Small sample bearing fault diagnosis method based on triple model
A fault diagnosis and triplet technology, applied in the mechanical field, can solve problems such as relying on training samples, bearing failures, and inability to obtain sample training models, and achieves the effect of high fault recognition rate
Pending Publication Date: 2022-06-07
ANHUI UNIV OF SCI & TECH
View PDF12 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
However, the working environment of these equipment is relatively harsh, and the bearings are prone to failure. Therefore, if the bearing failure can be found in time, a lot of manpower and material resources can be saved, and even major accidents can be avoided.
At present, with the development of artificial intelligence, deep learning has been widely used in bearing fault diagnosis and has achieved a good fault recognition rate, but most of the bearing fault diagnosis methods based on deep learning rely on a large number of training samples , to obtain a satisfactory result
However, in actual production practice, there are not enough samples to train the model, so a small-sample bearing fault diagnosis method based on triplet model is proposed
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment
[0046] This example uses the bearing data set of Jiangnan University. The data set collects noise signals of four states: bearing health, bearing inner ring fault, bearing outer ring fault and bearing rolling element fault at 600, 800 and 1000 speeds respectively. The sampling frequency is 50kHz. In this example, the bearing data at 800 rotation speed is used to verify the fault diagnosis method of small sample bearing based on triple model. Specific steps are as follows:
[0047] Step 1: Obtain the vibration timing signal of the bearing and divide it into a training set and a test set;
[0048] Step 2: Preprocess the signals of the training set and the test set respectively, and convert the one-dimensional time series signal into a two-dimensional signal, as shown in Table 1
[0049] Table 1 Bearing fault dataset
[0050]
[0051] Step 3: Randomly select 5, 10, 15, 30, 50, 80, and 100 samples from each bearing fault in the training set to verify the recognition accuracy o...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses a small sample bearing fault diagnosis method based on a triple model, and the method comprises the following steps: carrying out the preprocessing of an obtained one-dimensional time sequence signal of an original bearing, and obtaining a two-dimensional time-frequency image; randomly selecting samples to train the input triple model and iteratively updating model parameters; extracting feature vectors of the training samples and calculating feature vector mean values of various bearing faults; and judging the fault category of the test set sample to calculate the performance of the model. According to the method, the triad model is used for learning the intra-class and inter-class features of the bearing faults, and the high bearing fault recognition rate can be obtained under the condition that the number of training samples is small by using the model.
Description
technical field [0001] The invention belongs to the technical field of machinery, and specifically designs a small sample bearing fault diagnosis method based on a triple model. Background technique [0002] With the rapid development of industrialization, a large number of intelligent rotating machinery equipment has been adopted in many fields such as aerospace, electric power and manufacturing industries. Bearings are an important component of these rotating machinery, and the health of bearings is related to the normal operation of these equipment. However, the working environment of these equipments is harsh, and the bearings are prone to failure. Therefore, if the failure of the bearing can be found in time, a lot of manpower and material resources can be saved, and even major accidents can be avoided. At present, with the development of artificial intelligence, deep learning has been widely used in bearing fault diagnosis, and obtained a good fault recognition rate, b...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G01M13/045
CPCG01M13/045Y02T90/00
Inventor 谢由生朱国庆
Owner ANHUI UNIV OF SCI & TECH