Файл: Ufimtsev P. Fundamentals of the physical theory of diffraction (Wiley 2007)(348s) PEo .pdf

ВУЗ: Не указан

Категория: Не указан

Дисциплина: Не указана

Добавлен: 28.06.2024

Просмотров: 960

Скачиваний: 0

ВНИМАНИЕ! Если данный файл нарушает Ваши авторские права, то обязательно сообщите нам.

232 Chapter 9 Multiple Diffraction of Edge Waves

Figure 9.2 An edge wave arising at edge L2 propagates in the direction ˆ1 =

R

21

and undergoes

ki

 

the next diffraction at edge L1. The unit vectors ˆt1 and ˆt2 are tangents to the edges L1 and L2. (Reprinted from Ufimtsev (1989) with the permission of the Journal of Acoustical Society of America).

The amplitude u02 of the equivalent canonical wave is defined by equating the normal derivatives of the real and equivalent incident waves at the diffraction point z1 = r1 = 0:

1

 

∂v (R2, ϕ2, ϕ02)

 

 

ikR

 

 

 

 

 

 

 

 

 

 

 

 

1

∂u2eq

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

= =

=

 

 

R21 sin γ02

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

= r1

 

 

 

 

 

 

 

 

2

∂ϕ2

e

 

2

 

R

2

=

R

21

, ϕ

=

ϕ

 

 

∂ϕ1

 

r1 0, ϕ1

 

(9.5)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2

 

02

 

 

 

 

z1

 

π .

According to this equation,

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1

 

 

 

 

∂v (R21, ϕ2, ϕ02)

 

 

 

 

 

 

 

u02 = −

 

 

 

 

2

 

 

 

 

 

 

 

 

 

eikR21 ,

 

 

with ϕ2 = ϕ02, (9.6)

ikR21 sin γ01 sin γ02

 

 

 

 

 

∂ϕ2

 

 

 

 

 

 

 

or

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

u02 = w02eikR21

 

 

 

 

 

 

(9.7)

with

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1

 

 

 

 

 

 

 

 

 

∂v (R21, ϕ2, ϕ02)

 

=

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

w02 = −

ikR21 sin γ01 sin γ02

 

 

 

 

 

 

∂ϕ2

 

 

ϕ2 ϕ02 .

 

(9.8)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Now, we notice that the Helmholtz equation governing the wave field and its diffraction can be differentiated with respect to the parameter ϕ01 without changing the type of this equation. This means that the derivative (with respect to a free parameter) of any solution of the Helmholtz equation is also a solution of the same equation. The boundary conditions also admit differentiation with respect to ϕ01. In Chapter 7, we found the edge diffracted field generated by the incident plane wave (7.2). The incident

TEAM LinG


9.2 Grazing Diffraction 233

wave (9.4) is the derivative of the wave (7.2). Therefore, the diffracted field generated

by the wave (9.4) can be found by the differentiation of the edge diffracted fields found in Chapter 7, if we replace uinc(ζ ) by u02 = w02 exp(ikR21). Before completing this

procedure, however, we make another observation.

The incident wave propagating in the grazing direction to the plate creates the identical scattering sources jh(0) = 2uinc on both sides of the plate. According to Equation (1.10), the field generated by the sources induced on one side of the plate is completely cancelled by the field generated by the identical sources induced on the opposite side of the same plate. Therefore, in this particular case of the grazing incidence, the field uh(0) equals zero and uh(1) uh(t).

In view of this observation and the one made in the previous paragraph, the elementary edge wave generated at edge L1 is found by the differentiation of Equation (7.97) with the simultaneous replacement of uinc(ζ ) by w02(ζ ) exp(ikR21):

 

 

(t)

 

dζ

 

(t)

 

 

 

 

eik[R1(ζ )+R21(ζ )]

 

 

duh

=

 

w02(ζ )

 

 

Fh , mˆ ) ϕ01=π ·

 

 

.

(9.9)

 

2π

∂ϕ01

R1(ζ )

 

The diffracted wave diverging from the whole

edge L1 is determined respectively

by the integral

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

(t)

1

 

 

 

 

(t)

 

 

 

 

eik R1(ζ )+R21(ζ )]

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

uh

= 2π

 

 

w02(ζ ) ∂ϕ01

Fh

 

 

=π

·

[

R1(ζ )

 

 

 

L1

 

, mˆ ) ϕ01

dζ .

(9.10)

The ray asymptotic of this field is found by application of the stationary-phase technique described in Section 8.1. We omit all intermediate details and present the final expression

 

 

(t)

 

 

1

 

 

 

 

 

eiπ/4

 

 

∂g(ϕ

1

, ϕ

01

, α )

eik(R1+R21) ϕ01=π ,

 

 

uh

= w02st )

R1(1 + R11) sin γ012π k

∂ϕ01

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

(9.11)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

where α1 = 2π,

R1(1 + R11)

> 0 if (1 + R11) > 0, and

 

R1(1 + R11)

=

 

 

 

 

if (1 + R11) < 0. Here, all variable parameters and coordinates

i

 

R1(1 + R11)

are

the functions of the stationary point ζst , for example, γ01st ), ϕ1st ). The caustic

parameter

ρ1st ) is determined according to Equation (8.23) as

 

 

 

 

 

 

 

 

 

 

1

 

 

1

 

 

 

dγ01

 

 

 

ks

 

v

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

ˆ1

· ˆ1

,

 

 

 

(9.12)

 

 

 

 

 

 

 

 

 

= sin γ01

dζ

a1 sin γ01

 

 

 

 

 

 

 

 

 

ρ1

 

 

 

 

 

 

TEAM LinG


234 Chapter 9 Multiple Diffraction of Edge Waves

 

where the unit vector

 

vˆ1 = a1

t1

(9.13)

dζ

 

is the principal normal to edge L1 and a1 is the radius of curvature of this edge at the

ˆs

stationary point ζst . The unit vector k1 shows the directions of the diffracted rays (9.11) that form the diffraction cone. In accordance with Equation (8.7), this vector is defined by the equation

ks

· ˆ

=

ki

· ˆ

= −

cos γ

 

.

(9.14)

t

1

 

t

1

 

01

ˆ

 

 

ˆ1

 

 

 

 

 

Asymptotic expression (9.11) can also be written in the form of Equation (8.13):

(t)

w02st )

∂g(ϕ1, ϕ01, α)

eiπ/4

eik(R1+R21)

,

(9.15)

uh

 

 

 

 

 

 

R1

∂ϕ01

2π k (ζst )

 

where st ) = R21st ) + R1st ). Note that the quantity (ζ ) = d2 (ζ )/dζ 2 is easier to calculate than the caustic parameter ρ1(ζ ) in Equation (9.11).

The above approximations (9.11) and (9.15) are the nonuniform asymptotics. They are singular in the directions ϕ1 = 0, 2π , because

 

 

1 cos

ϕ1

 

 

g(ϕ1, ϕ01, α1)|ϕ01=π = −

2

 

 

→ ∞,

with ϕ1 → 0, 2π . (9.16)

∂ϕ01

2

 

sin2

ϕ1

 

 

 

 

 

 

 

 

2

 

 

 

 

This singularity is a consequence of the singularity of function g(ϕ1, ϕ01, α1) in the directions ϕ1 = π ± ϕ0. It can be treated as shown in Section 7.9.

It is worth noting that, in view of Equation (9.8), the wave (9.11), (9.15) arising due to the grazing/slope diffraction is less in magnitude by a small factor of 1/kR21 compared to the wave (8.29) generated by the ordinary edge diffraction.

9.2.2Electromagnetic Waves

A similar problem for the grazing diffraction of electromagnetic waves was investigated in the paper by Ufimtsev (1991). Its solution is found in the same way as for

acoustic waves. Here, it is supposed that the edge wave traveling from edge L2 to edge

tot

L1 is polarized perpendicularly to the plate S1 (E2 S1) and its magnetic vector is parallel to this plate,

H2totz1 = v2(R2, ϕ2, ϕ02)eikR2 .

(9.17)

In the vicinity of edge L1, this wave is approximated by the equivalent wave (9.4),

eq = eq

where one should set u2 H2z . The quantity u02 in Equation (9.4) is defined by

1

Equation (9.6), and the equivalent wave is written as

eq

= w02eikR21 ,

(9.18)

H2z1

TEAM LinG


9.2 Grazing Diffraction 235

with the quantity w02 defined by Equation (9.8). The elementary edge wave arising at edge L1 is calculated by the differentiation of the EEW given in Equations (7.135) and (7.136), assuming that E0t = 0 and H0t = w02 exp(ikR21).

By the integration of EEWs over edge L1, one finds the wave diffracted at the edge L1:

 

 

 

(t)

 

1

 

 

 

ikR21

 

 

∂G(ζ )

 

 

 

 

 

eikR1

 

 

 

 

E21

=

 

 

Z0 L1

w02(ζ )e

 

 

 

 

 

 

 

 

 

 

dζ1

(9.19)

 

 

 

2π

 

 

 

∂ϕ01

 

 

 

 

R1

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

ϕ01

=

π

 

 

 

 

 

and

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

R1

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

H

(t)

=

1

 

 

 

w02(ζ )e

ikR21

×

∂G(ζ )

 

 

 

 

eikR1

dζ1.

(9.20)

21

 

2π

 

L1

 

 

∂ϕ01

 

 

 

 

R1

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

ϕ01

=

π

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Here are its ray asymptotics derived by the stationary-phase technique:

E

(t)

 

w02st ) ∂g(ϕ1, π , α1)

 

 

eik(R1+R21)+iπ/4

 

21

eϕ1 Z0

 

 

 

 

 

 

 

 

,

(9.21)

sin

2

γ01

 

∂ϕ01

 

 

 

 

 

 

 

 

= ˆ

 

 

2π kR1(1 + R11)

 

(t) = [ × (t)]

H21 R1 E /Z0,

where eˆϕ1 is the unit vector associated with the in the first asymptotic approximation,

(9.22)

polar angle ϕ1 (Fig. 9.1). Therefore,

(t)

w02st ) ∂g(ϕ1, π , α1)

 

 

eik(R1+R21)+iπ/4

 

Eϕ1

= −Z0Hϑ1 = Z0

 

 

 

 

 

.

(9.23)

sin2 γ01

 

 

 

∂ϕ01

 

2π kR1(1 + R11)

 

 

 

 

 

 

 

 

 

 

The caustic parameter ρ1 is defined by Equation (9.12). According to Equation (7.149) and (7.150),

(t)

(t)

/ sin γ01

 

Hϑ1

= −Ht1

(9.24)

and hence

(t)

1

∂g(ϕ1, π , α1)

 

 

eik(R1+R21)+iπ/4

 

Ht1

= w02st )

 

 

 

 

 

.

(9.25)

sin γ01

 

 

 

∂ϕ01

 

2π kR1(1 + R11)

 

 

 

 

 

 

 

 

 

 

This is exactly the same ray asymptotic (9.11) found above for the acoustic waves. Thus, the relationship

TEAM LinG


236 Chapter 9 Multiple Diffraction of Edge Waves

 

 

∂Hinc

 

∂uinc

 

Ht = uh,

if

t

=

 

(9.26)

∂n

∂n

at the scattering edge L1

exists between the acoustic and electromagnetic waves arising due to the grazing diffraction.

9.3 SLOPE DIFFRACTION IN THE CONFIGURATION OF FIGURE 9.1

The following relationship exists between acoustic and electromagnetic waves arising due to the slope diffraction at edge L2 (Fig. 9.1):

uh = Ht , if ∂uinc = ∂Htinc

∂n ∂n

at the diffraction point on edge L2.

Here, ˆt is the tangent to the edge L2 and nˆ is the normal to the plate S1 (Fig. 9.1).

9.3.1 Acoustic Waves

Suppose that an external wave

 

 

 

 

uext = u0eikφ0

(9.27)

generates at edge L1 the diffracted wave (of the type of Equation (8.29)):

 

u1hinc = v1(R1, ϕ1, ϕ01)eikR1

 

= u0

eiπ/4

1

g(ϕ1, ϕ01, α1)eikR1 ,

 

sin γ0

 

 

(9.28)

2π k

R1(1 + R11)

with γ0 = π γ01, cos γ01 = ˆt1 · φ0, and the angle γ01 is shown in Figure 9.2. This wave hits the edge L2 where it undergoes diffraction. That is why we denote it as u1inch . The function g(ϕ1, ϕ01, α1) is defined by Equation (2.64) and is equal to zero in the direction ϕ1 = π to edge L2.

TEAM LinG