CN117150376B - Classification method, device and equipment for fault characteristic samples of axial plunger pump - Google Patents
Classification method, device and equipment for fault characteristic samples of axial plunger pump Download PDFInfo
- Publication number
- CN117150376B CN117150376B CN202311420841.3A CN202311420841A CN117150376B CN 117150376 B CN117150376 B CN 117150376B CN 202311420841 A CN202311420841 A CN 202311420841A CN 117150376 B CN117150376 B CN 117150376B
- Authority
- CN
- China
- Prior art keywords
- axial plunger
- plunger pump
- fault
- sample
- classification
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 69
- 230000035945 sensitivity Effects 0.000 claims abstract description 130
- 238000011156 evaluation Methods 0.000 claims abstract description 59
- 230000009467 reduction Effects 0.000 claims abstract description 59
- 238000012937 correction Methods 0.000 claims abstract description 54
- 230000000694 effects Effects 0.000 claims abstract description 29
- 230000006870 function Effects 0.000 claims description 26
- 238000013507 mapping Methods 0.000 claims description 25
- 238000010276 construction Methods 0.000 claims description 12
- 238000004364 calculation method Methods 0.000 claims description 7
- 238000004590 computer program Methods 0.000 claims description 6
- 238000010606 normalization Methods 0.000 claims description 5
- 238000012163 sequencing technique Methods 0.000 claims 2
- 238000005299 abrasion Methods 0.000 abstract description 23
- 230000036541 health Effects 0.000 description 13
- 238000003745 diagnosis Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 6
- 230000002776 aggregation Effects 0.000 description 3
- 238000004220 aggregation Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000000926 separation method Methods 0.000 description 3
- 238000012512 characterization method Methods 0.000 description 2
- 230000007423 decrease Effects 0.000 description 2
- 241000223257 Thermomyces Species 0.000 description 1
- 238000013145 classification model Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000004880 explosion Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Landscapes
- Details Of Reciprocating Pumps (AREA)
Abstract
The application provides a classification sensitivity correction method, device and equipment. The classification sensitivity correction method provided by the application is applied to an axial plunger pump fault characteristic sample, wherein the axial plunger pump fault characteristic sample comprises multiple types of axial plunger pump fault characteristics, and the method comprises the following steps: calculating the distance between similar axial plunger pump fault characteristics in the axial plunger pump fault characteristic samples; determining the ratio of the median value to the maximum value of the distances between the fault characteristics of the similar axial plunger pumps as an evaluation factor; the product of the evaluation factor and the original classification sensitivity is determined as the corrected classification sensitivity, so that the problem that the dimension reduction classification effect of the fault characteristic sample of the axial plunger pump is poor when the problems of abrasion of a valve plate, abrasion of a sliding shoe, abrasion of a single plunger and bearing faults of the axial plunger pump occur can be solved, and the dimension reduction classification accuracy of the fault characteristic sample of the axial plunger pump can be improved.
Description
Technical Field
The application relates to the technical field of fault diagnosis, in particular to a method, a device and equipment for classifying fault characteristic samples of an axial plunger pump.
Background
The identification of the fault features has a direct relation with the dimension of the feature space, and the feature dimension is too low, so that the phenomenon of under fitting can be caused, and the fault diagnosis accuracy is poor. The high-dimensional characteristic can bring about an overfitting phenomenon generated by dimensional explosion, which is unfavorable for improving the accuracy of fault diagnosis.
At present, the fault diagnosis of the axial plunger pump has two diagnosis directions with different developments, one is to directly process high-dimensional characteristics to obtain a diagnosis model; the other method is to reprocess the extracted features, convert the high-dimensional feature space into low-dimensional new features by constructing a dimension-reducing mapping method for various features, and take the new features as diagnostic basis. These methods can effectively classify health and fault conditions, but still do not work well. Therefore, a method is needed to solve the problem that the prior method has poor dimension reduction classification effect on the fault characteristic sample of the axial plunger pump when the axial plunger pump pumps the problems of valve plate abrasion, sliding shoe abrasion, single plunger abrasion and bearing faults.
Disclosure of Invention
In view of the above, the present application provides a classification sensitivity correction method, apparatus and device, which are used to solve the problem that the existing method has poor dimension reduction classification effect on the fault characteristic sample of the axial plunger pump when the axial plunger pump is out of the valve plate, the sliding shoes are worn, the single plunger is worn and the bearing is faulty.
Specifically, the application is realized by the following technical scheme:
a first aspect of the present application provides a classification sensitivity correction method, which is characterized in that the method is applied to an axial plunger pump fault feature sample, where the axial plunger pump fault feature sample includes multiple types of axial plunger pump fault features, and different types of axial plunger pump fault feature fault generating positions, and the method includes:
calculating the distance between similar axial plunger pump fault characteristics in the axial plunger pump fault characteristic samples;
determining the ratio of the median value to the maximum value of the distances between the fault characteristics of the similar axial plunger pumps as an evaluation factor;
determining the product of the evaluation factor and the original classification sensitivity as the corrected classification sensitivity; wherein the classification sensitivity represents a ratio of an average distance between sample points in the different types of axial plunger pump failure feature samples to an average distance between sample points in the same type of axial plunger pump failure feature samples.
A second aspect of the present application provides a classification sensitivity correction device, wherein the device is applied to an axial plunger pump fault signature sample, the axial plunger pump fault signature sample including multiple types of axial plunger pump fault signatures, different types of axial plunger pump fault signature fault generating positions being different, the device comprising: a calculation module and a determination module; wherein,
The calculation module is used for calculating the distance between similar axial plunger pump fault characteristics in the axial plunger pump fault characteristic sample;
the determining module is used for determining the ratio of the median value to the maximum value of the distances between the fault characteristics of the similar axial plunger pumps as an evaluation factor;
the determining module is used for determining the product of the evaluation factor and the original classification sensitivity as the corrected classification sensitivity; wherein the classification sensitivity represents a ratio of an average distance between sample points in the different types of axial plunger pump failure feature samples to an average distance between sample points in the same type of axial plunger pump failure feature samples.
A third aspect of the present application provides a classification sensitivity correction apparatus comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps of any of the methods provided in the first aspect of the present application when the program is executed.
A fourth aspect of the present application provides a computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of any of the methods provided in the first aspect of the present application.
The classification sensitivity correction method, the device and the equipment are applied to an axial plunger pump fault characteristic sample, wherein the axial plunger pump fault characteristic sample comprises multiple types of axial plunger pump fault characteristics, the distances between similar axial plunger pump fault characteristics in the axial plunger pump fault characteristic sample are calculated, the ratio of the median value and the maximum value of the distances between the similar axial plunger pump fault characteristics is determined as an evaluation factor, and then the product of the evaluation factor and the original classification sensitivity is determined as corrected classification sensitivity; wherein the classification sensitivity represents a ratio of an average distance between sample points in the different types of axial plunger pump failure feature samples to an average distance between sample points in the same type of axial plunger pump failure feature samples. Therefore, the evaluation factor can be determined based on the distance between similar axial plunger pump fault characteristics in the axial plunger pump fault characteristic samples, classification sensitivity is corrected based on the evaluation factor, the problem that the dimension reduction classification effect of the axial plunger pump fault characteristic samples is poor when the axial plunger pump has problems of valve plate abrasion, sliding shoe abrasion, single plunger abrasion and bearing fault is solved, the dimension reduction classification effect of the axial plunger pump fault characteristic samples is improved, and the dimension reduction classification accuracy of the axial plunger pump fault characteristic samples can be improved.
Drawings
FIG. 1 is a flowchart of a classification sensitivity correction method according to an embodiment of the present application;
FIG. 2 is a diagram illustrating the significance of a classification sensitivity representation in accordance with an exemplary embodiment of the present application;
FIG. 3 is a graph showing the results of classification sensitivity according to an exemplary embodiment of the present application;
FIG. 4 is a graph showing data dimension reduction classification results for the first 10 features in descending order according to an exemplary embodiment of the present application;
FIG. 5 is a graph showing the data dimension reduction classification result for the first 10 features after descending order according to an exemplary embodiment of the present application;
FIG. 6 is a graph showing data dimension reduction classification results for the first 50 features in descending order according to an exemplary embodiment of the present application;
FIG. 7 is a graph showing data dimension reduction classification results for the first 50 features after descending order according to an exemplary embodiment of the present application;
FIG. 8 is a data dimension reduction classification result for the first 100 features in descending order, as shown in an exemplary embodiment of the present application;
FIG. 9 is a graph showing data dimension reduction classification results for the first 100 features after descending order according to an exemplary embodiment of the present application;
FIG. 10 is a graph showing data dimension reduction classification results for the first 150 features in descending order according to an exemplary embodiment of the present application;
FIG. 11 is a graph showing data dimension reduction classification results for the first 150 features after descending order according to an exemplary embodiment of the present application;
FIG. 12 is a diagram illustrating initial feature space dimensions and classification performance according to an exemplary embodiment of the present application;
FIG. 13 is a graph showing the impact of classifying results on data dimension reduction when the degree of confusion is 5, according to an exemplary embodiment of the present application;
FIG. 14 illustrates the impact on the data dimension reduction classification result when the degree of confusion is 15, according to an exemplary embodiment of the present application;
FIG. 15 illustrates the impact on the data dimension reduction classification result when the confusion is 25, according to an exemplary embodiment of the present application;
FIG. 16 is a graph showing the impact of classifying results on data dimension reduction when the degree of confusion is 35, according to an exemplary embodiment of the present application;
FIG. 17 is a graph showing the impact of classifying results on data dimension reduction when the degree of confusion is 45, according to an exemplary embodiment of the present application;
FIG. 18 is a graph showing the impact of classifying results on data dimension reduction when the degree of confusion is 55, according to an exemplary embodiment of the present application;
