This project deals with the problem of fault tolerant control (FTC) for the management of interconnected, nonlinear systems affected by multiple sensor faults. Sensor faults are of paramount importance due to the large number of sensors used for a) monitoring and control of large-scale systems (e.g. transportation systems, energy and power systems), and b) providing rich and redundant information for executing safety-critical tasks (e.g. aerospace systems, petrochemical processes). The number of sensors is expected to increase in the way of creating smart cities, a societal challenge of European Union. To this end, Europe 2020 strategy has set as key priority the enrichment of digital society that will provide `intelligence’ to a conventional city, using information and communication technologies integrated with sensors and sensor networks. This project proposes a methodology that can contribute in the reliability of these means, the safe system functioning and the protection of everyday life. The novelty of the proposed FTC method lies in its capacity to handle multiple sensor faults, and furthermore to compensate their effects on interconnected, nonlinear systems. This is realized by deploying a large number of agents in a non-centralized architecture, important for FTC in large-scale systems. The FTC is conducted based on the decision of a multiple fault diagnosis (MSFD) mechanism designed to detect and isolate multiple sensor faults in interconnected, multisensory controlled systems. Given that multiple sensor FTC can be significantly affected by the substandard performance of the MSFD mechanism and the network imperfections of the agent’s communication, derived methods will be supported by MSFD guarantees and communication protocol will be designed to manage communication problems. The strong experience of the fellow in MSFD of interconnected, nonlinear systems and supervisor’s expertise in multi-sensor FTC can ensure the success of this project.
Fields of science
Call for proposal
See other projects for this call