FUTURISM · Multiple sensor fault tolerant control for management of interconnected, nonlinear systems

Multiple sensor fault tolerant control for management of interconnected, nonlinear systems

The project `FUTuRISM' started on the 1st of May 2014 and ended on the 30th of April 2016. The main goals of the project were on one hand the research on the problem of multiple sensor fault tolerant control for the management of interconnected nonlinear systems, and on the other hand the training of the research fellow, Dr. V. Reppa towards launching an independent, successful career in the field of fault diagnosis (FD) and fault tolerant control (FTC). The `FUTuRISM' project and its actions were supervised by Prof. S. Olaru. The research objectives of the `FUTuRISM' project were addressed assuming the availability of a model describing the system(s), the boundedness of the modeling uncertainties, and the possible occurrence of multiple sensor faults. A summary of the research results can be classified as follows: Research Objective 1- Multiple sensor fault diagnosis (FD) with detection and isolation guarantees for nonlinear systems: To address this project objective, two research practices were followed. The first one was the design of a systematic set-theoretic methodology for multiple sensor FD of linear and nonlinear systems. The second one was the analysis of this methodology in order to provide guarantees for its performance with respect to robustness, sensor fault detectability and isolability. Both the design and analysis were realized by assuming the closed- loop operation of the system that ensures the boundedness of the system states and inputs. Research Objective 2- Multiple sensor fault tolerant control (FTC) with fault tolerance guarantees for nonlinear systems: Similarly to the previous objective, the research to meet this objective was pursued along two lines, the design and analysis. Specifically, the first line was the design of active FTC schemes exploiting the results of a multiple sensor FD mechanism. The second line involved the performance analysis of the developed active FTC schemes with respect to the control design characteristics and objectives. Research Objective 3- Multiple sensor FTC of interconnected nonlinear systems: The third research objective was the design of non-centralized sensor FTC architectures for inter-connected nonlinear systems. The research was focused mainly on the design of distributed and decentralized FD architectures for detecting and isolating of multiple sensor faults in inter- connected systems. Then, non-centralized FTC schemes were developed for compensating the effects of multiple sensor faults affecting a network of interconnected systems. Another line of research was the management of the developed FTC schemes under communication imperfections. The design of the non-centralized FD and FTC architectures was realized in conjunction with the analysis of their performance with respect to the diagnosability, stability and tracking control objectives. Research Objective 4- Application of multiple sensor FTC to complex systems: The last research objective of this project was the application of the developed sensor FD and FTC algorithms to complex systems such as intelligent transportation systems and intelligent buildings. At the same time, the project `FUTuRISM' aimed at training the research fellow towards launching an independent research career. The training objectives were: Training 1- Empowering general research skills in order to become a distinguished researcher in the field of FD and FTC: The IEF research fellow was trained to identify, formulate and solve some problems representing bottlenecks in the existing theories, and conduct innovative work by working methodologically and being open-minded. Training 2- Supervising a high quality doctoral dissertation: The objective was to enhance the skills of the research fellow through the active participation in the supervision of doctoral students. Supervision skills along with the research skills can significantly contribute to a successful academic career. Training 3- Publishing novel research results in high quality journals and international conferences: The objective was to build a very good track record of publication, including citations and number of publications, for launching an independent research career. To this end, the goal was to perform high-quality research work not only with the scientist-in-charge, but also with international experts in the field of FD and FTC. Training 4- Interacting with European and international experts: The objective was that the fellow will collaborate with world-known researchers in the field of FD and FTC, in order to benefit from their manifold way of thinking and working, and to establish new fruitful collaborations and networks, which can be further exploited participating in research projects or acquiring a research position in the future. Training 5- Transferring the scientific knowledge to industry: The goal was to enable the interaction of the fellow with the industry in order to understand the practical constraints with respect to computational complexity, cost and implementation time, as well as to get feedback for the research assumptions. Contact Details: Scientist in Charge: Sorin OLARU Professor, CentraleSupelec Laboratory of Signals and Systems 3 rue Joliot Curie, 91192 cedex Gif sur Yvette, France Email: sorin.olaru@centralesupelec.fr Fellow: Dr. Vasiliki Reppa Post-doctoral research associate, CentraleSupelec Laboratory of Signals and Systems 3 rue Joliot Curie, 91192 cedex Gif sur Yvette, France Email : Vasiliki.Reppa@centralesupelec.fr Website: http://www.l2s.centralesupelec.fr/en/perso/vasiliki.reppa

Data: CORDIS, © European Union