATLAB.
Precise dynamic models of processes are required for many applications, ranging from control engineering to the natural sciences and economics. Frequently, such precise models cannot be derived using theoretical considerations alone. Therefore, they must be determined experimentally. This book treats the determination of dynamic models based on measurements taken at the process, which is known as system identification or process identification. Both offline and online methods are presented, i.e. methods that post-process the measured data as well as methods that provide models during the measurement. The book is theory-oriented and application-oriented and most methods covered have been used successfully in practical applications for many different processes. Illustrative examples in this book with real measured data range from hydraulic and electric actuators up to combustion engines. Real experimental data is also provided on the Springer webpage, allowing readers to gather their first experience with the methods presented in this book. Among others, the book covers the following subjects: determination of the non-parametric frequency response, (fast) Fourier transform, correlation analysis, parameter estimation with a focus on the method of Least Squares and modifications, identification of time-variant processes, identification in closed-loop, identification of continuous time processes, and subspace methods. Some methods for nonlinear system identification are also considered, such as the Extended Kalman filter and neural networks. The different methods are compared by using a real three-mass oscillator process, a model of a drive train. For many identification methods, hints for the practical implementation and application are provided. The book is intended to meet the needs of students and practicing engineers working in research and development, design and manufacturing.
Sampled-data Models for Linear and Nonlinear Systems provides a fresh new look at a subject with which many researchers may think themselves familiar. Rather than emphasising the differences between sampled-data and continuous-time systems, the authors proceed from the premise that, with modern sampling rates being as high as they are, it is becoming more appropriate to emphasise connections and similarities. The text is driven by three motives: · the ubiquity of computers in modern control and signal-processing equipment means that sampling of systems that really evolve continuously is unavoidable; · although superficially straightforward, sampling can easily produce erroneous results when not treated properly; and · the need for a thorough understanding of many aspects of sampling among researchers and engineers dealing with applications to which they are central. The authors tackle many misconceptions which, although appearing reasonable at first sight, are in fact either partially or completely erroneous. They also deal with linear and nonlinear, deterministic and stochastic cases. The impact of the ideas presented on several standard problems in signals and systems is illustrated using a number of applications. Academic researchers and graduate students in systems, control and signal processing will find the ideas presented in Sampled-data Models for Linear and Nonlinear Systems to be a useful manual for dealing with sampled-data systems, clearing away mistaken ideas and bringing the subject thoroughly up to date. Researchers in statistics and economics will also derive benefit from the reworking of ideas relating a model derived from data sampling to an original continuous system.
This review volume reports the state-of-the-art in Linear Parameter Varying (LPV) system identification. It focuses on the most recent LPV identification methods for both discrete-time and continuous-time models--
This unique book provides an overview of continuous time modeling in the behavioral and related sciences. It argues that the use of discrete time models for processes that are in fact evolving in continuous time produces problems that make their application in practice highly questionable. One main issue is the dependence of discrete time parameter estimates on the chosen time interval, which leads to incomparability of results across different observation intervals. Continuous time modeling by means of differential equations offers a powerful approach for studying dynamic phenomena, yet the use of this approach in the behavioral and related sciences such as psychology, sociology, economics and medicine, is still rare. This is unfortunate, because in these fields often only a few discrete time (sampled) observations are available for analysis (e.g., daily, weekly, yearly, etc.). However, as emphasized by Rex Bergstrom, the pioneer of continuous-time modeling in econometrics, neither human beings nor the economy cease to exist in between observations. In 16 chapters, the book addresses a vast range of topics in continuous time modeling, from approaches that closely mimic traditional linear discrete time models to highly nonlinear state space modeling techniques. Each chapter describes the type of research questions and data that the approach is most suitable for, provides detailed statistical explanations of the models, and includes one or more applied examples. To allow readers to implement the various techniques directly, accompanying computer code is made available online. The book is intended as a reference work for students and scientists working with longitudinal data who have a Master's- or early PhD-level knowledge of statistics.
In view of the importance of system identification, the International Federation of Automatic Control (IFAC) and the International Federation of Operational Research Societies (IFORS) hold symposia on this topic every three years. Interest in continuous time approaches to system identification has been growing in recent years. This is evident from the fact that the of invited sessions on continuous time systems has increased from one in the 8th number Symposium that was held in Beijing in 1988 to three in the 9th Symposium in Budapest in 1991. It was during the 8th Symposium in August 1988 that the idea of bringing together important results on the topic of Identification of continuous time systems was conceived. Several distinguished colleagues, who were with us in Beijing at that time, encouraged us by promising on the spot to contribute to a comprehensive volume of collective work. Subsequently, we contacted colleagues all over the world, known for their work in this area, with a formal request to contribute to the proposed volume. The response was prompt and overwhelmingly encouraging. We sincerely thank all the authors for their valuable contributions covering various aspects of identification of continuous time systems.
Bringing together important advances in the field of continuous system identification, this book deals with both parametric and nonparametric methods. It pays special attention to the problem of retaining continuous model parameters in the estimation equations, to which all the existing techniques used in estimating discrete models may be applied. It is aimed at both the academic researcher and the control engineer in industry. The techniques covered range from certain simple numerical or graphical methods applicable to some of the frequently encountered model forms, to attractive recursive algorithms for continuous model identification suitable for real time implementation. These include the recent methods based on orthogonal functions such as those of Walsh and Poisson moment functionals. Some techniques based on stable model adaptation principles are also presented and illustrated.
Models of dynamical systems are required for various purposes in the field of systems and control. The models are handled either in discrete time (DT) or in continuous time (CT). Physical systems give rise to models only in CT because they are based on physical laws which are invariably in CT. In system identification, indirect methods provide DT models which are then converted into CT. Methods of directly identifying CT models are preferred to the indirect methods for various reasons. The direct methods involve a primary stage of signal processing, followed by a secondary stage of parameter estimation. In the primary stage, the measured signals are processed by a general linear dynamic operation—computational or realized through prefilters, to preserve the system parameters in their native CT form—and the literature is rich on this aspect. In this book: Identification of Continuous-Time Systems-Linear and Robust Parameter Estimation, Allamaraju Subrahmanyam and Ganti Prasada Rao consider CT system models that are linear in their unknown parameters and propose robust methods of estimation. This book complements the existing literature on the identification of CT systems by enhancing the secondary stage through linear and robust estimation. In this book, the authors provide an overview of CT system identification, consider Markov-parameter models and time-moment models as simple linear-in-parameters models for CT system identification, bring them into mainstream model parameterization via basis functions, present a methodology to robustify the recursive least squares algorithm for parameter estimation of linear regression models, suggest a simple off-line error quantification scheme to show that it is possible to quantify error even in the absence of informative priors, and indicate some directions for further research. This modest volume is intended to be a useful addition to the literature on identifying CT systems.
