Hello everyone, I have a question regarding creep modeling in Abaqus.

So, I want to make a creep model in Abaqus using the "Viscoelastic - Relaxation Test Data" method by using combined test data. I input both normalized shear and volumetric modulus as a function of the creep coefficient from the standard, as presented in Figure 1. I also inputted the elastic modulus and Poisson's ratio as shown in Figure 2 and didn't forget to change the moduli time scale to "instantaneous". I inputted the normalized shear and volumetric modulus as calculated from the equation in Figure 1 to the abaqus as seen in Figure 3, where the long-term normalized shear and volumetric modulus was taken from the latest value from the time case (393 days). I made the model as a 2D single-element plane stress (CPS4R).

I defined several steps to provide the analyzed time, as presented in Figure 4. The defined time was in seconds. The applied load was 10 MPa, and the boundary condition was displacement/rotation type by ticking all U1, U2, and UR3 to make it a fixed support, as shown in Figure 5. After the job was completed, I compared the Abaqus results with the numerical results from the standard, and I got the results as seen in Figures 6 and 7 for creep and total strain, respectively.

As you can see in the results, the creep strain at 28 days from Abaqus results is not 0 when it should be 0 according to the numerical calculation. It also affected the total strain result. Is it because I defined the time as 0 days? Because, as I have seen, the creep strain at 0 days is 0. Should I rearrange the analyzed time so 28 days will act like 0 days? But if I make it like that, the elastic strain at 0 days (which is actually 28 days) will be 0 as well when it should be stress/E_c28.

And then, I also tried to use the creep test data option as well, where the normalized shear and volumetric modulus are referred to in Figure 8. I inputted the normalized shear and volumetric modulus as presented in Figure 9, where the long-term normalized shear and volumetric modulus was taken from the latest value from the time case (393 days). The results for this model are presented in Figures 10 and 11 for creep and total strains, respectively.

As you can see, the results have the same problem as the relaxation test data model. And also, I wonder why the result is different between the relaxation and creep test data models. Is it because I set the allowable average root-mean-square error differently? As we know, the default value for the error is 0.01, and since I keep getting aborted, I increase the error value for each model until it is completed.

Thank you so much for the answers and insights from you all. Have a nice day!