default = [],
typecast = tuple,
message = "Liste de calculs supplémentaires à stocker et/ou effectuer",
- listval = ["APosterioriCovariance", "Innovation"]
+ listval = ["APosterioriCovariance", "BMA", "Innovation"]
+ )
+ self.defineRequiredParameter(
+ name = "EstimationType",
+ default = "State",
+ typecast = str,
+ message = "Estimation d'etat ou de parametres",
+ listval = ["State", "Parameters"],
+ )
+ self.defineRequiredParameter(
+ name = "StoreInternalVariables",
+ default = False,
+ typecast = bool,
+ message = "Stockage des variables internes ou intermédiaires du calcul",
)
def run(self, Xb=None, Y=None, U=None, HO=None, EM=None, CM=None, R=None, B=None, Q=None, Parameters=None):
# ----------------------
self.setParameters(Parameters)
#
+ if self._parameters["EstimationType"] == "Parameters":
+ self._parameters["StoreInternalVariables"] = True
+ #
# Opérateurs
# ----------
if B is None:
if R is None:
raise ValueError("Observation error covariance matrix has to be properly defined!")
#
- Ht = HO["Tangent"].asMatrix(None)
- Ha = HO["Adjoint"].asMatrix(None)
+ Ht = HO["Tangent"].asMatrix(Xb)
+ Ha = HO["Adjoint"].asMatrix(Xb)
#
- Mt = EM["Tangent"].asMatrix(None)
- Ma = EM["Adjoint"].asMatrix(None)
+ if self._parameters["EstimationType"] == "State":
+ Mt = EM["Tangent"].asMatrix(Xb)
+ Ma = EM["Adjoint"].asMatrix(Xb)
#
if CM is not None and CM.has_key("Tangent") and U is not None:
- Cm = CM["Tangent"].asMatrix(None)
+ Cm = CM["Tangent"].asMatrix(Xb)
else:
Cm = None
#
# Nombre de pas du Kalman identique au nombre de pas d'observations
# -----------------------------------------------------------------
- duration = Y.stepnumber()
+ if hasattr(Y,"stepnumber"):
+ duration = Y.stepnumber()
+ else:
+ duration = 2
+ #
+ # Précalcul des inversions de B et R
+ # ----------------------------------
+ if self._parameters["StoreInternalVariables"]:
+ if B is not None:
+ BI = B.I
+ elif self._parameters["B_scalar"] is not None:
+ BI = 1.0 / self._parameters["B_scalar"]
+ #
+ if R is not None:
+ RI = R.I
+ elif self._parameters["R_scalar"] is not None:
+ RI = 1.0 / self._parameters["R_scalar"]
#
# Initialisation
# --------------
Xn = Xb
Pn = B
+ #
self.StoredVariables["Analysis"].store( Xn.A1 )
if "APosterioriCovariance" in self._parameters["StoreSupplementaryCalculations"]:
self.StoredVariables["APosterioriCovariance"].store( Pn )
+ covarianceXa = Pn
+ Xa = Xn
+ previousJMinimum = numpy.finfo(float).max
#
for step in range(duration-1):
- Ynpu = numpy.asmatrix(numpy.ravel( Y[step+1] )).T
+ if hasattr(Y,"store"):
+ Ynpu = numpy.asmatrix(numpy.ravel( Y[step+1] )).T
+ else:
+ Ynpu = numpy.asmatrix(numpy.ravel( Y )).T
#
- if Cm is not None:
+ if U is not None:
if hasattr(U,"store") and len(U)>1:
Un = numpy.asmatrix(numpy.ravel( U[step] )).T
elif hasattr(U,"store") and len(U)==1:
Un = numpy.asmatrix(numpy.ravel( U[0] )).T
else:
Un = numpy.asmatrix(numpy.ravel( U )).T
- #
- Xn_predicted = Mt * Xn + Cm * Un
else:
Un = None
- #
+ #
+ if self._parameters["EstimationType"] == "State" and Cm is not None and Un is not None:
+ Xn_predicted = Mt * Xn + Cm * Un
+ Pn_predicted = Mt * Pn * Ma + Q
+ elif self._parameters["EstimationType"] == "State" and (Cm is None or Un is None):
Xn_predicted = Mt * Xn
- Pn_predicted = Mt * Pn * Ma + Q
+ Pn_predicted = Mt * Pn * Ma + Q
+ elif self._parameters["EstimationType"] == "Parameters":
+ # Xn_predicted = Mt * Xn
+ # Pn_predicted = Mt * Pn * Ma + Q
+ # --- > Par principe, M = Id, Q = 0
+ Xn_predicted = Xn
+ Pn_predicted = Pn
+ #
+ if self._parameters["EstimationType"] == "Parameters" and Cm is not None and Un is not None:
+ d = Ynpu - Ht * Xn_predicted - Cm * Un
+ else:
+ d = Ynpu - Ht * Xn_predicted
#
- d = Ynpu - Ht * Xn_predicted
K = Pn_predicted * Ha * (Ht * Pn_predicted * Ha + R).I
Xn = Xn_predicted + K * d
Pn = Pn_predicted - K * Ht * Pn_predicted
#
self.StoredVariables["Analysis"].store( Xn.A1 )
+ if "APosterioriCovariance" in self._parameters["StoreSupplementaryCalculations"]:
+ self.StoredVariables["APosterioriCovariance"].store( Pn )
if "Innovation" in self._parameters["StoreSupplementaryCalculations"]:
self.StoredVariables["Innovation"].store( numpy.ravel( d.A1 ) )
+ if self._parameters["StoreInternalVariables"]:
+ Jb = 0.5 * (Xn - Xb).T * BI * (Xn - Xb)
+ Jo = 0.5 * d.T * RI * d
+ J = float( Jb ) + float( Jo )
+ self.StoredVariables["CurrentState"].store( Xn.A1 )
+ self.StoredVariables["CostFunctionJb"].store( Jb )
+ self.StoredVariables["CostFunctionJo"].store( Jo )
+ self.StoredVariables["CostFunctionJ" ].store( J )
+ if J < previousJMinimum:
+ previousJMinimum = J
+ Xa = Xn
+ if "APosterioriCovariance" in self._parameters["StoreSupplementaryCalculations"]:
+ covarianceXa = Pn
+ else:
+ Xa = Xn
+ #
+ #
+ # Stockage supplementaire de l'optimum en estimation de parametres
+ # ----------------------------------------------------------------
+ if self._parameters["EstimationType"] == "Parameters":
+ self.StoredVariables["Analysis"].store( Xa.A1 )
if "APosterioriCovariance" in self._parameters["StoreSupplementaryCalculations"]:
- self.StoredVariables["APosterioriCovariance"].store( Pn )
+ self.StoredVariables["APosterioriCovariance"].store( covarianceXa )
+ #
+ if "BMA" in self._parameters["StoreSupplementaryCalculations"]:
+ self.StoredVariables["BMA"].store( numpy.ravel(Xb) - numpy.ravel(Xa) )
#
logging.debug("%s Taille mémoire utilisée de %.1f Mo"%(self._name, m.getUsedMemory("M")))
logging.debug("%s Terminé"%self._name)
