def test_ReliabilityBenchmarkProblem(self):
limitStateFunction = ot.SymbolicFunction(["R", "S"], ["R - S"])
muR = 4.0
sigmaR = 1.0
muS = 2.0
sigmaS = 1.0
threshold = 0.0
R = ot.Normal(muR, sigmaR)
S = ot.Normal(muS, sigmaS)
myDistribution = ot.ComposedDistribution([R, S])
inputRandomVector = ot.RandomVector(myDistribution)
outputRandomVector = ot.CompositeRandomVector(limitStateFunction,
inputRandomVector)
thresholdEvent = ot.ThresholdEvent(outputRandomVector, ot.Less(),
threshold)
name = "R-S"
probability = 0.123456789
problem = otb.ReliabilityBenchmarkProblem(name, thresholdEvent,
probability)
#
print(problem)
print(problem.toFullString())
p = problem.getProbability()
assert p == probability
s = problem.getName()
assert s == name
def __init__(self, threshold=0.0, expo=1):
"""
Creates a reliability problem RP54.
The event is {g(X) < threshold} where
X = (x1, x2, ...., x20)
g(X) = (x1 + x2 + ... + x20) - 8.951
Parameters
----------
threshold : float
The threshold.
expo : float
Rate parameter of the Xi Exponential distribution
for i in {1, 2, ..., 20}.
"""
formula = "x1+x2+x3+x4+x5+x6+x7+x8+x9+x10"
formula = formula + " + x11+x12+x13+x14+x15+x16+x17+x18+x19+x20-8.951"
print(formula)
limitStateFunction = ot.SymbolicFunction(
[
"x1",
"x2",
"x3",
"x4",
"x5",
"x6",
"x7",
"x8",
"x9",
"x10",
"x11",
"x12",
"x13",
"x14",
"x15",
"x16",
"x17",
"x18",
"x19",
"x20",
],
[formula],
)
X = [ot.Exponential(expo) for i in range(20)]
myDistribution = ot.ComposedDistribution(X)
inputRandomVector = ot.RandomVector(myDistribution)
outputRandomVector = ot.CompositeRandomVector(
limitStateFunction, inputRandomVector
)
thresholdEvent = ot.ThresholdEvent(outputRandomVector, ot.Less(), threshold)
name = "RP54"
probability = 0.000998
super(ReliabilityProblem54, self).__init__(name, thresholdEvent, probability)
return None
def __init__(self, threshold=0.0, mu=[0.0] * 2, sigma=[1.0] * 2):
"""
Creates a reliability problem RP35.
The event is {g(X) < threshold} where
g(x1, x2) = min(g1, g2) with
g1 = 2 - x2 + exp(-0.1 * x1^2) + (0.2 * x1) ^ 4
g2 = 4.5 - x1 * x2
We have x1 ~ Normal(mu[0], sigma[0]) and x2 ~ Normal(mu[1], sigma[1]).
Parameters
----------
threshold : float
The threshold.
mu : sequence of floats
The list of two items representing the means of the gaussian distributions.
sigma : float
The list of two items representing the standard deviations of
the gaussian distributions.
"""
equations = ["var g1 := 2 - x2 + exp(-0.1 * x1^2) + (0.2 * x1) ^ 4"]
equations.append("var g2 := 4.5 - x1 * x2")
equations.append("gsys := min(g1, g2)")
formula = ";".join(equations)
limitStateFunction = ot.SymbolicFunction(["x1", "x2"], ["gsys"],
formula)
inputDimension = len(mu)
if inputDimension != 2:
raise Exception(
"The dimension of mu is %d, but the expected dimension is 2." %
(inputDimension))
inputDimension = len(sigma)
if inputDimension != 2:
raise Exception(
"The dimension of sigma is %d, but the expected dimension is 2."
% (inputDimension))
X1 = ot.Normal(mu[0], sigma[0])
X1.setDescription(["X1"])
X2 = ot.Normal(mu[1], sigma[1])
X2.setDescription(["X2"])
myDistribution = ot.ComposedDistribution([X1, X2])
inputRandomVector = ot.RandomVector(myDistribution)
outputRandomVector = ot.CompositeRandomVector(limitStateFunction,
inputRandomVector)
thresholdEvent = ot.ThresholdEvent(outputRandomVector, ot.Less(),
threshold)
name = "RP35"
probability = 0.00354
super(ReliabilityProblem35, self).__init__(name, thresholdEvent,
probability)
return None
def __init__(self, threshold=0.0, muR=4.0, sigmaR=1.0, muS=2.0, sigmaS=1.0):
"""
Create a R-S reliability problem.
The event is {g(X) < threshold} where
g(R, S) = R - S
We have R ~ Normal(muR, sigmaR) and S ~ Normal(muS, sigmaS).
Waarts (2000) uses muR = 7.0 and muS = 2.0 leading to
beta = 3.54.
Parameters
----------
threshold : float
The threshold.
muR : float
The mean of the R gaussian distribution.
sigmaR : float
The standard deviation of the R gaussian distribution.
muS : float
The mean of the S gaussian distribution.
sigmaS : float
The standard deviation of the S gaussian distribution.
Example
-------
problem = RminusSReliabilityBenchmarkProblem()
"""
limitStateFunction = ot.SymbolicFunction(["R", "S"], ["R - S"])
R = ot.Normal(muR, sigmaR)
R.setDescription("R")
S = ot.Normal(muS, sigmaS)
S.setDescription("S")
myDistribution = ot.ComposedDistribution([R, S])
inputRandomVector = ot.RandomVector(myDistribution)
outputRandomVector = ot.CompositeRandomVector(
limitStateFunction, inputRandomVector
)
thresholdEvent = ot.ThresholdEvent(outputRandomVector, ot.Less(), threshold)
name = "R-S"
diff = R - S
probability = diff.computeCDF(threshold)
super(RminusSReliabilityBenchmarkProblem, self).__init__(
name, thresholdEvent, probability
)
return None
def __init__(self,
threshold=0.0,
mu1=1.5,
sigma1=1.0,
mu2=2.5,
sigma2=1.0):
"""
Creates a reliability problem RP53.
The event is {g(X) < threshold} where
g(X1, X2) = sin(5 * X1 / 2) + 2 - (X1^2 + 4) * (X2 - 1) / 20
We have X1 ~ Normal(mu1, sigma1) and X2 ~ Normal(mu2, sigma2).
Parameters
----------
threshold : float
The threshold.
mu1 : float
The mean of the X1 gaussian distribution.
sigma1 : float
The standard deviation of the X1 gaussian distribution.
mu2 : float
The mean of the X2 gaussian distribution.
sigma2 : float
The standard deviation of the X2 gaussian distribution.
"""
formula = "sin(5 * x1 / 2) + 2 - ( x1 * x1 + 4 ) * ( x2 - 1 ) / 20"
limitStateFunction = ot.SymbolicFunction(["x1", "x2"], [formula])
print("sin(5 * x1 / 2) + 2 - ( x1 * x1 + 4 ) * ( x2 - 1 ) / 20")
X1 = ot.Normal(mu1, sigma1)
X1.setDescription(["X1"])
X2 = ot.Normal(mu2, sigma2)
X2.setDescription(["X2"])
myDistribution = ot.ComposedDistribution([X1, X2])
inputRandomVector = ot.RandomVector(myDistribution)
outputRandomVector = ot.CompositeRandomVector(limitStateFunction,
inputRandomVector)
thresholdEvent = ot.ThresholdEvent(outputRandomVector, ot.Less(),
threshold)
name = "RP53"
probability = 0.0313
super(ReliabilityProblem53, self).__init__(name, thresholdEvent,
probability)
return None
def __init__(self,
threshold=0.0,
mu1=0.0,
sigma1=1.0,
mu2=0.0,
sigma2=1.0):
"""
Creates a reliability problem RP25.
The event is {g(X) < threshold} where
---
g(x1, x2) = max(x1^2 -8 * x2 + 16, -16 * x1 + x2 + 32)
---
We have x1 ~ Normal(mu1, sigma1) and x2 ~ Normal(mu2, sigma2).
---
Parameters
----------
threshold : float
The threshold.
***
mu1 : float
The mean of the X1 gaussian distribution.
***
sigma1 : float
The standard deviation of the X1 gaussian distribution.
***
mu2 : float
The mean of the X2 gaussian distribution.
***
sigma2 : float
The standard deviation of the X2 gaussian distribution.
"""
equations = ["var g1 := x1^2 -8 * x2 + 16"]
equations.append("var g2 := -16 * x1 + x2 + 32")
equations.append("gsys := max(g1, g2)")
formula = ";".join(equations)
limitStateFunction = ot.SymbolicFunction(["x1", "x2"], ["gsys"],
formula)
print(formula)
X1 = ot.Normal(mu1, sigma1)
X1.setDescription(["X1"])
X2 = ot.Normal(mu2, sigma2)
X2.setDescription(["X2"])
myDistribution = ot.ComposedDistribution([X1, X2])
inputRandomVector = ot.RandomVector(myDistribution)
outputRandomVector = ot.CompositeRandomVector(limitStateFunction,
inputRandomVector)
thresholdEvent = ot.ThresholdEvent(outputRandomVector, ot.Less(),
threshold)
name = "RP25"
probability = 0.00000614
super(ReliabilityProblem25, self).__init__(name, thresholdEvent,
probability)
return None
def getXEvent(b, t, mu_S, covariance, R, delta_t):
full = buildCrossing(b, t, mu_S, covariance, R, delta_t)
X = ot.RandomVector(full)
f1 = ot.SymbolicFunction(["R", "X1", "X2"], ["X1 - R"])
e1 = ot.ThresholdEvent(ot.CompositeRandomVector(f1, X), ot.Less(), 0.0)
f2 = ot.SymbolicFunction(["R", "X1", "X2"], ["X2 - R"])
e2 = ot.ThresholdEvent(ot.CompositeRandomVector(f2, X),
ot.GreaterOrEqual(), 0.0)
event = ot.IntersectionEvent([e1, e2])
return X, event
def __init__(self,
threshold=0.0,
mu1=0.0,
sigma1=1.0,
mu2=0.0,
sigma2=1.0):
"""
Creates a reliability problem RP57.
