Page 469 - Handbook Of Integral Equations
P. 469

where n =[µ] + 1, and*
                                         (k)
                                        f n–µ (a)=0,  k =0, 1, ... , n – 1.                (16)
                                                                                         µ
                   Let µ > 0. We say that a function f(x) ∈ L 1 has an integrable fractional derivative D a+ f if
                n–µ         n
               I a+ f(x) ∈ AC , where n =[µ]+1.
                   In other words, this definition introduces a notion involving only the first of the two condi-
                                                µ
               tions (15) and (16) describing the class I a+ (L 1 ).
                   Let µ > 0. In this case the relation
                                                 µ  µ
                                               D I ϕ(x)= ϕ(x)                              (17)
                                                 a+ a+
               holds for any integrable function ϕ(x), and the relation
                                                   µ
                                                µ
                                               I D f(x)= f(x)                              (18)
                                                a+
                                                   a+
               holds for any function f(x) such that
                                                       µ
                                                 f(x) ∈ I (L 1 ).                          (19)
                                                       a+
               If we replace (19) by the condition that the function f(x) ∈ L 1 (a, b) has an integrable deriva-
                    µ
               tive D a+ f(x), then relation (18) fails in general and must be replaced by the formula
                                                    n–1      µ–k–1
                                                        (x – a)    (n–k–1)
                                    µ
                                       µ
                                   I D f(x)= f(x) –               f n–µ  (a),              (20)
                                       a+
                                    a+
                                                         Γ(µ – k)
                                                    k=0
                                           n–µ
               where n =[µ] + 1 and f n–µ (x)= I a+ f(x). In particular, for 0 < µ < 1 we have
                                                        f 1–µ (a)   µ–1
                                        µ
                                           µ
                                       I D f(x)= f(x) –       (x – a)  .                   (21)
                                        a+  a+
                                                         Γ(µ)
                 8.5-4. The Solution of the Generalized Abel Equation
               Consider the Abel integral equation
                                                x
                                                   y(t)

                                                        dt = f(x),                         (22)
                                                 (x – t) µ
                                               a
               where 0 < µ < 1. Suppose that x ∈ [a, b], f(x) ∈ AC, and y(t) ∈ L 1 , and apply the technique of
               fractional differentiation. We divide Eq. (22) by Γ(1 – µ), and, by virtue of (1), rewrite this equation
               as follows:
                                                     f(x)
                                           1–µ
                                          I a+  y(x)=     ,    x > a.                      (23)
                                                   Γ(1 – µ)
                                                             1–µ
               Let us apply the operator of fractional differentiation D a+ to (23). Using the properties of the
               operators of fractional integration and differentiation, we obtain
                                                       1–µ
                                                      D a+ f(x)
                                                y(x)=         ,                            (24)
                                                      Γ(1 – µ)
                 * From now on in Section 8.5, by f  (n) (x) we mean the nth derivative of f(x) with respect to x and f  (n) (a) ≡ f (n)  (x)      .
                                                                                           x=a



                 © 1998 by CRC Press LLC









               © 1998 by CRC Press LLC
                                                                                                             Page 451
   464   465   466   467   468   469   470   471   472   473   474