CN116226719A - Bearing fault diagnosis method based on multidimensional steady-state vibration characteristics and related components - Google Patents
Bearing fault diagnosis method based on multidimensional steady-state vibration characteristics and related components Download PDFInfo
- Publication number
- CN116226719A CN116226719A CN202310228866.7A CN202310228866A CN116226719A CN 116226719 A CN116226719 A CN 116226719A CN 202310228866 A CN202310228866 A CN 202310228866A CN 116226719 A CN116226719 A CN 116226719A
- Authority
- CN
- China
- Prior art keywords
- vibration
- steady
- state
- bearing
- multidimensional
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M13/00—Testing of machine parts
- G01M13/04—Bearings
- G01M13/045—Acoustic or vibration analysis
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E10/00—Energy generation through renewable energy sources
- Y02E10/70—Wind energy
- Y02E10/72—Wind turbines with rotation axis in wind direction
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computing Systems (AREA)
- Acoustics & Sound (AREA)
- Algebra (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The application discloses a bearing fault diagnosis method and related components based on multidimensional steady-state vibration characteristics, and relates to the technical field of equipment diagnosis. The method comprises the following steps: determining a multidimensional steady-state characteristic corresponding to a target bearing according to a vibration signal of the target bearing in a history normal working condition, and constructing a steady-state vibration characteristic matrix based on the multidimensional steady-state characteristic; the multidimensional steady-state features include time domain features and frequency domain features; determining a current vibration characteristic vector according to the current vibration signal of the target bearing, and determining an abnormal vibration characteristic and an abnormal characteristic vector corresponding to the abnormal vibration characteristic according to the difference between the steady-state vibration characteristic matrix and the vibration characteristic vector; and taking the abnormal vibration characteristics as an abnormal monitoring index, and carrying out bearing fault diagnosis by utilizing a preset fault diagnosis rule based on the abnormal characteristic vector. The abnormal state of the bearing can be sharply identified, the bearing fault can be rapidly positioned, and the bearing fault diagnosis capability is improved.
Description
Technical Field
The invention relates to the technical field of equipment diagnosis, in particular to a bearing fault diagnosis method, device, equipment and storage medium based on multidimensional steady-state vibration characteristics.
Background
Currently, with the development of modern industry and the continuous improvement of the scientific and technological level, the electromechanical devices are continuously developed towards the directions of large-scale, high-speed, continuous, centralized, automatic and precise, and the compositions and structures thereof are also becoming more and more complex, which directly lead to the increase of failure rate and abnormal difficulty of diagnosis, wherein key components such as rolling bearings, and certain slight damage faults or abnormalities can cause the failure, paralysis and even catastrophic consequences of the whole system if not detected and removed in time.
In the prior art, fault diagnosis of a bearing is mainly based on a basic theory of vibration signal processing, a time domain vibration signal of bearing operation is measured, and a time frequency analysis means is utilized to infer the bearing fault type. In actual operation, the early abnormal signal representation and degree of bearing operation are different due to the differences of load, rotation speed, lubrication and environmental factors, the judgment of fault characteristics is greatly dependent on the vibration analysis experience of technicians, in addition, the fault location is excessively dependent on subjective factors of analysts due to the reasons of rotation speed fluctuation, load fluctuation and the like of the actual bearing operation, and the missed judgment and the false judgment are easily caused.
Disclosure of Invention
In view of the above, the invention aims to provide a bearing fault diagnosis method, device, equipment and medium based on multidimensional steady-state vibration characteristics, which can be used for rapidly identifying abnormal states of a bearing, rapidly positioning the bearing fault and improving the bearing fault diagnosis capability. The specific scheme is as follows:
in a first aspect, the present application discloses a bearing fault diagnosis method based on multidimensional steady-state vibration characteristics, comprising:
determining a multidimensional steady-state characteristic corresponding to a target bearing according to a vibration signal of the target bearing in a history normal working condition, and constructing a steady-state vibration characteristic matrix based on the multidimensional steady-state characteristic; the multidimensional steady-state features include time domain features and frequency domain features;
determining a current vibration characteristic vector according to the current vibration signal of the target bearing, and determining an abnormal vibration characteristic and an abnormal characteristic vector corresponding to the abnormal vibration characteristic according to the difference between the steady-state vibration characteristic matrix and the vibration characteristic vector;
and taking the abnormal vibration characteristics as an abnormal monitoring index, and carrying out bearing fault diagnosis by utilizing a preset fault diagnosis rule based on the abnormal characteristic vector.
Optionally, the determining the multidimensional steady-state feature corresponding to the target bearing according to the vibration signal of the target bearing during the historical normal working condition includes:
acquiring original vibration signals of a target bearing collected in a history way, and screening out vibration signals of the target bearing under different normal working conditions from the original vibration signals;
and taking vibration signals of the target bearing under different normal working conditions as a steady-state sample set, and extracting features of the steady-state sample set to obtain the multidimensional steady-state features.
Optionally, the determining the multidimensional steady-state feature corresponding to the target bearing according to the vibration signal of the target bearing during the historical normal working condition includes:
extracting time domain features from the vibration signals during the historical normal working conditions; the time domain features include root mean square values, peaks and variances;
performing signal transformation on the vibration signal in the history normal working condition to obtain a corresponding frequency domain signal, and extracting frequency domain characteristics from the frequency domain signal; the frequency domain features include envelope spectrum features, normal spectrum features, and empirical mode decomposition signal features.