The event is {g(X) < threshold} where
g(x1, x2) = min(max(g1, g2), g3) with
g1 = -x1^2 + x2^3 + 3
g2 = 2 - x1 - 8 * x2
g3 = (x1 + 3)^2 + (x2 + 3)^2 - 4
We have x1 ~ Normal(mu1, sigma1) and x2 ~ Normal(mu2, sigma2).
***
Parameters
----------
threshold : float
The threshold.
mu1 : float
The mean of the X1 gaussian distribution.
sigma1 : float
The standard deviation of the X1 gaussian distribution.
mu2 : float
The mean of the X2 gaussian distribution.
sigma2 : float
The standard deviation of the X2 gaussian distribution.
"""
equations = ["var g1 := -x1^2 + x2^3 + 3"]
equations.append("var g2 := 2 - x1 - 8 * x2")
equations.append("var g3 := (x1 + 3)^2 + (x2 + 3)^2 - 4")
equations.append("gsys := min(max(g1, g2), g3) ")
formula = ";".join(equations)
limitStateFunction = ot.SymbolicFunction(["x1", "x2"], ["gsys"],
formula)
print(formula)
X1 = ot.Normal(mu1, sigma1)
X1.setDescription(["X1"])
X2 = ot.Normal(mu2, sigma2)
X2.setDescription(["X2"])
myDistribution = ot.ComposedDistribution([X1, X2])
inputRandomVector = ot.RandomVector(myDistribution)
outputRandomVector = ot.CompositeRandomVector(limitStateFunction,
inputRandomVector)
thresholdEvent = ot.ThresholdEvent(outputRandomVector, ot.Less(),
threshold)
name = "RP57"
probability = 0.0284
super(ReliabilityProblem57, self).__init__(name, thresholdEvent,
probability)
return None
def __init__(self, threshold=0.0, mu=[10.0] * 2, sigma=[3.0] * 2):
"""
Creates a reliability problem RP24.
The event is {g(X) < threshold} where
g(x1, x2) = 2.5 - 0.2357 * (x1 - x2) + 0.00463 * (x1 + x2 - 20)^4
We have x1 ~ Normal(mu[0], sigma[0]) and x2 ~ Normal(mu[1], sigma[1]).
Parameters
----------
threshold : float
The threshold.
mu : sequence of floats
The list of two items representing the means of the gaussian distributions.
sigma : float
The list of two items representing the standard deviations of
the gaussian distributions.
"""
formula = "2.5 - 0.2357 * (x1 - x2) + 0.00463 * (x1 + x2 - 20)^4"
limitStateFunction = ot.SymbolicFunction(["x1", "x2"], [formula])
inputDimension = len(mu)
if inputDimension != 2:
raise Exception(
"The dimension of mu is %d, but the expected dimension is 2." %
(inputDimension))
inputDimension = len(sigma)
if inputDimension != 2:
raise Exception(
"The dimension of sigma is %d, but the expected dimension is 2."
% (inputDimension))
X1 = ot.Normal(mu[0], sigma[0])
X1.setDescription(["X1"])
X2 = ot.Normal(mu[1], sigma[1])
X2.setDescription(["X2"])
myDistribution = ot.ComposedDistribution([X1, X2])
inputRandomVector = ot.RandomVector(myDistribution)
outputRandomVector = ot.CompositeRandomVector(limitStateFunction,
inputRandomVector)
thresholdEvent = ot.ThresholdEvent(outputRandomVector, ot.Less(),
threshold)
name = "RP24"
probability = 0.00286
super(ReliabilityProblem24, self).__init__(name, thresholdEvent,
probability)
return None
def __init__(self):
"""
Creates the four-branch serial system from Waarts.
References
----------
Waarts, P.-H. (2000). Structural reliability using finite element
methods: an appraisal of DARS: Directional Adaptive Response Surface
Sampling. Ph. D. thesis, Technical University of Delft, The
Netherlands. Pages 58, 69, 160.
Thèse Vincent Dubourg 2011, Méta-modèles adaptatifs pour l’analyse
de fiabilité et l’optimisation sous contrainte fiabiliste,
section "A two-dimensional four-branch serial system", page 182
Parameters
----------
None.
Example
-------
problem = FourBranchSerialSystemReliabilityBenchmarkProblem()
"""
formulaList = [
"var y0 := 3 + 0.1 * (x0 - x1)^2 - (x0 + x1) / sqrt(2)",
"var y1 := 3 + 0.1 * (x0 - x1)^2 + (x0 + x1) / sqrt(2)",
"var y2 := x0 - x1 + 7 / sqrt(2)",
"var y3 := x1 - x0 + 7 / sqrt(2)",
"y := min(y0,y1,y2,y3)",
]
formula = ";".join(formulaList)
limitStateFunction = ot.SymbolicFunction(["x0", "x1"], ["y"], formula)
x0 = ot.Normal(0.0, 1.0)
x1 = ot.Normal(0.0, 1.0)
inputDistribution = ot.ComposedDistribution((x0, x1))
inputRandomVector = ot.RandomVector(inputDistribution)
outputRandomVector = ot.CompositeRandomVector(
limitStateFunction, inputRandomVector
)
thresholdEvent = ot.ThresholdEvent(outputRandomVector, ot.Less(), 0.0)
name = "Four-branch serial system (Waarts, 2000)"
beta = 2.85
probability = ot.Normal().computeComplementaryCDF(beta)
super(FourBranchSerialSystemReliabilityBenchmarkProblem, self).__init__(
name, thresholdEvent, probability
)
return None
def __init__(self,
threshold=146.14,
mu1=78064.0,
sigma1=11710.0,
mu2=0.0104,
sigma2=0.00156):
"""
Creates a reliability problem RP28.
