What cause the sudden drop of from my FEA model

[n39]

I was tring to do a FEA static general analysis using ABAQUS on pile to pile cap retrofitted by steel jacket. I did the experiment test earlier and get the results (the blue line). When I was tryng to model it, I use hard contact with both tangential and normal behavior on my model (The red line), but the results shows there is a sudden drop that I don't know why. When I use tie interaction on the steel jacket retrofitting, the results also doesn't give a good results (the gray line). Based on the experiment a hard contact interaction should represent the actual experiment, but my FEA shows otherwise. What was the cause of that sudden drop (the red line) and what can I do to fix it?

 

[FEA way]

It’s hard to say based on just this plot. Plots like this should be interpreted together with other results such as contour plots, animations of the deformed shape and so on. The setup of the analysis (geometry, mesh, BCs, loads, interactions) is also important.

 

[n39]

Can you help me check my model sir? I need a consultation about my model, it's for my thesis.