Thermal displacements of concrete dams: Accounting for water temperature in statistical models

Measurements of concrete dam displacements are influenced by various factors such as hydrostatic load, thermal effect and irreversible phenomena (creep, swelling, etc.). To interpret measurements and improve the assessment of irreversible effects, splitting the different influences is necessary. For this purpose, models based on statistics and physics are commonly employed in engineering studies. Although they are efficient in most cases to analyse the displacement of concrete dams, these models are built on a certain number of hypotheses, necessary to write “simple” mathematical relationships, but leading to uncertainties. To evaluate the suitability of these physico-statistical models (importance of the hypotheses) and to improve it, a 2D finite element (FE) model has been developed as a heuristic case. This study shows the importance of water temperature and temperature gradient in the assessment of the thermal displacements. So, a new physico-statistical model is proposed to account for these phenomena. The evaluation of its performance on both the FE heuristic case and real cases shows that the improved assessment of thermal effects on reversible phenomena leads to a reduced uncertainty on residuals. Thus, the proposed approach yields to a better assessment of irreversible trends.