An Extended Active Resonance Suppression Scheme Based on a Dual-Layer Network for High-Performance Double-Inertia Drive System

Reducing the system order and requiring precise knowledge of the system states and mechanical parameters are of crucial significance for suppressing the mechanical resonance of double-inertia servo systems. Committed to this article, both an extended active resonance suppression (EARS) scheme and a dual-layer network using five improved sliding-mode observers (ISMOs) are presented in this article. An integral-proportional controller with two feedback loops as well as two feedforward items is developed for facilitating our EARS structure to suppress the mechanical resonance more thoroughly. Interestingly, based on the pole-zero cancellation, two adapting factors concerning two feedforward items are ingeniously exploited for diminishing the system order to improve the dynamic response and the system stability. Furthermore, for the zero-point shift phenomenon, our EARS scheme can cope with it instinctively. In the dual-layer network, an inline gain with a reaching law in each ISMO is designed to make the identification parameters converge exponentially to their actual values. It is worth noting that the reaching law, consisting of a convergent factor and a chattering-suppression factor, enables to adjust the convergent rate with independent-adaption suppression of the sliding-mode chatter. Abundant simulations and experimental results are rendered to verify the viability.