FIG. 19 is a graph showing the impact on the data dimension reduction classification result when the degree of confusion is 100, according to an exemplary embodiment of the present application;
FIG. 20 illustrates the impact on the data dimension reduction classification result when the degree of confusion is 200, according to an exemplary embodiment of the present application;
FIG. 21 is a graph illustrating the relationship of cluster radius and confusion for clusters according to an exemplary embodiment of the present application;
FIG. 22 is a graph showing the relationship between average distance between cluster center points and distance between all sample points according to an exemplary embodiment of the present application;
FIG. 23 is a graph showing classification sensitivity before and after correction in accordance with an exemplary embodiment of the present application;
FIG. 24 is a flowchart of a second embodiment of a classification sensitivity correction method for an axial plunger pump failure feature sample provided herein;
FIG. 25 is a graph showing the results of a PCA dimension reduction classification method according to an exemplary embodiment of the present application;
FIG. 26 is a graph showing the results of an LDA dimension reduction classification method according to an exemplary embodiment of the present application;
FIG. 27 is a graph showing the results of a LLE dimension reduction classification method according to an exemplary embodiment of the present application;
FIG. 28 is a graph showing the results of the dimension reduction classification method provided herein according to an exemplary embodiment of the present application;
FIG. 29 is a hardware block diagram of a device for classifying sensitivity correction of an axial plunger pump failure feature sample where the device for classifying sensitivity correction of an axial plunger pump failure feature sample provided in the present application is located;
FIG. 30 is a flowchart of a first embodiment of a classification sensitivity correction device for axial plunger pump failure feature samples provided herein;
fig. 31 is a flowchart of a second embodiment of a classification sensitivity correction device for an axial plunger pump fault feature sample provided in the present application.
Detailed Description
Reference will now be made in detail to exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, the same numbers in different drawings refer to the same or similar elements, unless otherwise indicated. The implementations described in the following exemplary examples are not representative of all implementations consistent with the present application. Rather, they are merely examples of apparatus and methods consistent with some aspects of the present application as detailed in the accompanying claims.
The terminology used in the present application is for the purpose of describing particular embodiments only and is not intended to be limiting of the present application. As used in this application and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the term "and/or" as used herein refers to and encompasses any or all possible combinations of one or more of the associated listed items.
It should be understood that although the terms first, second, third, etc. may be used herein to describe various information, these information should not be limited by these terms. These terms are only used to distinguish one type of information from another. For example, a first message may also be referred to as a second message, and similarly, a second message may also be referred to as a first message, without departing from the scope of the present application. The word "if" as used herein may be interpreted as "at … …" or "at … …" or "responsive to a determination", depending on the context.
The application provides a classification sensitivity correction method, device and equipment, which are used for solving the problem that the prior method has poor dimension reduction classification effect on an axial plunger pump fault characteristic sample when the axial plunger pump is worn out by a valve plate, a sliding shoe, a single plunger and a bearing are in fault.
The classification sensitivity correction method, the device and the equipment are applied to an axial plunger pump fault characteristic sample, wherein the axial plunger pump fault characteristic sample comprises multiple types of axial plunger pump fault characteristics, the distances between similar axial plunger pump fault characteristics in the axial plunger pump fault characteristic sample are calculated, the ratio of the median value and the maximum value of the distances between the similar axial plunger pump fault characteristics is determined as an evaluation factor, and then the product of the evaluation factor and the original classification sensitivity is determined as corrected classification sensitivity; wherein the classification sensitivity represents a ratio of an average distance between sample points in the different types of axial plunger pump failure feature samples to an average distance between sample points in the same type of axial plunger pump failure feature samples. Therefore, the evaluation factor can be determined based on the distance between similar axial plunger pump fault characteristics in the axial plunger pump fault characteristic samples, classification sensitivity is corrected based on the evaluation factor, the problem that the dimension reduction classification effect of the axial plunger pump fault characteristic samples is poor when the axial plunger pump has problems of valve plate abrasion, sliding shoe abrasion, single plunger abrasion and bearing fault is solved, the dimension reduction classification effect of the axial plunger pump fault characteristic samples is improved, and the dimension reduction classification accuracy of the axial plunger pump fault characteristic samples can be improved.
Specific examples are given below to describe the technical solutions of the present application in detail.
Fig. 1 is a flowchart of a classification sensitivity correction method according to an embodiment of the present application. Referring to fig. 1, the method provided in this embodiment is applied to an axial plunger pump fault feature sample, where the axial plunger pump fault feature sample includes multiple types of axial plunger pump fault features, and different types of axial plunger pump fault feature faults generate different positions, and the method includes:
s101, calculating the distance between similar axial plunger pump fault characteristics in the axial plunger pump fault characteristic sample.
