DIESEL FUEL FILTER ELEMENT

With the trend of mechanical equipment towards large-scale, intelligent and high-precision, the use of rotating parts, such as roller bearing, has been significantly improved to achieve power transmission, position fixation and other purposes. When they are damaged or fail, the mechanical equipment operation safety and production benefit will be impacted. However, due to the special installation position of these rotating parts, it is more difficult to research and judge the health status of the equipment, and previous methods relying on humans or experience can no longer work. Therefore, developing intelligent detection and diagnosis method to implement equipment health monitoring have become a hot research topic.

With the rapid development of artificial intelligent, more and more machine learning methods make mechanical equipment intelligent diagnosis come true and prosper, such as reinforcement learning (RL) [1], [2], generative adversarial networks (GAN) [3], autoencoder (AE) [4] and support vector machine (SVM) [5], [6], [47]. Among them, SVM is a classification algorithm based on statistical learning, which is not easy to fall into local minima and separates training data through optimal hyperplane while training data can be mapped to high-dimensional features through nonlinear mapping methods, such as polynomial functions and radial basis functions. In addition, SVM can provide accurate decision hyperplane under limited samples, and has good generalization ability. In view of its excellent performance, SVM has been widely used in many fields. Wang et al. proposed an intelligent fault diagnosis method based on the combination of generalized composite multi-scale weighted permutation entropy (GCMWPE) and SVM [7], which can extract bearing features from multiple scales to construct high-dimensional feature collection. Bayati et al. proposed a fault location method for DC microgrid based on SVM [8]. By using the local measured value at one end of each line, the accurate location of high impedance fault can be located, and the experimental results show that the scheme is robust to noise and other disturbances. Ref. [9] proposed an intelligent fault diagnosis method for lithium-ion battery based on support vector machine, which uses discrete cosine filtering to eliminate noise.