MPCO recorder for moment curvature analysis

[mike_oleary]

Hi, I am trying to use the mpco recorder for a moment curvature fibre section analysis. I am not using this via stko but rather just TCL files. I have had good success with the recorder for models but I'm struggling with how to use it for moment curvature analysis. Unfortunately the documentation on the OpenSees wiki has nothing, so I'm a little stuck with what to do. Is there any documentation on using the recorder? (i.e. what are all the options available and how to use for a moment curvature analysis). If not it's there someone here that has used it for moment curvature analysis?

[STKO Team]

You can find the doc here:
https://asdea.eu/software/stko-documentation/

If you have issues, attach here your tcl file and we will have a look

[mike_oleary]

I've attached a screen shot of the error I am getting and also an example file. Note that if the mpco recorder is removed and the node recorder used instead it runs fine (I couldn't seem to attach a text file).

Code: Select all

wipe

# Model

model Basic -ndm 3 -ndf 6

# Materials

uniaxialMaterial Concrete04 1 30000000 0.002 0.003 800000 0 0 0
uniaxialMaterial ElasticBilin 2 200000000000 0 0.0025 200000000000 0 0.0025

# Sections

section Fiber 1 -GJ 1000000 {
	fiber 0 -0.475 0.05 1
	fiber 0 -0.425 0.05 1
	fiber 0 -0.375 0.05 1
	fiber 0 -0.32499999999999996 0.05 1
	fiber 0 -0.27499999999999997 0.05 1
	fiber 0 -0.225 0.05 1
	fiber 0 -0.17499999999999996 0.05 1
	fiber 0 -0.12499999999999997 0.05 1
	fiber 0 -0.07499999999999998 0.05 1
	fiber 0 -0.024999999999999988 0.05 1
	fiber 0 0.025 0.05 1
	fiber 0 0.07500000000000004 0.05 1
	fiber 0 0.12500000000000008 0.05 1
	fiber 0 0.17500000000000002 0.05 1
	fiber 0 0.22500000000000006 0.05 1
	fiber 0 0.275 0.05 1
	fiber 0 0.32500000000000007 0.05 1
	fiber 0 0.3750000000000001 0.05 1
	fiber 0 0.42500000000000004 0.05 1
	fiber 0 0.4750000000000001 0.05 1
	fiber -0.4 -0.4 0.0012566370614359172 2
	fiber 0.4 -0.4 0.0012566370614359172 2
}

# Nodes

node 1 0 0 0
node 2 0 0 0

# Supports

fix 1 1 1 1 1 1 1
fix 2 0 1 1 1 1 0

# Elements

element zeroLengthSection 1 1 2 1 -doRayleigh False

recorder mpco D:\\temp\\nodempco.out -N displacement
# recorder Node -file D:\\temp\\node.out -precision 6 -time -node 2 -dof 6 disp

# Time Series

timeSeries Constant 1 -factor 1

# Time Patterns

pattern Plain 1 1 1 {
	load 2 1000 0 0 0 0 0
}

constraints Plain
numberer Plain
system UmfPack
test EnergyIncr 1E-08 10 0 2
algorithm Newton
integrator LoadControl 0 1 0 0
analysis Static
analyze 1    


# Time Series

timeSeries Linear 2 -factor 1

# Time Patterns

pattern Plain 2 2 1 {
	load 2 0 0 0 0 0 1
}

constraints Plain
numberer Plain
system UmfPack
test EnergyIncr 1E-08 10 0 2
algorithm Newton
integrator DisplacementControl 2 6 2.5E-05 1 2.5E-05 2.5E-05
analysis Static
analyze 100    

[STKO Team]

It's a problem with the zeroLengthSection: it does not format the output in such a way that the MPCO Recorder is able to understand the shape of the fiber section. We just changed the code and will do a Pull Request to the OpenSees Source code.

In the meantime, instead of a zeroLength section, use a unit-length beam element. In this way its rotation will be equal tot the curvature

[mike_oleary]

Thanks! Sorry I'm late to reply, I've been busy on other stuff since I first tried this and now am going to look back into switching over to the mpco recorder. I'll learn more once I use it, but maybe you'll have the answer for me. I'm curious how the stress/strains of fibres are recorded. If I recall correctly setting up some of the standard recorders to record fibres either requires an tag (which is not known in the TCL files) or using x,y coordinates (which I don't love either). Will reach fibre be recorded with some metadata about the fibre to provide some way to collerate where in the section it belongs?

[kesavapraba]

MPCO recorder option sends the results into a HDF5 database (you can see these files after the analysis) for the chosen options. For example, In the case of section.fiber.stress, the recorder identifies internally all the fibers using its algorithm and stores the data into HDF5 file. So, you are not required to provide either coordinates or fiber tag.

[mike_oleary]

Perfect, thanks

[STKO Team]

You're welcome