Chapter 45
Identification of Acoustic Emission
Signals for Bearings Fault Based
on Neural Network of Cuckoo Search
Algorithm
Ping Yang, Yun Li, and Yang Yu
Abstract As the key component of rotating machinery, rolling bearings often fail
to the wear during the long-term operation which directly affects the safe and stable
operation of the units. In this paper, the acoustic emission equipment is used to
collect signals from the rolling bearing fault, and an intelligent fault diagnosis
method based on Cuckoo search (CS) Algorithm optimization neural network is
proposed. The acoustic emission signal analysis from the measured rolling bearing
shows that the stability and accuracy are high by this algorithm, and the average
accuracy reached 93% in the classification. The effectiveness of the newly proposed
algorithm is verified by examples of roller bearing fault diagnosis.
45.1 Introduction
Rolling bearings are one of the important components of rotating machinery. In
rotating machinery, rolling bearing failure can account for up to 30% [1]. The failure
information can be obtained by collecting the state data of rolling bearing, which
can effectively reduce economic losses and accidents. Acoustic emission technology
can obtain early failure information of rolling bearings.
Acoustic emission (AE) is the phenomenon of parts of materials emitting instantaneous elastic waves due to quick release of energy [2]. Acoustic emission characteristic parameters have amplitude, duration, energy count, rise time, count and so
on. The characteristic parameters of acoustic emission are different due to different
faults of the rolling bearing.
The fault diagnosis of rolling bearing has always been concerned by researchers.
Therefore, bearing fault identification has great significance, and the recognition
accuracy of the feature set to the bearing fault is improved while reducing the time
consumption. According to research findings at home and abroad, BP neural network
is the most used and most mature training algorithm for neural network training [3,
P. Yang · Y. Li (B) · Y. Yu
Shenyang University of Technology, Shenyang 110870, China
e-mail: 1786308837@qq.com
© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021
G. Shen et al. (eds.), Advances in Acoustic Emission Technology, Springer Proceedings
in Physics 259, https://doi.org/10.1007/978-981-15-9837-1_45
517
Identification of Acoustic Emission
Signals for Bearings Fault Based
on Neural Network of Cuckoo Search
Algorithm
Ping Yang, Yun Li, and Yang Yu
Abstract As the key component of rotating machinery, rolling bearings often fail
to the wear during the long-term operation which directly affects the safe and stable
operation of the units. In this paper, the acoustic emission equipment is used to
collect signals from the rolling bearing fault, and an intelligent fault diagnosis
method based on Cuckoo search (CS) Algorithm optimization neural network is
proposed. The acoustic emission signal analysis from the measured rolling bearing
shows that the stability and accuracy are high by this algorithm, and the average
accuracy reached 93% in the classification. The effectiveness of the newly proposed
algorithm is verified by examples of roller bearing fault diagnosis.
45.1 Introduction
Rolling bearings are one of the important components of rotating machinery. In
rotating machinery, rolling bearing failure can account for up to 30% [1]. The failure
information can be obtained by collecting the state data of rolling bearing, which
can effectively reduce economic losses and accidents. Acoustic emission technology
can obtain early failure information of rolling bearings.
Acoustic emission (AE) is the phenomenon of parts of materials emitting instantaneous elastic waves due to quick release of energy [2]. Acoustic emission characteristic parameters have amplitude, duration, energy count, rise time, count and so
on. The characteristic parameters of acoustic emission are different due to different
faults of the rolling bearing.
The fault diagnosis of rolling bearing has always been concerned by researchers.
Therefore, bearing fault identification has great significance, and the recognition
accuracy of the feature set to the bearing fault is improved while reducing the time
consumption. According to research findings at home and abroad, BP neural network
is the most used and most mature training algorithm for neural network training [3,
P. Yang · Y. Li (B) · Y. Yu
Shenyang University of Technology, Shenyang 110870, China
e-mail: 1786308837@qq.com
© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021
G. Shen et al. (eds.), Advances in Acoustic Emission Technology, Springer Proceedings
in Physics 259, https://doi.org/10.1007/978-981-15-9837-1_45
517