Přístupnostní navigace
E-application
Search Search Close
Doctoral Thesis
Author of thesis: Ing. Bořek Ščerba
Acad. year: 2025/2026
Supervisor: doc. Ing. Tomáš Návrat, Ph.D.
Reviewers: prof. Ing. Radim Halama, Ph.D., prof. Ing. Libor Pantělejev, Ph.D.
Stopping fatigue tests for visual readings is labor-intensive and conflicts with uninterrupted acquisition recommended in fatigue crack growth rate standards. Furthermore, measuring fatigue crack length during ongoing tests reduces operator workload and enables automation. This thesis develops a digital image correlation (DIC)–based methodology that replaces visual inspection with a traveling microscope (VIM) and validates it against VIM data across materials, specimen sizes and types, and load levels, while also comparing it with a DIC thresholding approach, the Strain threshold method (STM). Sensitivity to input parameters is analyzed, limitations are identified, and user recommendations are formulated. The core estimator is a physics-based Inflection point method (IPM) and its fast, filtered implementation (fIPMf). The workflow is as follows. DIC provides displacements via software-defined “virtual extensometers” laid out along the expected crack path. The resulting curve is a projection of crack opening displacements. Gaussian Process Regression—used here to mitigate spatial noise—fits this curve, and the inflection point of the fitted curve is taken as the crack tip (IPM). Its fast implementation (fIPM) restructures processing and reduces computation time by 55%; the filtered variant (fIPMf) adds an adaptive Kalman filter with a soft-monotonicity safeguard to stabilize the time series, reducing temporal noise by about an order of magnitude to sub-micrometer levels. Validation used compact tension and middle tension specimens, with all the methods (VIM, IPM, and STM) applied to the same images acquired from the same side of the specimen, removing through-thickness bias from unequal crack lengths. The first validation compared crack-length results on images acquired during test stoppages. In this setting, IPM matched the VIM baseline with a root-mean-square error (RMSE) of 31 μm, while STM reached 274 μm. The most important sensitivity finding was the negligible load dependence of IPM, in contrast to STM. An additional validation compared crack-growth-rate results from VIM and fIPMf on the same samples measured sequentially. VIM data were fitted with the NASGRO equation and used as a baseline; fIPMf results deviated by approximately 20-30%, which is comparable to the spread among VIM datasets themselves. For real-time use, fIPMf offers the same data-processing frequency as STM, which is more than 10× higher than for full-field DIC. In summary, the proposed methodology enables non-interruptive, VIM-comparable crack-length and crack-growth evaluation during running tests and outperforms STM by a large margin. It offers a processing speed of about 20 frames/s. Key benefits include a threshold-free, physics-based estimator; robust adaptive fitting and filtering that learn during the test and react to changing conditions; and negligible load-level sensitivity, enabling K-decreasing tests to identify the stress intensity factor threshold—something difficult with STM.
Crack opening displacement (COD) profile; Digital image correlation (DIC) for fatigue testing; Fatigue crack length measurement in metals; Fatigue crack growth rate evaluation; Inflection point method for crack-tip localization (IPM); Strain threshold method (STM); Virtual extensometers for crack monitoring; Visual inspection with a traveling microscope (VIM).
Date of defence
12.03.2026
Result of the defence
Defended (thesis was successfully defended)
Process of defence
Autorem navržená metodika určování délky trhliny při cyklickém zatěžování pomocí DIC představuje originální příspěvek k rozvoji experimentálních metod lomové mechaniky. Autor ověřil její použitelnost na několika typech kovových materiálů a prokázal možnosti jejího perspektivního praktického využití.
Language of thesis
English
Faculty
Fakulta strojního inženýrství
Department
Institute of Solid Mechanics, Mechatronics and Biomechanics
Study programme
Applied Mechanics (D-IME-P)
Composition of Committee
prof. Ing. Jindřich Petruška, CSc. (předseda) prof. Ing. Libor Pantělejev, Ph.D. (člen) prof. Ing. Radim Halama, Ph.D. (člen) Ing. Lubomír Junek, Ph.D. (člen) prof. RNDr. Michal Kotoul, DrSc. (člen) prof. RNDr. Ludvík Kunz, CSc., dr. h. c. (člen) Ing. Oldřich Ševeček, Ph.D. (člen)
Supervisor’s reportdoc. Ing. Tomáš Návrat, Ph.D.
Reviewer’s reportprof. Ing. Radim Halama, Ph.D.
Reviewer’s reportprof. Ing. Libor Pantělejev, Ph.D.
Grade proposed by reviewer: C
Responsibility: Mgr. et Mgr. Hana Odstrčilová