src/daComposant/daAlgorithms/ExtendedKalmanFilter.py

   1 # -*- coding: utf-8 -*-
   2 #
   3 # Copyright (C) 2008-2017 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", "CostFunctionJb", "CostFunctionJo", "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(Parameters)
  65         #
  66         if self._parameters["EstimationOf"] == "Parameters":
  67             self._parameters["StoreInternalVariables"] = True
  68         #
  69         # Opérateurs
  70         # ----------
  71         if B is None:
  72             raise ValueError("Background error covariance matrix has to be properly defined!")
  73         if R is None:
  74             raise ValueError("Observation error covariance matrix has to be properly defined!")
  75         #
  76         H = HO["Direct"].appliedControledFormTo
  77         #
  78         if self._parameters["EstimationOf"] == "State":
  79             M = EM["Direct"].appliedControledFormTo
  80         #
  81         if CM is not None and "Tangent" in CM and U is not None:
  82             Cm = CM["Tangent"].asMatrix(Xb)
  83         else:
  84             Cm = None
  85         #
  86         # Nombre de pas identique au nombre de pas d'observations
  87         # -------------------------------------------------------
  88         if hasattr(Y,"stepnumber"):
  89             duration = Y.stepnumber()
  90         else:
  91             duration = 2
  92         #
  93         # Précalcul des inversions de B et R
  94         # ----------------------------------
  95         if self._parameters["StoreInternalVariables"]:
  96             BI = B.getI()
  97             RI = R.getI()
  98         #
  99         # Initialisation
 100         # --------------
 101         Xn = Xb
 102         Pn = B
 103         #
 104         self.StoredVariables["Analysis"].store( Xn.A1 )
 105         if "APosterioriCovariance" in self._parameters["StoreSupplementaryCalculations"]:
 106             self.StoredVariables["APosterioriCovariance"].store( Pn.asfullmatrix(Xn.size) )
 107             covarianceXa = Pn
 108         Xa               = Xn
 109         previousJMinimum = numpy.finfo(float).max
 110         #
 111         for step in range(duration-1):
 112             if hasattr(Y,"store"):
 113                 Ynpu = numpy.asmatrix(numpy.ravel( Y[step+1] )).T
 114             else:
 115                 Ynpu = numpy.asmatrix(numpy.ravel( Y )).T
 116             #
 117             Ht = HO["Tangent"].asMatrix(ValueForMethodForm = Xn)
 118             Ht = Ht.reshape(Ynpu.size,Xn.size) # ADAO & check shape
 119             Ha = HO["Adjoint"].asMatrix(ValueForMethodForm = Xn)
 120             Ha = Ha.reshape(Xn.size,Ynpu.size) # ADAO & check shape
 121             #
 122             if self._parameters["EstimationOf"] == "State":
 123                 Mt = EM["Tangent"].asMatrix(ValueForMethodForm = Xn)
 124                 Mt = Mt.reshape(Xn.size,Xn.size) # ADAO & check shape
 125                 Ma = EM["Adjoint"].asMatrix(ValueForMethodForm = Xn)
 126                 Ma = Ma.reshape(Xn.size,Xn.size) # ADAO & check shape
 127             #
 128             if U is not None:
 129                 if hasattr(U,"store") and len(U)>1:
 130                     Un = numpy.asmatrix(numpy.ravel( U[step] )).T
 131                 elif hasattr(U,"store") and len(U)==1:
 132                     Un = numpy.asmatrix(numpy.ravel( U[0] )).T
 133                 else:
 134                     Un = numpy.asmatrix(numpy.ravel( U )).T
 135             else:
 136                 Un = None
 137             #
 138             if self._parameters["EstimationOf"] == "State":
 139                 Xn_predicted = numpy.asmatrix(numpy.ravel( M( (Xn, Un) ) )).T
 140                 if Cm is not None and Un is not None: # Attention : si Cm est aussi dans M, doublon !
