| Abstract | Primjena računalne tomografije (CT) u dimenzionalnom mjeriteljstvu metoda je kojom je omogućeno mjerenje vanjskih i unutarnjih geometrija predmeta mjerenja skeniranjem predmeta rendgenskim zračenjem. Riječ je o nerazornoj metodi ispitivanja koja omogućava trodimenzionalno mjerenje unutarnje geometrije predmeta što je čini interesantnom u širokoj primjeni. Jedan od preduvjeta za implementaciju metode računalne tomografije u području dimenzionalnog mjeriteljstva jest osiguravanje sljedivosti rezultata mjerenja. Kako se radi o izrazito složenom mjernom sustavu s velikim brojem utjecajnih parametara, mjerna nesigurnost rezultata mjerenja često nije procijenjena, a mjeriteljska sljedivost u općem slučaju nije osigurana. U cilju uspostavljanja sljedivosti kod računalne tomografije u dimenzionalnom mjeriteljstvu u radu su provedena teorijska i eksperimentalna istraživanja utjecajnih parametara procesa mjerenja računalnom tomografijom na rezultate dimenzionalnih i geometrijskih značajki na predmetima jednostavne i složenije geometrije. U teorijskom dijelu rada objašnjeni su parametri koji nastaju kao rezultat nesavršenosti mjernog sustava i njegovih komponenata. Predložena je klasifikacija utjecajnih parametara s obzirom na tijek procesa provedbe dimenzionalnih mjerenja računalnom tomografijom. S obzirom na nedostatnosti postojećih referentnih predmeta, a na osnovu rezultata istraživanja značajki postojećih referentnih predmeta predložen je model novog referentnog predmeta kod primjene računalne tomografije u području dimenzionalnog mjeriteljstva. Zbog velikog broja parametara koji utječu na rezultat mjerenja, eksperimentalna istraživanja započeta su provedbom djelomičnog faktorskog plana pokusa pri čemu je izdvojen manji broj značajnijih parametara. Daljnja istraživanja provedena su mjerenjima računalnom tomografijom te provedbom simulacija procesa skeniranja. U cilju praćenja i eliminacije sustavnih pogrešaka koje nastaju kao rezultat djelovanja ulaznih parametara, istražen je utjecaj pogrešaka nastalih u procesu skeniranja predmeta rendgenskim zračenjem, pogrešaka nastalih u procesu rekonstrukcije poprečnih presjeka i 3D modela te utjecaj predmeta mjerenja. U radu je procijenjena mjerna nesigurnost rezultata mjerenja sukladno normi ISO 15530-3:2011, odnosno smjernicama VDI/VDE 2630 Part 2.1, te kombinacijom metode Monte Carlo sukladno normi JCGM 101:2008 i virtualnog CT sustava korištenjem softverskog paketa aRTist. Usporedba dobivenih rezultata mjerenja s referentnim vrijednostima provedena je izračunom faktora slaganja En. |
| Abstract (english) | Application of Computed Tomography (CT) in dimensional metrology is a method which enables measurement of both external and internal geometry of measured objects, by scanning objects using X-ray. It is a non-destructive measuring method that allows three-dimensional measurement of the internal geometry of an object. This feature makes the method interesting for a wide range of applications. A prerequisite for implementation of computed tomography in dimensional measurement is assuring metrological traceability of the measurement results. Considering the fact that computed tomography is a very complex system with a large number of influence parameters, neither measurement uncertainty nor measurement traceability can generally be assured. With the goal of establishing traceability of computed tomography in dimensional measurements, theoretical and experimental research was carried out on objects of simple and complex geometries. The research was carried out in several phases, which are summarized in the following chapters. Chapter 1 Introduction The first chapter elaborates on the need for conducting research in the field of traceability assurance. Additionally, through a review of the existing literature the current state of this scientific field is found to be encountering difficulties, primarily in the area of conducting assurance of measurement results traceability through the application of the CT measurement system. The current state of the field, ascertained by the review of the existing literature motivated this doctoral thesis which explores the possibilities of conducting dimensional metrology using the method of computed tomography. Further, the hypothesis and the main aim of this research are presented, followed by the research plan and expected scientific contribution. Chapter 2 Computed tomography in dimensional metrology In the second chapter of the thesis, parameters influencing measurement system are classified and explained. At the beginning of the second chapter application of industrial computed tomography in the field of dimensional measurements is described. This is followed by a schematic display of the process through which the measurement is conducted. The resulting 2D data is described mathematically using the Radon transform, which represents the mathematical background for connecting Cartesian and projection coordinates. In this part of the thesis, a detailed description of the filtration process of raw CT data is given. The artefacts analyzed are beam hardening artefacts, scatter artefacts, ring artefacts, metal artefacts, motion artefacts and partial volume artefacts. They appear in the process of scanning with cone beam x-ray, as a consequence of the imperfections in measurement system. Chapter 3 Traceability assurance of computed tomography in dimensional metrology In third chapter the existing reference objects used for systematic error correction are summarized. This is conducted according to the purpose of the reference object in terms of correcting different kinds of errors as well as according to the geometry of the object. Due to the insufficiency of the existing reference objects and based on the results of the research where the features and geometry of existing reference objects was taken into account, a new reference object for dimensional measurements with computed tomography is proposed. The proposal of the new reference object follows guidelines given in VDI/VDE 2630 - Part 1.3. This part of the paper also develops methods for assessing measurement uncertainty of measurement results as well as provides an analysis of the results of interlaboratory comparisons in the area of using CT for dimensional measurement. A description of the model for assessing measurement uncertainty in accordance with the Guide to the expression of uncertainty in measurement (JCGM 100:2008) is provided. Furthermore, the model is described in accordance with the ISO 15530-3:2011 as well as the guidelines given in VDI/VDE 2630 Part 2.1. Finally, the description of the model which includes the use of the Monte Carlo simulation (MCS) method according to JCGM 101:2008 and the implementation of the virtual CT system through the use of the software package aRTist is given. Chapter 4 Experimental research In the fourth chapter experimental research on influence parameters in the overall measurement process are conducted. In the first phase of experimental research preliminary research using the partial design of experiment was conducted. In further research, the preliminary research results were used to analyse the influence of number of projections, the influence of geometrical magnification and the influence of beam hardening corrections. Additionally, the research also includes investigation on the influence of the positioning of the measurement object in relation to rotation axis as well as the influence of noise reduction and object surface roughness. The research was conducted on homogeneous objects of both simple and complex geometry as well as objects with different material density. Results obtained with CT are compared to results obtained using a tactile CMM. Chapter 5 Evaluation of measurement uncertainty In the fifth chapter measurement uncertainty of results obtained from measuring an aluminium and polymer cylinder with complex geometry was estimated using a combination of virtual computed tomography and Monte Carlo simulations according to the JCGM 101:2008 standard. The values of extended measurement uncertainties, given with the coverage factor k = 2 and probability P = 95 %, are in range from 16 μm to 52 μm, depending on the measured dimensional characteristic. The results were validated through the comparison between measured results and reference values which was achieved by calculating the En number. Furthermore, a substitution measurement method in accordance with the ISO 15530-3:2011 and the VDI/VDE 2630 Part 2.1 guidelines was conducted. The values of extended measurement uncertainties, given with the coverage factor k = 2 and probability P = 95 %, are in range from 6 μm to 34 μm, depending on the measured dimensional characteristic. Chapter 6 Conclusion In the sixth chapter of the doctoral thesis, an overview of the complete conducted research is provided, final conclusions are drawn and guidelines for further research are provided. |