The event is {g(X) < threshold} where
g(x1, x2) = x1 * x2 - threshold
with threshold = 146.14.
We have x1 ~ Normal(mu1, sigma1) and x2 ~ Normal(mu2, sigma2).
Parameters
----------
threshold : float
The threshold.
mu1 : float
The mean of the X1 gaussian distribution.
sigma1 : float
The standard deviation of the X1 gaussian distribution.
mu2 : float
The mean of the X2 gaussian distribution.
sigma2 : float
The standard deviation of the X2 gaussian distribution.
"""
formula = "x1 * x2"
limitStateFunction = ot.SymbolicFunction(["x1", "x2"], [formula])
X1 = ot.Normal(mu1, sigma1)
X1.setDescription(["X1"])
X2 = ot.Normal(mu2, sigma2)
X2.setDescription(["X2"])
myDistribution = ot.ComposedDistribution([X1, X2])
inputRandomVector = ot.RandomVector(myDistribution)
outputRandomVector = ot.CompositeRandomVector(limitStateFunction,
inputRandomVector)
thresholdEvent = ot.ThresholdEvent(outputRandomVector, ot.Less(),
threshold)
name = "RP28"
Y = X1 * X2
probability = Y.computeCDF(threshold)
super(ReliabilityProblem28, self).__init__(name, thresholdEvent,
probability)
return None
def __init__(self, threshold=0.0):
"""
Create a axial stressed beam reliability problem.
The inputs are R, the Yield strength, and F, the traction load.
The event is {g(X) < threshold} where
g(R, F) = R - F/(pi_ * 100.0)
We have R ~ LogNormalMuSigma() and F ~ Normal().
Parameters
----------
threshold : float
The threshold.
Example
-------
problem = AxialStressedBeamReliability()
"""
limitStateFunction = ot.SymbolicFunction(["R", "F"], ["R - F/(pi_ * 100.0)"])
R_dist = ot.LogNormalMuSigma(300.0, 30.0, 0.0).getDistribution()
R_dist.setName("Yield strength")
R_dist.setDescription("R")
F_dist = ot.Normal(75000.0, 5000.0)
F_dist.setName("Traction load")
F_dist.setDescription("F")
myDistribution = ot.ComposedDistribution([R_dist, F_dist])
inputRandomVector = ot.RandomVector(myDistribution)
outputRandomVector = ot.CompositeRandomVector(
limitStateFunction, inputRandomVector
)
thresholdEvent = ot.ThresholdEvent(outputRandomVector, ot.Less(), 0.0)
name = "Axial stressed beam"
diff = R_dist - F_dist / (np.pi * 100.0)
probability = diff.computeCDF(threshold)
super(AxialStressedBeamReliability, self).__init__(
name, thresholdEvent, probability
)
return None
def __init__(self,
threshold=0.0,
mu1=10.0,
sigma1=3.0,
mu2=10.0,
sigma2=3.0):
"""
Creates a reliability problem RP24.
The event is {g(X) < threshold} where
g(x1, x2) = 2.5 - 0.2357 * (x1 - x2) + 0.00463 * (x1 + x2 - 20)^4
We have x1 ~ Normal(mu1, sigma1) and x2 ~ Normal(mu2, sigma2).
***
Parameters
----------
threshold : float
The threshold.
mu1 : float
The mean of the X1 gaussian distribution.
sigma1 : float
The standard deviation of the X1 gaussian distribution.
mu2 : float
The mean of the X2 gaussian distribution.
sigma2 : float
The standard deviation of the X2 gaussian distribution.
"""
formula = "2.5 - 0.2357 * (x1 - x2) + 0.00463 * (x1 + x2 - 20)^4"
limitStateFunction = ot.SymbolicFunction(["x1", "x2"], [formula])
print(formula)
X1 = ot.Normal(mu1, sigma1)
X1.setDescription(["X1"])
X2 = ot.Normal(mu2, sigma2)
X2.setDescription(["X2"])
myDistribution = ot.ComposedDistribution([X1, X2])
inputRandomVector = ot.RandomVector(myDistribution)
outputRandomVector = ot.CompositeRandomVector(limitStateFunction,
inputRandomVector)
thresholdEvent = ot.ThresholdEvent(outputRandomVector, ot.Less(),
threshold)
name = "RP24"
probability = 0.00286
super(ReliabilityProblem24, self).__init__(name, thresholdEvent,
probability)
return None
def __init__(self, threshold=0.0, mu=0, sigma=1):
"""
Creates a reliability problem RP107.
