Divjak, Alan (2021) Proceduralno modeliranje prostorno nehomogenih mrežastih objekata za aditivnu proizvodnju = Procedural modeling of spatially inhomogeneous mesh objects for additive manufacturing. Dissertation (PhD) thesis. Grafički fakultet. [Mentor: Modrić, Damir].
|
PDF
Doktorski rad Divjak Alan.pdf Download (34MB) | Preview |
Abstract
Upotreba replika ljudskih kosti u istraživanjima česta je u brojnim područjima medicine, ali riječ je o pojednostavljenim modelima čija mehanička svojstva ne odgovaraju stvarnim kostima. Motiv izrade ove disertacije definiranje je procesa izrade replika ljudskih kosti realnog raspona mehaničkih svojstava. Replike su napravljene od funkcionalno gradiranog materijala temeljenog na gradijentu poroznosti koji su realizirani putem prostorno nehomogenih mrežastih struktura. Oralna implantologija jedno od područja medicine gdje je potreba za takvim replikama iznimno izražena, stoga je istraživanje procesa izrade usmjereno na ljudsku čeljust. Proces izrade koristi proceduralno 3D modeliranje i aditivnu proizvodnju, stoga je provedeno istraživanje koje utvrđuje optimalne parametre upotrebe navedenih tehnologija. Provedena je mehanička karakterizacija korištenog materijala za 3D tisak. Utvrđena je optimalna konfiguracija proceduralno generiranih mrežastih struktura za postizanje najvećeg mogućeg raspona mehaničkih svojstava. Napravljena je analiza računalne zahtjevnosti izrade modela replika i izvedivosti njihove pripreme za 3D tisak. Istražene su tehničke mogućnosti izrade mrežastih struktura na korištenom 3D printeru. Provedena je virtualna mehanička karakterizacija uzoraka mrežastih struktura pomoću metode konačnih elemenata. Na kraju su rezultati istraživanja ujedinjeni kako bi se izradio konačni model replike dijela ljudske čeljusti. Rezultati istraživanja potvrđuju definirane hipoteze disertacije i valjanost predloženog procesa izrade replika. Korišteni softveri za modeliranje i pripremu modela predstavljaju jedine problematične komponente u procesu izrade. Demonstrirali su značajna ograničenja koja rezultiraju izradom modela dijela kosti najvećeg volumena od 1 cm3, što predstavlja neupotrebljivo malen uzorak. Stoga je napravljen vlastiti program koji ujedno generira mrežaste strukture i radi pripremu za 3D tisak. Testiranje mogućnosti programa pokazuje kako je sposoban proceduralno modelirati i pripremiti za 3D tisak replike bilo koje kosti u ljudskom tijelu. Zaključak je kako predloženi proces izrade omogućava izradu replika ljudskih kosti realnog raspona mehaničkih svojstava. Replike kosti mogu se izraditi brzo i jednostavno korištenjem potrošačkih LCD stereolitografskih 3D printera niske cijene, što njihovu upotrebu u medicinskim istraživanjima čini iznimno pogodnom.
| Item Type: | Dissertation (PhD) thesis |
|---|---|
| Mentor name: | Modrić, Damir |
| Thesis Committee: | Pap, Klaudio and Matijević, Mile and Schauperl, Zdravko |
| Defence date: | 1 October 2021 |
| Abstract in english: | Treatment of human bones is a common procedure in medicine, as is medical research that deals with them. Planning and preparation of bone treatment, and research of new treatment procedures are ideally based on real human bones so that the outcomes of planning and research give realistic results. Doctors can rarely use human bones for these purposes. An additional problem is the need to use the bones of patients of a specific age group, gender, or bone disease. These activities would significantly be facilitated by the existence of artificial models of human bones that are identical in shape and mechanical properties to real ones. One of the areas of medicine where the need for such artificial bones is extremely pronounced is oral implantology. Oral implantology is a field of dental medicine that deals with the treatment of patients with a lack of teeth. The goal of treatment is the implantation of dental implants that return the mouth to a functionally and aesthetically correct condition. The need for jaw models that are a realistic representation of key morphological and qualitative aspects of real jaws is the reason why the field of oral implantology was chosen as the focus of this dissertation. Creating a model of the jaw of a realistic shape is not a problem. This procedure has long been perfected and is based on a three-dimensional reconstruction of the surface of a volumetric bone model created through radiological segmentation. Achieving realistic mechanical properties is a much more complex task, primarily because bone density and its mechanical properties vary spatially. Precisely this aspect of model making is problematic, and the question of how to realize spatially graded variations in density and mechanical properties arises. The goal of this dissertation is to solve this problem by making a bone model with a realistic distribution of mechanical properties. The cost of making the model must be low, as well as the price of the equipment used, so they can be accessible to the widest possible range of users in various institutions such as colleges, hospitals, and institutes. Although the research focuses on oral implantology, the same approach can be applied to any bone in the body, human or animal. Achieving a completely realistic mechanical behavior of the jaw requires detailed knowledge of the two components. The first is the spatial distribution of bone tissue density that determines spatial distribution of mechanical properties. This is precisely the purpose of this dissertation - the production of material that has an appropriate spatial distribution ofthe required mechanical properties. The second component is the material used to make the jaw model. Human bone has certain mechanical properties and the task of the material used is to mimic them as well as possible. The selection of such material was not the aim of this research as it falls within the material science and is the subject of further research. Materials that spatially change certain properties are called functionally graded materials (FGM). The aim of this dissertation is to investigate and define the process of making functionally graded materials, and to produce a part of the jaw from it using additive manufacturing technology. Produced part must posses a realistic distribution of mechanical properties, while the cost of equipment and manufacturing must be low. To achieve this goal, three key technologies were used: finite element method, procedural 3D modeling, and additive manufacturing. The finite element method is used to determine the mechanical properties of material samples made from three-dimensional mesh structures of different densities. Based on this data, it is possible to define the relationship between the material density and mechanical properties, which is crucial for a realistic imitation of mechanical properties of bone tissue of different densities. Procedural 3D modeling is a set of techniques that allows automated creation of extremely complex 3D models by defining a set of rules based on which the model is generated. Jaw models made of FGM are an example of extremely complex 3D models, so the use of procedural 3D modeling is the only purposeful way to make them. Additive production was chosen as the production method because of the numerous advantages it provides over traditional methods of making FGM, which are expensive and complex. There are several additive manufacturing processes that are actively used in the production of FGM. 3D printers based on these processes are currently very expensive and have shortcomings that make them incompatible with one of the default production criteria set in this paper, which is the low cost of the equipment and material. The solution is to use an alternative additive manufacturing process that is rarely used to make FGM, and that is stereolithography (SLA). The use of this process limits the production of models to FGM based on the porosity gradient, but by combining procedural 3D modeling and appropriate properties of stereolithography it is possible to achieve the desired results.The path from the radiological image of the patient to the final model consists of a series of steps, so the experimental part is divided into seven parts that deal with determining the optimal solution for each step of the process. The following is a brief description of the experimental part and the results obtained. The first part is the mechanical characterization of the material used in the 3D printing process to make the jaw model. Knowledge of the mechanical properties of the material used is necessary in order to later determine the mechanical properties of FGM samples of different densities using the finite element method. Mechanical three-point bending testing is performed on a computer-controlled universal testing machine. Additively produced test samples made from the material used to make the final jaw part model are used. Equal mechanical testing is performed in a virtual environment using the finite element method. The real test data obtained on the Hegewald & Peschke INSPEKT 20-1 universal testing machine are used and the mechanical properties of the material are iteratively changed until the virtual specimen is deformed in the same way as the real one, thus obtaining the actual mechanical properties. The second part deals with determining the resolution of the 3D printer used. It is necessary to produce smallest pores possible when making FGM based on porosity, which is achieved by making the thinnest possible struts that form edges of the pores. The pores must be as thin as possible to make the interface between the implant and the model as realistic as possible. Simple 3D models with different strut thicknesses are made and made on a 3D printer. The minimum strut thickness is defined by the model with the thinnest struts made without errors. This data is the basis for further development of FGM. The third part deals with the process of making the mesh structures that make up FGM and their optimal shape. Four essential items are examined: optimal structure, density variability, computational complexity, and feasibility of preparation for 3D printing. The research results for the first two items are positive, while the other two showed that the planned way of making a 3D model and its preparation for 3D printing is not currently possible due to the extreme geometric complexity of the models and consequently enormous demands on computer resources. The fourth part deals with addressing the negative research outcome of the aforementioned two items. A method of generating 3D model and 3D print preparationdirectly from a very simple data set generated in the early stages of model generation is proposed. The testing show that the proposed algorithm gives satisfactory results. The fifth part is concerned with production of FGM samples of different densities. FGM density can in theory range from 0 to 100%, but in reality, a 3D printer can make models with a smaller density range. Determining this range is the goal of this part of the research. The sixth part deals with the mechanical characterization of FGM samples of different densities. For the needs of research, 3D models of FGM of different densities are created. The models are transferred to FEA software and virtual mechanical testing is performed, This established the relationship between the density of the samples and their mechanical properties. The seventh and final part of the research is the creation of the final model of the jaw part using FGM. This is the culmination of all previous research, and the goal is to show that it is possible to model and prepare such a complex model for 3D printing and make it using a 3D printer. The results show that the defined goals were achieved, the expected scientific contributions were obtained, and the set hypotheses were fully confirmed: H1 Different geometric configurations of spatial mesh structures can simulate objects of different mechanical characteristics H2 Commercial stereolithographic printing technology can be used to produce spatial mesh structures H1 Different geometric configurations of spatial mesh structures can simulate objects of different mechanical characteristics |
| Uncontrolled Keywords: | oralna implatologija, proceduralno modeliranje, aditivna proizvodnja, stereolitografija, metoda konačnih elemenata |
| Keywords in english: | oral implantology, procedural modeling, additive manufacturing, stereolithography, finite element method |
| Subjects: | TECHNICAL SCIENCES > Graphic Technology |
| Institution: | Grafički fakultet |
| City: | Zagreb |
| Number of Pages: | 297 |
| Callnumber: | 655.3-023.5:616.31:004 DIV p |
| Inventory number: | 10431 |
| Depositing User: | Nina Jelača |
| Status: | Unpublished |
| Date Deposited: | 20 Jun 2022 09:48 |
| Last Modified: | 20 Jun 2022 09:48 |
| URI: | http://eprints.grf.unizg.hr/id/eprint/3357 |
Actions (login required)
 |
View Item |
