2 Copyright (C) 2008-2015 EDF R&D
4 This file is part of SALOME ADAO module.
6 This library is free software; you can redistribute it and/or
7 modify it under the terms of the GNU Lesser General Public
8 License as published by the Free Software Foundation; either
9 version 2.1 of the License, or (at your option) any later version.
11 This library is distributed in the hope that it will be useful,
12 but WITHOUT ANY WARRANTY; without even the implied warranty of
13 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
14 Lesser General Public License for more details.
16 You should have received a copy of the GNU Lesser General Public
17 License along with this library; if not, write to the Free Software
18 Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
20 See http://www.salome-platform.org/ or email : webmaster.salome@opencascade.com
22 Author: Jean-Philippe Argaud, jean-philippe.argaud@edf.fr, EDF R&D
24 .. index:: single: Blue
25 .. _section_ref_algorithm_Blue:
27 Calculation algorithm "*Blue*"
28 ------------------------------
33 This algorithm realizes a BLUE (Best Linear Unbiased Estimator) type estimation
34 of the state of a system. More precisely, it is an Aitken estimator.
36 This algorithm is always the fastest of all the assimilation algorithms of ADAO.
37 It is theoretically reserved for observation operator cases which are linear,
38 even if it sometimes works in "slightly" non-linear cases. One can verify the
39 linearity of the observation operator with the help of the
40 :ref:`section_ref_algorithm_LinearityTest`.
42 In case of non-linearity, even slightly marked, it will be easily prefered the
43 :ref:`section_ref_algorithm_ExtendedBlue` or the
44 :ref:`section_ref_algorithm_3DVAR`.
46 Optional and required commands
47 ++++++++++++++++++++++++++++++
49 .. index:: single: Background
50 .. index:: single: BackgroundError
51 .. index:: single: Observation
52 .. index:: single: ObservationError
53 .. index:: single: ObservationOperator
54 .. index:: single: StoreSupplementaryCalculations
55 .. index:: single: Quantiles
56 .. index:: single: SetSeed
57 .. index:: single: NumberOfSamplesForQuantiles
58 .. index:: single: SimulationForQuantiles
60 The general required commands, available in the editing user interface, are the
64 *Required command*. This indicates the background or initial vector used,
65 previously noted as :math:`\mathbf{x}^b`. Its value is defined as a
66 "*Vector*" or a *VectorSerie*" type object.
69 *Required command*. This indicates the background error covariance matrix,
70 previously noted as :math:`\mathbf{B}`. Its value is defined as a "*Matrix*"
71 type object, a "*ScalarSparseMatrix*" type object, or a
72 "*DiagonalSparseMatrix*" type object.
75 *Required command*. This indicates the observation vector used for data
76 assimilation or optimization, previously noted as :math:`\mathbf{y}^o`. It
77 is defined as a "*Vector*" or a *VectorSerie* type object.
80 *Required command*. This indicates the observation error covariance matrix,
81 previously noted as :math:`\mathbf{R}`. It is defined as a "*Matrix*" type
82 object, a "*ScalarSparseMatrix*" type object, or a "*DiagonalSparseMatrix*"
86 *Required command*. This indicates the observation operator, previously
87 noted :math:`H`, which transforms the input parameters :math:`\mathbf{x}` to
88 results :math:`\mathbf{y}` to be compared to observations
89 :math:`\mathbf{y}^o`. Its value is defined as a "*Function*" type object or
90 a "*Matrix*" type one. In the case of "*Function*" type, different
91 functional forms can be used, as described in the section
92 :ref:`section_ref_operator_requirements`. If there is some control :math:`U`
93 included in the observation, the operator has to be applied to a pair
96 The general optional commands, available in the editing user interface, are
97 indicated in :ref:`section_ref_assimilation_keywords`. In particular, the
98 optional command "*AlgorithmParameters*" allows to choose the specific options,
99 described hereafter, of the algorithm. See
100 :ref:`section_ref_options_AlgorithmParameters` for the good use of this command.
102 The options of the algorithm are the following:
104 StoreSupplementaryCalculations
105 This list indicates the names of the supplementary variables that can be
106 available at the end of the algorithm. It involves potentially costly
107 calculations or memory consumptions. The default is a void list, none of
108 these variables being calculated and stored by default. The possible names
109 are in the following list: ["APosterioriCovariance", "BMA", "CostFunctionJ",
110 "OMA", "OMB", "Innovation", "SigmaBck2", "SigmaObs2",
111 "MahalanobisConsistency", "SimulatedObservationAtBackground",
112 "SimulatedObservationAtOptimum", "SimulationQuantiles"].
114 Example : ``{"StoreSupplementaryCalculations":["BMA","Innovation"]}``
117 This list indicates the values of quantile, between 0 and 1, to be estimated
118 by simulation around the optimal state. The sampling uses a multivariate
119 gaussian random sampling, directed by the *a posteriori* covariance matrix.
120 This option is useful only if the supplementary calculation
121 "SimulationQuantiles" has been chosen. The default is a void list.
