Piotr Oziabło – Wydział Informatyki Politechniki Białostockiej
Absolwent Informatyki oraz Automatyki i Robotyki na Politechnice Białostockiej. Doktorant na Wydziale Informatyki Politechniki Białostockiej. W ramach pracy naukowej zajmuje się badaniami nad regulatorami PID ułamkowego, zmiennego rzędu, matematyczną teorią sterowania oraz algorytmami optymalizacji. Autor i współautor kilkunastu prac z dziedziny teorii sterowania i regulacji ułamkowego, zmiennego rzędu.
Wykład PDM2021:
- MATLAB Toolbox for fractional-variable-order closed loop systems, Piotr Oziabło/14.12.2021, godz. 16-17
One of the most common control approaches in modern industry is a closed loop PID control. PID controllers as an input take the control error which is a difference between expected and actual (measured) value of the control variable. The control error is in this case multiplied by the constant value, integrated and differentiated which corresponds to the proportional (P), integral (I) and derivative (D) terms of PID controller. As a result of the mentioned operations the control signal is generated which is then send to the controlled object/process. A generalization of PID is a fractional-order PID (FOPID) controller where the integral and derivative terms utilize fractional-order calculus to calculate the output control signal. In the other words FOPID controllers instead of calculating first order integral/difference of the control error, perform the operation of fractional order integration and differentiation. FOPID controllers proved their usefulness in many applications outperforming standard PID approach. Further generalization of FOPID control are fractional-variable-order controllers (FVOPID). In this case the integral/difference terms of the controller are not only fractional but are also functions which values change during the control process. This could potentially result in more robust, efficient, and flexible control processes in comparison to the systems which only perform constant order operations. Since FVOPID concept is relatively new there are not so many tools which support engineers and scientists in the study of such systems. In this work a MATLAB Toolbox is presented that allows to simulate and tune closed-loop systems with fractional-variable-order digital PID (FVOPID) controllers. The toolbox provides MATLAB Simulink blocks of FVOPID based on two different implementations. An additional part of the toolbox is an application with a GUI interface which main purpose is to simulate and find the optimal parameters of FVOPID for a given closed-loop system. The tuning process, in this case, is performed by different optimization methods which are set to minimize the preselected objective function.