src/daComposant/daAlgorithms/ExtendedKalmanFilter.py

   1 # -*- coding: utf-8 -*-
   2 #
   3 # Copyright (C) 2008-2019 EDF R&D
   4 #
   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.
   9 #
  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.
  14 #
  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
  18 #
  19 # See http://www.salome-platform.org/ or email : webmaster.salome@opencascade.com
  20 #
  21 # Author: Jean-Philippe Argaud, jean-philippe.argaud@edf.fr, EDF R&D
  22
  23 import logging
  24 from daCore import BasicObjects
  25 import numpy
  26
  27 # ==============================================================================
  28 class ElementaryAlgorithm(BasicObjects.Algorithm):
  29     def __init__(self):
  30         BasicObjects.Algorithm.__init__(self, "EXTENDEDKALMANFILTER")
  31         self.defineRequiredParameter(
  32             name     = "ConstrainedBy",
  33             default  = "EstimateProjection",
  34             typecast = str,
  35             message  = "Prise en compte des contraintes",
  36             listval  = ["EstimateProjection"],
  37             )
  38         self.defineRequiredParameter(
  39             name     = "EstimationOf",
  40             default  = "State",
  41             typecast = str,
  42             message  = "Estimation d'etat ou de parametres",
  43             listval  = ["State", "Parameters"],
  44             )
  45         self.defineRequiredParameter(
  46             name     = "StoreInternalVariables",
  47             default  = False,
  48             typecast = bool,
  49             message  = "Stockage des variables internes ou intermédiaires du calcul",
  50             )
  51         self.defineRequiredParameter(
  52             name     = "StoreSupplementaryCalculations",
  53             default  = [],
  54             typecast = tuple,
  55             message  = "Liste de calculs supplémentaires à stocker et/ou effectuer",
  56             listval  = [
  57                 "Analysis",
  58                 "APosterioriCorrelations",
  59                 "APosterioriCovariance",
  60                 "APosterioriStandardDeviations",
  61                 "APosterioriVariances",
  62                 "BMA",
  63                 "CostFunctionJ",
  64                 "CostFunctionJAtCurrentOptimum",
  65                 "CostFunctionJb",
  66                 "CostFunctionJbAtCurrentOptimum",
  67                 "CostFunctionJo",
  68                 "CostFunctionJoAtCurrentOptimum",
  69                 "CurrentOptimum",
  70                 "CurrentState",
  71                 "IndexOfOptimum",
  72                 "InnovationAtCurrentAnalysis",
  73                 "InnovationAtCurrentState",
  74                 "PredictedState",
  75                 "SimulatedObservationAtCurrentAnalysis",
  76                 "SimulatedObservationAtCurrentOptimum",
  77                 "SimulatedObservationAtCurrentState",
  78                 ]
  79             )
  80         self.defineRequiredParameter( # Pas de type
  81             name     = "Bounds",
  82             message  = "Liste des valeurs de bornes",
  83             )
  84         self.requireInputArguments(
  85             mandatory= ("Xb", "Y", "HO", "R", "B"),
  86             optional = ("U", "EM", "CM", "Q"),
  87             )
  88
  89     def run(self, Xb=None, Y=None, U=None, HO=None, EM=None, CM=None, R=None, B=None, Q=None, Parameters=None):
  90         self._pre_run(Parameters, Xb, Y, R, B, Q)
  91         #
  92         if self._parameters["EstimationOf"] == "Parameters":
  93             self._parameters["StoreInternalVariables"] = True
  94         #
  95         # Opérateurs
  96         # ----------
  97         Hm = HO["Direct"].appliedControledFormTo
  98         #
  99         if self._parameters["EstimationOf"] == "State":
 100             Mm = EM["Direct"].appliedControledFormTo
 101         #
 102         if CM is not None and "Tangent" in CM and U is not None:
 103             Cm = CM["Tangent"].asMatrix(Xb)
 104         else:
 105             Cm = None
 106         #
 107         # Nombre de pas identique au nombre de pas d'observations
 108         # -------------------------------------------------------
 109         if hasattr(Y,"stepnumber"):
