1 # -*- coding: utf-8 -*-
3 # Copyright (C) 2008-2021 EDF R&D
5 # This library is free software; you can redistribute it and/or
6 # modify it under the terms of the GNU Lesser General Public
7 # License as published by the Free Software Foundation; either
8 # version 2.1 of the License.
10 # This library is distributed in the hope that it will be useful,
11 # but WITHOUT ANY WARRANTY; without even the implied warranty of
12 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
13 # Lesser General Public License for more details.
15 # You should have received a copy of the GNU Lesser General Public
16 # License along with this library; if not, write to the Free Software
17 # Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
19 # See http://www.salome-platform.org/ or email : webmaster.salome@opencascade.com
21 # Author: Jean-Philippe Argaud, jean-philippe.argaud@edf.fr, EDF R&D
24 from daCore import BasicObjects
27 # ==============================================================================
28 class ElementaryAlgorithm(BasicObjects.Algorithm):
30 BasicObjects.Algorithm.__init__(self, "TABUSEARCH")
31 self.defineRequiredParameter(
32 name = "MaximumNumberOfSteps",
35 message = "Nombre maximal de pas d'optimisation",
38 self.defineRequiredParameter(
40 typecast = numpy.random.seed,
41 message = "Graine fixée pour le générateur aléatoire",
43 self.defineRequiredParameter(
44 name = "LengthOfTabuList",
47 message = "Longueur de la liste tabou",
50 self.defineRequiredParameter(
51 name = "NumberOfElementaryPerturbations",
54 message = "Nombre de perturbations élémentaires pour choisir une perturbation d'état",
57 self.defineRequiredParameter(
58 name = "NoiseDistribution",
61 message = "Distribution pour générer les perturbations d'état",
62 listval = ["Gaussian","Uniform"],
64 self.defineRequiredParameter(
65 name = "QualityCriterion",
66 default = "AugmentedWeightedLeastSquares",
68 message = "Critère de qualité utilisé",
69 listval = ["AugmentedWeightedLeastSquares","AWLS","DA",
70 "WeightedLeastSquares","WLS",
71 "LeastSquares","LS","L2",
75 self.defineRequiredParameter(
76 name = "NoiseHalfRange",
78 typecast = numpy.matrix,
79 message = "Demi-amplitude des perturbations uniformes centrées d'état pour chaque composante de l'état",
81 self.defineRequiredParameter(
82 name = "StandardDeviation",
84 typecast = numpy.matrix,
85 message = "Ecart-type des perturbations gaussiennes d'état pour chaque composante de l'état",
87 self.defineRequiredParameter(
88 name = "NoiseAddingProbability",
91 message = "Probabilité de perturbation d'une composante de l'état",
95 self.defineRequiredParameter(
96 name = "StoreInternalVariables",
99 message = "Stockage des variables internes ou intermédiaires du calcul",
101 self.defineRequiredParameter(
102 name = "StoreSupplementaryCalculations",
105 message = "Liste de calculs supplémentaires à stocker et/ou effectuer",
112 "CurrentIterationNumber",
117 "SimulatedObservationAtBackground",
118 "SimulatedObservationAtCurrentState",
119 "SimulatedObservationAtOptimum",
122 self.defineRequiredParameter( # Pas de type
124 message = "Liste des valeurs de bornes",
126 self.requireInputArguments(
127 mandatory= ("Xb", "Y", "HO", "R", "B"),
129 self.setAttributes(tags=(
135 def run(self, Xb=None, Y=None, U=None, HO=None, EM=None, CM=None, R=None, B=None, Q=None, Parameters=None):
136 self._pre_run(Parameters, Xb, Y, U, HO, EM, CM, R, B, Q)
138 if self._parameters["NoiseDistribution"] == "Uniform":
139 nrange = numpy.ravel(self._parameters["NoiseHalfRange"]) # Vecteur
140 if nrange.size != Xb.size:
141 raise ValueError("Noise generation by Uniform distribution requires range for all variable increments. The actual noise half range vector is:\n%s"%nrange)
142 elif self._parameters["NoiseDistribution"] == "Gaussian":
143 sigma = numpy.ravel(self._parameters["StandardDeviation"]) # Vecteur
144 if sigma.size != Xb.size:
145 raise ValueError("Noise generation by Gaussian distribution requires standard deviation for all variable increments. The actual standard deviation vector is:\n%s"%sigma)
147 # Opérateur d'observation
148 # -----------------------
149 Hm = HO["Direct"].appliedTo
151 # Précalcul des inversions de B et R
152 # ----------------------------------
156 # Définition de la fonction de deplacement
157 # ----------------------------------------
