3. Operands#
3.1. Operand RESULTAT#
♦ RESULTAT
Name of the result object to be enriched if the calculation is continued (see also ETAT_INIT).
3.2. Operand MODELE#
♦ MODELE = mo
Name of the model whose elements are the subject of thermal calculation.
3.3. Operand CHAM_MATER#
♦ CHAM_MATER = chmat
The name of the material field that is assigned on the model.
3.4. Operand CARA_ELEM#
◊ CARA_ELEM = character
This keyword is only necessary in the case of a non-linear thermal resolution with orthotropic material (THER_NL_ORTH). It then makes it possible to transmit to the resolution operator the definition of the local axes of the orthotropy coordinate system in the 2D or 3D environment (command AFFE_CARA_ELEM, keyword MASSIF).
3.5. Keyword EXCIT#
♦ EXCIT =
Keyword factor used to define multiple loads. For each occurrence of the keyword factor, a load is defined, possibly multiplied by a function of time.
3.5.1. Operand CHARGE#
Charge-like concept produced by AFFE_CHAR_THER or by AFFE_CHAR_THER_F [U4.44.02] and by AFFE_CHAR_CINE or by AFFE_CHAR_CINE_F [U4.44.03].
Important note:
For each occurrence of the factor keyword EXCITles different concepts char used must be built on the same model mo.
3.5.2. Operand FONC_MULT#
Multiplicative coefficient as a function of time (concept such as function, sheet or formula) applied to the load.
Important note:
The concomitant use of FONC_MULT with a load containing temperature-dependent thermal loads is prohibited; that is, for loads ECHANGE_, RAYONNEMENT, , SOUR_NLou * FLUNL.
3.6. Operand TYPR_CALCUL#
◊ TYPE_CALCUL =/'TRAN', [DEFAUT]
/”STAT”,
The calculation is scalable if TYPE_CALCUL = “TRAN” or stationary if TYPE_CALCUL =” STAT “. In the case of an evolutionary calculation, the calculation of the initial state must be specified.
3.7. Keyword ETAT_INIT#
If TYPE_CALCUL = “TRAN”
◊ ETAT_INIT =
Allows you to define the initial field from which the evolutionary calculation is carried out. The initial field is assigned the order number 0 and the initial instant takes as value the first real in the instant list as defined by INCREMENT.
Note:
If the keyword ETAT_INIT is absent, we only perform the stationary calculation at the first moment defined under the keyword INCREMENT.
3.7.1. Operand STAT#
/STAT = “OUI”
The initial value is then the result of a previous stationary calculation. This calculation takes into account the boundary conditions defined under the keyword CHARGE and the material characteristics at a given temperature. By default, this temperature is zero. However, it is possible to take another initial temperature field using VALE, CHAM_NO, or EVOL_THER.
3.7.2. Operand VALE#
/VALE = tinit
The initial temperature value is taken to be constant throughout the structure.
3.7.3. Operand CHAM_NO#
/CHAM_NO = finish
The initial value is defined by a temperature_no field (result of operator CREA_CHAMP [U4.72.04]).
3.7.4. Operand EVOL_THER#
/EVOL_THER = time
The initial value is extracted from an evol_ther data structure.
3.7.5. Operand NUME_ORDRE/INST#
/INST = instinci_evol
Order number of the field to be extracted from this data structure. Extraction of the initial thermal state in evol_ther_temp from archive number NUME_ORDRE or archive time INST to continue the calculation. If NUME_ORDRE or INST are not filled in, the last existing archived number in evol is taken.
Note:
Attention, this is the order number in the data structure read again by the previous keyword EVOL_THER. If this data structure was calculated with a list of times different from the one used under the keyword factor INCREMENTde the current resolution, it is imperative to fill in NUME_ORDRE under INCREMENT, the same order number value corresponding to different physical times. In the case where the two lists of moments are identical, it is possible to avoid entering the same one NUME_ORDRE twice, sub ETAT_INIT and sub INCREMENT.
