NSGA-II (Non-dominated Sorting Genetic Algorithm II) can be effectively combined with process operability analysis to optimize the design and control of chemical processes by Including the operability set and its defining criteria (achievable inputs/outputs, desired operating regions, disturbances) as constraints in the NSGA-II optimization. NSGA-II will generate a set of Pareto optimal solutions that represent different trade-offs between design and operability objectives at specific variables. I believe that combining NSGA-II with process operability analysis, you can design and control chemical processes that are not only optimal in terms of desired objectives but also robust and operable under practical conditions. This integrated approach helps to avoid the pitfalls of sequential design and control and leads to more sustainable and efficient chemical production.

Remove the "borders"

Hello,

When I try to run the jupyter notebook for the shower example, I get the following error. I'm not sure why this is or what to do about it. Any help would be greatly appreciated! Specifically, this is from operability_index_shower.ipynb and from the following line of code:

AOS_region = multimodel_rep(model, AIS_bounds, AIS_resolution)

Thanks!

See below for the error.

TypeError Traceback (most recent call last)
Cell In[9], line 1
----> 1 AOS_region = multimodel_rep(model, AIS_bounds, AIS_resolution)

File ~\Anaconda3\envs\opyr2\lib\site-packages\opyrability.py:205, in multimodel_rep(model, bounds, resolution, polytopic_trace, perspective, plot, EDS_bound, EDS_resolution)
202 AS_coords = np.concatenate(Vertices_list, axis=0)
203 for i in range(len(mapped_region)):
--> 205 polyplot = _get_patch(mapped_region[i], linestyle="solid",
206 edgecolor=EDGES_COLORS, linewidth=EDGES_WIDTH,
207 facecolor=AS_COLOR)
208 ax.add_patch(polyplot)
211 lower_xaxis = mapped_region.bounding_box[0][0]

File ~\Anaconda3\envs\opyr2\lib\site-packages\polytope\polytope.py:2319, in _get_patch(poly1, **kwargs)
2317 ind = np.argsort(angle)
2318 # create patch
-> 2319 patch = mpl.patches.Polygon(V[ind, :], True, **kwargs)
2320 patch.set_zorder(0)
2321 return patch

TypeError: init() takes 2 positional arguments but 3 were given

Add cyipopt, JAX, matplotlib, glpk to the initial install

@jkitchin and @sanqdinh,

I have tested opyrability's installation, importing and usage on win64/linux64 sucessfully. Unfortunately, I don't have a macOS system. Does any of you do? If yes, do you mind testing the installation methods from https://github.com/CODES-group/opyrability#installation?

The pip one is a two-step procedure due to cyipopt's availability only on conda-forge. The conda "method" installs everything that is needed in a single terminal command (all is described in the link above).

Thanks in advance!

When space is 1D, OI_eval breaks due to

  File "...\shower_operability.py", line 24, in main
    OI = opy.OI_eval(TDS_region, EDS_bounds, perspective='inputs')
         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "...\opyrability.py", line 423, in OI_eval 
    volumes_i.append(sp.spatial.ConvexHull(v_intersect).volume)
                     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "_qhull.pyx", line 2431, in scipy.spatial._qhull.ConvexHull.__init__
  File "_qhull.pyx", line 279, in scipy.spatial._qhull._Qhull.__init__
ValueError: Need at least 2-D data

Either

• Correct plot warning when dimension is not 2 to display non-2-dimensional instead of 3 or greater
• Add verbosity to give different warnings when when dimensions are either 1 or >=3

pip install opyrability fails when environment is using newer than python 3.10 due to a failure with numpy

Managed to install multiple versions somehow

opyrability         1.4.7
opyrability         1.4.9
opyrability         1.4.9