[1287.5.1] In [20] we obtained the analytical solution of a fractional differential equation of rational order, which we use to analyze our fitting results for model A and model B. [1287.5.2] For a general solution of equations (11) and (12) with arbitrary real see [35]. [1287.5.3] The restriction to rational is not a drawback, because we can approximate and by a rational value on a grid between 0 and 1. [1287.5.4] This number of grid points is chosen to be 20, which keeps computation times reasonably limited as the computing time increases quadratically with the lowest common denominator of with 1.
[page 1288, §1] [1288.1.1] The solution for for model B is a sum of Mittag-Leffler type functions:
(14) |
where is the smallest number for which both and are integers. [1288.1.2] The coefficients are the zeros of the characteristic polynomial
(15) |
the function is defined as [36]
(16) |
[1288.1.3] The coefficients are the solutions of the linear system of equations
(17) | ||||
(18) | ||||
(19) |
[1288.1.4] This solution is only valid if all the roots of the characteristic polynomial in (15) are distinct, which is checked in the computations. [1288.1.5] Because the linear system of equations (17)-(19) is underdetermined we choose one fundamental solution for and a multiplication factor for such that .
[page 1289, §1] [1289.1.1] The analytical solutions are plotted for glycerol at (Fig. 7). [1289.1.2] The fitting values for for model A are and . [1289.1.3] Both values lie in the time interval where the relaxation occurs, which confirms the interpretation of these fitting parameters as relaxation times. [1289.1.4] For model B the fitted times are and . [1289.1.5] So marks the onset of the relaxation and the end.
[1289.2.1] We note that the fractional derivatives appearing in the initial value problem (7) can be generalized to fractional derivatives of arbitrary type introduced in [29] and defined as
(20) |
[1289.2.2] For the case it reduces to the Riemann-Liouville fractional derivative, while for to the Liouville-Caputo-type derivative [33]. [1289.2.3] Because
(21) |
the solution of our initial value problem does not change by replacing the Riemann-Liouville fractional derivatives with these generalized Riemann-Liouville fractional derivatives of type .