3. Modeling A#

3.1. Characteristics of modeling A#

Models POU_D_E and DIS_TR

Number of knots: 19

Number of stitches: 19 i.e. 18 SEG2 and 1 POI1

Node group:

\(\mathrm{PALIER}\text{\_}A\)

\(\mathrm{PALIER}\text{\_}B\)

3.2. Tested sizes and results#

  • First modal calculation: it is of type GEP **.We solve it*via* the operator CALC_MODES + SOLVEUR_MODAL =_F (METHODE =” SORENSEN “) (concept MODES).

\(N°\)

\(\mathrm{FREQ}(\mathrm{Hz})\)

displayed in the .mess

Tolerance

2

124.231

10-6

3

124.231

10-6

4

498.302

10-6

5

498.302

10-6

6

1118.15

10-6

7

1118.15

10-6

8

1993.47

10-6

9

1993.47

10-6

10

2021.39

10-6

11

2850.72

10-6

  • We are also testing the INFO_MODE command. Since GEP is standard (real symmetric matrices) its eigenvalues belong only to the real axis. In this case, we can therefore compare the two counting methods (COMPTAGE/METHODE =” STURM “and” APM “) and check that they actually give the same results.

We thus determine the number of eigenvalues (NB_FREQ) contained strictly in a frequency band [FREQ_MIN, FREQ_MAX] (if Sturm) or in the disk with center FREQ_CENTRE and radius, in frequency, \(\frac{\sqrt{\text{RAYON\_CONTOUR}}}{2\pi }\) (if APM). We specify the counting method used (Sturm or APM).

Concept

FREQ_MIN/ CENTRE_CONTOUR

FREQ_MAX/

RAYON_CONTOUR

NB_FREQ

Counting method

NBMOD01

-1.0

120.0

1 We count \({\lambda }_{1}\).

Sturm

NBMOD02

-1.0

130.0

3 We count \({({\lambda }_{i})}_{\text{i=1,3}}\).

Sturm

NBMOD03

-1.0

1200.0

7 We count \({({\lambda }_{i})}_{\text{i=1,7}}\).

Sturm

NBMOD11

0.0+0.0j

5.684 105

(= \({(\mathrm{120x2}\pi )}^{2}\)) »

1 Same NBMOD01

APM

NBMOD12

0.0+0.0j

6.671 105

(= \({(\mathrm{130x2}\pi )}^{2}\)) »

3 Same NBMOD02

APM

NBMOD13

0.0+0.0j

5.684 107

(= \({(\mathrm{1200x2}\pi )}^{2}\)) »

7 Same NBMOD03

APM

  • Second modal calculation: it is of type QEP *.**It is solved*via using the CALC_MODES + SOLVEUR_MODAL =_F (METHODE =”QZ”) operator (=”QZ”) (concept MODEQ).

\(N°\) [1] _

\(\mathrm{FREQ}(\mathrm{Hz})\) displayed in .mess (= \(\frac{\mathrm{\Im }({\lambda }_{i})}{2\pi }\))

\(\mathit{AMORTISSEMENT}\) displayed in .mess (= \(\frac{\mathrm{-}\mathrm{\Re }({\lambda }_{i})}{\mathrm{\mid }{\lambda }_{i}\mathrm{\mid }}\))

Eigenvalue module (= \(\mathrm{\mid }{\lambda }_{i}\mathrm{\mid }\))

Tolerance

Not applicable

Not included in Code_Aster because true eigenvalue

Not applicable

0

Not applicable

Not applicable

Not included in Code_Aster because true eigenvalue

Not applicable

0

Not applicable

1

123.915 + the conjugate complex

10-11

778.5

0.5

2

124.546 + the conjugate complex

10-09

782.5

0.5

10

2850.72 + the conjugate complex

10-15

18849.5

0.5

11

3099.17 + the conjugate complex

10-11

19472.6

0.5

41

21273.2 + the conjugate complex

10-12

133663.4

0.5

42

21380.2 + the conjugate complex

10-12

134335.7

0.5

  • We are also testing the INFO_MODE command. Since it is a QEP with real matrices, its eigenvalues are either real or complex conjugate. So we can only use the APM method here. It determines the number of eigenvalues (NB_FREQ) contained here strictly in the disk with center CENTRE_CONTOUR and radius RAYON_CONTOUR.

Concept

CENTRE_CONTOUR

RAYON_CONTOUR

NB_FREQ

Enumeration method

NBMOD04

0.0+0.0j

779.114 (= \(\mathrm{124x2}\pi\))

4 We count the 2 zero values + the \(({\lambda }_{\mathrm{1,}}\stackrel{ˉ}{{\lambda }_{1}})\) couple.

APM

NBMOD05

0.0+779.114j (= \(0.0+\mathrm{124x2}\pi j\))

7

2 We count the 2 values \({\lambda }_{1}\) and \({\lambda }_{2}\) without their conjugate.

APM

NBMOD06

0.0+0.0j

1.884 104

(= \(\mathrm{3000x2}\pi\)) »

22 We count the 2 zero values + the \({({\lambda }_{i},\stackrel{ˉ}{{\lambda }_{i}})}_{\text{i=1,10}}\) couples.

APM

NBMOD07

0.0+0.0j

1.338 105

(= \(\mathrm{21300x2}\pi\)) »

84 We count the 2 zero values + the \({({\lambda }_{i},\stackrel{ˉ}{{\lambda }_{i}})}_{\text{i=1,41}}\) couples.

APM

NBMOD08

779.114 (1.0+j) (= \(\mathrm{124x2}\pi (1.0+j)\))

701.203 (= \(\mathrm{0.9x}\mathrm{124x2}\pi\))

0

APM