Specifically, a sample space of an axial plunger pump failure feature sample is defined. Assuming that the sample space contains M types of health states of the axial plunger pump, each type of health state contains C m There are s features in each sample point. At this time, the sample space of the axial plunger pump failure feature sample can be expressed as follows:
. Normalizing the features in the sample space to obtain normalized features +.>。
Wherein,，/>sample space for the axial plunger pump failure feature sample.
Further, the distance between the similar axial plunger pump failure features may be expressed as a first formula:
，
Wherein,distance between fault features of similar axial plunger pumps; c (C) m The number of samples contained for the axial plunger pump fault feature samples; />Is the characteristic obtained after normalization;。
s102, determining the ratio of the median value to the maximum value of the distances between the fault features of the similar axial plunger pumps as an evaluation factor.
In particular, the evaluation factor is used to correct for classification sensitivity. The evaluation factor may be determined according to a second formula, wherein the second formula is:
,
wherein U is the evaluation factor;a median value of distances between fault features of the similar axial plunger pumps; />Is the maximum value of the distance between the fault characteristics of the similar axial plunger pumps.
According to the method provided by the embodiment, the ratio of the median value to the maximum value of the distances between the fault characteristics of the similar axial plunger pumps is determined as the evaluation factor, so that the fault characteristics of different types of axial plunger pumps with higher similarity can be distinguished, and the accuracy of the follow-up classification sensitivity correction is improved. Compared with the prior art, which can only distinguish health and fault states and cannot distinguish faults among different parts, the method provided by the invention corrects the classification sensitivity by utilizing the distance information among the features of one type, wherein the distance information among the features of different types is different, namely two classification sensitivity values with smaller original gap are respectively multiplied by different corrected evaluation factors, so that the difference of the classification sensitivity is enlarged, and the degree of distinction among the features of different types can be improved.
S103, determining the product of the evaluation factor and the original classification sensitivity as the corrected classification sensitivity; wherein the classification sensitivity represents a ratio of an average distance between sample points in the different types of axial plunger pump failure feature samples to an average distance between sample points in the same type of axial plunger pump failure feature samples.
In particular, the classification sensitivity indicates the ratio of the average distance between the fault features of different types of axial plunger pumps to the average distance between the fault features of the same type of axial plunger pumps, and the higher the classification sensitivity, the better the classification characteristic of the feature. FIG. 2 is a diagram illustrating the meaning of a classification sensitivity representation in accordance with an exemplary embodiment of the present application.
Specifically, according to a first formula, the average distance between fault features of different types of axial plunger pumps can be calculated:
，
wherein,an average distance between fault features of the different types of axial plunger pumps; m represents the category number of the health states contained in the axial plunger pump; />Is the distance between the fault characteristics of the similar axial plunger pumps.
The sample points of the fault characteristics of the similar axial plunger pumps are represented by the average value of the characteristics:
，
wherein,sample points which are the fault characteristics of the similar axial plunger pumps and are represented by the average value of the characteristics; c (C) m The number of samples contained in the axial plunger pump fault signature samples; />Is the characteristic obtained after normalization.
For the same feature, the average distance between sample points in different types of axial plunger pump failure feature samples can be expressed as:
，
wherein,the average distance between sample points in fault characteristic samples of different types of axial plunger pumps is set; m represents the category number of the health states contained in the axial plunger pump; />Is a sample point of the fault characteristics of the similar axial plunger pump expressed by the mean value of the characteristics.
Further, the classification sensitivity can be expressed as:
，
wherein,sensitivity to the classification; />Average distance between fault features of different axial plunger pumps;is the average distance between the fault characteristics of the same kind of axial plunger pump. FIG. 3 is a graph illustrating the results of classification sensitivity according to an exemplary embodiment of the present application. Referring to FIG. 3, it can be seen that the original letterClassification sensitivity of the number, IMF1, and IMF2 features is greater. The features are arranged according to the descending order of classification sensitivity, and the data dimension reduction classification results of the first 10, the first 50, the first 100 and the first 150 of the adjusted features are calculated in sequence, wherein the data dimension reduction classification results of the first 10 features are shown in fig. 4 and 5. The data dimension reduction classification results of the first 50 features are shown in fig. 6 and 7. The data dimension reduction classification results of the first 100 features are shown in fig. 8 and 9. The data dimension reduction classification results of the first 150 features are shown in fig. 10 and 11. Referring to fig. 4 to 11, it can be seen that the dimension reduction effect of the axial plunger pump fault feature samples can be significantly improved after sorting the classification sensitivities. When only the first ten features are selected for sorting, dimension reduction results before sorting can only effectively distinguish bearing faults and health states, and cannot effectively treat sliding shoe faults, plunger faults and valve plate abrasion faults; the dimension reduction is carried out on the 10 features with highest classification sensitivity, the obtained result is obviously improved, the classification boundaries among different faults are clearly visible, and the aliasing phenomenon of the boundaries still exists. When the first 50 features are taken for dimension reduction, as shown in fig. 7, in a 2-dimensional feature space obtained by dimension reduction of the features after classification sensitivity sorting, 5 types of faults can be effectively isolated, clusters of the same type of features are obvious, and if the original feature space is not sorted, the faults of the sliding shoes, the faults of the plungers and the abrasion faults of the valve plates can not be effectively separated. With the increase of the number of the feature spaces, features with poor classification sensitivity are integrated into the dimension-reduced mapping, and from the aspect of results, classification results after dimension reduction can be directly affected. Thus, the feature dimension reduction result is related to the number of features and the classification sensitivity of the features.
