To address the vibration interference from multiple devices in a floating-raft system, this study proposes a multichannel coupled hybrid adaptive algorithm (H∞-CNGHAFxLMS) based on H∞ control and normalized global error, aimed at enhancing the performance and stability of the traditional filtered-x least mean squares (FxLMS) algorithm. First, a simplified 3-DOF model of the horizontal-longitudinal dynamics of the floating-raft system is established, and the system's dynamic behavior with and without actuators is analyzed. Then, the three-channel FxLMS algorithm is improved by incorporating a feedback path and normalized global error. Simulation results show that the normalized global error–mixed adaptive algorithm (NGHAFxLMS) algorithm provides better vibration suppression performance across three channels compared to the conventional algorithms.To further improve the convergence speed and stability, H∞ control and multichannel coupling compensation are introduced. Simulation results demonstrate that the optimized H∞-CNGHAFxLMS algorithm achieves the lowest steady-state error under the conditions of Kc = 0.2 and γ = 1.0. Experimental validation on a test platform shows that the H∞-CNGHAFxLMS algorithm achieves RMS reductions of 5.52 dB and 5.21 dB under steady-state and non-steady-state conditions, respectively, demonstrating excellent vibration suppression, rapid convergence, and superior robustness.
Gao et al. (Wed,) studied this question.