Head of the research laboratory: Katalin M. Hangos, DSc, professor
Members of the laboratory
· Tibor Dulai, PhD, assistant professor
· Miklós Gerzson, PhD, associate professor
· Attila Magyar, DSc, professor
· Ágnes Werner-Stark, PhD, associate professor
PhD students
· Zsuzsanna Nagy (supervisor: Ágnes Werner-Stark)
· Wijaya Kurniawan (supervisors: Lőrinc Márton and Katalin Hangos)
· Ádám Brand (supervisors: Attila Magyar and Veronika Szücs)
Activity of the research laboratory
The research laboratory carries out research in the interdisciplinary areas of systems and control theory, and artificial intelligence. The research topics can be arranged along the following main directions.
1. Modelling and dynamic analysis of networked nonlinear positive systems, their diagnosis and design of their distributed control systems.
They focus on bio-, chemical- and ecological-inspired positive nonlinear systems that can be represented by kinetic and generalized Lotka-Volterra models. In addition to the static network topologies with loops they investigate networks with dynamic connections having distributed time delays, too. Passivity-based methods are used for designing distributed control systems for such networked systems, while observer-based approaches are developed for diagnosis.
2. Scheduling logistic, industrial and medical processes using artificial intelligence approaches.
The scheduling of large-scale and/or complex processes by exact solution methods is a time-consuming and sometimes unfeasible task. At the same time heuristic solution methods offer acceptable, near-optimal (sometimes even optimal) solutions requiring only a short execution time. Metaheuristic and problem-specific heuristic methods are being developed for scheduling logistic, industrial and medical processes.
3. Monitoring and improving business and industrial processes using process mining and artificial intelligence algorithms.
The following two priority areas are considered. (1) Conformance checking is a main type of process mining, which confronts the modelled behaviors with the observed behaviors of a process to detect, locate and explain the deviations between them. (2) The Capacitated Arc Routing Problem (CARP) is an NP-hard combinatorial optimization problem, which requires to determine the least cost route plans on a graph of a road network for vehicles subject to some constraints. The Dynamic CARP (DCARP), is a variation of the CARP that considers dynamic changes in the problem (e.g., a new task appears).
4. In the field of engineering application of intelligent control, they focus on two starting projects. (1) Dynamic analysis, control and diagnosis of smart lighting systems. Here the fuzzy rule-based techniques are applied that is capable of handling the uncertainties in these systems. (2) Stochastic Petri nets models for dynamic analysis of transportation systems.
Research results
The following research results have been achieved recently.
a) A method has been developed for physics-inspired modelling of networked systems with distributed time delays. It has been shown that complex balanced reaction networks remain stable with arbitrary distributed time delays [1].
b) A complex optimization-based method has been developed for designing stabilizing controllers for polynomial systems with time-delays using kinetic realizations [2].
c) A passivity-based design method has been proposed for distributed control systems of kinetic systems with time delays [3].
d) The design methods of artificial intelligence-based scheduling have been successfully applied for logistic, production and patient appointment planning processes [4], [5].
e) A multi-perspective online conformance checking technique has been developed, which aims to confront the modelled behaviors in form of a Data Petri net process model with a stream of events as the observed behaviors. Two existing techniques have been applied for calculating conformance checking: a prefix-alignment-based technique, which is able to compute conformance statistics for incomplete process executions by applying incremental heuristic search, and an alignment-based multi-perspective conformance checking technique, which is able to compute conformance statistics for complete process instances while focusing on multiple perspectives [6].
f) An Artificial Bee Colony Algorithm (ABC) algorithm has been developed for the CARP, and a new move operator, together with the sub-route plan operator have been proposed and utilized in the algorithm. The algorithm has been tested both as a CARP and a DCARP solver, too [7].
Selected publications
1. Lipták, Gy.; Pituk, M ; Hangos, K M: Modelling and stability analysis of complex balanced kinetic systems with distributed time delays, JOURNAL OF PROCESS CONTROL 84 pp. 13-23. , 11 p. (2019)
2. Lipták, Gy. ; Hangos, K M ; Szederkényi, G: Stabilizing feedback design for time delayed polynomial systems using kinetic realizations, EUROPEAN JOURNAL OF CONTROL 57 pp. 153-171. , 9 p. (2021)
3. Márton, L., Szederkényi G, Hangos KM: Modelling and control of networked kinetic systems with delayed interconnections, JOUNAL OF PROCESS CONTROL, 130 paper: 103084 (2023)
4. Auer, P; Dósa, Gy; Dulai, T; Fügenschuh, A; Näser, P; Ortner, R; Werner-Stark, Á: A new heuristic and an exact approach for a production planning problem. Central European Journal of Operations Research, pp. 1-35. (2020)
5. Süle, Z; Baumgartner, J; Leitold, D; Dulai, T; Orosz, Á; Vathy-Fogarassy, Á: Software Framework and Graph-based Methodology for Optimal Patient Appointment Planning. Procedia Computer Science, pp. 1169-1176. (2023)
6. Nagy, Zs.; Werner-Stark, Á: An Alignment-based Multi-Perspective Online Conformance Checking Technique, ACTA POLYTECHNICA HUNGARICA 19 : 4 pp. 105-127, 23 p. (2022)
7. Nagy, Zs.; Werner-Stark, Á.; Dulai, T.: An Artificial Bee Colony Algorithm for Static and Dynamic Capacitated Arc Routing Problems, MATHEMATICS 10 : 13 Paper: 2205, 38 p. (2022)
About the Head of laboratory
 |
Katalin M. Hangos has an MSc in chemistry (ELTE TTK, 197 6) and computer science (ELTE TTK, 1980). She has a DSc (1993), and habilitations (in chemical engineering in 1994, and in information technologies in 2000). She is a professor at the Department of Electrical Engineering and Information Systems of the University of Pannonia and a research professor at the HUN-REN Research Institute for Computer Science and Control, in its Systems and Control Laboratory. She is one of the few female professors of process control with a strong background in systems and control theory and computer science. Her main area of interest is dynamic modelling of process systems for control and diagnostic purposes. She is a (co-)author of more than 100 journal papers, 4 books and 200 conference papers on various aspects of modelling, diagnosis and control of process systems including nonlinear, stochastic, Petri net, qualitative and graph-based models. She is a member of the IFAC Technical Committee on Chemical Process Control, the Scientific Committee of the Hungarian Academy of Sciences on Information Technologies, and the Doctoral Council of the Hungarian Academy of Sciences. |