Defence of dissertation in the field of Automation, Systems and Control Engineering, M.Sc. (Tech.) Igor Buzhinskii

Automated devices continuously infiltrate our lives: while complex transport and power plant control systems are already common, smart home, factories and unmanned driving systems are becoming wide-spread. At the same time, all these systems interact with human beings and hence must be safe and correct. Mathematically justified formal approaches, which are applied in the dissertation, assure these properties more thoroughly than traditional testing and simulation.

The key idea behind the dissertation is to consider the models of the systems, and especially their physical (i.e., process and hardware) parts. Traditionally, even if such models are used in analysis, they are constructed manually, which is a laborious and an error-prone process. The dissertation shows how such models can be obtained automatically from recorded behavior examples of control systems and how such generated models can be used in the overall formal safety analysis.

From a practical point of view, this makes formal methods useful in some areas where they had been inapplicable previously, and simplifies their use otherwise. For example, the dissertation considers safety assessment of nuclear instrumentation and control (I&C) systems, which are responsible for safe operation of nuclear power plants (NPPs). Once the physical part (that is, the processes in the NPP) is taken into account, the analysis becomes simpler in terms of required computational resources, such as time and memory. Other domains where the obtained results can be applied are PLC programs, modular manufacturing systems and energy distribution systems (smart grids).

Opponent: Professor Jean-Jacques Lesage, ENS Paris-Saclay, France

Custos: Professor Valeriy Vyatkin, Aalto University School of Electrical Engineering, Department of Electrical Engineering and Automation.

Thesis webpage

Contact information: Igor Buzhinskii, Department of Electrical Engineering and Automation, [email protected], +358449125387