Relay feedback has attracted considerable research attention for more than a century but there has been no recent summary of the many newly-developed tools and results now available for this important area as a whole, those that have been published tending to focus on one process or controller type only. Relay Feedback is divided into three parts, the first of which is devoted to the analysis of relay feedback systems within a general setting with information on: existence of solutions; existence of limit cycles; local and global stability of limit cycles; limit cycles with more than two switchings per period; plants with time delay; relays with asymmetric hysteresis. The second part, on the improvement of process identification shows the reader how to: modify a standard relay to provide better excitation of a process at a number of important frequencies; devise new algorithms designed to make better use of information from relay feedback tests. The book's third part is a presentation of recent developments in control design providing: a unified framework for the design of internal-model, proportional-integral-derivative or general-single-loop controllers for SISO or MIMO systems with or without time delays; characterisation of time delays and non-minimum phase zeros for closed-loop systems. Relay Feedback presents a comprehensive, up-to-date and detailed treatment of relay feedback theory, the use of relay feedback for process identification and the use of identified models for general control design in a single volume. The work assumes only knowledge of linear system theory on the part of the reader and should therefore be of use to graduate students and practising engineers as well as to researchers.

This unique book is the only recent summary presenting a comprehensive, up-to-date and detailed treatment of relay feedback theory, the use of relay feedback for process identification and the use of identified models for general control design in a single volume.

The presence of considerable time delays in many industrial processes is well recognized and achievable performances of conventional unity feedback control systems are degraded if a process has a relatively large time delay compared to its time constants. In this case, dead time compensation is necessary in order to enhance the performances. The most popular scheme for such compensation is the Smith Predictor, but it is unsuitable for unstable or lightly damped processes because the compensated closed-loop system always contains the process poles themselves. An alternative scheme for delay elimination from the closed-loop is the finite spectrum assignment (FSA) strategy and it can arbitrarily assign the closed-loop spectrum. One may note that the Smith Predictor Control can be found in delay systems control books and many process control books, but the FSA control is rarely included in these books. It is therefore timely and desirable to fill this gap by writing a book which gives a comprehensive treatment of the FSA approach. This is useful and worthwhile since the FSA provides not only an alternative way but also certain advantages over the Smith-Predictor. The book presents the state-of-the-art of the finite spectrum assignment for time-delay systems in frequency domain. It mainly contains those works carried out recently by the authors in this field. Most of them have been published and others are awaiting publication. They are assembled together and reorganized in such a way that the presentation is logical, smooth and systematic."