TY - GEN
T1 - Research on acoustic emission source localization of carbon fiber composite plate based on wavelet neural network
AU - Wang, Yin Ling
AU - Li, Hua Cong
N1 - Publisher Copyright:
© 2019 IEEE.
PY - 2019/7
Y1 - 2019/7
N2 - The material will generate elastic waves during the damage, and analyze the damage signal to achieve accurate positioning of the material damage, which is called acoustic emission (AE) damage source localization. The carbon fiber composite material plate is an anisotropic material, and the propagation speeds of the acoustic emission waves are different in different directions, and it is difficult to achieve accurate positioning of the damage by conventional methods. In this paper, a method based on wavelet neural network for acoustic emission damage source location is proposed. Firstly, the characteristics of the mode wave and the wave velocity propagation characteristics in the carbon fiber composite board are analyzed. The appropriate acoustic emission signal mode is selected as the measurement signal. Then, according to the propagation velocity of the modal wave along different directions of the carbon fiber composite plate, the relationship between the velocity and the angle of the carbon fiber is fitted; the time difference between the arbitrary position of the carbon fiber composite plate and the four sensors is obtained by the fitting speed relationship formula, and the time difference is calculated. The relationship between the time difference of all positions and the positioning position; then some data is selected as the training data, and the wavelet is passed. The damage localization model of carbon fiber composite board was established by neural network. Finally, the acoustic emission source localization experiment of carbon fiber composite board was carried out to prove the correctness of this method.
AB - The material will generate elastic waves during the damage, and analyze the damage signal to achieve accurate positioning of the material damage, which is called acoustic emission (AE) damage source localization. The carbon fiber composite material plate is an anisotropic material, and the propagation speeds of the acoustic emission waves are different in different directions, and it is difficult to achieve accurate positioning of the damage by conventional methods. In this paper, a method based on wavelet neural network for acoustic emission damage source location is proposed. Firstly, the characteristics of the mode wave and the wave velocity propagation characteristics in the carbon fiber composite board are analyzed. The appropriate acoustic emission signal mode is selected as the measurement signal. Then, according to the propagation velocity of the modal wave along different directions of the carbon fiber composite plate, the relationship between the velocity and the angle of the carbon fiber is fitted; the time difference between the arbitrary position of the carbon fiber composite plate and the four sensors is obtained by the fitting speed relationship formula, and the time difference is calculated. The relationship between the time difference of all positions and the positioning position; then some data is selected as the training data, and the wavelet is passed. The damage localization model of carbon fiber composite board was established by neural network. Finally, the acoustic emission source localization experiment of carbon fiber composite board was carried out to prove the correctness of this method.
KW - Acoustic emission
KW - Lamb wave
KW - Source localization
KW - Time difference positioning
KW - Wavelet neural network
UR - http://www.scopus.com/inward/record.url?scp=85074816122&partnerID=8YFLogxK
U2 - 10.1109/ICMAE.2019.8880963
DO - 10.1109/ICMAE.2019.8880963
M3 - 会议稿件
AN - SCOPUS:85074816122
T3 - 2019 IEEE 10th International Conference on Mechanical and Aerospace Engineering, ICMAE 2019
SP - 302
EP - 305
BT - 2019 IEEE 10th International Conference on Mechanical and Aerospace Engineering, ICMAE 2019
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 10th IEEE International Conference on Mechanical and Aerospace Engineering, ICMAE 2019
Y2 - 22 July 2019 through 25 July 2019
ER -