Variables influence¶

In this example, we illustrate the use of the VariableInfluence 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.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¶

The VariableInfluence post-processing performs first-order variable influence analysis.

The method computes \(\frac{d f}{d x_i} \cdot \left(x_{i_*} - x_{initial_design}\right)\), where \(x_{initial_design}\) is the initial value of the variable and \(x_{i_*}\) is the optimal value of the variable.

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,
    "MDF",
    "y_4",
    design_space,
    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": "SLSQP", "max_iter": 10})

    INFO - 09:03:51:
    INFO - 09:03:51: *** Start MDOScenario execution ***
    INFO - 09:03:51: MDOScenario
    INFO - 09:03:51:    Disciplines: SobieskiAerodynamics SobieskiMission SobieskiPropulsion SobieskiStructure
    INFO - 09:03:51:    MDO formulation: MDF
    INFO - 09:03:51: Optimization problem:
    INFO - 09:03:51:    minimize -y_4(x_shared, x_1, x_2, x_3)
    INFO - 09:03:51:    with respect to x_1, x_2, x_3, x_shared
    INFO - 09:03:51:    subject to constraints:
    INFO - 09:03:51:       g_1(x_shared, x_1, x_2, x_3) <= 0.0
    INFO - 09:03:51:       g_2(x_shared, x_1, x_2, x_3) <= 0.0
    INFO - 09:03:51:       g_3(x_shared, x_1, x_2, x_3) <= 0.0
    INFO - 09:03:51:    over the design space:
    INFO - 09:03:51:       +-------------+-------------+-------+-------------+-------+
    INFO - 09:03:51:       | Name        | Lower bound | Value | Upper bound | Type  |
    INFO - 09:03:51:       +-------------+-------------+-------+-------------+-------+
    INFO - 09:03:51:       | x_shared[0] |     0.01    |  0.05 |     0.09    | float |
    INFO - 09:03:51:       | x_shared[1] |    30000    | 45000 |    60000    | float |
    INFO - 09:03:51:       | x_shared[2] |     1.4     |  1.6  |     1.8     | float |
    INFO - 09:03:51:       | x_shared[3] |     2.5     |  5.5  |     8.5     | float |
    INFO - 09:03:51:       | x_shared[4] |      40     |   55  |      70     | float |
    INFO - 09:03:51:       | x_shared[5] |     500     |  1000 |     1500    | float |
    INFO - 09:03:51:       | x_1[0]      |     0.1     |  0.25 |     0.4     | float |
    INFO - 09:03:51:       | x_1[1]      |     0.75    |   1   |     1.25    | float |
    INFO - 09:03:51:       | x_2         |     0.75    |   1   |     1.25    | float |
    INFO - 09:03:51:       | x_3         |     0.1     |  0.5  |      1      | float |
    INFO - 09:03:51:       +-------------+-------------+-------+-------------+-------+
    INFO - 09:03:51: Solving optimization problem with algorithm SLSQP:
    INFO - 09:03:51:     10%|█         | 1/10 [00:00<00:00,  9.11 it/sec, obj=-536]
    INFO - 09:03:51:     20%|██        | 2/10 [00:00<00:01,  6.46 it/sec, obj=-2.12e+3]
 WARNING - 09:03:51: MDAJacobi has reached its maximum number of iterations but the normed residual 1.7130677857005655e-05 is still above the tolerance 1e-06.
    INFO - 09:03:51:     30%|███       | 3/10 [00:00<00:01,  5.44 it/sec, obj=-3.75e+3]
    INFO - 09:03:52:     40%|████      | 4/10 [00:00<00:01,  5.22 it/sec, obj=-3.96e+3]
    INFO - 09:03:52:     50%|█████     | 5/10 [00:00<00:00,  5.08 it/sec, obj=-3.96e+3]
    INFO - 09:03:52: Optimization result:
    INFO - 09:03:52:    Optimizer info:
    INFO - 09:03:52:       Status: 8
    INFO - 09:03:52:       Message: Positive directional derivative for linesearch
    INFO - 09:03:52:       Number of calls to the objective function by the optimizer: 6
    INFO - 09:03:52:    Solution:
    INFO - 09:03:52:       The solution is feasible.
    INFO - 09:03:52:       Objective: -3963.408265187933
    INFO - 09:03:52:       Standardized constraints:
    INFO - 09:03:52:          g_1 = [-0.01806104 -0.03334642 -0.04424946 -0.0518346  -0.05732607 -0.13720865
    INFO - 09:03:52:  -0.10279135]
    INFO - 09:03:52:          g_2 = 3.333278582928756e-06
    INFO - 09:03:52:          g_3 = [-7.67181773e-01 -2.32818227e-01  8.30379541e-07 -1.83255000e-01]
    INFO - 09:03:52:       Design space:
    INFO - 09:03:52:          +-------------+-------------+---------------------+-------------+-------+
    INFO - 09:03:52:          | Name        | Lower bound |        Value        | Upper bound | Type  |
    INFO - 09:03:52:          +-------------+-------------+---------------------+-------------+-------+
    INFO - 09:03:52:          | x_shared[0] |     0.01    | 0.06000083331964572 |     0.09    | float |
    INFO - 09:03:52:          | x_shared[1] |    30000    |        60000        |    60000    | float |
    INFO - 09:03:52:          | x_shared[2] |     1.4     |         1.4         |     1.8     | float |
    INFO - 09:03:52:          | x_shared[3] |     2.5     |         2.5         |     8.5     | float |
    INFO - 09:03:52:          | x_shared[4] |      40     |          70         |      70     | float |
    INFO - 09:03:52:          | x_shared[5] |     500     |         1500        |     1500    | float |
    INFO - 09:03:52:          | x_1[0]      |     0.1     |         0.4         |     0.4     | float |
    INFO - 09:03:52:          | x_1[1]      |     0.75    |         0.75        |     1.25    | float |
    INFO - 09:03:52:          | x_2         |     0.75    |         0.75        |     1.25    | float |
    INFO - 09:03:52:          | x_3         |     0.1     |  0.1562448753887276 |      1      | float |
    INFO - 09:03:52:          +-------------+-------------+---------------------+-------------+-------+
    INFO - 09:03:52: *** End MDOScenario execution (time: 0:00:01.129678) ***

{'max_iter': 10, 'algo': 'SLSQP'}

Post-process scenario¶

Lastly, we post-process the scenario by means of the VariableInfluence plot.

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("VariableInfluence", fig_size=(20, 20), save=False, show=True)

Partial variation of the functions wrt design variables, 9 variables required to explain 99% of -y_4 variations, 5 variables required to explain 99% of g_1[0] variations, 5 variables required to explain 99% of g_1[1] variations, 5 variables required to explain 99% of g_1[2] variations, 5 variables required to explain 99% of g_1[3] variations, 5 variables required to explain 99% of g_1[4] variations, 4 variables required to explain 99% of g_1[5] variations, 4 variables required to explain 99% of g_1[6] variations, 1 variables required to explain 99% of g_2 variations, 7 variables required to explain 99% of g_3[0] variations, 7 variables required to explain 99% of g_3[1] variations, 3 variables required to explain 99% of g_3[2] variations, 3 variables required to explain 99% of g_3[3] variations

    INFO - 09:03:52: Output name; most influential variables to explain 0.99% of the output variation
    INFO - 09:03:52:    -y_4; x_1[1], x_2, x_3, x_shared[0], x_shared[1], x_shared[2], x_shared[3], x_shared[4], x_shared[5]
    INFO - 09:03:52:    g_1[0]; x_1[0], x_1[1], x_shared[0], x_shared[3], x_shared[5]
    INFO - 09:03:52:    g_1[1]; x_1[0], x_1[1], x_shared[0], x_shared[3], x_shared[5]
    INFO - 09:03:52:    g_1[2]; x_1[0], x_1[1], x_shared[0], x_shared[3], x_shared[5]
    INFO - 09:03:52:    g_1[3]; x_1[0], x_1[1], x_shared[0], x_shared[3], x_shared[5]
    INFO - 09:03:52:    g_1[4]; x_1[0], x_1[1], x_shared[0], x_shared[3], x_shared[5]
    INFO - 09:03:53:    g_1[5]; x_1[0], x_1[1], x_shared[3], x_shared[5]
    INFO - 09:03:53:    g_1[6]; x_1[0], x_1[1], x_shared[3], x_shared[5]
    INFO - 09:03:53:    g_2; x_shared[0]
    INFO - 09:03:53:    g_3[0]; x_2, x_3, x_shared[0], x_shared[1], x_shared[2], x_shared[4], x_shared[5]
    INFO - 09:03:53:    g_3[1]; x_2, x_3, x_shared[0], x_shared[1], x_shared[2], x_shared[4], x_shared[5]
    INFO - 09:03:53:    g_3[2]; x_3, x_shared[1], x_shared[2]
    INFO - 09:03:53:    g_3[3]; x_3, x_shared[1], x_shared[2]

<gemseo.post.variable_influence.VariableInfluence object at 0x7f8ad77714f0>

Total running time of the script: (0 minutes 2.728 seconds)

Download Jupyter notebook: plot_variable_influence.ipynb

Download Python source code: plot_variable_influence.py

Gallery generated by Sphinx-Gallery