Optionally, an evaluation index of classification sensitivity correction can be determined according to the ratio of the average distance to the maximum distance among the clusters of the fault characteristic samples of the various axial plunger pumps; the fault characteristic sample clustering characterization of the axial plunger pump is characterized by the phenomenon that the fault characteristic samples of the same type of axial plunger pump are clustered after the fault characteristics are sequenced according to classification sensitivity.
Specifically, the evaluation index of the classification sensitivity correction may be determined according to a fourth formula, where the fourth formula is:
，
wherein,the higher the evaluation index is, the better the effect of the classification sensitivity correction is represented; />The average distance among clusters of the fault characteristic samples of the axial plunger pumps is the average distance among clusters of the fault characteristic samples of the axial plunger pumps; />And clustering maximum distances among fault characteristic samples of the axial plunger pumps. FIG. 12 is a diagram illustrating initial feature space dimensions and classification performance according to an exemplary embodiment of the present application.
Further, the evaluation index also includes a degree of confusion of the axial plunger pump failure feature samples. The calculation formula of the confusion degree is as follows, and the essence of the calculation formula is that entropy values of sample point distribution in axial plunger pump fault characteristic samples in an original space are calculated:
，
Wherein,for sample dot->Is a confusing degree of (1); />For sample dot->Sample point->Is a similarity of (3).
And selecting the features of the first 63 orders after sorting to reduce the dimension, and obtaining the influence of the confusion degree on the dimension reduction classification result of the data as shown in fig. 13 to 20. Referring to fig. 13 to 20, it can be seen that the value of the confusion directly affects the distribution characteristics of the feature clusters. As the confusion value increases, there is a tendency for data clusters of different failure types to move away from each other, and data clusters of the same failure type exhibit a tendency to tightly aggregate. Wherein, the valve plate wear, the sliding shoe wear and the plunger wear faults show a trend from aggregation to separation to aggregation along with the increase of the confusion degree value, and when the confusion degree value is more than 300, the classification boundaries of three fault categories are overlapped; and the characteristic distribution of the health state and the bearing faults shows better separation effect. When the confusion degree is smaller, the fault sample can be broken into a plurality of clusters, and at the moment, the classification boundary can be determined among different types of fault clusters, but the obtained classification model is extremely complex, and the fault classification performance can be seriously affected. In order to select the optimal confusion, the classification effect of various fault samples of the two-dimensional plane under different confusion conditions needs to be evaluated.
Normalizing the obtained classification plane, mapping the value of the sample characteristic to [0,1], and calculating the clustering radius of each cluster under the same spatial scale:
，
wherein,clustering radius of clusters of the fault characteristic samples of the similar axial plunger pumps after normalization; />Maximum value of sample data; />Is acquisition point data; />Is the sample data minimum. Fig. 21 is a graph showing a relationship between cluster radius and confusion of clusters according to an exemplary embodiment of the present application. Referring to fig. 21, it can be seen that the cluster radius of the cluster decreases with an increase in the degree of confusion, and when the degree of confusion is less than 100, the cluster radius of the cluster decreases sharply, and when the degree of confusion is greater than 100, the cluster radius of the cluster starts to become constant. That is, an increase in the magnitude of the confusion will result in a tighter cluster of the same type of axial plunger pump failure feature samples. The average distance among clusters represents the cluster interval, and clusters of fault characteristic samples of different types of axial plunger pumps show a tendency of separation along with the increase of the confusion degree value, so that the classification of the fault characteristic samples of the axial plunger pumps is more facilitated theoretically. But the individual clusters are extremely spaced apart and the multiple similar clusters are aliased. The clustering center of the axial plunger pump fault characteristic sample is calculated through a k-Means method, and the relationship between the average distance between the clustering center points and the distance between all sample points is shown in fig. 22. Referring to fig. 22, it can be known that, with increasing confusion, the axial plunger pump fault feature samples obtained after dimension reduction have multiple kinds of aggregation phenomena, that is, although the average value of distances between clustering centers is basically unchanged, the maximum value of the distances between partial clustering centers is increased, and the minimum value of the distances between partial clustering centers is also reduced, so that the fault features of the internal parts of the axial plunger pump are integrated into one clustering block, and the bearing faults and the health states are separated from each other.
Specifically, the determining the product of the evaluation factor and the original classification sensitivity as the corrected classification sensitivity includes: determining the corrected classification sensitivity according to a third formula, wherein the third formula is:
，
wherein,the corrected classification sensitivity; />The original classification sensitivity is adopted; u is the evaluation factor;a median value of distances between fault features of the similar axial plunger pumps; />Is the maximum value of the distance between the fault characteristics of the similar axial plunger pumps. FIG. 23 is a graph showing classification sensitivity before and after correction in accordance with an exemplary embodiment of the present application.
According to the method provided by the embodiment, the product of the evaluation factor and the original classification sensitivity is determined as the corrected classification sensitivity, so that the corrected classification sensitivity has a higher value, the problem that the dimension reduction classification effect of the fault characteristic sample of the axial plunger pump is poor when the axial plunger pump pumps the problems of valve plate abrasion, sliding shoe abrasion, single plunger abrasion and bearing faults is solved, the dimension reduction classification effect of the fault characteristic sample of the axial plunger pump is improved, and the dimension reduction classification accuracy of the fault characteristic sample of the axial plunger pump can be improved.
The classification sensitivity correction method is applied to an axial plunger pump fault feature sample, wherein the axial plunger pump fault feature sample comprises multiple types of axial plunger pump fault features, the distances between similar axial plunger pump fault features in the axial plunger pump fault feature sample are calculated, the ratio of the median value to the maximum value of the distances between the similar axial plunger pump fault features is determined as an evaluation factor, and then the product of the evaluation factor and the original classification sensitivity is determined as corrected classification sensitivity; wherein the classification sensitivity represents a ratio of an average distance between sample points in the different types of axial plunger pump failure feature samples to an average distance between sample points in the same type of axial plunger pump failure feature samples. Therefore, the evaluation factors can be determined based on the distances between similar axial plunger pump fault characteristics in the axial plunger pump fault characteristic samples, classification sensitivity is corrected based on the evaluation factors, the problem that when the axial plunger pump has problems of valve plate abrasion, sliding shoe abrasion, single plunger abrasion and bearing fault, the dimension reduction classification effect of the axial plunger pump fault characteristic samples is poor is solved, the dimension reduction classification effect of the axial plunger pump fault characteristic samples is improved, and the dimension reduction classification accuracy of the axial plunger pump fault characteristic samples can be improved.
Fig. 24 is a flowchart of a second embodiment of a classification sensitivity correction method provided in the present application. Referring to fig. 24, in the method provided in this embodiment, after determining the product of the evaluation factor and the original classification sensitivity as the corrected classification sensitivity, the method further includes:
s2401, constructing probability distribution according to the similarity between sample points of the axial plunger pump fault characteristic samples in a high-dimensional space.
In particular, for a high-dimensional spaceSample Point->Sample point->The similarity of (2) is expressed as->If->The closer->，/>The larger. The similarity construction probability distribution between sample points of the axial plunger pump fault characteristic sample in the high-dimensional space can be expressed as follows:
，
wherein,for sample dot->Sample point->Similarity of (2); />To->Gaussian distribution variance as center;is a sample point in high dimensional space.
S2402, constructing probability distribution according to the similarity between sample points of the axial plunger pump fault characteristic samples in a low-dimensional space.
Specifically, the similarity construction probability distribution of the axial plunger pump fault characteristic sample between sample points in a low-dimensional space can be expressed as follows:
，
Wherein,for sample dot->Sample point->Similarity of (2); />、/>Is a low-dimensional spaceSample points in (a); in low-dimensional space->。
S2403, constructing a mapping function of similarity construction probability distribution between sample points in a high-dimensional space and similarity construction probability distribution between sample points in a low-dimensional space; the mapping function characterizes the classification sensitivity correction effect, and the smaller the value of the mapping function is, the better the classification sensitivity correction effect is.
Specifically, the constructed mapping function is:
，
wherein, C is the constructed mapping function;a probability distribution for similarity between sample points in a high dimensional space for the axial plunger pump fault signature samples; />A probability distribution for similarity between sample points in a low dimensional space for the axial plunger pump fault signature samples; />Similarity between sample point i and sample point j in the high-dimensional space; />Is the similarity between sample point i and sample point j in the low dimensional space.
Further, toObtaining a gradient:
，
wherein C is a well-constructedMapping functions;similarity between sample point i and sample point j in the high-dimensional space; />Similarity between sample point i and sample point j in the low-dimensional space; / >、/>Is a sample point in a low dimensional space. It follows that the mapping function is for +.>And->The values of (2) are as close as possible, but when +.>When larger, the person is in need of->Smaller, larger value of the mapping function; when->Less time, ->Larger, smaller values of the mapping function. Therefore, the mapping function needs to be optimized.
