Robustness#

In this example, we illustrate the use of the Robustness plot on the Sobieski's SSBJ problem.

from __future__ import annotations

from gemseo import configure_logger
from gemseo import create_discipline
from gemseo import create_scenario
from gemseo.problems.mdo.sobieski.core.design_space import SobieskiDesignSpace

Import#

The first step is to import some high-level functions and a method to get the design space.

configure_logger()
<RootLogger root (INFO)>

Description#

In the Robustness post-processing, the robustness of the optimum is represented by a box plot. Using the quadratic approximations of all the output functions, we propagate analytically a normal distribution with 1% standard deviation on all the design variables, assuming no cross-correlations of inputs, to obtain the mean and standard deviation of the resulting normal distribution. A series of samples are randomly generated from the resulting distribution, whose quartiles are plotted, relatively to the values of the function at the optimum. For each function (in abscissa), the plot shows the extreme values encountered in the samples (top and bottom bars). Then, 95% of the values are within the blue boxes. The average is given by the red bar.

Create disciplines#

At this point, we instantiate the disciplines of Sobieski's SSBJ problem: Propulsion, Aerodynamics, Structure and Mission

disciplines = create_discipline([
    "SobieskiPropulsion",
    "SobieskiAerodynamics",
    "SobieskiStructure",
    "SobieskiMission",
])

Create design space#

We also create the SobieskiDesignSpace.

design_space = SobieskiDesignSpace()

Create and execute scenario#

The next step is to build an MDO scenario in order to maximize the range, encoded 'y_4', with respect to the design parameters, while satisfying the inequality constraints 'g_1', 'g_2' and 'g_3'. We can use the MDF formulation, the SLSQP optimization algorithm and a maximum number of iterations equal to 100.

