This thesis presents the development and validation of a numerical simulation of grass cutting based on the Discrete Element Method (DEM), with the aim of accurately reproducing the mechanical behavior of grass stems during interaction with robotic lawnmower blades. Grass stems are modelled as flexible and breakable structures composed of bonded sphero-cylindrical particles, allowing the simulation to capture the bending behavior, frictional effects and contact forces during the cutting process. The first part of the thesis introduces the theoretical background of the DEM model and discusses its suitability for modeling bendable materials. The use of sphero-cylinders is investigated as an alternative to conventional spherical particles, highlighting their advantages in terms of computational efficiency and geometrical consistency with grass stems. The Hertz-Mindlin contact theory is adopted to describe the particle interactions. An extensive experimental test campaign is conducted to determine the mechanical and physical properties of grass stems of Festuca Arundinacea selected for the numerical simulation. These tests are: (i) the bending test used to evaluate the Young Modulus under self-weight conditions, (ii) the tensile test providing the tensile strength, (iii) the friction test quantifying the interaction between grass and steel blades and that between grass stems, (iv) the cutting test determining the shear strength required up to failure. Finally, a grass cutting simulation for the mechanical power evaluation is developed with the software STAR-CCM+. The result of the simulated cutting process is then compared with a physical cutting test performed with a robotic lawnmower, where electrical power absorption is measured and converted into mechanical power using motor efficiency curves. The comparison between experimental and numerical results shows good agreement, confirming the reliability of the proposed modelling approach. The outcomes of this thesis highlight the potential of DEM-based simulation as a powerful tool for analyzing grass cutting processes and for supporting the design and optimization of robotic lawnmower systems.
Questa tesi illustra lo sviluppo e la validazione di una simulazione numerica del taglio dell’erba basata sul Metodo degli Elementi Discreti (DEM), finalizzato a riprodurre accuratamente il comportamento meccanico degli steli d’erba durante l’interazione con le lame di un tosaerba robotizzato. Gli steli d’erba sono modellati come strutture flessibili composte da particelle sfero-cilindriche, consentendo alla simulazione di catturare il comportamento flessionale, gli effetti di attrito e le forze di contatto durante il processo di taglio. La prima parte della tesi introduce il quadro teorico del modello DEM e ne discute la sua idoneità per la modellazione di materiali flessibili. L’utilizzo degli sfero-cilindri è analizzato come alternativa alle particelle sferiche convenzionali, evidenziandone i vantaggi in termini di efficienza computazionale e coerenza geometrica con la forma degli steli d’erba. La teoria di Hertz-Mindlin è adottata per descrivere le interazioni tra le particelle. Diversi test sono stati condotti per determinare le proprietà meccaniche e fisiche degli steli d’erba di Festuca Arundinacea selezionati per svolgere la simulazione numerica. Questi test sono: (i) la prova di flessione utilizzata per valutare il Modulo di Young in condizioni di auto-peso, (ii) la prova di trazione che consente di determinare la resistenza a trazione, (iii) le prove di attrito che quantificano l’interazione tra l’erba e le lame in acciaio e quella tra gli steli d’erba, (iv) la prova di taglio che determina la resistenza a taglio fino a rottura. Infine, viene sviluppata una simulazione del taglio dell’erba per la valutazione della potenza meccanica con il software STAR-CCM+. Il risultato del processo di taglio simulato è quindi confrontato con una prova di taglio fisica eseguita con un tosaerba robotizzato, in cui l’assorbimento di potenza elettrica è misurato e convertito in potenza meccanica utilizzando le curve di efficienza del motore. Il confronto tra i risultati sperimentali e numerici mostra un buon accordo, confermando l’affidabilità dell’approccio di modellazione proposto. I risultati di questa tesi evidenziano il potenziale delle simulazioni basate sul modello DEM come uno strumento efficace per l’analisi dei processi di taglio dell’erba e per supportare la progettazione e l’ottimizzazione dei sistemi di tosaerba robotizzati.
Discrete element method simulation of grass cutting: characterization and calibration of mechanical parameters
ZANARDO, MARCO
2025/2026
Abstract
This thesis presents the development and validation of a numerical simulation of grass cutting based on the Discrete Element Method (DEM), with the aim of accurately reproducing the mechanical behavior of grass stems during interaction with robotic lawnmower blades. Grass stems are modelled as flexible and breakable structures composed of bonded sphero-cylindrical particles, allowing the simulation to capture the bending behavior, frictional effects and contact forces during the cutting process. The first part of the thesis introduces the theoretical background of the DEM model and discusses its suitability for modeling bendable materials. The use of sphero-cylinders is investigated as an alternative to conventional spherical particles, highlighting their advantages in terms of computational efficiency and geometrical consistency with grass stems. The Hertz-Mindlin contact theory is adopted to describe the particle interactions. An extensive experimental test campaign is conducted to determine the mechanical and physical properties of grass stems of Festuca Arundinacea selected for the numerical simulation. These tests are: (i) the bending test used to evaluate the Young Modulus under self-weight conditions, (ii) the tensile test providing the tensile strength, (iii) the friction test quantifying the interaction between grass and steel blades and that between grass stems, (iv) the cutting test determining the shear strength required up to failure. Finally, a grass cutting simulation for the mechanical power evaluation is developed with the software STAR-CCM+. The result of the simulated cutting process is then compared with a physical cutting test performed with a robotic lawnmower, where electrical power absorption is measured and converted into mechanical power using motor efficiency curves. The comparison between experimental and numerical results shows good agreement, confirming the reliability of the proposed modelling approach. The outcomes of this thesis highlight the potential of DEM-based simulation as a powerful tool for analyzing grass cutting processes and for supporting the design and optimization of robotic lawnmower systems.| File | Dimensione | Formato | |
|---|---|---|---|
|
Zanardo_Marco.pdf.pdf
Accesso riservato
Dimensione
5.5 MB
Formato
Adobe PDF
|
5.5 MB | Adobe PDF |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/108199