TU Darmstadt / ULB / TUbiblio

System dynamics improvement with state-of-the-art mechatronics : potentials and challenges

Jungblut, Jens ; Viehmann, Andreas ; Jardin, Philippe ; Rinderknecht, Stephan (2020)
System dynamics improvement with state-of-the-art mechatronics : potentials and challenges.
Neapel, Italien (17.02.2020-19.02.2020)
Konferenzveröffentlichung, Bibliographie

Kurzbeschreibung (Abstract)

Throughout the increasing ambition to optimize a mechanical system in respect to its weight andefficiency to match environmental requirements, the systems tend to be more sensitive to vibrations.These cause undesired audible noise, which can be countered by implementing passive componentssuch as dampers. These on the other hand lead to a decrease in efficiency and an increase in mass.Mechatronic systems open up a new design space, which have the potential to partially resolve thecompromise between efficiency, weight and the noise-vibration-harshness (NVH) behaviour. How-ever, this potential also comprises several challenges, which have to be mastered. Three differenttopics will be addressed in this talk.The first one is the active vibration control of narrow banded vibrations, where two systemsare discussed. The first system is an active piezoelectric bearing where the actuators are placed inthe load path. Piezoelectric actuators are well suited for this task because of their high stiffness.Classic control algorithms can be applied to achieve a decent vibration reduction. However, un-der consideration of system knowledge, control algorithms can be designed which show a betterperformance than the classic approaches without requiring a model. The second system is a plan-etary gearbox where piezoelectric inertial mass actuators are used to reduce vibrations. A modelfree control is implemented which can only be implemented with the knowledge of the excitationbehaviour.The second challenge focuses on jerking in powertrains of pure electric and hybrid electricvehicles with two electric motors. The use of two electric motors offers a broad variety of oper-ational modes, which allows an efficient operation of each motor. For high comfort gear shiftswithout interruption of traction force, torque blending between the two electric motors is neces-sary. If the torque at the driveshaft is not kept constant, jerking occurs which is uncomfortable forthe vehicle passengers. The minimization of unpleasant jerking throughout control of the electricmotors is the second topic to be addressed.Third, new arising methods throughout digitalization, which may lead to a transformationof classic mechatronics, are discussed. That is done with the example application of hybrid dy-namic system modelling approach where traditional physical modelling is used in combinationwith machine learning algorithms. From our perspective, the additional usage of available systemknowledge in these hybrid models will outerperform models which are solely based on big dataapproaches. That is because synergies between both worlds are used in such way, that the overallsystem performance is improved.

Typ des Eintrags: Konferenzveröffentlichung
Erschienen: 2020
Autor(en): Jungblut, Jens ; Viehmann, Andreas ; Jardin, Philippe ; Rinderknecht, Stephan
Art des Eintrags: Bibliographie
Titel: System dynamics improvement with state-of-the-art mechatronics : potentials and challenges
Sprache: Englisch
Publikationsjahr: 2 Februar 2020
Titel der Zeitschrift, Zeitung oder Schriftenreihe: MEDYNA 2020: 3rd Euro-Mediterranean Conference on Structural Dynamics and Vibroacoustics
Buchtitel: MEDYNA 2020: 3rd Euro-Mediterranean Conference on Structural Dynamics and Vibroacoustics
Veranstaltungsort: Neapel, Italien
Veranstaltungsdatum: 17.02.2020-19.02.2020
Kurzbeschreibung (Abstract):

Throughout the increasing ambition to optimize a mechanical system in respect to its weight andefficiency to match environmental requirements, the systems tend to be more sensitive to vibrations.These cause undesired audible noise, which can be countered by implementing passive componentssuch as dampers. These on the other hand lead to a decrease in efficiency and an increase in mass.Mechatronic systems open up a new design space, which have the potential to partially resolve thecompromise between efficiency, weight and the noise-vibration-harshness (NVH) behaviour. How-ever, this potential also comprises several challenges, which have to be mastered. Three differenttopics will be addressed in this talk.The first one is the active vibration control of narrow banded vibrations, where two systemsare discussed. The first system is an active piezoelectric bearing where the actuators are placed inthe load path. Piezoelectric actuators are well suited for this task because of their high stiffness.Classic control algorithms can be applied to achieve a decent vibration reduction. However, un-der consideration of system knowledge, control algorithms can be designed which show a betterperformance than the classic approaches without requiring a model. The second system is a plan-etary gearbox where piezoelectric inertial mass actuators are used to reduce vibrations. A modelfree control is implemented which can only be implemented with the knowledge of the excitationbehaviour.The second challenge focuses on jerking in powertrains of pure electric and hybrid electricvehicles with two electric motors. The use of two electric motors offers a broad variety of oper-ational modes, which allows an efficient operation of each motor. For high comfort gear shiftswithout interruption of traction force, torque blending between the two electric motors is neces-sary. If the torque at the driveshaft is not kept constant, jerking occurs which is uncomfortable forthe vehicle passengers. The minimization of unpleasant jerking throughout control of the electricmotors is the second topic to be addressed.Third, new arising methods throughout digitalization, which may lead to a transformationof classic mechatronics, are discussed. That is done with the example application of hybrid dy-namic system modelling approach where traditional physical modelling is used in combinationwith machine learning algorithms. From our perspective, the additional usage of available systemknowledge in these hybrid models will outerperform models which are solely based on big dataapproaches. That is because synergies between both worlds are used in such way, that the overallsystem performance is improved.

Fachbereich(e)/-gebiet(e): 16 Fachbereich Maschinenbau
16 Fachbereich Maschinenbau > Institut für Mechatronische Systeme im Maschinenbau (IMS)
Hinterlegungsdatum: 29 Jun 2020 05:29
Letzte Änderung: 29 Jun 2020 05:29
PPN:
Export:
Suche nach Titel in: TUfind oder in Google
Frage zum Eintrag Frage zum Eintrag

Optionen (nur für Redakteure)
Redaktionelle Details anzeigen Redaktionelle Details anzeigen