Metode neizrazitog i adaptivnog upravljanja za primjenu u slijednim sustavima

scientific master's theses

• Author: Mario Punčec
• Mentor: Zdenko Kovačić (mentor)
• Committee member: Stjepan Bogdan (predsjednik povjerenstva)
• Committee member: Zdenko Kovačić (član povjerenstva)
• Committee member: Mladen Crneković (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; (Department of Control and Computer Engineering); Zagreb
• Defense date and country: 2005-05-25, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Electrical Engineering; Automation and Robotics
• Universal decimal classification (UDC): 621.3 - Electrical engineering

Abstract

Neizrazito upravljanje se, zbog svoje povezanosti s ljudskim načinom zaključivanja i reagiranja, sve češće koristi kao napredniji i ljudima prirodniji način upravljanja. Ono jednostavno formira operaterovo iskustveno poznavanje sustava upravljanja u pravila upravljanja. Kod takvog načina upravljanja se pokazalo kao problem što takvih pravila može biti puno, jer broj pravila raste geometrijski s brojem promatranih varijabli, što znači i veliku potrebnu računalnu moć. U ovom radu je predstavljena metoda razdvajanja neizrazitog regulatora u kaskadni neizraziti regulator čime je znatno smanjen broj pravila. Stabilnost sustava upravljanja osigurana je proračunom parametara regulatora prema Lyapunovljevoj teoriji stabilnosti, koja je prilagođena radu s neizrazitim varijablama. Kaskadni neizraziti regulator testiran je najprije simulacijski u Matlab Simulinku, a zatim i na laboratorijskoj maketi: valjčić na letvi.
U ovom radu je također predstavljena metoda adaptivnog upravljanja s referentnim modelom i lead-lag kompenzatorom, koja ne zahtijeva prethodno detaljno poznavanje procesa, već je potrebno znati samo osnovne značajke dinamičkog ponašanja, koje se iskazuju preko referentnog modela drugog reda. Ovu metodu adaptacije moguće je nadograditi na postojeći sustav upravljanja, neovisno o postojećem regulatoru, jer se kompenzator ugrađuje na izlaz iz regulatora. Za provođenje ovog tipa adaptacije potreban je test signal, na temelju kojeg se zaključuje o nastalim promjenama o sustavu, te o akcijama koje treba poduzeti da bi se poništilo djelovanje nastale promjene na dinamičko ponašanje sustava. Ovaj algoritam adaptacije testiran je na modelu sinkronog motora s permanentnim magnetima.
Predstavljeni algoritam adaptacije, te klasični neizraziti regulator implementirani su kao mogućnost odabira u industrijski orijentirani program za off-line simulaciju robota - Leonardo.

Abstract (english)

The fuzzy control, with its tight relation to human decision making and reacting, is ever more often used in control applications because it is an advanced and for humans a more natural control method. It simply transforms human operator’s experience into control rules. A typical problem in fuzzy control arises with the number of control rules, because the number of control rules increases geometrically with the number of observed variables, thus increasing the required computing power. This thesis presents a method of separation of the fuzzy controller into cascade fuzzy controllers, significantly reducing the number of control rules. The control system stability is ensured by determining controller parameters according to Lyapunov stability theory, which has been adjusted to work with fuzzy variables. The cascade fuzzy controller was simulated in Matlab Simulink and then tested on a “ball and beam“.
This thesis also presents an adaptive control method based on reference model and lead-lag compensator. The presented method does not require detailed previous knowledge about the controlled process, but only the basic information about its dynamic behaviour expressed in the form of a second-order reference model. An existing control system can be upgraded with this adaptation method regardless of the existing controller, because the compensator is added to the controller output. The implementation of this adaptation type requires a test signal which is used for determination of system changes and actions required for reduction of the effect of these changes on the system dynamic behavior. This adaptation algorithm has been tested on a permanent magnet synchronous motor model.
cascade fuzzy control, adaptive control, intelligent controlThe presented adaptation algorithm, along with a classical fuzzy controller, was implemented as a feature in the software tool for off-line simulation of industrial robots – Leonardo.

• Keywords: Kaskadno neizrazito upravljanje; adaptivno upravljanje; inteligentno upravljanje
• Keywords (english): Cascade fuzzy control; adaptive control; intelligent control
• Language: croatian
• URN:NBN: urn:nbn:hr:168:595372
• Study programme: Title: Postgraduate master programme in electrical engineering; Study programme type: university; Study level: postgraduate; Academic / professional title: Magistar znanosti elektrotehnike (Magistar znanosti elektrotehnike)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2019-07-02 09:14:25