Fault diagnosis of rolling bearing in multi-dimensional entropy space

Rolling bearing is an important part of rotating machinery in various industries. Once it breaks down, it will affect the safety of the whole mechanical equipment. Therefore, the real-time monitoring and diagnosis of rolling bearing is of great significance to the whole industrial safety. Aiming at the problem of rolling bearing fault diagnosis, this paper presents a method of rolling bearing fault diagnosis based on multi-dimensional entropy space distance. First, empirical mode decomposition (EMD) is used to decompose the original vibration signal of the rolling bearing into multiple components of intrinsic mode function (IMF). The first five IMF components are taken and the improved information entropy (IIE), improved energy entropy (IEE) and improved permutation entropy (IPE) after linear transformation are calculated for each IMF component signal. Three kinds of entropy are used as coordinate axes to construct three-dimensional space, and then the space distance between the three-dimensional space composed of three kinds of entropy of test data and the three-dimensional space of sample data is calculated. The space distance between the three-dimensional space and a certain working condition is the smallest, and the bearing is such working condition. The experimental results show that this method can effectively diagnose the fault types of rolling bearings.