 141                     Cm = Cm.reshape(Xn.size,Un.size) # ADAO & check shape
 142                     Xn_predicted = Xn_predicted + Cm * Un
 143                 Pn_predicted = Q + Mt * Pn * Ma
 144             elif self._parameters["EstimationOf"] == "Parameters":
 145                 # --- > Par principe, M = Id, Q = 0
 146                 Xn_predicted = Xn
 147                 Pn_predicted = Pn
 148             #
 149             if self._parameters["Bounds"] is not None and self._parameters["ConstrainedBy"] == "EstimateProjection":
 150                 Xn_predicted = numpy.max(numpy.hstack((Xn_predicted,numpy.asmatrix(self._parameters["Bounds"])[:,0])),axis=1)
 151                 Xn_predicted = numpy.min(numpy.hstack((Xn_predicted,numpy.asmatrix(self._parameters["Bounds"])[:,1])),axis=1)
 152             #
 153             if self._parameters["EstimationOf"] == "State":
 154                 d  = Ynpu - numpy.asmatrix(numpy.ravel( H( (Xn_predicted, None) ) )).T
 155             elif self._parameters["EstimationOf"] == "Parameters":
 156                 d  = Ynpu - numpy.asmatrix(numpy.ravel( H( (Xn_predicted, Un) ) )).T
 157                 if Cm is not None and Un is not None: # Attention : si Cm est aussi dans H, doublon !
 158                     d = d - Cm * Un
 159             #
 160             _A = R + Ht * Pn_predicted * Ha
 161             _u = numpy.linalg.solve( _A , d )
 162             Xn = Xn_predicted + Pn_predicted * Ha * _u
 163             Kn = Pn_predicted * Ha * (R + Ht * Pn_predicted * Ha).I
 164             Pn = Pn_predicted - Kn * Ht * Pn_predicted
 165             #
 166             self.StoredVariables["Analysis"].store( Xn.A1 )
 167             if "APosterioriCovariance" in self._parameters["StoreSupplementaryCalculations"]:
 168                 self.StoredVariables["APosterioriCovariance"].store( Pn )
 169             if "Innovation" in self._parameters["StoreSupplementaryCalculations"]:
 170                 self.StoredVariables["Innovation"].store( numpy.ravel( d.A1 ) )
 171             if self._parameters["StoreInternalVariables"] or "CurrentState" in self._parameters["StoreSupplementaryCalculations"]:
 172                 self.StoredVariables["CurrentState"].store( Xn )
 173             if self._parameters["StoreInternalVariables"] \
 174                 or "CostFunctionJ" in self._parameters["StoreSupplementaryCalculations"] \
 175                 or "CostFunctionJb" in self._parameters["StoreSupplementaryCalculations"] \
 176                 or "CostFunctionJo" in self._parameters["StoreSupplementaryCalculations"]:
 177                 Jb  = 0.5 * (Xn - Xb).T * BI * (Xn - Xb)
 178                 Jo  = 0.5 * d.T * RI * d
 179                 J   = float( Jb ) + float( Jo )
 180                 self.StoredVariables["CostFunctionJb"].store( Jb )
 181                 self.StoredVariables["CostFunctionJo"].store( Jo )
 182                 self.StoredVariables["CostFunctionJ" ].store( J )
 183                 if J < previousJMinimum:
 184                     previousJMinimum  = J
 185                     Xa                = Xn
 186                     if "APosterioriCovariance" in self._parameters["StoreSupplementaryCalculations"]:
 187                         covarianceXa  = Pn
 188             else:
 189                 Xa = Xn
 190             #
 191         #
 192         # Stockage supplementaire de l'optimum en estimation de parametres
 193         # ----------------------------------------------------------------
 194         if self._parameters["EstimationOf"] == "Parameters":
 195             self.StoredVariables["Analysis"].store( Xa.A1 )
 196             if "APosterioriCovariance" in self._parameters["StoreSupplementaryCalculations"]:
 197                 self.StoredVariables["APosterioriCovariance"].store( covarianceXa )
 198         #
 199         if "BMA" in self._parameters["StoreSupplementaryCalculations"]:
 200             self.StoredVariables["BMA"].store( numpy.ravel(Xb) - numpy.ravel(Xa) )
 201         #
 202         self._post_run(HO)
 203         return 0
 204
 205 # ==============================================================================
 206 if __name__ == "__main__":
 207     print('\n AUTODIAGNOSTIC \n')