Optionally, the determining the abnormal vibration feature according to the difference between the steady-state vibration feature matrix and the vibration feature vector includes:
determining a steady-state offset of the target bearing on each feature according to the difference between the steady-state vibration feature matrix and the vibration feature vector;
and determining abnormal vibration characteristics according to each steady-state offset and the corresponding offset threshold.
Optionally, the determining the steady-state offset of the target bearing on each feature according to the difference between the steady-state vibration feature matrix and the vibration feature vector includes:
longitudinally splicing the steady-state vibration characteristic matrix and the vibration characteristic vector, and normalizing according to columns after splicing to obtain a normalized steady-state vibration characteristic matrix and a normalized vibration characteristic vector;
calculating the distance between the normalized steady-state vibration characteristic matrix and the normalized vibration characteristic vector to obtain a distance vector;
obtaining a target vector representing the approximation degree of the steady-state vibration characteristic matrix and the vibration characteristic vector according to the target distance vector;
and calculating the current steady-state offset of the target bearing on each feature based on the steady-state vibration feature matrix, the vibration feature vector and the target vector.
Optionally, the preset fault diagnosis rule is a fault diagnosis rule constructed according to the type of the fault to be detected and the type of the target bearing.
Optionally, the bearing fault diagnosis method based on the multidimensional steady-state vibration feature further includes:
and modifying the target parameters in the preset fault diagnosis rules according to the current running environment of the target bearing.
In a second aspect, the present application discloses a bearing fault diagnosis device based on multidimensional steady-state vibration characteristics, comprising:
the characteristic matrix construction module is used for determining a multidimensional steady-state characteristic corresponding to the target bearing according to a vibration signal of the target bearing in a history normal working condition period and constructing a steady-state vibration characteristic matrix based on the multidimensional steady-state characteristic; the multidimensional steady-state features include time domain features and frequency domain features;
the abnormal feature determining module is used for determining a current vibration feature vector according to the current vibration signal of the target bearing, determining an abnormal vibration feature according to the difference between the steady-state vibration feature matrix and the vibration feature vector and determining an abnormal feature vector corresponding to the abnormal vibration feature;
and the fault diagnosis module is used for carrying out bearing fault diagnosis by taking the abnormal vibration characteristics as abnormal monitoring indexes and utilizing a preset fault diagnosis rule based on the abnormal characteristic vector.
In a third aspect, the present application discloses an electronic device comprising:
a memory for storing a computer program;
and the processor is used for executing the computer program to realize the bearing fault diagnosis method based on the multidimensional steady-state vibration characteristics.
In a fourth aspect, the present application discloses a computer-readable storage medium for storing a computer program; wherein the computer program, when executed by the processor, implements the aforementioned bearing fault diagnosis method based on the multidimensional steady-state vibration characteristics.
In the method, a multidimensional steady-state characteristic corresponding to a target bearing is determined according to a vibration signal of the target bearing in a history normal working condition, and a steady-state vibration characteristic matrix is constructed based on the multidimensional steady-state characteristic; the multidimensional steady-state features include time domain features and frequency domain features; determining a current vibration characteristic vector according to the current vibration signal of the target bearing, and determining an abnormal vibration characteristic and an abnormal characteristic vector corresponding to the abnormal vibration characteristic according to the difference between the steady-state vibration characteristic matrix and the vibration characteristic vector; and taking the abnormal vibration characteristics as an abnormal monitoring index, and carrying out bearing fault diagnosis by utilizing a preset fault diagnosis rule based on the abnormal characteristic vector. Therefore, the characteristic matrix is established according to the multidimensional steady-state characteristics during the bearing operation, after the abnormal vibration characteristics are determined by comparing the current vibration characteristic vector with the characteristic matrix, the bearing fault judgment is carried out by utilizing the preset fault diagnosis rule, the complexity of the man-hour frequency analysis is reduced, the abnormal state of the bearing can be rapidly identified, the fault can be rapidly positioned by combining the diagnosis rule, and the problems that the early fault is not easy to find during the bearing fault diagnosis, and the fault characteristic extraction is influenced by the subjective factors of the personnel to cause missed judgment and misjudgment are solved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are required to be used in the embodiments or the description of the prior art will be briefly described below, and it is obvious that the drawings in the following description are only embodiments of the present invention, and that other drawings can be obtained according to the provided drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a bearing fault diagnosis method based on multidimensional steady-state vibration characteristics;
FIG. 2 is a flowchart of a specific method for diagnosing bearing faults based on multidimensional steady-state vibration characteristics;
FIG. 3 is a specific diagnostic flow chart for diagnosing using preset fault diagnosis rules provided herein;
FIG. 4 is a schematic structural diagram of a bearing fault diagnosis device based on multidimensional steady-state vibration characteristics;
fig. 5 is a block diagram of an electronic device provided in the present application.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
In the prior art, fault diagnosis of a bearing is mainly based on a basic theory of vibration signal processing, a time domain vibration signal of bearing operation is measured, and a time frequency analysis means is utilized to infer the bearing fault type. In actual operation, the early abnormal signal representation and degree of bearing operation are different due to the differences of load, rotation speed, lubrication and environmental factors, the judgment of fault characteristics is greatly dependent on the vibration analysis experience of technicians, in addition, the fault location is excessively dependent on subjective factors of analysts due to the reasons of rotation speed fluctuation, load fluctuation and the like of the actual bearing operation, and the missed judgment and the false judgment are easily caused. In order to overcome the technical problems, the application provides a bearing fault diagnosis method based on multidimensional steady-state vibration characteristics, which can be used for rapidly identifying abnormal states of a bearing, rapidly positioning the bearing fault and improving the bearing fault diagnosis capability.
