TY - GEN
T1 - VARIABLE STEP SIZE FXLMS ALGORITHM BASED ON NATURAL LOGARITHM FUNCTION FOR PIEZOELECTRIC ACTIVE VIBRATION CONTROL
AU - Zhu, Ximing
AU - Wang, Wei
AU - Li, Weiguang
AU - Yang, Zhichun
N1 - Publisher Copyright:
© 2023 Proceedings of the International Congress on Sound and Vibration. All rights reserved.
PY - 2023
Y1 - 2023
N2 - Filtered-x least mean square (FxLMS) algorithm is widely used in the field of active vibration control (AVC) because of its small computation and easy implementation. Meanwhile, structural vibration can be suppressed quickly due to its strong adaptability and fast convergence speed. However, there is a contradiction between the steady-state error and the convergence rate of the fixed-step FxLMS algorithm. In this paper, a natural logarithm variable-Step FxLMS (NLVSFxLMS) logarithm is proposed by logarithm function and constructs a nonlinear function between the algorithm step μ and the error signal e(n). In addition, the AVC simulation and experiment of cantilever beam are carried out using piezoelectric macro fiber composite (MFC).Simulation and experimental results show that compared with the traditional FxLMS algorithm and some existing variable step FxLMS (VSSFxLMS) algorithms, NLVSFxLMS algorithm has smaller steady-state error and large step in the beginning.
AB - Filtered-x least mean square (FxLMS) algorithm is widely used in the field of active vibration control (AVC) because of its small computation and easy implementation. Meanwhile, structural vibration can be suppressed quickly due to its strong adaptability and fast convergence speed. However, there is a contradiction between the steady-state error and the convergence rate of the fixed-step FxLMS algorithm. In this paper, a natural logarithm variable-Step FxLMS (NLVSFxLMS) logarithm is proposed by logarithm function and constructs a nonlinear function between the algorithm step μ and the error signal e(n). In addition, the AVC simulation and experiment of cantilever beam are carried out using piezoelectric macro fiber composite (MFC).Simulation and experimental results show that compared with the traditional FxLMS algorithm and some existing variable step FxLMS (VSSFxLMS) algorithms, NLVSFxLMS algorithm has smaller steady-state error and large step in the beginning.
KW - Active vibration control
KW - Piezoelectric macro fiber composite
KW - Variable step size FxLMS
UR - http://www.scopus.com/inward/record.url?scp=85170649324&partnerID=8YFLogxK
M3 - 会议稿件
AN - SCOPUS:85170649324
T3 - Proceedings of the International Congress on Sound and Vibration
BT - Proceedings of the 29th International Congress on Sound and Vibration, ICSV 2023
A2 - Carletti, Eleonora
PB - Society of Acoustics
T2 - 29th International Congress on Sound and Vibration, ICSV 2023
Y2 - 9 July 2023 through 13 July 2023
ER -