LENNART LJUNG

For practicing engineers who are faced with problems of modeling.Written by a recognized authority in the field of identification and control, this book draws together into a single volume the important aspects of system identification AND physical modelling. Explores techniques used to construct mathematical models of systems based on knowledge from physics, chemistry, biology, etc. (e.g., techniques with so called bond-graphs, as well those which use computer algebra for the modeling work).Explains system identification techniques used to infer knowledge about the behavior of dynamic systems based on observations of the various input and output signals that are available for measurement. Shows how both types of techniques need to be applied in any given practical modeling situation. Considers applications, primarily simulation.Features- brings together all basic techniques for modeling of dynamic systems - with even stress on physical modeling and system identification. - describes how to create physical models and - in cases where these physical models contain unknown parameters - how to estimate these from data from actual experiments of a system. - explains how to form models from basic physical equations. Shows in a specific example techniques that can be applied to a multitude of situations. - shows how to use data to develop nonlinear models in a simple fashion. - considers how to identify systems in a closed loop. Describes the problems that arise when data are collected under output feedback and describes how to deal with such a situation successfully.

This book is a comprehensive and coherent description of the theory, methodology and practice of System Identification-the science of building mathematical models of dynamic systems by observing input/output data. It puts the user in focus, giving the necessary background to understand theoretical foundation and emphasizing the practical aspects of the options and choices that face the user. The Second Edition has been updated to include material on subspace methods, non-linear black box models-such as neural networks-and methods that use frequency domain data.