The embodiment of the application discloses a bearing fault diagnosis method based on multidimensional steady-state vibration characteristics, which is shown in fig. 1, and can comprise the following steps:
step S11: determining a multidimensional steady-state characteristic corresponding to a target bearing according to a vibration signal of the target bearing in a history normal working condition, and constructing a steady-state vibration characteristic matrix based on the multidimensional steady-state characteristic; the multi-dimensional stationary features include time domain features and frequency domain features.
In this embodiment, firstly, a multi-dimensional steady-state feature corresponding to a target bearing is extracted according to a vibration signal of the target bearing during a historical normal working condition, the multi-dimensional steady-state feature includes a time domain feature and a frequency domain feature, and then a steady-state vibration feature matrix is constructed based on the multi-dimensional steady-state feature to obtain a steady-state model.
In this embodiment, the determining, according to the vibration signal of the target bearing during the historical normal working condition, the multidimensional steady-state characteristic corresponding to the target bearing may include: acquiring original vibration signals of a target bearing collected in a history way, and screening out vibration signals of the target bearing under different normal working conditions from the original vibration signals; and taking vibration signals of the target bearing under different normal working conditions as a steady-state sample set, and extracting features of the steady-state sample set to obtain the multidimensional steady-state features. It will be appreciated that the multi-dimensional steady state characteristic is characteristic of the bearing in normal operating conditions and therefore needs to be extracted from the vibration signal from different normal conditions. The vibration signals under different normal working conditions are historical vibration signal sets of the bearing to be monitored for long-term operation, and the vibration signals comprise the operation signals of the bearing under different normal working conditions and are not a single stable operation signal. The steady-state characteristic is that the bearing fault diagnosis rule relates to a set of all time-frequency domain characteristics, the vector form is presented, the time domain characteristics and the frequency domain characteristics are contained, the characteristic range is formulated according to the rolling bearing expert diagnosis rule, and the characteristic of a certain signal component or a converted signal is not adopted.
For example, an original vibration signal sample set is collected for the bearing, each sample is a one-dimensional time domain vibration signal, parameters such as sampling frequency, sampling process rotating speed and the like of each sample are known, a sample set during steady-state running of the bearing is screened out according to operation and maintenance records or equipment maintainer experience, and the steady-state sample set is set as X SET ＝[X 1 ,X 2 ,X 3 ...X n ] T The method comprises the steps of carrying out a first treatment on the surface of the Steady state feature extraction, i.e. for X j A time domain feature and a frequency domain feature are calculated. The extracted time-frequency domain features (m) are combined into a sample X j Steady state feature vector F of (2) j The steady state vibration characteristic matrix can be expressed as:
in this embodiment, the determining, according to the vibration signal of the target bearing during the historical normal working condition, the multidimensional steady-state characteristic corresponding to the target bearing may include: extracting time domain features from the vibration signals during the historical normal working conditions; the time domain features include root mean square values, peaks and variances; performing signal transformation on the vibration signal in the history normal working condition to obtain a corresponding frequency domain signal, and extracting frequency domain characteristics from the frequency domain signal; the frequency domain features include envelope spectrum features, normal spectrum features, and empirical mode decomposition signal features. Namely extracting time domain features from vibration signals during historical normal working conditions, including but not limited to root mean square values, peaks, variances and the like; the vibration signal is subjected to signal transformation, such as EMD (Empirical Mode Decomposition ) decomposition, signal envelope calculation, order transformation and the like, the transformed signal is subjected to FFT (Fast Fourier Transform, fast Fourier transformation) to obtain a corresponding frequency domain signal and frequency domain characteristic, and the frequency domain characteristic, such as spectrum peak value, spectrum mean value, 0-1000 Hz total vibration of an envelope spectrum, bearing characteristic frequency amplitude and the like, is calculated on the frequency domain data, wherein the frequency domain characteristic comprises, but is not limited to, envelope spectrum characteristic, common frequency spectrum characteristic, EMD decomposition signal characteristic and the like.
Step S12: and determining a current vibration characteristic vector according to the current vibration signal of the target bearing, and determining an abnormal vibration characteristic and an abnormal characteristic vector corresponding to the abnormal vibration characteristic according to the difference between the steady-state vibration characteristic matrix and the vibration characteristic vector.
In this embodiment, for example, as shown in fig. 2, a current vibration signal of the target bearing is obtained, a current vibration feature vector is extracted from the vibration signal, then the current vibration feature vector is compared with a steady vibration feature matrix, an abnormal vibration feature is determined according to the compared difference, and a vibration signal corresponding to the abnormal vibration feature is used as an abnormal feature vector. In the process of abnormality detection, whether a certain feature is abnormal is comprehensively judged according to the correlation of the multidimensional features, and the single comparison of the certain dimensional feature is not performed, namely, the vibration feature vector of the bearing to be detected is compared with the historical condition as a whole, and then the abnormal feature is found according to the deviation. The reason for doing so is that even though the bearings of the same model are different in performance of a certain feature under different running environments, the running result is in a normal state, and the multi-dimensional vibration feature is combined to detect by comparing with the self steady-state feature, so that erroneous judgment and missed judgment can be reduced.