3.7.6. Operand INST_ETAT_INIT#
An instant value istetaini can be associated with this initial state. By default:
when the initial state is defined by the data in the fields, there is no associated moment.
when the state is given by an evol_noli concept, it is the moment in the previous calculation (istetaini = instinci_evol).
3.7.7. Operand PRECISION/CRITERE#
Confer [U4.71.00].
3.8. Keyword INCREMENT#
♦ INCREMENT =_F
Defines the list of calculation times. The operands for the INCREMENT keyword have the same meaning as in document [U4.51.03].
Note:
the list of moments must contain at least two moments, otherwise an error message appears. However, this obligation disappears when a stationary calculation is made (keyword STATIONNAIREdans ETAT_INIT). In this case the calculation result will only contain the initial instant, calculated with the assumption of stationarity.
3.9. Keyword COMPORTEMENT#
◊ COMPORTEMENT =
The drying resolution was added in*Code_Aster* due to the analogy of the thermal and drying equations. This implies assimilating the drying calculation variable, the water concentration, to a variable of the type “TEMP” during the resolution.
By default, the resolution performed will be non-linear thermal. This key word factor therefore makes it possible to distinguish drying resolution from thermal resolution. In addition, the drying equation is characterized by a diffusion coefficient that can be expressed in various forms. This keyword factor also makes it possible to choose one of the drying equations, defined by the expression of its diffusion coefficient, available in Aster. To perform a non-linear thermal calculation, this keyword becomes optional, and the concept of behavior is transparent for the user.
Note:
If the keyword COMPORTEMENTest is absent, a « standard » nonlinear thermal calculation is performed .
3.9.1. Operand RELATION#
♦ RELATION:/'THER_NL' [DEFAUT]
/”THER_HYDR” /”SECH_GRANGER” /”SECH_MENSI” /”SECH_BAZANT” /”SECH_NAPPE”
The syntax and treatment of this keyword are analogous to using the keyword with the same name in the STAT_NON_LINE operator.
/”THER_NL”
Standard nonlinear thermal resolution.
Supported models:
3D continuous media: 3D
2D continuous media: 2D, AXIS
/”THER_HYDR”
Solving the heat equation with an additional source term: \(Q\dot{\xi }\)
\(Q\) is the heat of hydration, which is assumed to be constant. The hydration variable \(\xi\) is a solution of the nonlinear law of evolution, solved simultaneously with the thermal problem:
\(\dot{\xi }\mathrm{=}A(\xi ){e}^{\mathrm{-}E\mathrm{/}\mathit{RT}}\)
Refer to the documentation for operator DEFI_MATERIAU for the meaning of the various parameters.
At the initial moment of the calculation, the hydration variable takes as value the field found under the name “HYDR_ELNO” in the evol_ther data structure defined under ETAT_INIT. By default, the initial water state is taken virgin \(\xi =0\).
Supported models:
3D continuous media: 3D
2D continuous media: 2D, AXIS
/”SECH_GRANGER”
Drying resolution characterized by equation \(\frac{\partial C}{\partial t}-\text{Div}\left[D(C,T)\nabla C\right]=0\)
It is the nonlinear heat equation where the drying variable
plays the role of temperature. The choice of the behavioral relationship makes it possible to define the diffusion coefficient \(D(C,T)\) according to various usual forms in the literature. The Granger formulation of the diffusion coefficient is given by the expression:
\(D(C,T)=A\mathrm{exp}(\mathrm{BC})\frac{T}{{T}_{0}}\mathrm{.}\mathrm{exp}(\frac{{Q}_{s}}{R})(\frac{1}{T}-\frac{1}{{T}_{0}})\)
Refer to the documentation for operator DEFI_MATERIAU for the meaning of the various parameters. In the case of using this law SECH_GRANGER, it is necessary to ensure consistency between the material used and the law of behavior: that is to say that the keyword SECH_GRANGER has been filled in in DEFI_MATERIAU for the material used.
Supported models:
3D continuous media: 3D
2D continuous media: 2D, AXIS
As the drying is resolved by a thermal operator, the supported models are thermal models, but which then have only geometric conceptual value.
/”SECH_MENSI”
Resolution of drying characterized by the law of MENSI.
In the case of using this law SECH_MENSI, it is necessary to ensure consistency between the material used and the law of behavior: that is to say that the keyword SECH_MENSI has been filled in in DEFI_MATERIAU for the material used. Supported models: similar to SECH_GRANGER.
/”SECH_BAZANT”
Resolution of drying characterized by the law of BAZANT.
In the case of using this law SECH_BAZANT, it is necessary to ensure consistency between the material used and the law of behavior: that is to say that the keyword SECH_BAZANT has been filled in in DEFI_MATERIAU for the material used. Supported models: analogous to SECH_GRANGER.
/”SECH_NAPPE”
Drying resolution with a diffusion coefficient defined by an Aster sheet.
In the case of using this law SECH_NAPPE, it is necessary to ensure consistency between the material used and the law of behavior: that is to say that the keyword SECH_NAPPE has been filled in in DEFI_MATERIAU for the material used. Supported models: analogous to SECH_GRANGER.
3.9.2. Operands TOUT/GROUP_MA#
◊/TOUT = 'OUI'
/GROUP_MA = l_grmail
Specifies the meshes on which the behavior relationship is applied. In a manner similar to mechanics, it is possible to use several different drying laws, applied to groups of complementary meshes. On the other hand, thermal cannot be mixed with drying. The “THER_NL” behavior is necessarily applied to the entire mesh, option TOUT: “OUI”, option by default, which is in fact, in general, transparent to the user.
3.10. Operand EVOL_THER_SECH#
◊ EVOL_THER_SECH = to resuther
This operand is specific to the resolution of drying. Drying is solved after a preliminary thermal calculation in the general case, (calculation not coupled but linked thermal/drying), the thermal field intervening as an auxiliary variable, making it possible to calculate the diffusion coefficient of certain laws. It is an input data for the drying calculation, which must be an evol_ther data structure. This keyword is mandatory only for laws” SECH_GRANGER “and” SECH_NAPPE “, whose diffusion coefficient depends on temperature. The thermal evolution data structure given here will have been obtained by a previous execution of a thermal operator, linear or not.
3.11. Operand METHODE#
◊ METHODE = /' NEWTON '
/” NEWTON_KRYLOV “ /” MODELE_REDUIT “
Allows you to choose the method for solving the nonlinear incremental problem.
/' NEWTON '
The Newton-Raphson algorithm is used to solve the problem (see [R5.03.01]).
/' NEWTON_KRYLOV '
An inaccurate version of the Newton-Raphson algorithm is used; the precision of the resolutions of linear systems by an iterative method is adapted during each loading step (see [R5.03.01]).
/' MODELE_REDUIT '
A model reduction method is used to do the non-linear calculation (see [R5.01.05]). It is necessary to have built a reduced base beforehand (command DEFI_BASE_REDUITE).
3.12. Keyword MODELE_REDUIT#
Specify the characteristics of the method for solving the nonlinear incremental problem using a model reduction method (see [R5.01.05]).
/it ♦ BASE_PRIMAL = fashion_empi
66/ DOMAINE_REDUIT = /” NON “[DEFAUT]
/” OUI “
If DOMAINE_REDUIT = “OUI”
♦ GROUP_NO_INTERF = big
The keywords MATRICE, PREDICTION, REAC_ITER, and REAC_INCR have the same meaning and use as in the key word factor NEWTON (see § Error: Reference source not found).
It is necessary to provide an empirical base built on travel (thanks to the operator DEFI_BASE_REDUITE). This base must have been built on the same model and the same mesh as calculation THER_NON_LINE.
The model reduction is not compatible with the following features:
With linear search
With dualized boundary conditions (AFFE_CHAR_THER, AFFE_CHAR_CINE must be used)
It is possible to activate hyper-reduction using a DEIM method. In this case, the calculation is reduced to a restricted mesh area (called RID) and constructed using the DEFI_DOMAINE_REDUIT operator. You must give the group of nodes on which the interface is defined using the keyword GROUP_NO_INTERF.
