On-line estimation of the Caputo fractional derivatives with application to PIµD? control

Salim Ibrir

Research output: Chapter in Book/Report/Conference proceedingConference contribution


This paper proposes new procedures for calculation of the Caputo derivative of model-free measured signals. The evaluation of the non-integer derivative is realized by integrating a set of ordinary differential equations and convolution. The derivative of order ? (0 < ? < 2) is seen as an output of a linear-time-varying system driven by a time-dependent known signal. Two procedures are proposed depending on the variation range of the non-integer differentiation order. The proposed formulations facilitate the estimation of the fractional derivatives when they are associated to dynamical systems represented by integer-order differential equations. The efficiency of the developed numerical procedures are validated and compared to exact fractional derivatives for different values of ?. It is shown that PIµD? controllers can be easily realized by system augmentation and convolution. The advantages, the straightforwardness and the main features of the proposed design are given.
Original languageEnglish (US)
Title of host publicationIFAC-PapersOnLine
PublisherElsevier BV
Number of pages6
StatePublished - Apr 14 2021
Externally publishedYes

Bibliographical note

KAUST Repository Item: Exported on 2022-07-01
Acknowledgements: The author thanks King Fahd University of Petroleum and Minerals for supporting this research and acknowledge the support of King Abdulaziz City for Science and Technology (KACST) Technology Innovation Center (TIC) for Solid-State Lighting (SSL) grant EE002381, which is sub-awarded to KFUPM, from the primary grant KACST TIC R2-FP-008 awarded to King Abdullah University of Science and Technology (KAUST).
This publication acknowledges KAUST support, but has no KAUST affiliated authors.


Dive into the research topics of 'On-line estimation of the Caputo fractional derivatives with application to PIµD? control'. Together they form a unique fingerprint.

Cite this