The event is {g(X) < threshold} where
X = (x1, x2, ...., x10)
g(X) = 5*sqrt(10) -(x1 + x2 + x3 + x4 + x5 + x6 + x7 + x8 + x9 + x10).
We have xi ~ Normal(0, 1) for i in {1, 2, ...,10}
Parameters
----------
threshold : float
The threshold.
mu : float
The mean of the Xi Normal distribution for i in {1, 2, ..., 10}.
sigma : float
The standard deviation of the Xi Normal distribution
for i in {1, 2, ..., 10}.
"""
formula = "5*sqrt(10)-(x1 + x2 + x3 + x4 + x5 + x6 + x7 + x8 + x9 + x10)"
limitStateFunction = ot.SymbolicFunction(
["x1", "x2", "x3", "x4", "x5", "x6", "x7", "x8", "x9", "x10"],
[formula])
X = [ot.Normal(mu, sigma) for i in range(10)]
myDistribution = ot.ComposedDistribution(X)
inputRandomVector = ot.RandomVector(myDistribution)
outputRandomVector = ot.CompositeRandomVector(limitStateFunction,
inputRandomVector)
thresholdEvent = ot.ThresholdEvent(outputRandomVector, ot.Less(),
threshold)
name = "RP107"
probability = 0.000000292
super(ReliabilityProblem107, self).__init__(name, thresholdEvent,
probability)
return None
def __init__(self, threshold=0.0, a1=-1.0, b1=1.0, a2=-1.0, b2=1.0):
"""
Creates a reliability problem RP55.
The event is {g(X) < threshold} where
We have x1 ~ Uniform(a1, b1) and x2 ~ Uniform(a2, b2).
***
Parameters
----------
threshold : float
The threshold.
a1 , b1 : float
Parameters of the X1 uniform distribution.
a2 , b2 : float
Parameters of the X2 uniform distribution.
"""
equations = ["var g1 := 0.2 + 0.6 * (x0 - x1)^4 - (x0 - x1) / sqrt(2)"]
equations.append(
"var g2 := 0.2 + 0.6 * (x0 - x1)^4 + (x0 - x1) / sqrt(2)")
equations.append("var g3 := (x0 - x1) + 5 / sqrt(2) - 2.2")
equations.append("var g4 := (x1 - x0) + 5 / sqrt(2) - 2.2")
equations.append("gsys := min(g1, g2, g3, g4)")
formula = ";".join(equations)
limitStateFunction = ot.SymbolicFunction(["x0", "x1"], ["gsys"],
formula)
print(formula)
X1 = ot.Uniform(a1, b1)
X1.setDescription(["X1"])
X2 = ot.Uniform(a2, b2)
X2.setDescription(["X2"])
myDistribution = ot.ComposedDistribution([X1, X2])
inputRandomVector = ot.RandomVector(myDistribution)
outputRandomVector = ot.CompositeRandomVector(limitStateFunction,
inputRandomVector)
thresholdEvent = ot.ThresholdEvent(outputRandomVector, ot.Less(),
threshold)
name = "RP55"
probability = 0.36
super(ReliabilityProblem55, self).__init__(name, thresholdEvent,
probability)
return None
mean[0] = 7. mean[1] = 2. sigma = ot.Point(dim, 1.0) R = ot.IdentityMatrix(dim) myDistribution = ot.Normal(mean, sigma, R) # # Limit state # vect = ot.RandomVector(myDistribution) output = ot.CompositeRandomVector(limitState, vect) myEvent = ot.ThresholdEvent(output, ot.Less(), 0.0) # # Computation # bs = 1 # Monte Carlo experiment = ot.MonteCarloExperiment() myMC = ot.ProbabilitySimulationAlgorithm(myEvent, experiment) myMC.setMaximumOuterSampling(int(1e6) // bs) myMC.setBlockSize(bs) myMC.setMaximumCoefficientOfVariation(-1.0) myMC.run() #
# %%
# We consider three functions `f1`, `f2` and `f3` :
f1 = ot.SymbolicFunction(['x0', 'x1'], ['x0'])
f2 = ot.SymbolicFunction(['x0', 'x1'], ['x1'])
f3 = ot.SymbolicFunction(['x0', 'x1'], ['x0+x1'])
# %%
# We build :class:`~openturns.CompositeRandomVector` from these functions and the initial distribution.
Y1 = ot.CompositeRandomVector(f1, X)
Y2 = ot.CompositeRandomVector(f2, X)
Y3 = ot.CompositeRandomVector(f3, X)
# %%
# We define three basic events :math:`E_1=\{(x_0,x_1)~:~x_0 < 0 \}`, :math:`E_2=\{(x_0,x_1)~:~x_1 > 0 \}` and :math:`E_3=\{(x_0,x_1)~:~x_0+x_1>0 \}`.