3.13. Keyword CONVERGENCE#
◊ CONVERGENCE:
Allows you to define the values associated with the convergence criteria.
Note:
If none of the following two operands are present, then everything is as if: RESI_GLOB_RELA = 1.E-6.
3.13.1. Operand RESI_GLOB_RELA#
◊ RESI_GLOB_RELA =/1.e-6
/test
The algorithm continues external iterations as long as the relative residue is greater than testr.
\({(\underset{i=\mathrm{1,}\mathrm{...},\mathrm{nb}\mathrm{ddl}}{\Sigma }{({F}_{i}^{n})}^{2})}^{1/2}/{(\underset{i=\mathrm{1,}\mathrm{...},\mathrm{nb}\mathrm{ddl}}{\Sigma }{({S}_{i})}^{2})}^{1/2}>\mathrm{testr}\)
where \({F}_{i}\) refers to the residue and \({S}_{i}\) the thermal load, the index \(n\) designates the \({n}^{\mathrm{ième}}\) iteration.
3.13.2. Operand RESI_GLOB_MAXI#
/testl
The algorithm continues external iterations as long as the absolute residue is greater than testl.
\(\underset{i=\mathrm{1,}\mathrm{...},\mathrm{nb}\mathrm{ddl}}{\mathrm{max}}\mid {F}_{i}^{n}\mid >\mathrm{test1}\)
where \({F}_{i}\) refers to the residue, the index \(n\) refers to the \({n}^{\mathrm{ième}}\) iteration.
3.13.3. Operand ITER_GLOB_MAXI#
◊ ITER_GLOB_MAXI =/10
/Iterl
The algorithm continues iterations as long as their number is less than iterl.
3.14. Operand SCHEMA_TEMPS#
This operand is only available in the case TYPE_CALCUL = “TRAN”.
◊ THETA =/theta, [R]
/0.57, [DEFAUT]
The theta argument is the parameter of the theta method applied to the evolutionary problem. It must be between 0 (explicit method) and 1 (completely implicit method). In the absence of the keyword, the value used is \(\theta \mathrm{=}0.57\), a little higher than \(\theta \mathrm{=}0.5\) corresponds to the Crank-Nicholson schema. The impact of the choice of theta on the stability of the method is detailed in [R5.02.02].
3.15. Keyword SOLVEUR#
◊ SOLVEUR =
This factor keyword is optional: it allows you to choose another system resolution solver.
This operand is common to all global commands [U4.50.01].
3.16. Keyword ARCHIVAGE#
◊ ARCHIVAGE =
This keyword is optional: by default, all fields calculated for all calculated steps are archived in the result concept resulting from the order. It is used to store certain order numbers in a result data structure and/or to exclude certain fields from storage.
This keyword is identical to its equivalent for the STAT_NON_LINE operator, refer to the documentation [U4.51.03] for the description of the sub-keywords.
Note:
In case of stopping the calculation due to lack of time CPU, the time steps previously calculated are saved in the database.
3.17. Keyword OBSERVATION#
◊ OBSERVATION =
This keyword makes it possible to post-process certain fields to nodes or elements on parts of the model at times of a list (called observation) that is generally more refined than the list of archived moments defined in the keyword ARCHIVAGE [§3.13] (where we store all the fields over the entire model). It is mainly used to save on storage.
This keyword is repeatable and allows the creation of an observation table with the same name as the concept resulting from THER_NON_LINE.
For a description of the syntax of this factor keyword, refer to the documentation for the DYNA_NON_LINE [U4.53.01] equivalent keyword.
3.18. Keyword AFFICHAGE#
◊ AFFICHAGE =
This factor keyword allows you to customize the display of the convergence table.
If this keyword is not entered, the table is built according to the various calculation options and with INFO_RESIDU =” NON “.
For a description of the syntax of this factor keyword, refer to the documentation for the STAT_NON_LINE [U4.51.03] equivalent keyword.
3.19. Operand TITRE#
◊ TITRE = title
Title we want to give to the result stored in temper, an evol_ther-type data structure [U4.03.01].