, Duration: 11:25
Materials: [ Cód.: PIDcontrolAPP1.0.zip ]
This video constitutes ‘Part 2’ of the trial-and-error heuristic tuning of a PID
controller for a double-integrator (motion control) plant; PD control was
discussed in video [
First, it is intuitively justified that the final value in response to reference changes does achieve a stationary position error equal to zero (justified by the fact that when one is in place and at zero speed, then zero force is required to stay there, which is exactly what a PD regulator does). However, in response to disturbances there is a constant position error... the explanation is by analogy with a tilted platform: then at the reference point zero force is not the required value to achieve equilibrium, so PD has an offset when equilibrium is reached. This error is corrected by the integral action.
The final part of the video discusses that integral action to optimize the response to disturbances (unmodeled input forces) worsens the response to reference and that, to avoid overshoot due to error accumulation, the proportional part can be ‘fooled’ by lowering a certain parameter ‘b’.... Saturation also requires antiwindup (to compensate for excessive integral accumulation in the initial phases of the transient).
With all this, a PID regulator with derivative filter and with ‘2 degrees of freedom’ and ‘antiwindup’ modifications is finally designed (in a completely ‘math-free’ way) that achieves a satisfactory response to step, ramp and unmodeled step force changes (input disturbances).
‘Intuitive’ tuning strategies are extremely important in practice, because, in simple processes, it could be a quick option to solve a problem by a technician not specialized in control theory.
*Link to my [ whole collection] of videos in English. Link to larger [ Colección completa] in Spanish.