It can be understood that when the rolling bearing vibration signal is used for fault diagnosis, the original signal is usually required to be subjected to feature extraction and identification, and then the fault of the bearing is comprehensively judged according to the feature expression. Because the actual analysis is influenced by factors such as signal noise, lubrication environment, load and the like, even if the types of the bearings are the same, the early fault characterization and the early fault degree of the bearings are different in different application scenes. Therefore, the state evaluation model based on the multidimensional steady state characteristics can be established by utilizing the specific bearing normal operation historical data, the model can reflect the bearing health state in the current environment, and whether each characteristic is abnormal or not is comprehensively judged according to the relevance of the multidimensional characteristics, so that the early abnormal generation is sharply identified.
In this embodiment, the determining the abnormal vibration feature according to the difference between the steady-state vibration feature matrix and the vibration feature vector may include: determining a steady-state offset of the target bearing on each feature according to the difference between the steady-state vibration feature matrix and the vibration feature vector; and determining abnormal vibration characteristics according to each steady-state offset and the corresponding offset threshold. The steady-state offset of the target bearing on each feature can be determined according to the difference between the steady-state vibration feature matrix and the vibration feature vector, and whether the feature is an abnormal vibration feature can be judged according to the steady-state offset corresponding to the feature and the offset threshold corresponding to the feature.
In this embodiment, the determining, according to the difference between the steady-state vibration feature matrix and the vibration feature vector, the steady-state offset of the target bearing on each feature may include: longitudinally splicing the steady-state vibration characteristic matrix and the vibration characteristic vector, and normalizing according to columns after splicing to obtain a normalized steady-state vibration characteristic matrix and a normalized vibration characteristic vector; calculating the distance between the normalized steady-state vibration characteristic matrix and the normalized vibration characteristic vector to obtain a distance vector; obtaining a target vector representing the approximation degree of the steady-state vibration characteristic matrix and the vibration characteristic vector according to the target distance vector; and calculating the current steady-state offset of the target bearing on each feature based on the steady-state vibration feature matrix, the vibration feature vector and the target vector.
In this embodiment, the obtaining, according to the target distance vector, a target vector indicating an approximation degree of the steady-state vibration feature matrix and the vibration feature vector may include: and calculating a target vector representing the approximation degree of the steady-state vibration characteristic matrix and the vibration characteristic vector according to the target distance vector by using a radial basis function.
For example, for the bearing vibration signal at the current moment, the corresponding m feature component real-time feature vectors f are extracted now The method comprises the steps of carrying out a first treatment on the surface of the Longitudinally splicing the steady-state feature matrix and the real-time feature vector to obtain
And normalizing according to the columns, and marking as F 'after normalization' SET And f' now . Separately computing each steady-state feature vector F' j And f' now To obtain a distance vector D now ＝[d 1 ,d 2 ,d 3 ...d n ] T D is to now Substitution of the radial basis function +.>
Calculated to obtain the representation F SET And f now Vector w= [ w ] of approximation degree 1 ,w 2 ,w 3 ...w n ] T . Calculating the offset of the current bearing state and the steady state, expressed as
Steady state offset e for each feature for the current bearing i And setting a threshold value, and detecting that the bearing is abnormal at the current moment when the characteristic exceeds (or is lower than) the threshold value.
Step S13: and taking the abnormal vibration characteristics as an abnormal monitoring index, and carrying out bearing fault diagnosis by utilizing a preset fault diagnosis rule based on the abnormal characteristic vector.
After the abnormal vibration characteristics are determined, the abnormal vibration characteristics are used as abnormal monitoring indexes, and bearing fault diagnosis is carried out by using preset fault diagnosis rules. In this embodiment, the preset fault diagnosis rule may be a fault diagnosis rule constructed according to a type of the fault to be detected and a type of the target bearing, that is, different types of bearings correspond to different fault diagnosis rules for different types of the fault to be detected. For example, if there are k abnormal features that deviate from the threshold range, the corresponding abnormal vibration feature is denoted as f abn ＝[f 1 ,f 2 ,...f k ] T According to the motion mechanism of the bearing, the expert fault diagnosis rule of the fault type fault to be detected can be established, for example, f is calculated according to the fault types such as inner ring abrasion, outer ring abrasion, ball damage and the like abn And substituting the rule to automatically diagnose the current bearing fault. The method utilizes a set of cured diagnosis flow to receive the detected abnormal characteristic vector, automatically completes diagnosis and identification of various faults, and does not need manual analysis.
In this embodiment, the method for diagnosing a bearing fault based on multidimensional steady-state vibration features may further include: and modifying the target parameters in the preset fault diagnosis rules according to the current running environment of the target bearing. Namely, the judgment threshold value of each rule can be an empirical value, and can be properly adjusted according to the running environment of the bearing to be monitored and diagnosed, which does not belong to f abn The features of the (2) represent some basic features to be calculated in the vibration analysis process, and the calculation mode and the range of the features are fixed, generally auxiliary features and are not used as abnormality monitoring indexes.
