Carlos Andres Rivera Villarreyes Global Product Specialist - FEFLOW

We have noticed this problem in the past and was mainly associated to the scaling value given in the Display Settings in Windows 10.

Hi,
Without knowing the model settings, I think the problem can be managed by setting different weights for the observations. You can run FePEST with the NOPT option (Termination Criteria) to estimate one full Jacobian matrix. This can give you some insights about the sensitivities.

Cheers, Carlos

There are two types of equations in FEFLOW: those used for the assignment and those, which will dynamically be evaluated through the model run. In the case of boundary conditions, the expressions are only supported for assignments (before the model run). As soon as you click the buttom "Assign" the expression is evaluated and longer no more.
The second case of dynamic expression is supported for several material propeties. This is an option implemented in one of the latest release. The overview of parameters supported is below:

http://www.feflow.info/html/help74/feflow/09_Parameters/user_defined_equations.html

For your specific question, the easiest manner to control the boundary is via a simple Python function, which can be stored in the FEM file and call it through the run. The Support team (mike.de@dhigroup.com) could give you some hints about the usage.

Cheers
Carlos Rivera

Probably the current FePEST is missing some results. Typically, you will require to have a Jacobian matrix (*.jco file) computed from an Estimation run.

Regards, Carlos Rivera

Yes, this is technically possible. You can create an elemental distribution and link this to a specific parameter, e.g. In/out-flow on top/bottom. There are two manners to do this task: 1) using FEFLOW Expression Editor and/or 2) Python function.

Carlos

Dear Emilia,

The 3D mesh smooth option is mainy intented for 3D unstructured meshes. If you are having a mesh with prisms, I would recommend you to use "2D Mesh Smooth (with X-Y inheritance in 3D".

Best regards
Carlos Rivera

Hi Jimmy,

You have several options depending on the type of domain (2D / 3D) and mesh (quad, triangular, prism, tetrahedra, etc.). All these are available through the Data panel under the section "Auxiliary parameters".

Cheers
Carlos

Hi swhit

The answer really depends on the application case. In the case of M21 - FEFLOW coupling, you provide a framework for fully-hydrodynamical coupling. MIKE 21 FM is responsible to solve the 2D hydraulic and send this feedback to FEFLOW. A typical application would be the assessment of flooding events and their interaction with groundwater. Coupling supports mass and heat transport also.
In the case of MSHE - FEFLOW, MSHE looks the problem at the catchment hydrology and supports information not available in FEFLOW, e.g. role of land-use and vegetation in the net recharge estimation, etc. A common application is to estimate groundwater recharge through MSHE and this includes all hydrological cycle. Subsequently such information can be used in a detailed groundwater model in FEFLOW.
We have applied several coupling mechanisms in multiple projects. Let me know if you need more details.

Regards
Carlos Rivera

Hi Jimmy,
Unfortunately, the error message is not very clear. Maybe you are trying to assign a parameter type unsupported by PEST. I would recommend you to get in touch with the technical support (mike.de@dhigroup.com). They can investigate the problem deeply and give you a solution.

Best regards
Carlos Rivera

Hi,

You can use the following script to understand the syntaxis with MLW.

    ## Define Boundary Conditions for Flow
    # NoBC = 0
    # Dirichlet (1st kind) =1
    # Neumann (2nd kind) = 2
    # Cauchy (3rd kind) = 3
    # Single well (4th kind) = 4
    # Gradient-type (only for unsaturated problems) = 5
    BcFlowType = 4

    #Load documents
    doc=ifm.loadDocument(FEM_FILE)

    #Print all Multilayer Well attributes
    for nNode in range(0,doc.getNumberOfNodes()):

        if doc.getBcFlowType(nNode) == BcFlowType:
            if doc.queryMultiLayerWellInfo(nNode) == None:
                print(Normal Well BC found")
            else:
                MLWInfo=doc.queryMultiLayerWellInfo(nNode)
                # Print MLW ID, Name, Top Elevation, Bottom Elevation, Top Node, Bottom Node
                print(MLWInfo.getId(), MLWInfo.getName(), MLWInfo.getTopElevation(),
                    MLWInfo.getBottomElevation(), MLWInfo.getTopNode(), MLWInfo.getBottomNode())