Title:
Coupled Pore Relative Humidity Model for Concrete Shrinkage and Creep
Author(s):
Brock D. Hedegaard, Timothy J. Clement, Mija H. Hubler
Publication:
CRC
Volume:
Issue:
Appears on pages(s):
52
Keywords:
DOI:
Date:
1/1/2022
Abstract:
A new semi-empirical concrete shrinkage and creep model, called the CPRH Model, is proposed and calibrated. The new model proposes a coupling between autogenous and drying shrinkage using a volume-average pore relative humidity and treats drying creep as an additional stress-dependent shrinkage, linking together all these phenomena. The proposed expressions are designed to facilitate traditional integral-type analysis, but also uniquely support rate-type calculations that can be leveraged by analysis software. Model calibration uses the Northwestern University (NU) database of creep and shrinkage tests to determine new model parameters. The proposed model uses minimal inputs that are often known or may be assumed by the design engineer. Comparison of the proposed model to historical time-dependent models indicates that the new model provides a superior fit over a wider range of inputs. A New Way to Predict Creep and Shrinkage Traditional time-dependent analysis has relied on definition of a compliance function or creep coefficient. For time-varying stresses, the strain may be approximated (for example, through the age-adjusted modulus method) or computed using integral-type analysis. Rate-type analysis does not require computation of an integral over the entire stress history; thus, it is more efficient and accurate for more complex structural analysis. Previously, no existing time dependent model other than the B3/B4 basic creep expression had a convenient form for performing rate-type analysis. The model developed over the course of this research project changes this reality. The new model has closed form expressions for the compliance function and compliance rate, uniquely supporting both analysis approaches. These features place the model on the cutting edge of time-dependent structural analysis. The calibration was conducted in three steps: 1. Database management and preparation. 2. Identification of appropriate input parameters and calibration by nonlinear optimization; and 3. Statistical comparison and final model selection per information theory.