For example, as shown in fig. 3, a process of diagnosing a failure such as inner ring wear, outer ring wear, ball damage, etc. by using a preset failure diagnosis rule is shown. Wherein SQRT_Y is a spectral root mean square value within the range of 0-2000 hz; enve_y is the maximum amplitude of the envelope spectrum; BPFI_Y is the maximum amplitude of the fault frequency multiple of the bearing inner ring, and the frequency multiplication range is generally 1-5 times; BPFO_Y is MAX [ (1X-5X) BPFO ], the maximum amplitude of the frequency multiplication of the fault frequency of the bearing outer ring is 1-5 times of the frequency multiplication range; BSF_Y is MAX [ (1X-5X) BSF ], the bearing ball fault frequency is multiplied by the maximum amplitude, and the frequency multiplication range is generally 1-5 times; SIDE IY is the maximum amplitude of the SIDE band of the fault frequency of the inner ring, namely MAX [ BPFI_Y (+/-) (1X-2X) frequency conversion ]; SIDE_SY is the maximum amplitude of the SIDE band of the fault frequency of the ball, namely MAX [ BSF_Y (+/-) (1X-2X) frequency conversion ]; SQRT_ IYS is the root mean square value of the spectrum in the non-BPFI_Y+ -1.5 Hz interval and the non-sideband+ -1.5 Hz interval.
Taking a wind power bearing fault data set as an example, the data set comprises 761 groups of fault data of various 3 fault types of a bearing (model 6332) and 217 groups of normal bearing data, the sampling frequency of a vibration signal is 25600Hz, and the fault types are divided into an inner ring fault, an outer ring fault and a ball fault. The bearing fault diagnosis based on the steady-state vibration characteristics and the rules is described, and the abnormality detection and fault diagnosis flow of the current vibration state is described by using the multidimensional steady-state vibration characteristics and the fault diagnosis rules:
1. aligning the lengths of 217 normal sample time domain signals, and intercepting the 2s signals before sampling as original vibration signals;
2. time domain transforming, calculating the envelope signal of each sample;
3. performing time-frequency conversion, namely performing FFT (fast Fourier transform) on the envelope signal and the original vibration signal, and obtaining a 2-channel frequency spectrum signal by each sample;
4. and (3) steady-state feature extraction, namely, modeling the 2-channel frequency domain signal by respectively acquiring 36 features such as frequency conversion, inner ring fault frequency, outer ring fault frequency, roller fault frequency (the maximum frequency multiplication screening range is 5 frequency multiplication), bearing fault frequency +/-2X frequency conversion, time domain peak value, time domain root mean square value and the like according to the fault type of the data set.
5. Constructing a steady-state vibration characteristic matrix, wherein the shape of the matrix is 217 x 36; and setting a steady-state offset threshold for each dimension of the features according to the feature trend of the normal sample.
6. And (3) detecting the abnormality, namely intercepting the 2s vibration signals before 761 fault samples in a verification case, respectively extracting the corresponding 36 time-frequency domain features, substituting the feature vectors of each sample into a steady-state vibration feature matrix for calculation, and comparing the feature offset thresholds to obtain the abnormal feature vectors.
7. Fault diagnosis, substituting the abnormal feature vector into a preset fault diagnosis rule, and performing fault diagnosis on each sample. And finally, the fault diagnosis accuracy of 761 fault samples reaches more than 92%.
From the above, in this embodiment, the multidimensional steady-state feature corresponding to the target bearing is determined according to the vibration signal of the target bearing during the historical normal working condition, and a steady-state vibration feature matrix is constructed based on the multidimensional steady-state feature; the multidimensional steady-state features include time domain features and frequency domain features; determining a current vibration characteristic vector according to the current vibration signal of the target bearing, and determining an abnormal vibration characteristic and an abnormal characteristic vector corresponding to the abnormal vibration characteristic according to the difference between the steady-state vibration characteristic matrix and the vibration characteristic vector; and taking the abnormal vibration characteristics as an abnormal monitoring index, and carrying out bearing fault diagnosis by utilizing a preset fault diagnosis rule based on the abnormal characteristic vector. Therefore, the characteristic matrix is established according to the multidimensional steady-state characteristics during the bearing operation, after the abnormal vibration characteristics are determined by comparing the current vibration characteristic vector with the characteristic matrix, the bearing fault judgment is carried out by utilizing the preset fault diagnosis rule, the complexity of the man-hour frequency analysis is reduced, the abnormal state of the bearing can be rapidly identified, the fault can be rapidly positioned by combining the diagnosis rule, and the problems that the early fault is not easy to find during the bearing fault diagnosis, and the fault characteristic extraction is influenced by the subjective factors of the personnel to cause missed judgment and misjudgment are solved.
Correspondingly, the embodiment of the application also discloses a bearing fault diagnosis device based on multidimensional steady-state vibration characteristics, as shown in fig. 4, the device comprises:
the characteristic matrix construction module 11 is used for determining a multi-dimensional steady-state characteristic corresponding to a target bearing according to a vibration signal of the target bearing in a history normal working condition period and constructing a steady-state vibration characteristic matrix based on the multi-dimensional steady-state characteristic; the multidimensional steady-state features include time domain features and frequency domain features;
an abnormal feature determining module 12, configured to determine a current vibration feature vector according to the current vibration signal of the target bearing, determine an abnormal vibration feature according to a difference between the steady-state vibration feature matrix and the vibration feature vector, and determine an abnormal feature vector corresponding to the abnormal vibration feature;
the fault diagnosis module 13 is configured to perform bearing fault diagnosis using a preset fault diagnosis rule based on the abnormal feature vector by using the abnormal vibration feature as an abnormal monitoring index.
