Using Corrosion in Seastate analysis


Product(s):SACS
Version(s):All
Area:Seastate

Community Question:

Hello, I am working on a Fatigue analysis, and when I review my results in the Deterministic wave/Transfer function analysis the override list of elements set in the card GRPOVAL is reported. I have reported this, but the developers haven't done anything to repair

Answer 

First of all, we need to keep in mind that when we apply corrosion during our analysis, the weight and sections calculated by the Seastate module for the effect of the corrosion environment will persist in all analyses, once you have included the card CORR in your analysis, you can't ignore this effect unless it is a static analysis.

PSI Analysis

Next, it shows the Modules that SACS active when a PSI analysis is executed and then, the workflow followed in which we can see the moment of the process in which the overrading is applied on elements. it's important to mention that in the early stage of the development of SACS, the corrosion option (CURR card) was thought to be used only in the static analysis (in place).

When you run static analysis, you can see that the next modules are active.

Below the Workflow in a PSI analysis is presented.

If you implement the CORR and GRPOVAL card in your model and execute a static analysis, the Seastate module will be activated and, in this step, the sections will be updated considering the effects of corrosion and they will be stored in a temporary seastate.oci file. Later it will be reported in the post-processor validating the sections and members in override and finally, with this, the weight and sections to be reported in the list will be calculated.

Extraction Shape Analysis

Next, to understand the inner process during a Dynamic analysis below is presented the modules active during Extract mode shapes. 

When you run a Dynamic - Extract Mode Shape, SACS active the modules in the next screenshot.

Below is presented the workflow followed by SACS during an "Extract Mode Shape Analysis" to explain the inner process when the card CORR is active. In the next Flowchart, we can see that Dynpac receives updated section dimensions for corroded members, and it uses those dimensions to create members' mass and then their weight. 

Dynamic-Deterministic Wave / Transfer Function  Analysis

When you run a Dynamic-Deterministic Wave / Transfer Function, SACS activates the other options.

We feed the program with these files; dynmod and dynmass which already consider the effect of corrosion in elements. These two files' data will be used to calculate the new sections and weight

As you can see, the Dynpac module is the 4th module active when you run Extract Mode Shape, here Dynpac receives updated section dimensions for corroded members, and it uses those dimensions to create members' mass and then their weight. If you set a list of override elements this won't be considered because the sections, come from the result of the Seastate module, and the validation is performed in the post-processor module.

Why does SACS do that?

From the beginning, the workflow thought was to consider the use of this card (CORR and GRPOVALonly in the Seastate module, in fact, all modules of SACS work in an isolated way, a good workflow is considered the element's weight as a load in elements and considers the corrosion only to review the section (without use the card DEAD. This will avoid variation in the weight of the structure.

As conclusion,

  1. SeaState does not communicate directly with Dynpac. The order of modules is SeaState > Pre-Processor > Solver > Dynpac. This causes the dynpac module takes the results of corroded sections directly from the seastate module which considers the effects of corrosion.
  2. SeaState calculates the model weight as a load applied to the model. The weight is a static load for static analysis, not the mass matrix for performing a dynamic analysis like wave response or earthquake analysis. 
  3. The Dynpac module calculates mass matrices, needed for mode extraction and dynamic analyses.

 
Example

simple structure comparison of the reported self-weight between static and dynamic analysis -

The results obtained in the static analysis and dynamic analysis isolated without using Corrosion, the upper listing (static listing) matches with the lower listing (Dynpac listing) -
 

Now we have the comparison for the same model using Corrosion in the analysis and considering a group of elements to override the corrosion using the card GRPOVAL 

Conclusion:

It does not matter if you set a list of elements using the card GRPOVAL as long as the result has to go through the dynamic module the effect of this will be null.