158 def Tweak( x, NoiseDistribution, NoiseAddingProbability ):
159 _X = numpy.matrix(numpy.ravel( x )).T
160 if NoiseDistribution == "Uniform":
161 for i in range(_X.size):
162 if NoiseAddingProbability >= numpy.random.uniform():
163 _increment = numpy.random.uniform(low=-nrange[i], high=nrange[i])
164 # On ne traite pas encore le dépassement des bornes ici
166 elif NoiseDistribution == "Gaussian":
167 for i in range(_X.size):
168 if NoiseAddingProbability >= numpy.random.uniform():
169 _increment = numpy.random.normal(loc=0., scale=sigma[i])
170 # On ne traite pas encore le dépassement des bornes ici
175 def StateInList( x, TL ):
176 _X = numpy.ravel( x )
179 if numpy.all(numpy.abs( _X - numpy.ravel(state) ) <= 1e-16*numpy.abs(_X)):
181 # if _xInList: import sys ; sys.exit()
184 # Minimisation de la fonctionnelle
185 # --------------------------------
188 # _qualityS = CostFunction( _S, self._parameters["QualityCriterion"] )
189 _qualityS = BasicObjects.CostFunction3D(
196 _SSC = self._parameters["StoreSupplementaryCalculations"],
197 _QM = self._parameters["QualityCriterion"],
198 _SSV = self.StoredVariables,
201 _Best, _qualityBest = _S, _qualityS
203 _TabuList.append( _S )
204 while _n < self._parameters["MaximumNumberOfSteps"]:
206 if len(_TabuList) > self._parameters["LengthOfTabuList"]:
208 _R = Tweak( _S, self._parameters["NoiseDistribution"], self._parameters["NoiseAddingProbability"] )
209 # _qualityR = CostFunction( _R, self._parameters["QualityCriterion"] )
210 _qualityR = BasicObjects.CostFunction3D(
217 _SSC = self._parameters["StoreSupplementaryCalculations"],
218 _QM = self._parameters["QualityCriterion"],
219 _SSV = self.StoredVariables,
222 for nbt in range(self._parameters["NumberOfElementaryPerturbations"]-1):
223 _W = Tweak( _S, self._parameters["NoiseDistribution"], self._parameters["NoiseAddingProbability"] )
224 # _qualityW = CostFunction( _W, self._parameters["QualityCriterion"] )
225 _qualityW = BasicObjects.CostFunction3D(
232 _SSC = self._parameters["StoreSupplementaryCalculations"],
233 _QM = self._parameters["QualityCriterion"],
234 _SSV = self.StoredVariables,
237 if (not StateInList(_W, _TabuList)) and ( (_qualityW < _qualityR) or StateInList(_R,_TabuList) ):
238 _R, _qualityR = _W, _qualityW
239 if (not StateInList( _R, _TabuList )) and (_qualityR < _qualityS):
240 _S, _qualityS = _R, _qualityR
241 _TabuList.append( _S )
242 if _qualityS < _qualityBest:
243 _Best, _qualityBest = _S, _qualityS
245 if self._parameters["StoreInternalVariables"] or self._toStore("CurrentState"):
246 self.StoredVariables["CurrentState"].store( _Best )
247 if self._toStore("SimulatedObservationAtCurrentState"):
248 _HmX = Hm( numpy.asmatrix(numpy.ravel( _Best )).T )
249 _HmX = numpy.asmatrix(numpy.ravel( _HmX )).T
250 self.StoredVariables["SimulatedObservationAtCurrentState"].store( _HmX )
251 self.StoredVariables["CurrentIterationNumber"].store( len(self.StoredVariables["CostFunctionJ"]) )
252 self.StoredVariables["CostFunctionJb"].store( 0. )
253 self.StoredVariables["CostFunctionJo"].store( 0. )
254 self.StoredVariables["CostFunctionJ" ].store( _qualityBest )
256 # Obtention de l'analyse
257 # ----------------------
258 Xa = numpy.asmatrix(numpy.ravel( _Best )).T
260 self.StoredVariables["Analysis"].store( Xa.A1 )
262 if self._toStore("Innovation") or \
263 self._toStore("OMB") or \
264 self._toStore("SimulatedObservationAtBackground"):
267 if self._toStore("OMA") or \
268 self._toStore("SimulatedObservationAtOptimum"):
271 # Calculs et/ou stockages supplémentaires
272 # ---------------------------------------
273 if self._toStore("Innovation"):
274 self.StoredVariables["Innovation"].store( numpy.ravel(d) )
275 if self._toStore("BMA"):
276 self.StoredVariables["BMA"].store( numpy.ravel(Xb) - numpy.ravel(Xa) )
277 if self._toStore("OMA"):
278 self.StoredVariables["OMA"].store( numpy.ravel(Y) - numpy.ravel(HXa) )
279 if self._toStore("OMB"):
280 self.StoredVariables["OMB"].store( numpy.ravel(d) )
281 if self._toStore("SimulatedObservationAtBackground"):
282 self.StoredVariables["SimulatedObservationAtBackground"].store( numpy.ravel(HXb) )
283 if self._toStore("SimulatedObservationAtOptimum"):
284 self.StoredVariables["SimulatedObservationAtOptimum"].store( numpy.ravel(HXa) )
289 # ==============================================================================
290 if __name__ == "__main__":
291 print('\n AUTODIAGNOSTIC\n')