From the above, in this embodiment, the multidimensional steady-state feature corresponding to the target bearing is determined according to the vibration signal of the target bearing during the historical normal working condition, and a steady-state vibration feature matrix is constructed based on the multidimensional steady-state feature; the multidimensional steady-state features include time domain features and frequency domain features; determining a current vibration characteristic vector according to the current vibration signal of the target bearing, and determining an abnormal vibration characteristic and an abnormal characteristic vector corresponding to the abnormal vibration characteristic according to the difference between the steady-state vibration characteristic matrix and the vibration characteristic vector; and taking the abnormal vibration characteristics as an abnormal monitoring index, and carrying out bearing fault diagnosis by utilizing a preset fault diagnosis rule based on the abnormal characteristic vector. Therefore, the characteristic matrix is established according to the multidimensional steady-state characteristics during the bearing operation, after the abnormal vibration characteristics are determined by comparing the current vibration characteristic vector with the characteristic matrix, the bearing fault judgment is carried out by utilizing the preset fault diagnosis rule, the complexity of the man-hour frequency analysis is reduced, the abnormal state of the bearing can be rapidly identified, the fault can be rapidly positioned by combining the diagnosis rule, and the problems that the early fault is not easy to find during the bearing fault diagnosis, and the fault characteristic extraction is influenced by the subjective factors of the personnel to cause missed judgment and misjudgment are solved.
In some specific embodiments, the feature matrix construction module 11 may specifically include:
the vibration signal extraction unit is used for obtaining original vibration signals of the target bearing collected in a history mode and screening vibration signals of the target bearing under different normal working conditions from the original vibration signals;
and the multidimensional steady-state feature extraction unit is used for taking the vibration signals of the target bearing under different normal working conditions as a steady-state sample set, and extracting features of the steady-state sample set to obtain the multidimensional steady-state features.
In some specific embodiments, the feature matrix construction module 11 may specifically include:
the time domain feature extraction unit is used for extracting time domain features from the vibration signals in the history normal working condition period; the time domain features include root mean square values, peaks and variances;
the frequency domain feature extraction unit is used for carrying out signal transformation on the vibration signals in the history normal working condition period to obtain corresponding frequency domain signals, and extracting frequency domain features from the frequency domain signals; the frequency domain features include envelope spectrum features, normal spectrum features, and empirical mode decomposition signal features.
In some embodiments, the abnormal feature determination module 12 may specifically include:
a steady-state offset determining unit, configured to determine a steady-state offset of the target bearing on each feature currently according to a difference between the steady-state vibration feature matrix and the vibration feature vector;
and the abnormal feature determining unit is used for determining abnormal vibration features according to each steady-state offset and the corresponding offset threshold.
In some specific embodiments, the steady-state offset determination unit may specifically include:
the normalization unit is used for longitudinally splicing the steady-state vibration characteristic matrix and the vibration characteristic vector, normalizing the steady-state vibration characteristic matrix and the vibration characteristic vector according to columns after splicing to obtain a normalized steady-state vibration characteristic matrix and a normalized vibration characteristic vector;
the distance calculation unit is used for calculating the distance between the normalized steady-state vibration characteristic matrix and the normalized vibration characteristic vector to obtain a distance vector;
the target vector determining unit is used for obtaining a target vector representing the approximation degree of the steady-state vibration characteristic matrix and the vibration characteristic vector according to the target distance vector;
and the steady-state offset calculating unit is used for calculating the current steady-state offset of the target bearing on each characteristic based on the steady-state vibration characteristic matrix, the vibration characteristic vector and the target vector.
In some specific embodiments, the preset fault diagnosis rule may specifically be a fault diagnosis rule constructed according to a type of a fault to be detected and a type of the target bearing.
In some specific embodiments, the bearing fault diagnosis apparatus based on the multidimensional steady-state vibration characteristic may specifically include:
and the diagnosis rule modification unit is used for modifying the target parameters in the preset fault diagnosis rule according to the current running environment of the target bearing.
Further, the embodiment of the application further discloses an electronic device, and referring to fig. 5, the content in the drawing should not be considered as any limitation on the scope of use of the application.
Fig. 5 is a schematic structural diagram of an electronic device 20 according to an embodiment of the present application. The electronic device 20 may specifically include: at least one processor 21, at least one memory 22, a power supply 23, a communication interface 24, an input output interface 25, and a communication bus 26. Wherein the memory 22 is used for storing a computer program, which is loaded and executed by the processor 21 to implement the relevant steps in the bearing fault diagnosis method based on the multidimensional steady-state vibration feature disclosed in any of the foregoing embodiments.
In this embodiment, the power supply 23 is configured to provide an operating voltage for each hardware device on the electronic device 20; the communication interface 24 can create a data transmission channel between the electronic device 20 and an external device, and the communication protocol to be followed is any communication protocol applicable to the technical solution of the present application, which is not specifically limited herein; the input/output interface 25 is used for acquiring external input data or outputting external output data, and the specific interface type thereof may be selected according to the specific application requirement, which is not limited herein.
The memory 22 may be a carrier for storing resources, such as a read-only memory, a random access memory, a magnetic disk, or an optical disk, and the resources stored thereon include an operating system 221, a computer program 222, and data 223 including multidimensional steady-state characteristics, and the storage may be temporary storage or permanent storage.
