| Abstract |
The paper presents a novel, modified adaptive control algorithm for use in the position loop of servo drives. Speed loop and the drive are assumed to comprise a fast, robust, dominantly single-pole system. The algorithm is a high-gain model-reference control scheme with added resetting and anti wind-up mechanisms. The closed-loop system follows the reference model in a sliding mode fashion during the initial part of a positioning transient, while in the final part, it behaves like a linear high-gain system. This results in the robustness of a sliding mode control, with the steady state chattering avoided. The simulations for 10:1 change in drlve's moment of inertia show the small signal bandwidth practically unchanged, while the overall transient time is considerably reduced compared to conventional design. |