Specifically, the probability density function of the t distribution is:
，
wherein,-as a function of said probability density; />Is a degree of freedom when the degree of freedom is +>When the distribution of t is Gaussian distribution; when the degree of freedom is 1, the t distribution is the Cauchy distribution, ">。
Data that is gaussian in the high-dimensional space corresponds to t in the low-dimensional space. For similarity points in high-dimensional space, to satisfyThe distance between the data in the low-dimensional space needs to be smaller; conversely, in order to satisfy data having a large distance between data in a high-dimensional spaceA larger distance between data in a low-dimensional space is required. While the t distribution happens to meet this requirement. Definition of ∈Thermomyces>：
，
Wherein,is a sample point in a low dimensional space; />The probability of t distribution for a low dimensional data space.
The similarity gradient from gaussian to t-distribution can be expressed as:
，
wherein the method comprises the steps of，Is a sample point in a low dimensional space; / >Representing a similarity gradient of the gaussian distribution mapping to the t distribution.
In combination with the above example, the dimension reduction classification effect of the method provided by the application is verified by taking the class 4 typical faults and the health state of the axial plunger pump as test data, and the results are respectively compared with dimension reduction classification results of PCA, LDA, LLE and other methods, and are shown in fig. 25 to 28. Referring to fig. 25-28, it can be seen that the faults diagnosed include plunger pump health, valve plate wear, shoe wear, single-piston wear, and bearing faults, the number of samples is [600, 300, 300, 300, 300], for a total of 1800 samples, and the initial characteristic dimension is 162. From the dimension reduction classification result, the PCA and LLE methods have poor effect when processing the high-dimension characteristics of the axial plunger pump, and all samples are mixed together, so that the subsequent classification is difficult; the LDA can divide the characteristics of bearing faults and health states relatively effectively, and obvious aliasing exists in the classification boundary; the method provided by the application can effectively classify the health state and the bearing fault when the 162-dimensional characteristics are directly processed.
Corresponding to the embodiment of the classification sensitivity correction method, the application also provides an embodiment of the classification sensitivity correction device.
The embodiment of the classification sensitivity correction device can be applied to classification sensitivity correction equipment. The apparatus embodiments may be implemented by software, or may be implemented by hardware or a combination of hardware and software. Taking software implementation as an example, the device in a logic sense is formed by reading corresponding computer program instructions in the nonvolatile memory into the memory by the processor of the classification sensitivity correction device where the device is located. In terms of hardware, as shown in fig. 29, a hardware structure diagram of a classification sensitivity correction device where the classification sensitivity correction device provided in the present application is shown, except for a processor, a memory, a network interface, and a nonvolatile memory shown in fig. 29, the classification sensitivity correction device where the device is in an embodiment generally includes other hardware according to the actual function of the classification sensitivity correction device, which is not described herein again.
FIG. 30 is a flowchart of a classification sensitivity correction apparatus according to an embodiment of the present application. Referring to fig. 30, the apparatus provided in this embodiment is applied to an axial plunger pump failure feature sample of an axial plunger pump, and includes: a calculation module 3210 and a determination module 3220; wherein,
The calculating module 3210 is used for calculating the distance between similar axial plunger pump fault characteristics in the axial plunger pump fault characteristic sample;
the determining module 3220 is configured to determine, as an evaluation factor, a ratio of a median value to a maximum value of distances between fault features of the similar axial plunger pumps;
the determining module 3220 is further configured to determine a product of the evaluation factor and the original classification sensitivity as a corrected classification sensitivity; wherein the classification sensitivity represents a ratio of an average distance between sample points in the different types of axial plunger pump failure feature samples to an average distance between sample points in the same type of axial plunger pump failure feature samples.
The device provided in this embodiment may be used to implement the technical method of the method embodiment shown in fig. 1, and its implementation principle and technical effects are similar, and will not be described here again.
The classification sensitivity correction device is applied to an axial plunger pump fault feature sample, wherein the axial plunger pump fault feature sample comprises multiple types of axial plunger pump fault features, the distances between similar axial plunger pump fault features in the axial plunger pump fault feature sample are calculated, the ratio of the median value and the maximum value of the distances between the similar axial plunger pump fault features is determined as an evaluation factor, and then the product of the evaluation factor and the original classification sensitivity is determined as corrected classification sensitivity; wherein the classification sensitivity represents a ratio of an average distance between sample points in the different types of axial plunger pump failure feature samples to an average distance between sample points in the same type of axial plunger pump failure feature samples. Therefore, the evaluation factor can be determined based on the distance between similar axial plunger pump fault characteristics in the axial plunger pump fault characteristic samples, classification sensitivity is corrected based on the evaluation factor, the problem that the dimension reduction classification effect of the axial plunger pump fault characteristic samples is poor when the axial plunger pump has problems of valve plate abrasion, sliding shoe abrasion, single plunger abrasion and bearing fault is solved, the dimension reduction classification effect of the axial plunger pump fault characteristic samples is improved, and the dimension reduction classification accuracy of the axial plunger pump fault characteristic samples can be improved.
