dc.contributor.advisor | Sørby, Knut | |
dc.contributor.advisor | Brøtan, Vegard | |
dc.contributor.author | Chelishchev, Petr | |
dc.date.accessioned | 2020-09-09T11:36:52Z | |
dc.date.available | 2020-09-09T11:36:52Z | |
dc.date.issued | 2020 | |
dc.identifier.isbn | 978-82-326-4867-2 | |
dc.identifier.issn | 1503-8181 | |
dc.identifier.uri | https://hdl.handle.net/11250/2677053 | |
dc.description.abstract | The PhD thesis proposes new approaches and algorithms related to the field of Geometrical Dimensioning and Tolerancing (GD&T) inspection based on ISO standards on Geometrical Product Specifications (GPS). The main purpose of this work is to improve reliability and accuracy of measurement strategies in GD&T inspection (tolerance verification), which plays an important role in manufacturing industry. The results of this work provide contributions for further development and standardization of measurement strategies and procedures with optimized sample strategies to provide a desired level of the measurement uncertainty.
The objective of the PhD project is to identify key factors that influence the measurement uncertainty and confidence level, try to clarify effects of these influences to improve the current state of measurement strategies in GD&T inspection. Generally, the measurement uncertainty is affected by many factors, e.g. choice of measuring equipment, calibration, control of environment conditions, workpiece orientation and clamping. In measurements with Coordinate Measurement Machines (CMMs), addition factors like stylus system qualification, choice of probe configuration, probe deflection, traveling speed, approach vector, coordinate system alignment, and datum system definition will influence on the measurement uncertainty.
Most of the measurements involved in this thesis were performed in a Leitz PMMC-600 CMM, with an analogue measuring probe. All factors determining the uncertainty of CMM measurements can be divided into four main categories: equipment, environment, workpiece, and operator influence. The last category is mostly associated with the measurement strategy and procedure performed by the operator, which is of particular importance for the uncertainty in CMM measurements. Thereby, the following objectives were defined for this PhD project:
• development and investigation of sample strategies with optimal sample size;
• investigation of outlier detection methods;
• development of algorithms for calculation of substitute (reference) elements.
The choice of sample strategies and algorithms should be based on the tolerance requirements and the expected workpiece geometry deviations. The number of measurement points that are necessary to discover the decisive area of the workpiece surface for reliable geometrical verification is an important research question. In spite of many efforts of previous research, it remains an unsolved problem. Uniform guidelines based on international standards, which could determine criteria for optimal choice of the sample size has not been established so far.
Taking into account all aspects mentioned above, the main contributions of this PhD research are the following:
• An approach to define a confidence level for statistical tolerance intervals of various types of distributions and different sample sizes of measured variables based on CMM measurements of circular profiles after turning and milling machine operations (associated with ISO16269-6)
• Classification of data outliers and an investigation of outliers detecting procedures according to ISO16269-4
• An approach for optimizing the sample size for two-point diameter verification according to ISO14405-1
• An approach based on Artificial Neural Networks (ANN) to evaluate the maximum estimated error of the form deviation related to ISO1101
• New methods for defining substitute elements in estimation of the minimal distance between planes of cuboid objects by using the Minimum Volume Bounding Box (MVBB) principle
The contributions are presented in six papers, where five papers have been presented at conferences and conference proceedings, and one paper have been submitted for publication in a scientific journal. The contributions are addressed to quality engineers, metrology specialists and other researchers in metrology and GD&T inspection field. | en_US |
dc.language.iso | eng | en_US |
dc.publisher | NTNU | en_US |
dc.relation.ispartofseries | Doctoral theses at NTNU;2020:259 | |
dc.relation.haspart | Paper 1: Chelishchev, Petr; Popov, Aleksandr; Sørby, Knut. Robust estimation of optimal sample size for CMM measurements with statistical tolerance limits. MATEC Web of Conferences 2018 ;Volum 220. s. 1-5 | en_US |
dc.relation.haspart | Paper 2: Chelishchev, Petr; Popov, Aleksandr; Sørby, Knut. An investigation of outlier detection procedures for CMM measurement data. MATEC Web of Conferences 2018 ;Volum 220. s. 1-6 | en_US |
dc.relation.haspart | Paper 3: Chelishchev, Petr; Sørby, Knut. Optimization of Sample Size for Two-Point Diameter Verification in Coordinate Measurements. Lecture Notes in Electrical Engineering 2019 ;Volum 484.(1) s. 313-321
https://doi.org/10.1007/978-981-13-2375-1_39 | en_US |
dc.relation.haspart | Paper 4: Chelishchev, Petr; Sørby, Knut. Simulation Algorithm of Sample Strategy for CMM Based on Neural Network Approach. I: Advanced Manufacturing and Automation IX Conference proceedings IWAMA 2019. Springer 2020. s. 434-441
https://doi.org/10.1007/978-981-15-2341-0_54 | en_US |
dc.relation.haspart | Paper 5: Chelishchev, Petr; Sørby, Knut; Privalov, Vadim. Perspectives for appliance and accuracy improvement of coordinate measurements with laser technique. 2019 IEEE International Conference on Electrical Engineering and Photonics; 2019-10-17 - 2019-10-18
https://doi.org/10.1109/EExPolytech.2019.8906848 | en_US |
dc.relation.haspart | Paper 6:
Chelishchev, Petr; Sørby, Knut.
Estimation of Minimum Volume of Bounding Box for Geometrical Metrology | en_US |
dc.title | Sample Strategies, Data Processing and Algorithms for Coordinate Measurements | en_US |
dc.type | Doctoral thesis | en_US |
dc.subject.nsi | VDP::Technology: 500::Mechanical engineering: 570 | en_US |