The operating system 221 is used for managing and controlling various hardware devices on the electronic device 20 and the computer program 222, so as to implement the operation and processing of the processor 21 on the mass data 223 in the memory 22, which may be Windows Server, netware, unix, linux, etc. The computer program 222 may further include a computer program that can be used to perform other specific tasks in addition to the computer program that can be used to perform the bearing fault diagnosis method based on the multi-dimensional steady state vibration characteristics performed by the electronic device 20 as disclosed in any of the previous embodiments.
Further, the embodiment of the application also discloses a computer storage medium, wherein the computer storage medium stores computer executable instructions, and when the computer executable instructions are loaded and executed by a processor, the steps of the bearing fault diagnosis method based on the multidimensional steady-state vibration characteristics disclosed in any embodiment are realized.
In this specification, each embodiment is described in a progressive manner, and each embodiment is mainly described in a different point from other embodiments, so that the same or similar parts between the embodiments are referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant points refer to the description of the method section.
The steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. The software modules may be disposed in Random Access Memory (RAM), memory, read Only Memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.
Finally, it is further noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
The bearing fault diagnosis method, device, equipment and medium based on the multidimensional steady-state vibration characteristic provided by the invention are described in detail, and specific examples are applied to the principle and implementation of the invention, and the description of the examples is only used for helping to understand the method and core ideas of the invention; meanwhile, as those skilled in the art will have variations in the specific embodiments and application scope in accordance with the ideas of the present invention, the present description should not be construed as limiting the present invention in view of the above.
Claims (10)
1. A bearing fault diagnosis method based on multidimensional steady-state vibration characteristics is characterized by comprising the following steps:
determining a multidimensional steady-state characteristic corresponding to a target bearing according to a vibration signal of the target bearing in a history normal working condition, and constructing a steady-state vibration characteristic matrix based on the multidimensional steady-state characteristic; the multidimensional steady-state features include time domain features and frequency domain features;
determining a current vibration characteristic vector according to the current vibration signal of the target bearing, and determining an abnormal vibration characteristic and an abnormal characteristic vector corresponding to the abnormal vibration characteristic according to the difference between the steady-state vibration characteristic matrix and the vibration characteristic vector;
and taking the abnormal vibration characteristics as an abnormal monitoring index, and carrying out bearing fault diagnosis by utilizing a preset fault diagnosis rule based on the abnormal characteristic vector.
2. The method for diagnosing a bearing failure based on a multidimensional steady state vibration signature as recited in claim 1, wherein said determining a multidimensional steady state signature corresponding to a target bearing based on vibration signals of the target bearing during historical normal operating conditions includes:
acquiring original vibration signals of a target bearing collected in a history way, and screening out vibration signals of the target bearing under different normal working conditions from the original vibration signals;
and taking vibration signals of the target bearing under different normal working conditions as a steady-state sample set, and extracting features of the steady-state sample set to obtain the multidimensional steady-state features.
3. The method for diagnosing a bearing failure based on a multidimensional steady state vibration signature as recited in claim 1, wherein said determining a multidimensional steady state signature corresponding to a target bearing based on vibration signals of the target bearing during historical normal operating conditions includes:
extracting time domain features from the vibration signals during the historical normal working conditions; the time domain features include root mean square values, peaks and variances;
performing signal transformation on the vibration signal in the history normal working condition to obtain a corresponding frequency domain signal, and extracting frequency domain characteristics from the frequency domain signal; the frequency domain features include envelope spectrum features, normal spectrum features, and empirical mode decomposition signal features.
4. The method for diagnosing a bearing failure based on a multidimensional steady-state vibration feature as recited in claim 1, wherein said determining an abnormal vibration feature based on a difference between the steady-state vibration feature matrix and the vibration feature vector includes:
determining a steady-state offset of the target bearing on each feature according to the difference between the steady-state vibration feature matrix and the vibration feature vector;
and determining abnormal vibration characteristics according to each steady-state offset and the corresponding offset threshold.
5. The method for diagnosing a bearing failure based on multi-dimensional steady state vibration characteristics as set forth in claim 4, wherein said determining a steady state offset of the target bearing on each characteristic based on a difference between the steady state vibration characteristic matrix and the vibration characteristic vector includes:
longitudinally splicing the steady-state vibration characteristic matrix and the vibration characteristic vector, and normalizing according to columns after splicing to obtain a normalized steady-state vibration characteristic matrix and a normalized vibration characteristic vector;
calculating the distance between the normalized steady-state vibration characteristic matrix and the normalized vibration characteristic vector to obtain a distance vector;
obtaining a target vector representing the approximation degree of the steady-state vibration characteristic matrix and the vibration characteristic vector according to the target distance vector;
and calculating the current steady-state offset of the target bearing on each feature based on the steady-state vibration feature matrix, the vibration feature vector and the target vector.
6. The bearing fault diagnosis method based on the multidimensional steady-state vibration characteristics according to any one of claims 1 to 5, wherein the preset fault diagnosis rule is a fault diagnosis rule constructed according to a type of a fault to be detected and a type of the target bearing.
7. The method for bearing failure diagnosis based on multidimensional steady-state vibration characteristics of claim 6, further comprising:
and modifying the target parameters in the preset fault diagnosis rules according to the current running environment of the target bearing.
