Research on Fault Feature Extraction and Recognition of Rolling Bearings
- PDF / 913,894 Bytes
- 11 Pages / 595.276 x 790.866 pts Page_size
- 80 Downloads / 277 Views
Research on Fault Feature Extraction and Recognition of Rolling Bearings Fan Shi 1 & Guochun Xu 1
# Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract In the field of system health management, the quality of rolling equipment is very important. Therefore, the fault diagnosis of rolling bearings has become a hot research topic. In this paper, the traditional fault feature extraction method is used to optimize the non-linear and non-stationary characteristics of the bearing vibration signal. Furthermore, in order to improve the performance of the fault diagnosis, a novel signal fingerprint is proposed to recognize the fault type. The simulation result show that the new method is successful and effective, and the recognition rate can be improved up to 95.33%, which is better than the traditional methods. Keywords Fault diagnosis . Feature extraction . Feature selection . Fault recognition . Signal fingerprint
1 Introduction Today, industrial equipment, transportation equipment, household appliances, robots, etc. usually have a large number of rotating machinery, and rolling bearings are extremely important components. Once the rolling bearing is damaged, it will cause damage to the entire mechanical system, resulting in unpredictable consequences, such as accidental safety accidents, affecting production and economic losses [1–3] . More seriously, in practical applications, bearings are often the most vulnerable parts due to the harsh working environment. It is necessary to monitor the failure of the rolling bearing during the work, detect the working state of the bearing in real time, and find the fault and the type of the fault in time when an abnormality is found. Therefore, the field of investigation of mechanical failure mechanisms and the field of joint research on system health work cycle, Prognostics and Health Management (PHM) or System Health Management (SHM), is becoming more and more popular. The fault diagnosis of rolling bearings has been successful in various fields, and in practice it has demonstrated its powerful functions, such as
* Guochun Xu [email protected] Fan Shi [email protected] 1
Ningbo University, Ningbo, China
various energy systems [4], UAVs [5], robots [6], turbine systems and so on [7]. At the end of the last century, scholars conducted research on bearing fault detection [8]. This paper will also study the detection technology to ensure the stable and healthy work of rotating machinery. In general, fault detection of bearings can be based on a variety of information carriers, such as vibration signals at fault, temperature information, acoustic information, and the like. Because the vibration signal is easier to collect than other signals, it carries more fault information, and the signal analysis methods are many and mature, so it can be used to analyze the fault condition of the bearing [9, 10]. The information fusion method has very good performance in the field of signal analysis [11]. Fault diagnosis techniques usually involve two processes: the e
Data Loading...
