Gå til hovedinnhold
0
Jump to main content

Hybrid modellering og kontroll av mekatronikk-systemer gjennom PWA-system

Philipp Thomas Pasolli of the Faculty of Engineering and Science has submitted his thesis “Hybrid modeling and control of mechatronic systems using piecewise affine dynamics approach”, and will defend the thesis for the PhD-degree Wednesday 27 May 2020. (Photo: Private)

Mechatronic systems are dynamical systems which can have both continuous and discrete dynamic behavior, meaning the system states can both flow (continuously) and jump (discretely). A mechatronic system with both dynamic behaviors is called a hybrid system.

Philipp Thomas Pasolli

Ph.d.-kandidat og R&D Engineer

Philipp Thomas Pasolli disputerer for ph.d.-graden med avhandlingen “Hybrid modeling and control of mechatronic systems using piecewise affine dynamics approach” onsdag 27. mai 2020.

Han har fulgt doktorgradsprogrammet ved Fakultet for teknologi og realfag, med spesialisering i mekatronikk.

Doktorgradsarbeidet er har mottatt finansiering av EUs Clover-prosjekt, i tillegg til den ordinære norske stipendiatfinansieringen.

Slik oppsummerer Philipp Thomas Pasolli avhandlingen:

Hybrid modeling and control of mechatronic systems using a piecewise affine dynamics approach

Mechatronic systems are dynamical systems which can have both continuous and discrete dynamic behavior, meaning the system states can both flow (continuously) and jump (discretely).

A mechatronic system with both dynamic behaviors is called a hybrid system.

One of the subclasses of hybrid systems is called piecewise affine (PWA) systems.

PWA systems are the “simplest” extension to linear systems capable of modelling non-linear and non-smooth processes with arbitrary accuracy.

Simulation closer to real life

The industry standard, still today, is to create models of non-linear systems which are linearized around one systems operational point, cf. Figure 1.

Deviating from this point results in an error and therefore reducing the congruence and accuracy between model and system.

Using several linearization points, cf. Figure 2, for modelling a non-linear system increases the accuracy and congruence of the model and its representation of the system and decreases the error between them.

This results in simulation which are closer to the real system/real life, which is highly desirable. To be able to create such a model with multiple linearization’s, PWA modelling approach can be used.

Figure 1: Single linearization point (left) and Figure 2: Multiple linearization points

Figure 1: Single linearization point (til venstre) og Figure 2: Multiple linearization points

Electric and hydraulic models

In this thesis the PWA approach was used to create models which are based on two experimental setups of electric and hydraulic nature.

The electric setup included only one non-linear element (Tunnel-diode), resulting in a rather simple PWA model. The hydraulic setup, due to its hydraulic nature and the components installed, includes several non-linearities, therefore increasing the complexity of the system and the model.

For both setups, PWA model-based simulations, as well as measurements on the experimental setups were performed and compared to one another, confirming that the PWA modeling approach yields results which are better than just using a single linearization approach and which are comparable to the measurements of the real system.

Additionally, for the hydraulic setup an automatically switched position/force control approach was investigated. Traditionally a “nested” control approach is used for systems which have to be both position and force controlled, resulting in performance deficiencies in both controllers. Using a switched approach, as in this thesis, allows for tuning each controller individually, without them interfering with each other, resulting in an overall improved performance.

This switched control approach was included into the existing PWA model and simulations and experiments were performed, verifying the performance of this control approach.

  

Disputasfakta:

Kandidaten: Philipp Thomas Pasolli (1988 Bregenz, Østerrike) BSc Mechatronics Vorarlberg University of Applied Sciences Austria (2012), MSc Mechatronics Vorarlberg University of Applied Sciences, Austria and University of Agder, Norway (2014). Arbeidet som Test engineer at Thyssenkrupp Presta, Liechtenstein (2014 – 2016), arbeider nå som R&D Engineer ved Macgregor i Arendal.

Prøveforelesning og disputas finner sted digitalt på internett i konferanseprogrammet Zoom (lenke under) onsdag 27. mai 2020.

Disputasen blir ledet av dekan, professor Michael Rygaard Hansen Fakultet for teknologi og realfag, UiA.

Prøveforelesning kl 10:15

Disputas kl 12:15

Oppgitt emne for prøveforelesning: “Duality Principle Between the Design of Linear State Feedback Controllers and State Observers”

Tittel på avhandling: “Hybrid modeling and control of mechatronic systems using piecewise affine dynamics approach”

Søk etter avhandlingen i AURA - Agder University Research Archive, som er et digitalt arkiv for vitenskapelige artikler, avhandlinger og masteroppgaver fra ansatte og studenter ved Universitetet i Agder. AURA blir jevnlig oppdatert.

Opponenter:

Førsteopponent: Professor Bengt Lennartson, Chalmers Tekniska Høgskola, Sverige

Annenopponent: PD Dr.-Ing.habil Andreas Rauh, Universität Rostock - Lehrstuhl für Mechatronik, Tyskland

Bedømmelseskomitéen er ledet av førsteamanuensis Margrethe Wold,  Institutt for ingeniørvitenskap, UiA

Veiledere i doktorgradsarbeidet var førsteamanuensis Michael Ruderman (hovedveileder) og professor dr. Yuri Orlov, CICESE Centro de Investigación Científica y de Educación Superior de Ensenada, Baja California, Mexico (medveileder)

Opponent ex auditorio:

Disputasleder inviterer til spørsmål ex auditorio i innledningen i disputasen, med tidsfrister. Disputasleders e-post er tilgjengelig i chat-funksjonen under disputasen. Spørsmål om ex auditorio kan sendes til disputasleder Michael Rygaard Hansen på e-post michael.r.hansen@uia.no

Avhandlingen er tilgjengelig her: