A Novel Variable Step-Size Active Low-Frequency Train Noise Control Algorithm and Its Verification

Focusing on the problem of reducing the low-frequency noise of trains, this paper studies a new variable step-size active noise control algorithm for train noise reduction, and evaluates its application. This algorithm optimizes the variable step-size convergence coefficient by constructing a feedback error variable signal, and by eliminating the non-zero offset problem of the system. Simulation results show that the proposed algorithm can effectively improve the balance between convergence speed, tracking rate, and steady-state error. The experimental results show that the EV-FxLMS algorithm has a strong noise reduction ability for the single-frequency or mixed-frequency noise of trains below 500 Hz.