e1 = ot.ThresholdEvent(Y1, ot.Less(), 0.0)
e2 = ot.ThresholdEvent(Y2, ot.Greater(), 0.0)
e3 = ot.ThresholdEvent(Y3, ot.Greater(), 0.0)
# %%
# The restriction of the domain :math:`E_1` to :math:`[-4,4] \times [-4, 4]` is the grey area.
myGraph = ot.Graph(r'Representation of the event $E_1$', r'$x_1$', r'$x_2$',
True, '')
data = [[-4, -4], [0, -4], [0, 4], [-4, 4]]
myPolygon = ot.Polygon(data)
myPolygon.setColor('grey')
myPolygon.setEdgeColor('black')
myGraph.add(myPolygon)
view = otv.View(myGraph)
axes = view.getAxes()
_ = axes[0].set_xlim(-4.0, 4.0)
inVars[i] = "x" + str(i)
model = ot.SymbolicFunction(inVars, inVars)
# The output vector
Y = ot.CompositeRandomVector(model, X)
# The domain: [0, 1]^dim
domain = ot.Interval(dim)
# The event
event = ot.DomainEvent(Y, domain)
print("sample=", event.getSample(10))
#
# Case 2: process based event
#
# The input process
X = ot.WhiteNoise(distribution)
# The domain: [0, 1]^dim
domain = ot.Interval(dim)
# The event
event = ot.ProcessEvent(X, domain)
print("sample=", event.getSample(10))
# 3. from distribution
antecedent = ot.RandomVector(ot.Normal(2))
domain = ot.LevelSet(ot.SymbolicFunction(['x', 'y'], ['x^2+y^2']), ot.Less(),
1.0)
event = ot.DomainEvent(antecedent, domain)
print('sample=', event.getSample(10))
#Create a copula : IndependentCopula (no correlation)
aCopula = ot.IndependentCopula(dim)
aCopula.setName('Independent copula')
#Instanciate one distribution object
myDistribution = ot.ComposedDistribution([R_dist, F_dist], aCopula)
myDistribution.setName('myDist')
#We create a 'usual' RandomVector from the Distribution
vect = ot.RandomVector(myDistribution)
#We create a composite random vector
G = ot.RandomVector(limitState, vect)
#We create an Event from this RandomVector
myEvent = ot.Event(G, ot.Less(), 0.0)
#Using Monte Carlo simulations
cv = 0.05
NbSim = 100000
experiment = ot.MonteCarloExperiment()
algoMC = ot.ProbabilitySimulationAlgorithm(myEvent, experiment)
algoMC.setMaximumOuterSampling(NbSim)
algoMC.setBlockSize(1)
algoMC.setMaximumCoefficientOfVariation(cv)
#For statistics about the algorithm
initialNumberOfCall = limitState.getEvaluationCallsNumber()
#Perform the analysis
model = ot.MemoizeFunction(model) # %% # Remove all the values stored in the history mechanism. # Care : it is done regardless the status of the History mechanism. # %% model.clearHistory() # %% # Create the event whose probability we want to estimate. # %% vect = ot.RandomVector(distribution) G = ot.CompositeRandomVector(model, vect) event = ot.ThresholdEvent(G, ot.Less(), 0.0) # %% # Create a Monte Carlo algorithm. # %% experiment = ot.MonteCarloExperiment() algo = ot.ProbabilitySimulationAlgorithm(event, experiment) algo.setMaximumCoefficientOfVariation(0.1) algo.setMaximumStandardDeviation(0.001) algo.setMaximumOuterSampling(int(1e4)) # %% # Define the HistoryStrategy to store the values of :math:`P_n` and :math:`\sigma_n` used ot draw the convergence graph. # Compact strategy : N points
function = hyperplane(linear)
function.setName('H3')
functions.append(function)
for ih in range(len(functions)):
function = functions[ih]
distribution = ot.Normal(dim)
randomVector = ot.RandomVector(distribution)
composite = ot.CompositeRandomVector(function, randomVector)
for pft in [1e-4, 1e-6, 1e-8][1:2]:
k = ot.Normal().computeQuantile(pft)[0] * ot.Point(
linears[ih]).norm()
event = ot.ThresholdEvent(composite, ot.Less(), k)
print('--------------------')
print('model H' + str(ih) + ' dim=%d' % dim, 'pft=%.2e' % pft,
'k=%g' % k)
for n in [100, 1000][1:]:
for gamma1 in [0.25, 0.5, 0.75][1:2]:
# algo = ot.MonteCarlo(event)
# algo.setMaximumOuterSampling(100*n)
# algo.setMaximumCoefficientOfVariation(-1.)
# algo.run()
# result = algo.getResult()
# print result
algo = ot.AdaptiveDirectionalSampling(event)
algo.setMaximumOuterSampling(n)
def __init__(
self,
threshold=0.0,
mu=[0.0] * 3,
sigma=[1.0] * 3,
):
"""
Creates a reliability problem RP33.
The event is {g(X) < threshold} where
g(x1, x2, x3) = min(g1, g2) with
g1 = -x1 - x2 - x3 + 3 * sqrt(3)
g2 = -x3 + 3
We have :
x1 ~ Normal(mu[0], sigma[0])
x2 ~ Normal(mu[1], sigma[1])
x3 ~ Normal(mu[2], sigma[2]).
