Abstract
<jats:p>The article addresses the problem of numerically modeling creep deformations in modern high-strength modified self-compacting concretes. The relevance of the research is driven by the imperfections in the regulatory framework, specifically SP 63.13330.2018, which offers an overly simplified and conservative approach to accounting for creep. This approach fails to consider the specific characteristics of modern concretes and does not enable reliable forecasting. Conversely, the existing, more detailed domestic recommendations from 1984 are technically obsolete and inapplicable to high-strength concretes. The aim of the study is to assess the applicability of modern and outdated computational creep models for predicting deformations in high-strength concrete across different humidity conditions. The research was conducted using numerical modeling in the ATENA software complex, with verification of the results against experimental data. The results demonstrated that the direct application of even the most modern models (B3, MC 2010) leads to significant errors (ranging from 13% to 130%), which vary depending on humidity. Models from the 1970s-80s (exemplified by ACI-78) proved entirely inapplicable, showing discrepancies of up to 600% compared to experimental data. The necessity of modifying models not only for the specific concrete composition but also for its curing conditions and humidity levels is demonstrated. The study concludes that the verification and calibration of computational models based on experimental data are critically important before modeling complex stress-strain states. An action algorithm is proposed to enhance the accuracy of creep prediction under multiaxial loads, which includes mandatory experimental verification of boundary conditions.</jats:p>