Abstract
This article presents an adaptive modified equivalent-input-disturbance (AMEID) approach for feedback active noise control (ANC) to enhance low-frequency noise rejection. The key idea is to represent low-frequency noise as an equivalent input disturbance (EID) and suppress it through feedback control. The AMEID framework consists of two components. The first is a model-based component that embeds an internal model of the expected EID dynamics to ensure convergence. The second is a data-driven component that improves disturbance-rejection performance. Controller synthesis is performed by solving two discrete-time algebraic Riccati equations subject to a robust stability condition. Compared with conventional model-based EID and data-driven ANC schemes, the AMEID approach achieves faster transient response and greater robustness. Experimental results verify the effectiveness and superiority of the approach.

This publication has 30 references indexed in Scilit: