Miklos Gerzson, CSc, Associate Professor
Detection and identification of errors in technological system using process mining technics. The Petri net model referring to recent work of the system can be developed based on the measured values and event sequences originating from the observation of the technological systems using process mining technics. The aim of our research is the study of artificial intelligence based methods for the automatic generation of technological models. A further task is the comparison of models describing the planned and the real operation in the technological system. This can be done using graph structure comparison algorithms. The introduction of the arc weights during the generation of reachability graphs enables the calculation of the probability of the occurrence of given states on the reachability graph.
Diagnostic Methods Based on Reliability Analysis (FMEA and HAZOP)
The reliability analysis methods which are widely used in the industrial practice yields the results in table form in which the reasons and consequences of faults and deviations in the measured signals are summarized. The aim of the research is the development of such methods with which these tables can be generated automatically, for real time diagnostic purposes.
Students can participate in the research in the frame of individually instructed subjects such as Project Laboratory, Engineering Design or Thesis Work.