How to create a N2 Diagram

[Mr-PDM]

Hello guys,

I'm working on a Optimization project for college where we are optimizing a horizontal stabilizer in an aircraft. I want to create de N2 Diagram for this project, but I'm a bit lost about how to do it, there are a lot of scripts and documents and I don't know where to start. I have defined the variables and the different connections between them, but I don't know how to write them so the scripts can recognize them or how to properly run the codes to obtain the N2 diagram.

Sorry, it is a vague question but I'm not very familiar with this level of programming.
Thanks for understanding and for your help, and good work because this repository is amazing :)

[JustinSGray]

Were you able to follow the docs on creating a basic N2 diagram? If not, where did you get confused?

The general workflow is to define a model, then use the openmdao command line tools to create an n2. You save your model script, than point the command line tools at it like:

openmdao n2 <your_run_script.py>

alternatively, you can generate the N2 directly in the script itself, by just passing your problem instance to the om.n2() function --- after you've called setup on the problem:

p.setup()

om.n2(p)

[naylor-b]

To keep things simple, just copy all of the code shown in the example (including Resistor, Diode, etc) into the same python file and call it 'my_script.py' for example, and then run the command line openmdao n2 my_script.py

[robfalck]

There are a few different ways to create an N2 diagram. Those that @JustinSGray listed work, but just calling problem.run_model() on a problem instance will, by default, generate an N2 diagram in a subdirectory named "reports".

Of course, this means that to get an N2 diagram from OpenMDAO you have to have built a model...or at least the skeleton of a model where various components have inputs and outputs that are interconnected - the compute methods don't have to be functional at this point.

If you're trying to demonstrate data flow between disciplines of your model, XDSM diagrams are also an option. The pyxdsm allows these to be developed and they're a bit more lightweight than N2 diagrams. Which you choose to use is just a matter of user preference and the situation, but I think XDSM diagrams are better for communicating the model functionality, while N2 diagrams are a more in-depth model debugging tool.

[Mr-PDM]

Thanks to everyone for your help, and I'm sorry couldn't be more concise in my question. I think the XDSM diagram could fit more in this situation, but I will keep working in the N2 Diagram if it's necessary. <3

[Mr-PDM]

Speaking about XDSM, i find this error when i run the examples, I can't find what I am missing

Traceback (most recent call last):
File "c:\Users-\OneDrive\Escritorio\pyXDSM-main\examples\mdf.py", line 31, in
x.write("mdf")
File "C:\Users-\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\LocalCache\local-packages\Python310\site-packages\pyxdsm\XDSM.py", line 703, in write
subprocess.run(command, check=True)
File "C:\Program Files\WindowsApps\PythonSoftwareFoundation.Python.3.10_3.10.3056.0_x64__qbz5n2kfra8p0\lib\subprocess.py", line 503, in run
with Popen(*popenargs, **kwargs) as process:
File "C:\Program Files\WindowsApps\PythonSoftwareFoundation.Python.3.10_3.10.3056.0_x64__qbz5n2kfra8p0\lib\subprocess.py", line 971, in init
self._execute_child(args, executable, preexec_fn, close_fds,
File "C:\Program Files\WindowsApps\PythonSoftwareFoundation.Python.3.10_3.10.3056.0_x64__qbz5n2kfra8p0\lib\subprocess.py", line 1456, in _execute_child
hp, ht, pid, tid = _winapi.CreateProcess(executable, args,
FileNotFoundError: [WinError 2] El sistema no puede encontrar el archivo especificado

[JustinSGray]

you should post an bug report with a sample script to that repo. Supporting pyXDSM is outside the scope of what can be solved here.

[JustinSGray]

If you want an XDSM instead, you can check out the pyXDSM repo.

I actually highly encourage folks to build XDSMs using pyXDSM before starting to build an OpenMDAO model. Alternatively, if you find the N2 more useful then you can build a skeleton OpenMDAO model that just has components with I/O and connections. But no actual compute code.