Title of article :
Approximations of higher-order fractional differentiators and integrators using indirect discretization
Author/Authors :
YADAV, Richa Netaji Subhas Institute of Technology - Advanced Electronics Lab, Division of Electronics and Communication Engineering, India , GUPTA, Maneesha Netaji Subhas Institute of Technology - Advanced Electronics Lab, Division of Electronics and Communication Engineering, India
Abstract :
This paper describes new approximations of fractional order integrators (FOIs) and fractional order differentiators (FODs) by using a continued fraction expansion-based indirect discretization scheme. Different tenth-order fractional blocks have been derived by applying three different s-to-z transforms described earlier by Al-Alaoui, namely new two-segment, four-segment, and new optimized four-segment operators. A new addition has been done in the new optimized four-segment operator by modifying it by the zero reflection method. All proposed half (s ±1/2) and one-fourth (s ±1/4) differentiator and integrator models fulfill the stability criterion. The tenth-order fractional differ-integrators (s ±α) based on the modified new optimized four-segment rule show tremendously improved results with relative magni- tude errors (dB) of ≤ –15 dB for α = 1/2 and ≤ –20 dB for α = 1/4 in the full range of Nyquist frequency so these have been further analyzed. The main contribution of this paper lies in the reduction of these tenth-order blocks into four new fifth-order blocks of half and one-fourth order models of FODs and FOIs. The analyses of magnitude and phase responses show that the proposed new fifth-order half and one-fourth differ-integrators closely approximate their ideal counterparts and outperform the existing ones.
Keywords :
Fractional order differentiators , fractional order integrators , half and one , fourth order differentiators and integrators , continued fraction expansion
Journal title :
Turkish Journal of Electrical Engineering and Computer Sciences
Journal title :
Turkish Journal of Electrical Engineering and Computer Sciences