8. A bearing fault diagnosis device based on multidimensional steady-state vibration characteristics, comprising:
the characteristic matrix construction module is used for determining a multidimensional steady-state characteristic corresponding to the target bearing according to a vibration signal of the target bearing in a history normal working condition period and constructing a steady-state vibration characteristic matrix based on the multidimensional steady-state characteristic; the multidimensional steady-state features include time domain features and frequency domain features;
the abnormal feature determining module is used for determining a current vibration feature vector according to the current vibration signal of the target bearing, determining an abnormal vibration feature according to the difference between the steady-state vibration feature matrix and the vibration feature vector and determining an abnormal feature vector corresponding to the abnormal vibration feature;
and the fault diagnosis module is used for carrying out bearing fault diagnosis by taking the abnormal vibration characteristics as abnormal monitoring indexes and utilizing a preset fault diagnosis rule based on the abnormal characteristic vector.
9. An electronic device, comprising:
a memory for storing a computer program;
a processor for executing the computer program to implement the bearing fault diagnosis method based on the multidimensional steady-state vibration characteristics as claimed in any one of claims 1 to 7.
10. A computer-readable storage medium storing a computer program; wherein the computer program when executed by the processor implements the bearing fault diagnosis method based on the multidimensional steady-state vibration characteristics as claimed in any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310228866.7A CN116226719A (en) | 2023-03-10 | 2023-03-10 | Bearing fault diagnosis method based on multidimensional steady-state vibration characteristics and related components |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310228866.7A CN116226719A (en) | 2023-03-10 | 2023-03-10 | Bearing fault diagnosis method based on multidimensional steady-state vibration characteristics and related components |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116226719A true CN116226719A (en) | 2023-06-06 |
Family
ID=86588969
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310228866.7A Pending CN116226719A (en) | 2023-03-10 | 2023-03-10 | Bearing fault diagnosis method based on multidimensional steady-state vibration characteristics and related components |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116226719A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116861544B (en) * | 2023-09-04 | 2024-01-09 | 深圳大学 | A method and related equipment for locating abnormal building vibration sources based on edge-cloud collaboration |
| CN119714860A (en) * | 2025-02-25 | 2025-03-28 | 上海凯士比泵有限公司 | Method, device and medium for fault detection of rotating equipment |
| WO2025194992A1 (en) * | 2024-03-22 | 2025-09-25 | 华为技术有限公司 | Fault identification system and method |
-
2023
- 2023-03-10 CN CN202310228866.7A patent/CN116226719A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116861544B (en) * | 2023-09-04 | 2024-01-09 | 深圳大学 | A method and related equipment for locating abnormal building vibration sources based on edge-cloud collaboration |
| WO2025194992A1 (en) * | 2024-03-22 | 2025-09-25 | 华为技术有限公司 | Fault identification system and method |
| CN119714860A (en) * | 2025-02-25 | 2025-03-28 | 上海凯士比泵有限公司 | Method, device and medium for fault detection of rotating equipment |
| CN119714860B (en) * | 2025-02-25 | 2025-06-20 | 上海凯士比泵有限公司 | Method, apparatus and medium for fault detection of rotating equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wang et al. | A novel statistical time-frequency analysis for rotating machine condition monitoring | |
| CN116226719A (en) | Bearing fault diagnosis method based on multidimensional steady-state vibration characteristics and related components | |
| JP7558418B2 (en) | Method and apparatus for identifying anomalies in a mechanical device or mechanical component - Patents.com | |
| CN113947017A (en) | Method for predicting residual service life of rolling bearing | |
| CN108398265A (en) | A kind of online fault detection method of rolling bearing | |
| CN116434372A (en) | Intelligent data acquisition system and working condition identification system for equipment with variable working conditions | |
| US20240219267A1 (en) | Strong-robustness method for extracting early degradation features of signals and monitoring operational status of device | |
| CN118673305B (en) | Wind turbine generator vibration fault on-line diagnosis system based on feature extraction | |
| Lu et al. | CEEMD-assisted bearing degradation assessment using tight clustering | |
| EP2135144B1 (en) | Machine condition monitoring using pattern rules | |
| CN117454283A (en) | State evaluation method for wind turbine generator operation detection data | |
| CN118052141A (en) | Operation data driven wind driven generator gear box fault early warning method and system | |
| CN117350377A (en) | An equipment fault diagnosis method and device driven by knowledge graph | |
| CN117451347A (en) | Portal crane reduction gearbox intelligent health monitoring system based on fault characteristic model | |
| CN119779679A (en) | A bearing fault diagnosis method and system | |
| CN112621381A (en) | Intelligent health state evaluation method and device for machine tool feeding system | |
| CN105445004A (en) | Vibration curve normalized average life prediction method of equipment components | |
| CN119202987A (en) | A bearing fault diagnosis method and device | |
| Zheng et al. | Wavelet packet decomposition and neural network based fault diagnosis for elevator excessive vibration | |
| CN115656700B (en) | Detection method, training method, electric appliance, monitoring system and storage medium | |
| CN117786461A (en) | Water pump fault diagnosis method, control device and storage medium thereof | |
| CN117168740A (en) | Component fault detection method and device, storage medium and electronic equipment | |
| CN115754821A (en) | Method and device for diagnosing mechanical instability of converter transformer based on vibration voiceprint correlation | |
| JP2021076564A (en) | Current signal, sequential circulation type statistic information filter, and method and system for diagnosing rotary machine by deep learning | |
| CN120279944B (en) | Substation equipment operation state monitoring method and system based on voiceprint recognition |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |