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: 950

Скачиваний: 0

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

4.1 Integrals and Asymptotics 75

face ϕ = α):

+f (1),

g(1)

 

 

 

 

 

 

1

sin

π

 

 

 

 

 

 

 

 

 

n

 

=

 

 

 

 

 

n

 

+

cos

π

 

 

cos

ϕ ϕ0

 

 

 

n

 

 

 

 

 

 

 

 

 

n

 

 

 

 

 

 

sin ϕ0

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

cos ϕ + cos ϕ0

.

+

 

 

sin ϕ

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

cos ϕ + cos ϕ0

 

1

cotϕ0) ±

1

cot

π

 

2

2n

n

(4.19)

One should note that the functions f (1) and g(1) are singular in the two special directions

ϕ = 0,

when ϕ0 = π

(4.20)

and

 

 

ϕ = α,

when ϕ0 = α π ,

(4.21)

which relate to the grazing reflections from the faces under the grazing incidence. This is a special case when the integrand in (4.9) cannot be expanded into the Taylor

Figure 4.1 Directivity patterns of edge waves radiated by nonuniform components of the surface sources. The function f (1)(g(1)) corresponds to the case of the acoustically soft (hard) wedge; they also describe the Ez (Hz ) component of the electromagnetic wave scattered at the perfectly conducting wedge. Reprinted from Ufimtsev (1957) with permission of Zhurnal Tekhnicheskoi Fiziki.

TEAM LinG


76 Chapter 4 Radiation by the Nonuniform Component of Surface Sources

Figure 4.2 Directivity patterns of edge waves radiated by nonuniform components of the surface sources. The function f (1)(g(1)) corresponds to the case of the acoustically soft (hard) half-plane; they also describe the Ez (Hz ) component of the electromagnetic wave scattered at the perfectly conducting half-plane. Reprinted from Ufimtsev (1957) with permission of Zhurnal Tekhnicheskoi Fiziki.

series because its terms become infinite. Section 7.9 develops a special version of PTD that is free from the grazing singularity.

Figures 4.1 and 4.2 illustrate the behavior and beauty of functions f (1) and g(1).

4.2INTEGRAL FORM OF FUNCTIONS f (1) AND g(1)

It is well known in antenna and scattering theories that the directivity pattern of the far field can be considered as a conformal Fourier transform of the radiating/scattering sources distributed over antennas/scatterers. This is clearly seen in Equation (1.19). In this section we will establish this type of relationship between the directivity patterns f (1), g(1) and their sources js(1), jh(1) at the wedge.

The geometry of the problem is shown in Figure 3.1 and the incident wave is given by Equation (3.1). The nonuniform components of the surface sources

js(1) = u0Js(1),

jh(1) = u0Jh(1)

(4.22)

radiate the field defined by Equation (1.10). We consider first the radiation from the wedge face ϕ = 0:

 

 

 

 

 

 

 

 

eik

 

 

 

 

(1)

= −

u0

(1)

(kξ , ϕ0)dξ

(xξ )2+y2+ζ 2

dζ ,

(4.23)

us

 

 

Js

 

 

 

4π

 

−∞

'

 

0

(x ξ )2 + y2 + ζ 2

TEAM LinG


4.2 Integral form of Functions f (1) and g(1) 77

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

eik

 

 

 

 

 

 

(1)

= −

u0

 

 

(1)

 

 

(xξ )2+y2+ζ 2

 

 

 

uh

 

 

 

 

 

 

 

 

 

 

Jh

(kξ , ϕ0)dξ

 

 

 

 

 

 

 

 

 

 

 

 

 

dζ .

(4.24)

 

4π ∂y

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

0

 

 

'(x ξ )

2

+ y

2

+ ζ

2

In view of Equation (3.7),

 

 

 

 

−∞

 

 

 

 

 

 

 

 

(1)

 

 

 

 

 

 

 

 

i

 

(1)

(1)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

us

 

= −u0

 

 

 

Js

 

(kξ , ϕ0)H0

 

k((x ξ )2

+ y2

dζ

(4.25)

 

 

 

 

4

0

 

 

and

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

i ∂

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

(1)

= −u0

(1)

 

(1)

k((x ξ )2 + y2

dζ .

(4.26)

 

uh

 

 

 

 

 

Jh

(kξ , ϕ0)H0

 

4

∂y

0

According to Equations (2.61) and (2.63), the functions Js(1) and Jh(1) decrease as (kξ )−3/2 and (kξ )−1/2, respectively, with increasing distance ξ from the edge. At a certain distance ξ = ξeffective, these functions are sufficiently small and can be approx-

imated by zero for ξ ξeff . In the far zone, where r kξeff2 , the Hankel function in Equations (4.25) and (4.26) can be replaced by its asymptotics (2.29). This leads to

the asymptotic expressions

 

u(1)

u

 

ei(kr+π/4) ξeff

J(1)(kξ , ϕ

 

)eikξ cos ϕ dξ

(4.27)

 

 

 

 

 

 

 

 

 

 

 

0 2

 

 

 

 

 

0

 

s

 

2π kr

0

 

 

s

 

 

 

 

and

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

(1)

u0ik sin ϕ

ei(kr+π/4)

ξeff

(1)

(kξ , ϕ0)eikξ cos ϕ dξ .

 

uh

2

 

 

 

 

Jh

(4.28)

2π kr

 

0

These expressions describe the field radiated from the face ϕ = 0. By the replacement of ϕ by α ϕ and ϕ0 by α ϕ0, one can find the field radiated from the face ϕ = α. The total field created by both faces must be equal to that of Equations (4.12) and (4.13). By equating them we obtain the useful relationships

 

 

 

 

1

 

 

ξeff

 

 

 

f (1), ϕ0, α) = −

 

 

 

Js(1)(kξ , ϕ0)eikξ cos ϕ dξ

 

 

 

2

 

0

 

 

 

+

 

ξeff

Js(1)(kξ , α ϕ0)eikξ cosϕ) dξ

 

(4.29)

 

0

 

and

 

 

 

 

 

 

 

 

 

 

 

 

ik

 

 

ξeff

 

 

 

 

g(1)

, ϕ0, α) = −

 

sin ϕ

 

Jh(1)(kξ , ϕ0)eikξ cos ϕ dξ

 

 

2

0

 

 

 

+ sinϕ)

ξeff

Jh(1)(kξ , α ϕ0)eikξ cosϕ) dξ

.

(4.30)

 

0

 

 

 

 

 

 

 

 

 

 

 

 

TEAM LinG


= Ez(t)

78 Chapter 4 Radiation by the Nonuniform Component of Surface Sources

These expressions show that the functions f (1) and g(1) can also be interpreted as the directivity patterns of elementary diffracted waves generated (in the plane normal to the edge) by the sources distributed at the wedge along the lines normal to the edge.

4.3 OBLIQUE INCIDENCE OF A PLANE WAVE AT A WEDGE

For an oblique incidence, the relationship us(t) exists for the diffracted rays generated by the total surface currents, j(t) = j(0) + j(1). However, us(0,1) is not equal to Ez(0,1),

because of the polarization coupling in the electromagnetic PO field.

The complete equivalence uh(0,1,t) = Hz(0,1,t) exists between the acoustic and electromagnetic diffracted rays.

Here, we extend the results of Section 4.1 to the general case when the incident wave propagates under the oblique direction to the edge (Fig. 4.3). It is given by the equation

uinc = u0eik(x cos α˜ +y cos β+z cos γ )

(4.31)

with 0 < γ π/2. We use the “tilde-hat” for the angle α˜ to avoid possible confusion with the external angle α of the wedge.

The boundary conditions on the wedge faces are shown in Equations (2.2) and (2.3). To satisfy these conditions, the diffracted field must have the same dependence on the coordinate z as the incident wave (4.31):

ud = u(r, ϕ)eikz cos γ .

(4.32)

The substitution of this function ud into the wave equation (2.4) leads to the equation for the function u(r, ϕ),

 

u(r, ϕ) + k12u(r, ϕ) = 0, with k1 = k sin γ ,

(4.33)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 4.3 Oblique incidence of a plane wave at a wedge.

TEAM LinG


4.3 Oblique Incidence of a Plane Wave at a Wedge 79

where the Laplacian operator is defined by Equation (2.5). It is also expedient to represent the incident wave (4.31) in the form of Equation (3.1):

uinc = u0eikz cos γ eik1(x cos ϕ0+y sin ϕ0),

(4.34)

where

 

 

 

 

 

sin γ cos ϕ0 = − cos α˜ ,

sin γ sin ϕ0 = − cos β

(4.35)

and

 

 

 

 

 

tan ϕ0

=

cos β

with 0 ≤ ϕ0 < π .

 

 

,

(4.36)

cos α

 

 

˜

 

 

Thus, we reduced the three-dimensional (3-D) diffraction problem for the oblique incidence to the 2-D problem for the normal incidence = π/2) considered in Chapter 2. The solution for the oblique incidence can be automatically found by

simple replacements in the solution for the normal incidence. Namely, the quantity u0 should be replaced by u0eikz cos γ , the wave number k by k1 = k sin γ , and the angle

ϕ0 by ϕ0 = arctan(cos β/ cos α)˜ .

This rule has been established here for the exact solution of the wedge diffraction problem and for its asymptotics. One can show that it is also valid for the PO part of the field. First, by the substitution of the incident wave (4.34) into Equation (1.31)

we find the PO surface sources

 

js(0) = −u0eikz cos γ 2ik1 sin ϕ0eik1x cos ϕ0 ,

(4.37)

jh(0) = u0eikz cos γ 2eik1x cos ϕ0 .

(4.38)

Comparison with Equation (3.2) shows that the sources (4.37) and (4.38) satisfy the above rule for the transition from normal to oblique incidence. If the sources of the field satisfy this rule, one can expect the generated field does too. To verify that, we substitute Equations (4.37) and (4.38) into the original expressions (1.32) for the PO field:

u(0)

=

u

ik1 sin ϕ0

eik1ξ cos ϕ0

0

 

 

 

 

 

s

 

 

2π

0

 

(0)

= −u0

1

eik1ξ cos ϕ0

uh

 

 

 

 

 

 

2π

∂y

0

 

 

 

 

eik

 

 

 

 

ikζ cos γ

(xξ )2+y2+(zζ )2

 

 

dξ

e

 

 

 

 

 

 

 

 

dζ ,

 

 

 

 

 

 

 

 

−∞

 

 

'(x ξ )

2

+ y2

+ (z

2

 

 

 

 

ζ )(4.39)

 

 

 

 

eik

 

 

 

 

ikζ cos γ

 

(xξ )2+y2+(zζ )2

 

 

dξ

e

 

 

 

 

 

 

 

 

dζ ,

 

 

 

 

 

 

 

 

−∞

 

 

'(x ξ )

2

+ y2

+ (z

2

 

 

 

 

ζ )(4.40)

TEAM LinG