scenario = create_scenario(
    disciplines,
    "y_4",
    design_space,
    formulation_name="MDF",
    maximize_objective=True,
)
scenario.set_differentiation_method()
for constraint in ["g_1", "g_2", "g_3"]:
    scenario.add_constraint(constraint, constraint_type="ineq")
scenario.execute(algo_name="SLSQP", max_iter=10)
WARNING - 00:56:27: Unsupported feature 'minItems' in JSONGrammar 'SobieskiMission_discipline_output' for property 'y_4' in conversion to SimpleGrammar.
WARNING - 00:56:27: Unsupported feature 'maxItems' in JSONGrammar 'SobieskiMission_discipline_output' for property 'y_4' in conversion to SimpleGrammar.
   INFO - 00:56:27:
   INFO - 00:56:27: *** Start MDOScenario execution ***
   INFO - 00:56:27: MDOScenario
   INFO - 00:56:27:    Disciplines: SobieskiAerodynamics SobieskiMission SobieskiPropulsion SobieskiStructure
   INFO - 00:56:27:    MDO formulation: MDF
   INFO - 00:56:27: Optimization problem:
   INFO - 00:56:27:    minimize -y_4(x_shared, x_1, x_2, x_3)
   INFO - 00:56:27:    with respect to x_1, x_2, x_3, x_shared
   INFO - 00:56:27:    subject to constraints:
   INFO - 00:56:27:       g_1(x_shared, x_1, x_2, x_3) <= 0
   INFO - 00:56:27:       g_2(x_shared, x_1, x_2, x_3) <= 0
   INFO - 00:56:27:       g_3(x_shared, x_1, x_2, x_3) <= 0
   INFO - 00:56:27:    over the design space:
   INFO - 00:56:27:       +-------------+-------------+-------+-------------+-------+
   INFO - 00:56:27:       | Name        | Lower bound | Value | Upper bound | Type  |
   INFO - 00:56:27:       +-------------+-------------+-------+-------------+-------+
   INFO - 00:56:27:       | x_shared[0] |     0.01    |  0.05 |     0.09    | float |
   INFO - 00:56:27:       | x_shared[1] |    30000    | 45000 |    60000    | float |
   INFO - 00:56:27:       | x_shared[2] |     1.4     |  1.6  |     1.8     | float |
   INFO - 00:56:27:       | x_shared[3] |     2.5     |  5.5  |     8.5     | float |
   INFO - 00:56:27:       | x_shared[4] |      40     |   55  |      70     | float |
   INFO - 00:56:27:       | x_shared[5] |     500     |  1000 |     1500    | float |
   INFO - 00:56:27:       | x_1[0]      |     0.1     |  0.25 |     0.4     | float |
   INFO - 00:56:27:       | x_1[1]      |     0.75    |   1   |     1.25    | float |
   INFO - 00:56:27:       | x_2         |     0.75    |   1   |     1.25    | float |
   INFO - 00:56:27:       | x_3         |     0.1     |  0.5  |      1      | float |
   INFO - 00:56:27:       +-------------+-------------+-------+-------------+-------+
   INFO - 00:56:27: Solving optimization problem with algorithm SLSQP:
   INFO - 00:56:27:     10%|█         | 1/10 [00:00<00:00, 24.25 it/sec, obj=-536]
   INFO - 00:56:27:     20%|██        | 2/10 [00:00<00:00, 18.71 it/sec, obj=-2.12e+3]
WARNING - 00:56:27: MDAJacobi has reached its maximum number of iterations, but the normalized residual norm 5.741449586530469e-06 is still above the tolerance 1e-06.
   INFO - 00:56:27:     30%|███       | 3/10 [00:00<00:00, 15.41 it/sec, obj=-3.46e+3]
   INFO - 00:56:27:     40%|████      | 4/10 [00:00<00:00, 14.75 it/sec, obj=-3.96e+3]
   INFO - 00:56:27:     50%|█████     | 5/10 [00:00<00:00, 15.11 it/sec, obj=-4.61e+3]
   INFO - 00:56:28:     60%|██████    | 6/10 [00:00<00:00, 16.12 it/sec, obj=-4.5e+3]
   INFO - 00:56:28:     70%|███████   | 7/10 [00:00<00:00, 16.69 it/sec, obj=-4.26e+3]
   INFO - 00:56:28:     80%|████████  | 8/10 [00:00<00:00, 17.16 it/sec, obj=-4.11e+3]
   INFO - 00:56:28:     90%|█████████ | 9/10 [00:00<00:00, 17.55 it/sec, obj=-4.02e+3]
   INFO - 00:56:28:    100%|██████████| 10/10 [00:00<00:00, 17.87 it/sec, obj=-3.99e+3]
   INFO - 00:56:28: Optimization result:
   INFO - 00:56:28:    Optimizer info:
   INFO - 00:56:28:       Status: None
   INFO - 00:56:28:       Message: Maximum number of iterations reached. GEMSEO stopped the driver.
   INFO - 00:56:28:       Number of calls to the objective function by the optimizer: 12
   INFO - 00:56:28:    Solution:
   INFO - 00:56:28:       The solution is feasible.
   INFO - 00:56:28:       Objective: -3463.120411437138
   INFO - 00:56:28:       Standardized constraints:
   INFO - 00:56:28:          g_1 = [-0.01112145 -0.02847064 -0.04049911 -0.04878943 -0.05476349 -0.14014207
   INFO - 00:56:28:  -0.09985793]
   INFO - 00:56:28:          g_2 = -0.0020925663903177405
   INFO - 00:56:28:          g_3 = [-0.71359843 -0.28640157 -0.05926796 -0.183255  ]
   INFO - 00:56:28:       Design space:
   INFO - 00:56:28:          +-------------+-------------+---------------------+-------------+-------+
   INFO - 00:56:28:          | Name        | Lower bound |        Value        | Upper bound | Type  |
   INFO - 00:56:28:          +-------------+-------------+---------------------+-------------+-------+
   INFO - 00:56:28:          | x_shared[0] |     0.01    | 0.05947685840242058 |     0.09    | float |
   INFO - 00:56:28:          | x_shared[1] |    30000    |   59246.692998739   |    60000    | float |
   INFO - 00:56:28:          | x_shared[2] |     1.4     |         1.4         |     1.8     | float |
   INFO - 00:56:28:          | x_shared[3] |     2.5     |   2.64097355362077  |     8.5     | float |
   INFO - 00:56:28:          | x_shared[4] |      40     |  69.32144380869019  |      70     | float |
   INFO - 00:56:28:          | x_shared[5] |     500     |  1478.031626737187  |     1500    | float |
   INFO - 00:56:28:          | x_1[0]      |     0.1     |         0.4         |     0.4     | float |
   INFO - 00:56:28:          | x_1[1]      |     0.75    |  0.7608797907508461 |     1.25    | float |
   INFO - 00:56:28:          | x_2         |     0.75    |  0.7607584987262048 |     1.25    | float |
   INFO - 00:56:28:          | x_3         |     0.1     |  0.1514057659459843 |      1      | float |
   INFO - 00:56:28:          +-------------+-------------+---------------------+-------------+-------+
   INFO - 00:56:28: *** End MDOScenario execution (time: 0:00:00.565534) ***

Post-process scenario#

Lastly, we post-process the scenario by means of the Robustness which plots any of the constraint or objective functions w.r.t. the optimization iterations or sampling snapshots.

Tip

Each post-processing method requires different inputs and offers a variety of customization options. Use the high-level function get_post_processing_options_schema() to print a table with the options for any post-processing algorithm. Or refer to our dedicated page: Post-processing algorithms.

scenario.post_process(post_name="Robustness", save=False, show=True)
Boxplot of the optimization functions with normalized stddev 0.01
<gemseo.post.robustness.Robustness object at 0x7f192f53ce00>

Total running time of the script: (0 minutes 0.717 seconds)

Gallery generated by Sphinx-Gallery