file: optimize.py
at about line 400 in method any_constraints_all

    # Minimum plate thickeness
    if chk[1]:
        if not obj.get_plate_thk()>calc_object[0].get_dnv_min_thickness(lat_press)/1000:
            return False

to be replaced by

    # Minimum plate thickeness
    if chk[1]:
        if not calc_object[0].get_plate_thk()>calc_object[0].get_dnv_min_thickness(lat_press)/1000:
            return False

Implement reading and writing point, lines and section to and from GeniE. Pressures too?

Hello,

I was testing the linux modifications on windows and came across the following issue:
When loading the built-in sections, the file is not found. This is because the .csv file with the sections is in ANYstructure_local, which is a sub-dir of the directory containing main.py.

A better testing framework must be implemented.

Hi @audunarn,

Is it possible the psi_x on line 1471 in calc_structure.py is missing the check for negative value in the sqrt?
a few lines further psi_y has this check.

Regards Frederik

Hi Audun,

I think there is a small bug in the code for unstiffened curved.

calc_structure.py:
In the function unstiffened_shell, the sign for the bending stress is set zero or the absolute value depending on the sign (line 2961). This is fine, but changes the value of smsd.

A bit further in the code, '3.4 unstiffened circular cylinders', the same is done for sm0sd (line 3135). However, this uses the value of smsd, which is changed earlier, resulting in the bending stress not correctly being taken into account.

I would suggest moving the calculation of sa0sd, sm0sd to the beginning of the function, only doing it once and used these values in the further calculations.

Hope my explanation is clear.
Regards,
Frederik

After implementing multiprocessing, the timer for optimizing a plate field is somewhat strange. Also it is located the wrong place.

The code contains a geometry optimization code. It will automatically calulate the best plate field length and number of plate fields for a selected distance. It is 70% complete and is not activated in the application. It will be a significat enhancement to complete it.

optimize_geometry.py

Need to make a short documetation.

Hello,

I get the following error when running the program (both via the GUI and If I run the code itself with example_data.py):

  File "/home/frederik/theScriptingEngineer/Python/ANYstructure/ANYstructure_local/calc_structure.py", line 1895, in stiffened_panel
    stf_pl_data['UF Plate side'] = res_iter_pl.fun[0]
TypeError: 'float' object is not subscriptable

Unfortunately I don't understand the code enough to propose a solution...

More insight into the results to be added. What fails and what is OK.

Undo works fine, but redo does not.

Exception in Tkinter callback
Traceback (most recent call last):
File "C:\ProgramData\Anaconda3\lib\tkinter_init_.py", line 1699, in call
return self.func(*args)
File "C:\Users\any\Code\PyStructural\ANY_structural\optimize_multiple_window.py", line 396, in run_optimizaion
fp['p_ext']['loaded'], fp['p_ext']['ballast'], fp['p_ext']['part'] = 0,0,0
KeyError: 'p_ext'
Exception in Tkinter callback
Traceback (most recent call last):
File "C:\ProgramData\Anaconda3\lib\tkinter_init_.py", line 1699, in call
return self.func(*args)
File "C:\Users\any\Code\PyStructural\ANY_structural\optimize_multiple_window.py", line 396, in run_optimizaion
fp['p_ext']['loaded'], fp['p_ext']['ballast'], fp['p_ext']['part'] = 0,0,0
KeyError: 'p_ext'

hello i am facing this error when i try to open program
Traceback (most recent call last):
File "PyInstaller\loader\pyimod03_ctypes.py", line 77, in init
File "ctypes_init_.py", line 376, in init
FileNotFoundError: Could not find module 'shcore' (or one of its dependencies). Try using the full path with constructor syntax.

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
File "main.py", line 23, in
File "main.py", line 11, in main
File "ctypes_init_.py", line 446, in getattr
File "PyInstaller\loader\pyimod03_ctypes.py", line 79, in init
pyimod03_ctypes.install..PyInstallerImportError: Failed to load dynlib/dll 'shcore'. Most likely this dynlib/dll was not found when the application was frozen.

Hello,

When installing the requirements from scratch and then running main.py, an error occurs because sklearn cannot be found.
This can be resolved by installing scikit-learn. So I think this needs to be added to the requirements.

The same applies when installing using pip install anystructure

Regards,
Frederik

Hello,

I wanted to run ANYstructure on linux and I managed to do so. I did need to implement the following 2 changes:

1. main_application.py, line 330: change \ to / I think using os.path.join() would solve the problem, or to keep the string concatenation os.sep might also work.
2. main_application.py, line 67: vista theme did not work, but changing to classic did the trick.

Kind regards

More info:
stackoverflow for 1

Many calculations can be vectorized.

The code is partially a mess. Should be cleaned up.

Variables uses a mix of ._ and just .
Identify the variables that is not private and remove the _

I'm asking this question because I'm running into this myself and am wondering if you or anyone else had given it some thought already.

Some day (soon) people will start using this executable and will be producing reports with it. But how can they be sure that what comes out is correct? Some software packages manage to get an "approved" stamp from some certification authority while others do not.

It would be nice if some sponsor (DNVGL...?) would review the code and give it an approval or a recommendation.

Results should be cached. That way recalculations is not required all the time.

Hi!

I see you made plans to shift to Kivy, is this still something you are considering?

Train a neural network on more complex buckling calculation tools (e.g. PULS) and implement a keras version of it in this application.

ISO, ship rules etc. can be implemented.

Hi @audunarn

In calc_structures: unstiffened_plate_buckling()
line 1515: sxRd = Cxfy/gammaM if not all([sig_x1<0, sig_x2<0]) else 1
should be:
sxRd = Cxfy/gammaM if not all([sig_x1<0, sig_x2<0]) else fy/gammaM
I think.

Resistance factor for tension is yield divided by material factor, not 1

Regards,
Frederik

Implement a non dimensional cost number. Should be dependent on weight and weld length. But how to choose which one matters the most.....