Application	PLAXIS 2D PLAXIS 3D
Version	PLAXIS 2D CONNECT Edition PLAXIS 3D CONNECT Edition
Date created	28 July 2022
Date modified	28 July 2022
Original author	Viviana Gonzalez - Bentley Technical Support Group

The PLAXIS Soiltest application is a tool in which you can simulate standard laboratory tests, like Triaxial, Oedoemeter, DSS, and CRS tests, to help you tweak the soil model parameters and match the laboratory tests. In the Soiltest application, PLAXIS also offers an optimization routine to quickly get an optimized set of parameters using real laboratory test results. This option can be selected from the toolbar of the SoilTest facility.

However, in some cases, we see that the optimisation does not give a result, and you are faced with the Small overlap curves message.

The following article describes the possible reasons related to the message “Small overlaps curves” on the Parameter Optimisation feature in PLAXIS SoilTest (please see the example in Fig.1).

Additionally, the following link explains how to use the Parameter Optimisation feature in SoilTest. 

Fig.1: Message “Small overlap curves” on the Parameter Optimisation feature in PLAXIS SoilTest.

The message “Small overlap curves” indicates that there is no appropriate match between the calculated and expected curves. This error happens when more than half of the computed points are outside the expected range.

We have identified some frequent causes that may be associated with the message: "Small overlaps curves." Below we will describe the most common causes and how to overcome these issues:

1. The optimised parameter does not directly influence the target curve e.g., it is not possible to find the optimised stiffness parameters from a p-q diagram.
   
   
2. The x- and/or y- axis variable is/are incorrect e.g., using strain values instead of stress values.
   
   
3. A reference test was not performed before the optimisation test. e.g., the user may go directly to Parameter Optimisation, load a consolidation test and try to optimise some parameters. However, the program has no synthetic test data to compare with the laboratory data. The appropriate procedure is explained in the video for SoilTest's Parameter Optimisation tool.
   
   
4. The sign of the stresses is incorrect. In PLAXIS a negative stresses increment implies compression and a positive stresses increment implies unloading or extension.
   
   
5. The values of the strains are incorrect. The strain is expressed as a unitless value and not as a percentage value.
   
   
6. There are too many experimental points/data provided for the optimisation. Removing some values can help overcome the error.
   
   
7. The correct part of the curve for a specific parameter is not selected. To mitigate this error, it is necessary to select a proper range of experimental data to be used for the optimisation of the selected input parameter (e.g., for the Hardening soil model, choose the first part for E50 and unloading/reloading part for Eur).
   
   
8. The original and optimised data are equivalent. Therefore, overlapping charts were created. The user may be optimising the model constants of their constitutive model using results produced by the same constitutive model, instead of experimental data. In order to solve this issue, it is necessary to have laboratory data and import it using the Import curve option.
   
   
9. The correct phase of the curve for a specific parameter has not been selected. In the case of an Oedometer, CRS or General test, the SoilTest facility allows for multiple phases. However, the parameter optimisation facility can only deal with one complete phase at a time. Therefore, after importing the test data, the desired calculation phase needs to be selected from the drop-down list above the test data curve, together with the corresponding part of the test data in the column on the right-hand side. Fig.2 presents an example of an oedometer test.

Fig.2: Example of selecting a phase of the curve in an oedometer test.