13. K modeling#

13.1. Characteristics of modeling#

A non-linear 3D thermal modeling is used.

This modeling tests the creation of empirical thermal modes by incremental POD. Parameters are taken in such a way as to find the same modes as in modeling A.

13.2. Characteristics of the mesh#

The mesh contains 27 elements of type HEXA8.

13.3. Tested sizes and results#

We test some values of the primal base (as they are empirical modes, the tests are* done at the nearest sign):

Identification

Reference type

Point \(A\) - TEMP -Mode 1

NON_REGRESSION

Point \(A\) - TEMP -Mode 2

NON_REGRESSION

Point \(A\) - TEMP -Mode 3

NON_REGRESSION

We test some values of the dual base (as they are empirical modes, the tests are done to the nearest sign):

Identification

Reference type

Point \(A\) - FLUX_NOEU/FLUX - Mode 1

NON_REGRESSION

Point \(A\) - FLUX_NOEU/FLUY -Mode 1

NON_REGRESSION

Point \(A\) - FLUX_NOEU/FLUX - Mode 2

NON_REGRESSION

Point \(A\) - FLUX_NOEU/FLUY -Mode 2

NON_REGRESSION

Point \(A\) - FLUX_NOEU/FLUX - Mode 3

NON_REGRESSION

Point \(A\) - FLUX_NOEU/FLUY -Mode 3

NON_REGRESSION

We test the values of the reduced coordinates (table COOR_REDUIT) \({a}_{m}^{T}\) calculated when extracting empirical modes. For the primal base (temperature):

Identification

Instant

Mode

Reference type

\({a}_{m=1,t=10s}^{T}\)

\(t=10s\)

1

NON_REGRESSION

\({a}_{m=1,t=4s}^{T}\)

\(t=4s\)

1

NON_REGRESSION

\({a}_{m=3,t=10s}^{T}\)

\(t=10s\)

3

NON_REGRESSION

\({a}_{m=3,t=4s}^{T}\)

\(t=4s\)

3

NON_REGRESSION

For the dual base (heat flow \({\Phi}\)):

Identification

Instant

Mode

Reference type

\({a}_{m=1,t=10s}^{{\Phi}}\)

\(t=10s\)

1

NON_REGRESSION

\({a}_{m=1,t=4s}^{{\Phi}}\)

\(t=4s\)

1

NON_REGRESSION

\({a}_{m=4,t=10s}^{{\Phi}}\)

\(t=10s\)

4

NON_REGRESSION

\({a}_{m=4,t=4s}^{{\Phi}}\)

\(t=4s\)

4

NON_REGRESSION

13.4. notes#

The calculated modes are strictly identical between POD and POD incremental.