PLIMean

class PLIMean(*args)

PLI based on a mean perturbation.

Parameters:
POD : KrigingPOD, AdaptiveSignalPOD or PolynomialChaosPOD

The POD object where the run method has been performed.

delta : 1d or 2d sequence of float

The new values of the mean or sigma coefficient. Either 1d if delta values are the same for all marginals, or 2d if delta values are defined independently for each marginal.

sigmaScaled : bool

Change the type of the applied mean shiftingfor all the variables. If False (default case), the given delta values are the new marginal means. If True, newMean = mean + sigma x delta, where sigma is the standard deviation of each marginals.

Methods

drawContourIndices(marginal[, label, name]) Draw a contour plot of the indices for a specific marginal
drawIndices(idefect[, confidenceLevel, …]) Draw the indices of all margins for a specific defect
getDefectSizes() Accessor to the defect size where the indices are computed.
getDistribution() Accessor to the parameters distribution.
getGaussKronrod() Accessor to the Gauss Kronrod algorithm used to compute integrals
getIndices([idelta, marginal, idefect]) Accessor to the indices
getPLIObject(idefect) Accessor to the PLI object for a specific defect.
getSamplingSize() Accessor to the Monte Carlo sampling size.
run() Compute the indices
setDefectSizes(size) Accessor to the defect size where the indices are computed.
setDistribution(distribution) Accessor to the parameters distribution.
setGaussKronrod(algo) Accessor to the Gauss Kronrod algorithm used to compute integrals
setSamplingSize(size) Accessor to the Monte Carlo sampling size.
drawContourIndices(marginal, label=None, name=None)

Draw a contour plot of the indices for a specific marginal

Parameters:
marginal : int

The indice of the perturbed marginal.

label : list of string

The labels of each parameters.

Returns:
fig : matplotlib.figure

Matplotlib figure object.

ax : matplotlib.axes

Matplotlib axes object.

drawIndices(idefect, confidenceLevel=0.95, label=None, hellinger=True, name=None)

Draw the indices of all margins for a specific defect

Parameters:
idefect : int

The indice of the defect in the given delta list.

confidenceLevel : 0 < float < 1 or None

The wanted confidence level to compute the interval. If set to ‘None’ only the indices are plotted.

label : list of string

The labels of each parameters.

hellinger : bool

If True, the indices are plotted with respect to the hellinger distance between the original PDF and the perturbed PDF. Default is True.

Returns:
fig : matplotlib.figure

Matplotlib figure object.

ax : matplotlib.axes

Matplotlib axes object.

getDefectSizes()

Accessor to the defect size where the indices are computed.

Returns:
defectSize : sequence of float

The defect sizes where the Monte Carlo simulation is performed to compute the POD.

getDistribution()

Accessor to the parameters distribution.

Returns:
distribution : openturns.ComposedDistribution

The input parameters distribution used for the Monte Carlo simulation. Default is a Uniform distribution for all parameters.

getGaussKronrod()

Accessor to the Gauss Kronrod algorithm used to compute integrals

getIndices(idelta=None, marginal=None, idefect=None)

Accessor to the indices

Parameters:
idelta : int

The indice of the delta in the given delta list. Default is None = all.

marginal : int

The indice of the perturbed marginal. Default is None = all.

idefect : int

The indice of the defect in the given delta list. Default is None = all.

Returns:
indices : float, 1d, 2d or 3d array.

The parameter order of the full matrix is delta, marginal and defect. The returned array depends on the given parameter values.

getPLIObject(idefect)

Accessor to the PLI object for a specific defect.

Parameters:
idefect : int

The indice of the defect in the given delta list.

Returns:
pli : PLI

The PLI base object from which more results can be obtained.

getSamplingSize()

Accessor to the Monte Carlo sampling size.

Returns:
size : int

The size of the Monte Carlo simulation used to compute the POD for each defect size.

run()

Compute the indices

Notes

Run the analysis:
  • run a Monte Carlo simulation
  • compute the indices for each defect size

If, for a defect size, the probability estimate is less than 1e-3 or greater than 0.999, then the indices are not computed.

setDefectSizes(size)

Accessor to the defect size where the indices are computed.

Parameters:
defectSize : sequence of float

The defect sizes where the Monte Carlo simulation is performed to compute the POD.

setDistribution(distribution)

Accessor to the parameters distribution.

Parameters:
distribution : openturns.ComposedDistribution

The input parameters distribution used for the Monte Carlo simulation.

setGaussKronrod(algo)

Accessor to the Gauss Kronrod algorithm used to compute integrals

Parameters:
algo : openturns.GaussKronrod

The algorithm

setSamplingSize(size)

Accessor to the Monte Carlo sampling size.

Parameters:
size : int

The size of the Monte Carlo simulation used to compute the POD for each defect size.