Complete Design of Fault Tolerant Control Systems

From the analysis of the book proposal form, the authors have reported that no supplementary material will be provided. However, as the book will regard mainly the application of fault tolerant control techniques to real processes, it would be interesting to have back-of-book CD-Rom or ftp/www links that provide the real examples and data shown in the monograph. The reviewer agrees with the authors regarding the categories that describe the book. It is designed for courses at post-graduate levels, as it is a monograph reporting a detailed treatment on a specific subject. However, it could be seen also as a manual, i.e. used by practitioners. Regarding book’s contents, the authors should try to provide a monograph that is a valuable resource for researchers from both academia and industry. It should be used also in graduate-level courses. Even if the book could try to cover quite different diagnosis and fault-tolerant control methods, the authors should achieve a coherent presentation of them. The reviewer hopes that most readers will find the presentation self-contained. Obviously, the material should be readable and clearly presented. The authors should avoid that points of a section are not very obvious (such as some sections of “Diagnosis and Fault-Tolerant Control” by Mogens Blanke, Michel Kinnaert, Jan Lunze, and Marcel Staroswiecki, see Chapter 4) or insufficiently elaborated (such as Chapter 7 of “Diagnosis and Fault-Tolerant Control”). However, the reader could pursue further topics of interest with the help of the detailed and updated bibliographical notes that the authors aim to provide at the end of the book (Section 7). According to the book’s provisional table of contents, the book seems to be organised in 6 main chapters and the 7th one is the bibliography. Chapter 1 should be introductory, presenting an overview of the main ideas of the book, examples, and the various types of models used in the book (the same as Sections 1 and 2 of Isermann’s textbook). Chapter 2 should try to present models and methods applicable at a higher-level of abstraction, in which the analytical details of the plant model are absent. This chapter should address briefly also fault modelling, the diagnosis and fault-tolerant control schemes. As claimed by the authors, the major contribution of the monograph seems to concern Chapters 3, 4 and 5, where the approaches presented in the book will be illustrated on several real application examples, namely the three-tank system (Chapter 3), the winding machine (Chapter 4) and the active suspension system (Chapter 5). The major approaches of the book will be illustrated on these examples in these chapters. For each application example, in Section 3, 4 and 5, the monitored system should be described in detail, as shown in the journal paper “Fault-Toleran Control in Dynamic Systems: Application to a Winding machine” by Noura, Sauter, Hamelin and Theillion. For each case, the fault diagnosis and the fault tolerant control should be designed and presented. The fault cases, their descriptions and examples should complete each application chapter. From the provisional Table of Contents, Appendices do not seem to be present, even if they could be added for providing useful background e.g., in linear algebra, system theory basics, and controller design, if necessary. As for Isermann’s book, the descriptions of the terminology of the book and a fault technical dictionary could be included e.g. in Chapter 1 or 2. In particular, Chapters 3, 4 and 5 seem to represent the main contribution of this monograph. In fact, the works regarding FDI and FTC, which will be recalled in Chapter 2, will be validated on real application and not only by simulation, as suggested by the book’s authors. A complete architecture of a fault-tolerant control system should be briefly illustrated in Chapter 2. Therefore, these FTC strategies should be described in more detail in Chapters 3, 4 and 5 by taking into account the system modelling, the nominal control, the fault diagnosis and the fault compensation or the modification of the control law, according to the results issued from the fault diagnosis module, also recalled in Chapter 2. It is worth noting that the presentation of real industrial problems in these fields can be difficult due to confidential aspects. According to the book proposal form and from Prof. Noura’s CV, for many years the authors have been conducted the analysis of this kind of FDT methods through their application to realistic systems, corresponding to subsystems in large-scale industrial processes. The authors suggested to present the theoretical methods in a pedagogical manner. These methods will be applied to the systems cited above. The results should be clearly illustrated and commented using plots and tables. The generic way to apply these methods to many systems will be shown, when the system model is available, depending on its accuracy and detail level. Finally, according to the book proposal form, the authors have stated that this book may be oriented to both academic and industrial communities. In particular, for industrial practising engineers (R & D industrial area), the various applications involved in the complete design of FTC and FDI systems could provide a useful tool for an immediate transfer including limits of applicability. For the University and academic communities, the monograph could be useful for post-graduate, Ph.D. students and researchers. In fact, the design examples could be used in courses regarding Automatic Control, Electrical or Mechanical Engineering.