Parameters
----------
threshold : float
The threshold.
mu : sequence of floats
The list of 3 items representing the means of the gaussian distributions.
sigma : float
The list of 3 items representing the standard deviations of
the gaussian distributions.
"""
formula = "min(-x1 - x2 - x3 + 3 * sqrt(3), -x3 + 3)"
limitStateFunction = ot.SymbolicFunction(["x1", "x2", "x3"], [formula])
inputDimension = len(mu)
if inputDimension != 3:
raise Exception(
"The dimension of mu is %d, but the expected dimension is 3." %
(inputDimension))
inputDimension = len(sigma)
if inputDimension != 3:
raise Exception(
"The dimension of sigma is %d, but the expected dimension is 3."
% (inputDimension))
X1 = ot.Normal(mu[0], sigma[0])
X1.setDescription(["X1"])
X2 = ot.Normal(mu[1], sigma[1])
X2.setDescription(["X2"])
X3 = ot.Normal(mu[2], sigma[2])
X3.setDescription(["X3"])
myDistribution = ot.ComposedDistribution([X1, X2, X3])
inputRandomVector = ot.RandomVector(myDistribution)
outputRandomVector = ot.CompositeRandomVector(limitStateFunction,
inputRandomVector)
thresholdEvent = ot.ThresholdEvent(outputRandomVector, ot.Less(),
threshold)
name = "RP33"
probability = 0.00257
super(ReliabilityProblem33, self).__init__(name, thresholdEvent,
probability)
return None
import openturns as ot
ot.TESTPREAMBLE()
# create a function
dim = 4
function = ot.SymbolicFunction(
['E', 'F', 'L', 'I'], ['F*L^3/(3.*E*I)'])
# create a distribution
distribution = ot.Normal(
[50., 1.0, 10.0, 5.0], [1.0] * dim,
ot.IdentityMatrix(dim))
vect = ot.RandomVector(distribution)
composite = ot.CompositeRandomVector(function, vect)
event = ot.Event(composite, ot.Less(), -3.0)
# create an ADS algorithm
n = int(1e4)
algo = ot.AdaptiveDirectionalSampling(event)
algo.setMaximumOuterSampling(n)
algo.setGamma([0.6, 0.4])
algo.run()
result = algo.getResult()
print(result)
# ADS-2+
algo2 = algo
algo2.setPartialStratification(True)
algo2.run()
graph = F_dist.drawPDF()
view = viewer.View(graph)
# %%
# These independent marginals define the joint distribution of the input parameters :
myDistribution = sm.distribution
# %%
# We create a `RandomVector` from the `Distribution`, then a composite random vector. Finally, we create a `ThresholdEvent` from this `RandomVector`.
# %%
inputRandomVector = ot.RandomVector(myDistribution)
outputRandomVector = ot.CompositeRandomVector(
limitStateFunction, inputRandomVector)
myEvent = ot.ThresholdEvent(outputRandomVector, ot.Less(), 0.0)
# %%
# Using Monte Carlo simulations
# -----------------------------
# %%
cv = 0.05
NbSim = 100000
experiment = ot.MonteCarloExperiment()
algoMC = ot.ProbabilitySimulationAlgorithm(myEvent, experiment)
algoMC.setMaximumOuterSampling(NbSim)
algoMC.setBlockSize(1)
algoMC.setMaximumCoefficientOfVariation(cv)
#create joint probability distribution heave
distribution_h = ot.ComposedDistribution(marginals_h, copula_h)
distribution_h.setDescription(['h_exit', 'D_cover', 'i_ch'])
#create joint probability distribution
distribution_p = ot.ComposedDistribution(marginals_p, copula_p)
distribution_p.setDescription(['h_exit', 'D_cover', 'L', 'D', 'm_p', 'k', 'd70'])
#create joint probability distribution
distribution = ot.ComposedDistribution(marginals, copula)
distribution.setDescription(['h_exit', 'D_cover', 'm_u', 'i_ch', 'L', 'D', 'm_p', 'k', 'd70'])
#create the event we want to estimate the probability uplift
vect_u = ot.RandomVector(distribution_u)
G_u = ot.CompositeRandomVector(dijk.Z_u_function, vect_u)
event_u = ot.Event(G_u, ot.Less(), 0.0)
event_u.setName('uplift failure')
#create the event we want to estimate the probability heave
vect_h = ot.RandomVector(distribution_h)
G_h = ot.CompositeRandomVector(dijk.Z_h_function, vect_h)
event_h = ot.Event(G_h, ot.Less(), 0.0)
event_h.setName('heave failure')
#create the event we want to estimate the probability piping
vect_p = ot.RandomVector(distribution_p)
G_p = ot.CompositeRandomVector(dijk.Z_p_function, vect_p)
event_p = ot.Event(G_p, ot.Less(), 0.0)
event_p.setName('piping failure')
#create the event we want to estimate the probability
# Model
X0 = persalys.Input('X0', ot.Normal(1, 1))
X1 = persalys.Input('X1', ot.Normal(1, 1))
Y0 = persalys.Output('Y0')
model = persalys.SymbolicPhysicalModel('aModelPhys', [X0, X1], [Y0],
['sin(X0) + 8*X1'])
myStudy.add(model)
# limit state ##
limitState = persalys.LimitState('aLimitState', model, 'Y0', ot.Greater(), 20.)