 110             duration = Y.stepnumber()
 111         else:
 112             duration = 2
 113         #
 114         # Précalcul des inversions de B et R
 115         # ----------------------------------
 116         if self._parameters["StoreInternalVariables"] \
 117             or self._toStore("CostFunctionJ") \
 118             or self._toStore("CostFunctionJb") \
 119             or self._toStore("CostFunctionJo") \
 120             or self._toStore("CurrentOptimum") \
 121             or self._toStore("APosterioriCovariance"):
 122             BI = B.getI()
 123             RI = R.getI()
 124         #
 125         # Initialisation
 126         # --------------
 127         Xn = Xb
 128         Pn = B
 129         #
 130         if len(self.StoredVariables["Analysis"])==0 or not self._parameters["nextStep"]:
 131             self.StoredVariables["Analysis"].store( numpy.ravel(Xn) )
 132             if self._toStore("APosterioriCovariance"):
 133                 self.StoredVariables["APosterioriCovariance"].store( Pn.asfullmatrix(Xn.size) )
 134                 covarianceXa = Pn
 135         #
 136         Xa               = Xn
 137         XaMin            = Xn
 138         previousJMinimum = numpy.finfo(float).max
 139         #
 140         for step in range(duration-1):
 141             if hasattr(Y,"store"):
 142                 Ynpu = numpy.asmatrix(numpy.ravel( Y[step+1] )).T
 143             else:
 144                 Ynpu = numpy.asmatrix(numpy.ravel( Y )).T
 145             #
 146             Ht = HO["Tangent"].asMatrix(ValueForMethodForm = Xn)
 147             Ht = Ht.reshape(Ynpu.size,Xn.size) # ADAO & check shape
 148             Ha = HO["Adjoint"].asMatrix(ValueForMethodForm = Xn)
 149             Ha = Ha.reshape(Xn.size,Ynpu.size) # ADAO & check shape
 150             #
 151             if self._parameters["EstimationOf"] == "State":
 152                 Mt = EM["Tangent"].asMatrix(ValueForMethodForm = Xn)
 153                 Mt = Mt.reshape(Xn.size,Xn.size) # ADAO & check shape
 154                 Ma = EM["Adjoint"].asMatrix(ValueForMethodForm = Xn)
 155                 Ma = Ma.reshape(Xn.size,Xn.size) # ADAO & check shape
 156             #
 157             if U is not None:
 158                 if hasattr(U,"store") and len(U)>1:
 159                     Un = numpy.asmatrix(numpy.ravel( U[step] )).T
 160                 elif hasattr(U,"store") and len(U)==1:
 161                     Un = numpy.asmatrix(numpy.ravel( U[0] )).T
 162                 else:
 163                     Un = numpy.asmatrix(numpy.ravel( U )).T
 164             else:
 165                 Un = None
 166             #
 167             if self._parameters["EstimationOf"] == "State":
 168                 Xn_predicted = numpy.asmatrix(numpy.ravel( Mm( (Xn, Un) ) )).T
 169                 if Cm is not None and Un is not None: # Attention : si Cm est aussi dans M, doublon !
 170                     Cm = Cm.reshape(Xn.size,Un.size) # ADAO & check shape
 171                     Xn_predicted = Xn_predicted + Cm * Un
 172                 Pn_predicted = Q + Mt * Pn * Ma
 173             elif self._parameters["EstimationOf"] == "Parameters":
 174                 # --- > Par principe, M = Id, Q = 0
 175                 Xn_predicted = Xn
 176                 Pn_predicted = Pn
 177             #
 178             if self._parameters["Bounds"] is not None and self._parameters["ConstrainedBy"] == "EstimateProjection":
 179                 Xn_predicted = numpy.max(numpy.hstack((Xn_predicted,numpy.asmatrix(self._parameters["Bounds"])[:,0])),axis=1)
 180                 Xn_predicted = numpy.min(numpy.hstack((Xn_predicted,numpy.asmatrix(self._parameters["Bounds"])[:,1])),axis=1)
 181             #
 182             if self._parameters["EstimationOf"] == "State":
 183                 _HX          = numpy.asmatrix(numpy.ravel( Hm( (Xn_predicted, None) ) )).T
 184                 _Innovation  = Ynpu - _HX
 185             elif self._parameters["EstimationOf"] == "Parameters":
 186                 _HX          = numpy.asmatrix(numpy.ravel( Hm( (Xn_predicted, Un) ) )).T
 187                 _Innovation  = Ynpu - _HX
 188                 if Cm is not None and Un is not None: # Attention : si Cm est aussi dans H, doublon !
