Stochastic evaluation of the fatigue safety of concrete subjected to compressive fatigue loading

• Applicant: Professor Dr.-Ing. Steffen Marx
• Subject Area: Structural Engineering, Building Informatics and Construction Operation
• Term: since 2025
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 571276042

The increasing use of high-strength and ultra-high-strength concretes is leading to a growing number of slender and therefore more fatigue-prone reinforced and prestressed concrete structures. At the same time, the number of structures subjected to high fatigue loads, e. g. wind turbines and slender bridges, is growing. As a result, fatigue verification is becoming increasingly important for design. The design approaches against concrete fatigue in Germany essentially consist of the S-N curves and the design fatigue reference strength for concrete in compression. Although there are numerous fatigue tests on concrete under uniaxial compressive fatigue load in the literature, there are hardly any test results in the range of N > 10^7. Existing S-N curves in this range are therefore only estimations and the actual safety in the current verification concept of concrete fatigue is unknown. In the range N < 10^7, the number of cycles to failure of concrete scatters very widely. First evaluations suggest that the cause of the scattering of the number of cycles to failure of concretes in single-stage compression tests lies in the relation of the maximum stress to the mean value of the scattering concrete compressive strength of the tested batch. Assuming that the scatter of the numbers of cycles to failure and the compressive strengths of specimens of a concrete batch correlate, the maximum stress level can be related to the actual concrete compressive strength of the tested specimen, thus essentially eliminating the scatter of the numbers of cycles to failure. In order to investigate this hypothesis, the influence of the scattering of the compressive strength on the fatigue resistance of concrete will be determined experimentally, described scientifically and transferred into a stochastic design approach in the proposed project. For this purpose, 355 cylindrical specimens will be produced from one concrete batch and tested non-destructively (WP 1). The destructively determined compressive strengths (WP 2) and cyclic tests (WP 3) are statistically analyzed and form the basis of an experimental S-N curve adjusted for the scatter of the compressive strength. This is used to test the hypothesis. For validation, additional fatigue tests are carried out with a limited number of cycles (WP 5). During the experimental tests in WP 3 and 5, the cumulative dissipation energy is monitored as a damage indicator to predict the failure state of the specimens (WP 4). The residual compressive strength of all specimens from WP 5 that reach the defined limit number of cycles is determined and used for the statistical evaluation (WP 6). The experimental S-N curve from WP 3 is also used to create an analytical model of the S-N curve slope (WP 7). This model represents a test for the hypothesis and allows statements about the fatigue resistance due to the change in slope depending on the stress level.

• DFG Programme: Research Grants