propagation of acoustic waves in junction of thin slots
play

Propagation of acoustic waves in junction of thin slots Adrien - PowerPoint PPT Presentation

Propagation of acoustic waves in junction of thin slots Adrien SEMIN (Team AN-EDP, University Paris XI ; and Team-project POems, INRIA Rocquencourt) Joint work with Patrick JOLY and Bertrand MAURY CEMRACS, Wednesday, August 13 th 2008 Outline


  1. Propagation of acoustic waves in junction of thin slots Adrien SEMIN (Team AN-EDP, University Paris XI ; and Team-project POems, INRIA Rocquencourt) Joint work with Patrick JOLY and Bertrand MAURY CEMRACS, Wednesday, August 13 th 2008

  2. Outline Introduction and motivations Mathematic modelling of a 2D model problem Mathematic modelling Matched asymptotic expansions From the 2D problem, giving a 1D simplier problem General idea and jump conditions The first cases Writing the 1D problem Numerical simulations Conclusions and perspectives

  3. Outline Introduction and motivations Mathematic modelling of a 2D model problem Mathematic modelling Matched asymptotic expansions From the 2D problem, giving a 1D simplier problem General idea and jump conditions The first cases Writing the 1D problem Numerical simulations Conclusions and perspectives

  4. Scientific context ◮ I’m currently on the first PhD year, with Patrick Joly (INRIA Rocquencourt) et Bertrand Maury (University Paris XI) ◮ This work is a continuation of the PhD of Sébastien Tordeux (study of the Helmholtz equation between an half-space of R n and a “one-dimensional” domain)

  5. Motivations ◮ Goal : study the propagation of an acoustic wave in a network of thin slots

  6. Motivations ◮ Goal : study the propagation of an acoustic wave in a network of thin slots

  7. Motivations ◮ Goal : study the propagation of an acoustic wave in a network of thin slots ◮ Issue : establish the propagation in the junctions (red circles)

  8. Motivations ◮ Goal : study the propagation of an acoustic wave in a network of thin slots ◮ Issue : establish the propagation in the junctions (red circles) ◮ Our geometry studied just below : two slots of same thickness and one junction

  9. Outline Introduction and motivations Mathematic modelling of a 2D model problem Mathematic modelling Matched asymptotic expansions From the 2D problem, giving a 1D simplier problem General idea and jump conditions The first cases Writing the 1D problem Numerical simulations Conclusions and perspectives

  10. Outline Introduction and motivations Mathematic modelling of a 2D model problem Mathematic modelling Matched asymptotic expansions From the 2D problem, giving a 1D simplier problem General idea and jump conditions The first cases Writing the 1D problem Numerical simulations Conclusions and perspectives

  11. The geometry ε 2 α L ′ ε L ◮ Ω ε is the blue colored domain. ◮ When ε → 0, Ω ε tends to a 1D domain (colored in red), which can be parametrized by s ∈ ] − L , 0 [ ∪ ] 0 , L ′ [ .

  12. The equation we study Find u ε ( t , x ) ∈ R + × Ω ε such that  ∂ 2 u ε  ∂ t 2 ( t , x ) − ∆ u ε ( t , x ) 0 in R + × Ω ε  =     ∂ u ε   0 on R + × ∂ Ω ε = ∂ n   u ε ( 0 , . )  f on Ω ε =    ∂ u ε   g on Ω ε ∂ t ( 0 , . ) = + natural hypothesis on the Cauchy data f and g . The associated energy is � � � 2 � � E ε ( t , u ) = 1 ∂ u � � + |∇ x u | 2 dx � � ε ∂ t Ω ε For the function u ε , we have E ε ( t , u ε ) constant.

  13. Exact 2D solution (computed with FreeFem++) We took the lengths of the slots L = L ′ = 8, and we compute over t ∈ [ 0 , 8 ] . Here, its very difficult to see, but there exist a reflexion phenomena on the left slot.

  14. Restriction on the left slot - t = 0 We can see the form of the initial signal f , which is a 1D Gaussian centered on − L / 2

  15. Restriction on the left slot - t = 8 We can see the form of the solution u ε , which seems to be a derivate of a Gaussian. Numerical simulations show that the amplitude of this signal is like ε .

  16. Limit problem (known since a very long time) 2 α 2 α ε → 0 When ε tends to zero : ◮ u ε tends (in a meaning not precised here) to a limit function u 0 which is 1D with respect to the space, ◮ u 0 satisfies the 1D time-domain equation ∂ 2 u 0 ∂ t 2 − ∂ 2 u 0 ∂ s 2 = 0 on each segment of the 1 D domain ◮ u 0 ( s = 0 − ) = u 0 ( s = 0 + ) and ∂ u 0 ∂ s ( s = 0 − ) = ∂ u 0 ∂ s ( s = 0 + ) (Kirchoff laws)

  17. Limit problem (known since a very long time) 2 α 2 α ε → 0 When ε tends to zero : ◮ u ε tends (in a meaning not precised here) to a limit function u 0 which is 1D with respect to the space, ◮ u 0 satisfies the 1D time-domain equation ∂ 2 u 0 ∂ t 2 − ∂ 2 u 0 ∂ s 2 = 0 on each segment of the 1 D domain ◮ u 0 ( s = 0 − ) = u 0 ( s = 0 + ) and ∂ u 0 ∂ s ( s = 0 − ) = ∂ u 0 ∂ s ( s = 0 + ) (Kirchoff laws) ⇒ u 0 does not depend on the parameter α , however u ε does.

  18. Limit problem (known since a very long time) 2 α 2 α ε → 0 When ε tends to zero : ◮ u ε tends (in a meaning not precised here) to a limit function u 0 which is 1D with respect to the space, ◮ u 0 satisfies the 1D time-domain equation ∂ 2 u 0 ∂ t 2 − ∂ 2 u 0 ∂ s 2 = 0 on each segment of the 1 D domain ◮ u 0 ( s = 0 − ) = u 0 ( s = 0 + ) and ∂ u 0 ∂ s ( s = 0 − ) = ∂ u 0 ∂ s ( s = 0 + ) (Kirchoff laws) ⇒ u 0 does not depend on the parameter α , however u ε does. Our goal : study more precisely the behaviour of u ε with respect to ε

  19. Outline Introduction and motivations Mathematic modelling of a 2D model problem Mathematic modelling Matched asymptotic expansions From the 2D problem, giving a 1D simplier problem General idea and jump conditions The first cases Writing the 1D problem Numerical simulations Conclusions and perspectives

  20. Generalities about this method There exists a lot of litterature about this method, and in a first approximation we can distinguish two schools (which seem to have only few cross-over references) : ◮ The British school (D.G. Crighton and al.) ◮ The Russian school (A.M. Il’in and al.) There exists an alternate method (the multiscale method), which leads to the same calculus and to the same conclusions.

  21. Using the method ◮ Use of a overlapping domain decomposition ◮ Use of an ansatz on each part of the domain decomposition ◮ Injection of the ansatz in the equations (formally) ◮ Use of matching conditions (formally) ◮ Justify the whole development, and give error estimates a posteriori (not treated here)

  22. Overlapping domain decomposition ε 2 α L ′ ε L ϕ − ( ε ) ϕ + ( ε ) Two functions ϕ − , ϕ + ∈ C 1 ( R , R ) such that ◮ 0 < ϕ − ( ε ) < ϕ + ( ε ) ◮ lim ε → 0 ϕ ± ( ε ) = 0 et lim ε → 0 ϕ ± ( ε ) /ε = + ∞

  23. Overlapping domain decomposition Slots zones ε 2 α L ′ − ϕ − ( ε ) ε L − ϕ − ( ε ) ◮ Left slot : Ω − ( ε ) = ( s , ν ) ∈ ] − L , − ϕ − ( ε )[ × ] − ε/ 2 , ε/ 2 [ ◮ Right slot : Ω + ( ε ) = ( s , ν ) ∈ ] ϕ − ( ε ) , L ′ [ × ] − ε/ 2 , ε/ 2 [ ◮ Coordinates scaling : ( s , ν ) �→ ( S , µ ) = ( s , ν/ε ) We note � Ω ± ( ε ) the set Ω ± ( ε ) in the scaled coordinates, and � Ω ± its limit when ε tends to 0.

  24. Overlapping domain decomposition Near-field zone ε 2 α y x ε ϕ + ( ε ) ϕ + ( ε ) ◮ Near-field zone : Ω I ( ε ) ◮ Coordinates scaling : ( x , y ) �→ ( X , Y ) = ( x /ε, y /ε )

  25. Near-field given with the new scaled coordinates 2 α 2 α Y 1 y ε X x ε ϕ + ( ε ) 1 ϕ + ( ε ) /ε scaling of ϕ + ( ε ) amplitude 1 /ε ϕ + ( ε ) /ε ◮ Ω I ( ε ) converges to the point ( 0 , 0 ) ◮ � Ω I ( ε ) converges to one semi-infinite canonical domain � Ω I (thanks to the fact that ϕ + ( ε ) /ε → + ∞ ).

  26. Ansatz (slots zones) Ansatz On the slots Ω + ( ε ) et Ω − ( ε ) , we are looking for u ε on the form ∞ � u ε ( t , S , εµ ) = ε k u k , ± ( t , S , µ ) + o ( ε ∞ ) k = 0 with u k , ± defined on R + × � Ω ±

  27. Equations (slots zones) Putting this previous ansatz on the wave equation with Neumann condition on µ = ± 1 / 2 gives that ◮ ∀ k ∈ N , u k , ± ( t , S , µ ) = u k , ± ( t , S ) ◮ ∀ k ∈ N , we have the following 1D time-domain equations ∂ 2 u k , − − ∂ 2 u k , − = 0 , ( t , s ) ∈ R + × ] − L , 0 [ ∂ t 2 ∂ s 2 ∂ 2 u k , + − ∂ 2 u k , + 0 , ( t , s ) ∈ R + × ] 0 , L ′ [ = ∂ t 2 ∂ s 2

  28. Ansatz (near-field zone) Ansatz On the near-field zone Ω I ( ε ) , we are looking for u ε on the form ∞ � u ε ( t , ε X , ε Y ) = ε k U k ( t , X , Y ) + o ( ε ∞ ) k = 0 with U k defined on R + × � Ω I

  29. Equations (near-field zone) Putting this previous ansatz on the wave equation with Neumann condition on ∂ � Ω I gives that 0 in R + × � ∆ X , Y U 0 = Ω I 0 in R + × � ∆ X , Y U 1 = Ω I ∂ 2 U k in R + × � ∀ k ∈ N , ∆ X , Y U k + 2 = Ω I ∂ t 2 0 on R + × ∂ � ∀ k ∈ N , ∇ X , Y U k .� n = Ω I

  30. Use of matching conditions ε 2 α L ′ ε L ϕ − ( ε ) ϕ + ( ε ) On the gray domains, we have, with evident notations : ∞ � u ε ( t , s , ν ) ε k u k , ± ( t , s ) + o ( ε ∞ ) = k = 0 ∞ � � � t , s ε, ν ε k U k = ε k = 0 for ± s ∈ ] ϕ − ( ε ) , ϕ + ( ε )[ and ν ∈ ] − ε/ 2 , ε/ 2 [

Recommend


More recommend