Current version: 1.03 (updated 2018-03-28)

**[Download]** **[Instructions]**

Useful? Please cite either [1] or [2].

(Update 2019-02-22) Octave and Matlab solvers (version 1.01) that can be applied for nonlinear problems available **here**. A detailed description on their usage is given in the respective .m files.

SubIval (the **subi**nter**val**-based method; first appearance in [1]) is a numerical method for computations of the fractional derivative in IVPs (initial value problems). In a computed time step its usage results in an implicit formula much like one that can be obtained then applying an implicit BDF (backward differentiation formula) for a first order derivative.

The formula resulting from SubIval is:

\mathrm{D}

_{t}

^{\mathrm{\alpha}}

x(t)

\approx

ax(t)

+ b

.

\)

(1)

SubIval works so far for the Riemann-Liouville and Caputo definitions of the fractional derivative.

### References

- M.Sowa: “A subinterval-based method for circuits with fractional order elements”. Bull. Pol. Acad. Sci., Tech. Sci. 2014 vol. 62 no. 3, pp. 449-454 (2014).
- M.Sowa: “Application of SubIval in solving initial value problems with fractional derivatives”, Applied Mathematics and Computation 319, 86-103 (2018).