Naslov
A new concept of a fuzzy ontology controller for a temperature regulation

Autori
Benić, Juraj ; Krznar, Matija ; Stipančić, Tomislav ; Šitum, Željko

Izvornik
Iranian Journal of Fuzzy Systems (1735-0654) 19 (2022), 4; 125-136

Vrsta, podvrsta i kategorija rada
Radovi u časopisima, članak, znanstveni

Ključne riječi
ontology ; linked open data ; fuzzy ontology control ; fuzzy logic control ; SPARQL ; control theory ; temperature regulation

Sažetak
In this paper, a new concept of the fuzzy logic controller is presented. The proposed concept is called fuzzy ontology controller and it's based on an ontology description of the Fuzzy Logic Controller (FLC). In this concept, the SPARQL query sent to the ontology replaced the fuzzification process, fuzzy interface, and rules of the fuzzy controller while defuzzification is done on an obtained response. Fuzzy ontology is designed with Prot'{;e};g'{;e}; and hosted in the Virtuoso database on a remote server. Simulation results obtained with the proposed fuzzy ontology controller are compared to ones obtained with the classical fuzzy logic controller. Both controllers obtained the same simulation results due to the same membership function and defuzzification process. The results showed that this concept can be used as a part of some IoT solution within the vision of Industry 4.0. Also, results confirmed that the ontology-based controller works properly and gives the same output as FLC. The experiment is carried out on a heat exchange process controlled with a PLC. Modbus TCP protocol is used for communication between the server PC and a PLC. Experimental results show that only slow processes are controllable for now due to slow communication between PLC and server and the time needed for execution of SPARQL queries.

Izvorni jezik
Engleski

Znanstvena područja
Matematika, Računarstvo, Strojarstvo, Informacijske i komunikacijske znanosti

POVEZANOST RADA