print(limitState)
myStudy.add(limitState)
limitState.setThreshold(10.)
limitState.setOperator(ot.Less())
print(limitState)
# limit state ##
limitState2 = persalys.LimitState('aLimitState2', model)
print(limitState2)
myStudy.add(limitState2)
limitState2.setThreshold(15.)
print(limitState2)
model.addOutput(persalys.Output('Y1'))
model.setFormula('Y1', '1 + sin(X0) + 8*X1')
limitState2.setOutputName('Y1')
print(limitState2)
# L
mean[2] = 10.0
# I
mean[3] = 5.0
sigma = [1.0] * dim
R = ot.IdentityMatrix(dim)
myDistribution = ot.Normal(mean, sigma, R)
# We create a 'usual' RandomVector from the Distribution
vect = ot.RandomVector(myDistribution)
# We create a composite random vector
output = ot.RandomVector(myFunction, vect)
# We create an Event from this RandomVector
myEvent = ot.Event(output, ot.Less(), -3.0)
# We create a Monte Carlo algorithm
myAlgo = ot.MonteCarlo(myEvent)
myAlgo.setMaximumOuterSampling(250)
myAlgo.setBlockSize(4)
myAlgo.setMaximumCoefficientOfVariation(0.1)
print("MonteCarlo=", myAlgo)
# Perform the simulation
myAlgo.run()
# Stream out the result
print("MonteCarlo result=", myAlgo.getResult())
ot.TESTPREAMBLE()
dim = 2
distribution = ot.Normal(dim)
X = ot.RandomVector(distribution)
# 1. Composite/Composite
f1 = ot.SymbolicFunction(['x' + str(i) for i in range(dim)], ['x0'])
f2 = ot.SymbolicFunction(['x' + str(i) for i in range(dim)], ['x1'])
Y1 = ot.CompositeRandomVector(f1, X)
Y2 = ot.CompositeRandomVector(f2, X)
e1 = ot.ThresholdEvent(Y1, ot.Less(), 0.0)
e2 = ot.ThresholdEvent(Y2, ot.Greater(), 0.0)
e3 = e1.intersect(e2)
#print('e3=', e3)
proba_e3 = e3.getSample(10000).computeMean()[0]
print("proba_e3 = %.3g" % proba_e3)
ott.assert_almost_equal(proba_e3, 0.25, 1e-2, 1e-2)
e4 = e1.join(e2)
#print('e4=', e4)
proba_e4 = e4.getSample(10000).computeMean()[0]
print("proba_e4 = %.3g" % proba_e4)
ott.assert_almost_equal(proba_e4, 0.75, 1e-2, 1e-2)
from matplotlib import pylab as plt import math as m import time ot.Log.Show(ot.Log.NONE) # %% # Define an event to compute a probability myFunction = ot.SymbolicFunction(['E', 'F', 'L', 'I'], ['-F*L^3/(3.0*E*I)']) dim = myFunction.getInputDimension() mean = [50.0, 1.0, 10.0, 5.0] sigma = [1.0] * dim R = ot.IdentityMatrix(dim) myDistribution = ot.Normal(mean, sigma, R) vect = ot.RandomVector(myDistribution) output = ot.CompositeRandomVector(myFunction, vect) myEvent = ot.ThresholdEvent(output, ot.Less(), -3.0) # %% # **Stop a FORM algorithm using a calls number limit** # # A FORM algorithm termination can be controlled by the maximum number of iterations # # of its underlying optimization solver, but not directly by a maximum number of evaluations. # %% # Create the optimization algorithm myCobyla = ot.Cobyla() myCobyla.setMaximumEvaluationNumber(400) myCobyla.setMaximumAbsoluteError(1.0e-10) myCobyla.setMaximumRelativeError(1.0e-10) myCobyla.setMaximumResidualError(1.0e-10)
from __future__ import print_function
import openturns as ot
ot.TESTPREAMBLE()
# Uncertain parameters
distribution = ot.Normal([1.0] * 3, [2.0] * 3, ot.CorrelationMatrix(3))
distribution.setName("Unnamed")
# Model
f = ot.SymbolicFunction(["x", "y", "z"], ["x-1.5*y+2*z"])
# Sampling
size = 100
inputSample = distribution.getSample(size)
outputSample = f(inputSample)
comparisonOperators = [
ot.Less(), ot.LessOrEqual(),
ot.Greater(),
ot.GreaterOrEqual()
]
threshold = 3.0
ot.ResourceMap.SetAsUnsignedInteger(
"SimulationSensitivityAnalysis-DefaultSampleMargin", 10)
for i in range(4):
# Analysis based on an event
X = ot.RandomVector(distribution)
Y = ot.CompositeRandomVector(f, X)
event = ot.ThresholdEvent(Y, comparisonOperators[i], threshold)
algo = ot.SimulationSensitivityAnalysis(event, inputSample, outputSample)
print("algo=", algo)
# Perform the analysis
print("Mean point in event domain=", algo.computeMeanPointInEventDomain())