123 Example : ``{"Quantiles":[0.1,0.9]}``
126 This key allow to give an integer in order to fix the seed of the random
127 generator used to generate the ensemble. A convenient value is for example
128 1000. By default, the seed is left uninitialized, and so use the default
129 initialization from the computer.
131 Example : ``{"SetSeed":1000}``
133 NumberOfSamplesForQuantiles
134 This key indicates the number of simulation to be done in order to estimate
135 the quantiles. This option is useful only if the supplementary calculation
136 "SimulationQuantiles" has been chosen. The default is 100, which is often
137 sufficient for correct estimation of common quantiles at 5%, 10%, 90% or
140 Example : ``{"NumberOfSamplesForQuantiles":100}``
142 SimulationForQuantiles
143 This key indicates the type of simulation, linear (with the tangent
144 observation operator applied to perturbation increments around the optimal
145 state) or non-linear (with standard observation operator applied to
146 perturbated states), one want to do for each perturbation. It changes mainly
147 the time of each elementary calculation, usually longer in non-linear than
148 in linear. This option is useful only if the supplementary calculation
149 "SimulationQuantiles" has been chosen. The default value is "Linear", and
150 the possible choices are "Linear" and "NonLinear".
152 Example : ``{"SimulationForQuantiles":"Linear"}``
154 Information and variables available at the end of the algorithm
155 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
157 At the output, after executing the algorithm, there are variables and
158 information originating from the calculation. The description of
159 :ref:`section_ref_output_variables` show the way to obtain them by the method
160 named ``get`` of the variable "*ADD*" of the post-processing. The input
161 variables, available to the user at the output in order to facilitate the
162 writing of post-processing procedures, are described in the
163 :ref:`subsection_r_o_v_Inventaire`.
165 The unconditional outputs of the algorithm are the following:
168 *List of vectors*. Each element is an optimal state :math:`\mathbf{x}*` in
169 optimization or an analysis :math:`\mathbf{x}^a` in data assimilation.
171 Example : ``Xa = ADD.get("Analysis")[-1]``
173 The conditional outputs of the algorithm are the following:
175 APosterioriCovariance
176 *List of matrices*. Each element is an *a posteriori* error covariance
177 matrix :math:`\mathbf{A}*` of the optimal state.
179 Example : ``A = ADD.get("APosterioriCovariance")[-1]``
182 *List of vectors*. Each element is a vector of difference between the
183 background and the optimal state.
185 Example : ``bma = ADD.get("BMA")[-1]``
188 *List of values*. Each element is a value of the error function :math:`J`.
190 Example : ``J = ADD.get("CostFunctionJ")[:]``
193 *List of values*. Each element is a value of the error function :math:`J^b`,
194 that is of the background difference part.
196 Example : ``Jb = ADD.get("CostFunctionJb")[:]``
199 *List of values*. Each element is a value of the error function :math:`J^o`,
200 that is of the observation difference part.
202 Example : ``Jo = ADD.get("CostFunctionJo")[:]``
205 *List of vectors*. Each element is an innovation vector, which is in static
206 the difference between the optimal and the background, and in dynamic the
209 Example : ``d = ADD.get("Innovation")[-1]``
211 MahalanobisConsistency
212 *List of values*. Each element is a value of the Mahalanobis quality
215 Example : ``m = ADD.get("MahalanobisConsistency")[-1]``
218 *List of vectors*. Each element is a vector of difference between the
219 observation and the optimal state in the observation space.
221 Example : ``oma = ADD.get("OMA")[-1]``
224 *List of vectors*. Each element is a vector of difference between the
225 observation and the background state in the observation space.
227 Example : ``omb = ADD.get("OMB")[-1]``
230 *List of values*. Each element is a value of the quality indicator
231 :math:`(\sigma^b)^2` of the background part.
233 Example : ``sb2 = ADD.get("SigmaBck")[-1]``
236 *List of values*. Each element is a value of the quality indicator
237 :math:`(\sigma^o)^2` of the observation part.
239 Example : ``so2 = ADD.get("SigmaObs")[-1]``
241 SimulatedObservationAtBackground
242 *List of vectors*. Each element is a vector of observation simulated from
243 the background :math:`\mathbf{x}^b`.
245 Example : ``hxb = ADD.get("SimulatedObservationAtBackground")[-1]``
247 SimulatedObservationAtOptimum
248 *List of vectors*. Each element is a vector of observation simulated from
249 the analysis or optimal state :math:`\mathbf{x}^a`.
251 Example : ``hxa = ADD.get("SimulatedObservationAtOptimum")[-1]``
254 *List of vectors*. Each element is a vector corresponding to the observed
255 state which realize the required quantile, in the same order than the
256 quantiles required by the user.
258 Example : ``sQuantiles = ADD.get("SimulationQuantiles")[:]``
263 References to other sections:
264 - :ref:`section_ref_algorithm_ExtendedBlue`
265 - :ref:`section_ref_algorithm_3DVAR`
266 - :ref:`section_ref_algorithm_LinearityTest`
268 Bibliographical references: