src/daComposant/daAlgorithms/ExtendedKalmanFilter.py

   1 #-*-coding:iso-8859-1-*-
   2 #
   3 # Copyright (C) 2008-2016 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  = ["APosterioriCorrelations", "APosterioriCovariance", "APosterioriStandardDeviations", "APosterioriVariances", "BMA", "CurrentState", "CostFunctionJ", "Innovation"]
  57             )
  58         self.defineRequiredParameter( # Pas de type
  59             name     = "Bounds",
  60             message  = "Liste des valeurs de bornes",
  61             )
  62
  63     def run(self, Xb=None, Y=None, U=None, HO=None, EM=None, CM=None, R=None, B=None, Q=None, Parameters=None):
  64         self._pre_run()
  65         #
  66         # Paramètres de pilotage
  67         # ----------------------
  68         self.setParameters(Parameters)
  69         #
  70         if self._parameters.has_key("Bounds") and (type(self._parameters["Bounds"]) is type([]) or type(self._parameters["Bounds"]) is type(())) and (len(self._parameters["Bounds"]) > 0):
  71             Bounds = self._parameters["Bounds"]
  72             logging.debug("%s Prise en compte des bornes effectuee"%(self._name,))
  73         else:
  74             Bounds = None
  75         if self._parameters["EstimationOf"] == "Parameters":
  76             self._parameters["StoreInternalVariables"] = True
  77         #
  78         # Opérateurs
  79         # ----------
  80         if B is None:
  81             raise ValueError("Background error covariance matrix has to be properly defined!")
  82         if R is None:
  83             raise ValueError("Observation error covariance matrix has to be properly defined!")
  84         #
  85         H = HO["Direct"].appliedControledFormTo
  86         #
  87         if self._parameters["EstimationOf"] == "State":
  88             M = EM["Direct"].appliedControledFormTo
  89         #
  90         if CM is not None and CM.has_key("Tangent") and U is not None:
  91             Cm = CM["Tangent"].asMatrix(Xb)
  92         else:
  93             Cm = None
  94         #
  95         # Nombre de pas identique au nombre de pas d'observations
  96         # -------------------------------------------------------
  97         if hasattr(Y,"stepnumber"):
  98             duration = Y.stepnumber()
  99         else:
 100             duration = 2
 101         #
 102         # Précalcul des inversions de B et R
 103         # ----------------------------------
 104         if self._parameters["StoreInternalVariables"]:
 105             BI = B.getI()
 106             RI = R.getI()
 107         #
 108         # Initialisation
 109         # --------------
 110         Xn = Xb
 111         Pn = B
 112         #
 113         self.StoredVariables["Analysis"].store( Xn.A1 )
 114         if "APosterioriCovariance" in self._parameters["StoreSupplementaryCalculations"]:
 115             self.StoredVariables["APosterioriCovariance"].store( Pn.asfullmatrix(Xn.size) )
 116             covarianceXa = Pn
 117         Xa               = Xn
 118         previousJMinimum = numpy.finfo(float).max
 119         #
 120         for step in range(duration-1):
 121             if hasattr(Y,"store"):
 122                 Ynpu = numpy.asmatrix(numpy.ravel( Y[step+1] )).T
 123             else:
 124                 Ynpu = numpy.asmatrix(numpy.ravel( Y )).T
 125             #
 126             Ht = HO["Tangent"].asMatrix(ValueForMethodForm = Xn)
 127             Ht = Ht.reshape(Ynpu.size,Xn.size) # ADAO & check shape
 128             Ha = HO["Adjoint"].asMatrix(ValueForMethodForm = Xn)
 129             Ha = Ha.reshape(Xn.size,Ynpu.size) # ADAO & check shape
 130             #
 131             if self._parameters["EstimationOf"] == "State":
 132                 Mt = EM["Tangent"].asMatrix(ValueForMethodForm = Xn)
 133                 Mt = Mt.reshape(Xn.size,Xn.size) # ADAO & check shape
 134                 Ma = EM["Adjoint"].asMatrix(ValueForMethodForm = Xn)
 135                 Ma = Ma.reshape(Xn.size,Xn.size) # ADAO & check shape
 136             #
 137             if U is not None:
 138                 if hasattr(U,"store") and len(U)>1:
 139                     Un = numpy.asmatrix(numpy.ravel( U[step] )).T
 140                 elif hasattr(U,"store") and len(U)==1:
 141                     Un = numpy.asmatrix(numpy.ravel( U[0] )).T
 142                 else:
 143                     Un = numpy.asmatrix(numpy.ravel( U )).T
 144             else:
 145                 Un = None
 146             #
 147             if self._parameters["EstimationOf"] == "State":
 148                 Xn_predicted = numpy.asmatrix(numpy.ravel( M( (Xn, Un) ) )).T
 149                 if Cm is not None and Un is not None: # Attention : si Cm est aussi dans M, doublon !
 150                     Cm = Cm.reshape(Xn.size,Un.size) # ADAO & check shape
 151                     Xn_predicted = Xn_predicted + Cm * Un
 152                 Pn_predicted = Q + Mt * Pn * Ma
 153             elif self._parameters["EstimationOf"] == "Parameters":
 154                 # --- > Par principe, M = Id, Q = 0
 155                 Xn_predicted = Xn
 156                 Pn_predicted = Pn
 157             #
 158             if Bounds is not None and self._parameters["ConstrainedBy"] == "EstimateProjection":
 159                 Xn_predicted = numpy.max(numpy.hstack((Xn_predicted,numpy.asmatrix(Bounds)[:,0])),axis=1)
 160                 Xn_predicted = numpy.min(numpy.hstack((Xn_predicted,numpy.asmatrix(Bounds)[:,1])),axis=1)
 161             #
 162             if self._parameters["EstimationOf"] == "State":
 163                 d  = Ynpu - numpy.asmatrix(numpy.ravel( H( (Xn_predicted, None) ) )).T
 164             elif self._parameters["EstimationOf"] == "Parameters":
 165                 d  = Ynpu - numpy.asmatrix(numpy.ravel( H( (Xn_predicted, Un) ) )).T
 166                 if Cm is not None and Un is not None: # Attention : si Cm est aussi dans H, doublon !
 167                     d = d - Cm * Un
 168             #
 169             _A = R + Ht * Pn_predicted * Ha
 170             _u = numpy.linalg.solve( _A , d )
 171             Xn = Xn_predicted + Pn_predicted * Ha * _u
 172             Kn = Pn_predicted * Ha * (R + Ht * Pn_predicted * Ha).I
 173             Pn = Pn_predicted - Kn * Ht * Pn_predicted
 174             #
 175             self.StoredVariables["Analysis"].store( Xn.A1 )
 176             if "APosterioriCovariance" in self._parameters["StoreSupplementaryCalculations"]:
 177                 self.StoredVariables["APosterioriCovariance"].store( Pn )
 178             if "Innovation" in self._parameters["StoreSupplementaryCalculations"]:
 179                 self.StoredVariables["Innovation"].store( numpy.ravel( d.A1 ) )
 180             if self._parameters["StoreInternalVariables"]:
 181                 Jb  = 0.5 * (Xn - Xb).T * BI * (Xn - Xb)
 182                 Jo  = 0.5 * d.T * RI * d
 183                 J   = float( Jb ) + float( Jo )
 184                 if self._parameters["StoreInternalVariables"] or "CurrentState" in self._parameters["StoreSupplementaryCalculations"]:
 185                     self.StoredVariables["CurrentState"].store( Xn )
 186                 self.StoredVariables["CostFunctionJb"].store( Jb )
 187                 self.StoredVariables["CostFunctionJo"].store( Jo )
 188                 self.StoredVariables["CostFunctionJ" ].store( J )
 189                 if J < previousJMinimum:
 190                     previousJMinimum  = J
 191                     Xa                = Xn
 192                     if "APosterioriCovariance" in self._parameters["StoreSupplementaryCalculations"]:
 193                         covarianceXa  = Pn
 194             else:
 195                 Xa = Xn
 196             #
 197         #
 198         # Stockage supplementaire de l'optimum en estimation de parametres
 199         # ----------------------------------------------------------------
 200         if self._parameters["EstimationOf"] == "Parameters":
 201             self.StoredVariables["Analysis"].store( Xa.A1 )
 202             if "APosterioriCovariance" in self._parameters["StoreSupplementaryCalculations"]:
 203                 self.StoredVariables["APosterioriCovariance"].store( covarianceXa )
 204         #
 205         if "BMA" in self._parameters["StoreSupplementaryCalculations"]:
 206             self.StoredVariables["BMA"].store( numpy.ravel(Xb) - numpy.ravel(Xa) )
 207         #
 208         self._post_run(HO)
 209         return 0
 210
 211 # ==============================================================================
 212 if __name__ == "__main__":
 213     print '\n AUTODIAGNOSTIC \n'