Rotor misalignment quantitative recognition method based on VMD and DBN
A quantitative recognition and rotor technology, applied in character and pattern recognition, instruments, computer parts, etc., can solve problems such as poor model generalization ability, unclear DBN input vector construction method, and influence on misalignment state recognition rate, etc.
Inactive Publication Date: 2019-09-24
HUNAN UNIV OF SCI & TECH
View PDF4 Cites 13 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
The main reason is that the construction method of DBN input vector is not clear, and unreasonable input vector will lead to poor generalization ability of the model and affect the recognition rate of the misalignment state. Therefore, further research is needed
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
[0105] The experiment was carried out on a multifunctional rotor bearing system dynamic vibration test platform produced by Suzhou Dongling Vibration Test Instrument Co., Ltd. The structure of the test platform and the schematic diagram of misalignment are shown in image 3 As shown, 1 is the sensor, 2 is the adjusting gasket, 3 is the inner rotor, 4 is the outer rotor, and 5 is the intermediate bearing. The rotor of the test bench is a double rotor with inner and outer concentric nesting, which is driven by two servo motors respectively, which can realize precise control of speed and steering. During the experiment, adjusting shims of different thicknesses were added to the left end of the multifunctional rotor vibration test bench to simulate the degree of rotor misalignment. The description of the experimental conditions is shown in Table 1. The acceleration sensor is arranged at the left end to support the vertical direction. During the test, the speed of the inner rotor i...
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 rotor misalignment quantitative identification method based on VMD and DBN, and the method comprises the following steps: carrying out the variational mode decomposition of an acquired acceleration vibration signal, obtaining a mode component, and carrying out the further processing of the mode component, so as to obtain a data sample; inputting the training group data into a deep confidence network, and performing greedy training layer by layer from a low layer to a high layer; according to the labels and the classification rules of the Soft-max classifier, gradually and reversely fine-adjusting the parameters from the highest layer to the lowest layer, and completing training of the deep belief network model;; and calculating a correct recognition rate of non-centering quantitative recognition. According to the invention, vibration acceleration signals under three misalignment quantities are collected; the vibration signal is decomposed by using variation modal decomposition, then the modal function is analyzed, the decomposition layer number of the VMD is determined according to the mutual information theory, and the modal signal is reconstructed as an input sample to train the DBN classification model, so that the fault recognition process is greatly simplified, and the recognition accuracy is higher.
Description
technical field [0001] The invention relates to a rotor misalignment quantitative identification method, in particular to a rotor misalignment quantitative identification method based on VMD and DBN, belonging to the field of rotating machinery fault diagnosis. Background technique [0002] The rotor system is the core part of major rotating mechanical equipment such as aero-engines and gas turbines. Due to installation errors, support wear and other reasons, the axis line of the rotor and the center line of the bearing are tilted or offset, and the misalignment of the rotor system can be manifested in many ways. Rotor coupling deflection angle misalignment, multi-span rotor coupling parallel misalignment, multi-span rotor coupling deflection angle parallel misalignment, inner and outer rotor support offset misalignment, inner and outer rotor parallel misalignment. The harsh operating environment of high temperature, high pressure and high speed will further aggravate the se...
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 IPC IPC(8): G06K9/00G06K9/62
CPCG06F2218/00G06F18/24G06F18/214
Inventor 杨大炼张帆宇苗晶晶李学军蒋玲莉郭帅平张宏献廖子豪
Owner HUNAN UNIV OF SCI & TECH