 189                     _Innovation = _Innovation - Cm * Un
 190             #
 191             _A = R + numpy.dot(Ht, Pn_predicted * Ha)
 192             _u = numpy.linalg.solve( _A , _Innovation )
 193             Xn = Xn_predicted + Pn_predicted * Ha * _u
 194             Kn = Pn_predicted * Ha * (R + numpy.dot(Ht, Pn_predicted * Ha)).I
 195             Pn = Pn_predicted - Kn * Ht * Pn_predicted
 196             Xa, _HXa = Xn, _HX # Pointeurs
 197             #
 198             # ---> avec analysis
 199             self.StoredVariables["Analysis"].store( Xa )
 200             if self._toStore("SimulatedObservationAtCurrentAnalysis"):
 201                 self.StoredVariables["SimulatedObservationAtCurrentAnalysis"].store( _HXa )
 202             if self._toStore("InnovationAtCurrentAnalysis"):
 203                 self.StoredVariables["InnovationAtCurrentAnalysis"].store( _Innovation )
 204             # ---> avec current state
 205             if self._parameters["StoreInternalVariables"] \
 206                 or self._toStore("CurrentState"):
 207                 self.StoredVariables["CurrentState"].store( Xn )
 208             if self._toStore("PredictedState"):
 209                 self.StoredVariables["PredictedState"].store( Xn_predicted )
 210             if self._toStore("BMA"):
 211                 self.StoredVariables["BMA"].store( Xn_predicted - Xa )
 212             if self._toStore("InnovationAtCurrentState"):
 213                 self.StoredVariables["InnovationAtCurrentState"].store( _Innovation )
 214             if self._toStore("SimulatedObservationAtCurrentState") \
 215                 or self._toStore("SimulatedObservationAtCurrentOptimum"):
 216                 self.StoredVariables["SimulatedObservationAtCurrentState"].store( _HX )
 217             # ---> autres
 218             if self._parameters["StoreInternalVariables"] \
 219                 or self._toStore("CostFunctionJ") \
 220                 or self._toStore("CostFunctionJb") \
 221                 or self._toStore("CostFunctionJo") \
 222                 or self._toStore("CurrentOptimum") \
 223                 or self._toStore("APosterioriCovariance"):
 224                 Jb  = float( 0.5 * (Xa - Xb).T * BI * (Xa - Xb) )
 225                 Jo  = float( 0.5 * _Innovation.T * RI * _Innovation )
 226                 J   = Jb + Jo
 227                 self.StoredVariables["CostFunctionJb"].store( Jb )
 228                 self.StoredVariables["CostFunctionJo"].store( Jo )
 229                 self.StoredVariables["CostFunctionJ" ].store( J )
 230                 #
 231                 if self._toStore("IndexOfOptimum") \
 232                     or self._toStore("CurrentOptimum") \
 233                     or self._toStore("CostFunctionJAtCurrentOptimum") \
 234                     or self._toStore("CostFunctionJbAtCurrentOptimum") \
 235                     or self._toStore("CostFunctionJoAtCurrentOptimum") \
 236                     or self._toStore("SimulatedObservationAtCurrentOptimum"):
 237                     IndexMin = numpy.argmin( self.StoredVariables["CostFunctionJ"][nbPreviousSteps:] ) + nbPreviousSteps
 238                 if self._toStore("IndexOfOptimum"):
 239                     self.StoredVariables["IndexOfOptimum"].store( IndexMin )
 240                 if self._toStore("CurrentOptimum"):
 241                     self.StoredVariables["CurrentOptimum"].store( self.StoredVariables["Analysis"][IndexMin] )
 242                 if self._toStore("SimulatedObservationAtCurrentOptimum"):
 243                     self.StoredVariables["SimulatedObservationAtCurrentOptimum"].store( self.StoredVariables["SimulatedObservationAtCurrentAnalysis"][IndexMin] )
 244                 if self._toStore("CostFunctionJbAtCurrentOptimum"):
 245                     self.StoredVariables["CostFunctionJbAtCurrentOptimum"].store( self.StoredVariables["CostFunctionJb"][IndexMin] )
 246                 if self._toStore("CostFunctionJoAtCurrentOptimum"):
 247                     self.StoredVariables["CostFunctionJoAtCurrentOptimum"].store( self.StoredVariables["CostFunctionJo"][IndexMin] )
 248                 if self._toStore("CostFunctionJAtCurrentOptimum"):
 249                     self.StoredVariables["CostFunctionJAtCurrentOptimum" ].store( self.StoredVariables["CostFunctionJ" ][IndexMin] )
 250             if self._toStore("APosterioriCovariance"):
 251                 self.StoredVariables["APosterioriCovariance"].store( Pn )
 252             if self._parameters["EstimationOf"] == "Parameters" \
 253                 and J < previousJMinimum:
 254                 previousJMinimum    = J
 255                 XaMin               = Xa
 256                 if self._toStore("APosterioriCovariance"):
 257                     covarianceXaMin = Pn
 258         #
 259         # Stockage final supplémentaire de l'optimum en estimation de paramètres
 260         # ----------------------------------------------------------------------
 261         if self._parameters["EstimationOf"] == "Parameters":
 262             self.StoredVariables["Analysis"].store( XaMin )
 263             if self._toStore("APosterioriCovariance"):
 264                 self.StoredVariables["APosterioriCovariance"].store( covarianceXaMin )
 265             if self._toStore("BMA"):
 266                 self.StoredVariables["BMA"].store( numpy.ravel(Xb) - numpy.ravel(XaMin) )
 267         #
 268         self._post_run(HO)
 269         return 0
 270
 271 # ==============================================================================
 272 if __name__ == "__main__":
 273     print('\n AUTODIAGNOSTIC\n')