Regresar

PID-2DOF-ACO speed controller for DC motor on ARM platform

Abstract:

The following document shows the development of a PID-2DOF controller tuned by Ant Colony Optimization-ACO (Ant Colony Optimization), to control the speed of a DC motor. A PID-1DOF controller does not have the same benefits as a PID-2DOF controller because the latter allows a good response to sudden setpoint changes and disturbances. These two PID configurations can be tuned with conventional methods such as Ziegler-Nichols, but the values obtained are not the best compared to ACO tuning. The ACO belongs to the family of swarm intelligence algorithms, the one used in this project is the Ants System-AS (Ant System), it needs a cost function that adjusts the optimization based on a performance index, as the Integral of the Absolute Error in Time-ITAE (Integral Time Absolute Error). The hardware used for the experimentation was a DC motor with the STM32F4 programming card, to obtain experimental values; and MATLAB software was the computational tool that combined ACO tuning with the PID-2DOF controller structure. The comparison results were made with the Wilcoxon method.