Theme A

Cracking of a concrete arch dam due to seasonal temperature variations

Formulators: Richard Malm (KTH/SWECO), Rikard Hellgren (KTH/WSP), Tomas Ekström (ÅF) and Chaoran Fu (ÅF)

Dams located in cold areas are often subjected to harsh environmental conditions and especially to large variations in temperatures between summer and winter. This variation in climate between summer and winter can be significant in the Northern countries, where for instance, in Sweden, many dams can be subjected to temperature variations of about 60 – 70 °C over a year. Previous studies have shown that many of these concrete dams have been subjected to cracking as a result. Hence, in Theme A of the 14th ICOLD Benchmark Workshop, participants are asked to predict the extent of cracking that may occur on a typical concrete arch dam subjected to the seasonal temperature variations in northern Sweden using numerical analyses.

In order to perform this task, thermal analyses are required to determine the temperature distribution within the dam body due to the environmental conditions. After this, mechanical analyses are performed to evaluate the induced stresses and strains based on these temperature distributions. The formulators request results from linear elastic analyses as well as results from analyses considering material nonlinearity. The participants can, however, chose to contribute to selected parts of the defined problem.

In the theme, the geometry, material properties and the loads have been defined and are delivered by the formulators. Some aspects of the numerical modelling are intentionally defined so that the participants have an opportunity make their own assumptions and choose suitable approaches to solve the problem. For instance, the participants may choose their own constitutive material models to describe non-linear material behaviour, how to perform the thermal analyses and the level of detail regarding boundary conditions and interactions. Thereby, by comparing the different solutions, it should possible to draw conclusions regarding how different assumptions and approaches influence the results.

The non-linear finite element method (NLFEM) relies on advanced numerical routines and may suffer from convergence and stability problems. The use such tools are furthermore often demanding for the practicing engineer, as the validity of the results relies on the accuracy of the models used. There is thus a need to evaluate the use of the NLFEM for the analysis of large concrete dams and to show the possibilities and restrictions associated with the method. The goal with this theme is that the outcome will give examples of best practices regarding non-linear modelling of cracking in concrete dams and how to consider the loads induced by seasonal temperature variation.


The instructions and data can be downloaded from the following link