Optionally, the calculating module 3210 is specifically configured to calculate a distance between fault features of the similar axial plunger pumps according to a first formula, where the first formula is:
，
wherein,distance between fault features of similar axial plunger pumps; c (C) m The number of samples contained for the axial plunger pump fault feature samples; />Is the characteristic obtained after normalization;。
optionally, the determining module 3220 is specifically configured to determine the evaluation factor according to a second formula, where the second formula is:
,
wherein U is the evaluation factor;for the same kindA median value of distances between axial plunger pump failure features; />Is the maximum value of the distance between the fault characteristics of the similar axial plunger pumps.
Optionally, the determining module 3220 is specifically configured to determine the corrected classification sensitivity according to a third formula, where the third formula is:
,
wherein,the corrected classification sensitivity; />The original classification sensitivity is adopted; u is the evaluation factor;a median value of distances between fault features of the similar axial plunger pumps; />Is the maximum value of the distance between the fault characteristics of the similar axial plunger pumps.
Fig. 31 is a flowchart of a second embodiment of a classification sensitivity correction device provided in the present application. Referring to fig. 31, the apparatus provided in this embodiment further includes a construction module 3230, where the construction module 3230 is configured to construct a probability distribution according to a similarity between sample points of the axial plunger pump fault characteristic sample in a high-dimensional space;
The construction module 3230 is further configured to construct a probability distribution according to a similarity between sample points of the axial plunger pump fault feature sample in a low-dimensional space;
the construction module 3230 is further configured to construct a mapping function of similarity construction probability distribution between sample points in the high-dimensional space to similarity construction probability distribution between sample points in the low-dimensional space; the mapping function characterizes the classification sensitivity correction effect, and the smaller the value of the mapping function is, the better the classification sensitivity correction effect is.
Optionally, the constructing module 3230 is specifically configured to construct a mapping function, where the constructed mapping function is:
，/>
wherein C is a constructed mapping function;a probability distribution for similarity between sample points in a high dimensional space for the axial plunger pump fault signature samples; />A probability distribution for similarity between sample points in a low dimensional space for the axial plunger pump fault signature samples; />Similarity between sample point i and sample point j in the high-dimensional space; />Is the similarity between sample point i and sample point j in the low dimensional space.
Optionally, determining an evaluation index of classification sensitivity correction according to the ratio of the average distance to the maximum distance among the clusters of the fault characteristic samples of the various axial plunger pumps; wherein, the fault characteristic sample clustering characterization of the axial plunger pump sorts the fault characteristics according to the classification sensitivity, and the phenomenon of similar axial plunger pump fault characteristic sample clustering is similar
Optionally, determining the evaluation index of the classification sensitivity correction according to a fourth formula, where the fourth formula is:
,
wherein,the higher the evaluation index is, the better the effect of the classification sensitivity correction is represented; />The average distance among clusters of the fault characteristic samples of the axial plunger pumps is the average distance among clusters of the fault characteristic samples of the axial plunger pumps; />And clustering maximum distances among fault characteristic samples of the axial plunger pumps.
With continued reference to fig. 29, the present application further provides a classification sensitivity correction apparatus, including a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor implementing the steps of any one of the methods provided in the first aspect of the present application when the program is executed.
Further, the present application also provides a computer readable storage medium having stored thereon a computer program which when executed by a processor implements the steps of any of the methods provided in the first aspect of the present application.
The implementation process of the functions and roles of each unit in the above device is specifically shown in the implementation process of the corresponding steps in the above method, and will not be described herein again.
For the device embodiments, reference is made to the description of the method embodiments for the relevant points, since they essentially correspond to the method embodiments. The apparatus embodiments described above are merely illustrative, wherein the elements illustrated as separate elements may or may not be physically separate, and the elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purposes of the present application. Those of ordinary skill in the art will understand and implement the present invention without undue burden.
The foregoing description of the preferred embodiments of the present invention is not intended to limit the invention to the precise form disclosed, and any modifications, equivalents, improvements and alternatives falling within the spirit and principles of the present invention are intended to be included within the scope of the present invention.
Claims (10)
1. A method for classifying an axial plunger pump failure feature sample, wherein the method for classifying an axial plunger pump failure feature sample is applied to an axial plunger pump failure feature sample, the axial plunger pump failure feature sample comprises multiple types of axial plunger pump failure features, and failure feature failure generation positions of different types of axial plunger pumps are different, the method comprising:
calculating the distance between similar axial plunger pump fault characteristics in the axial plunger pump fault characteristic samples;
determining the ratio of the median value to the maximum value of the distances between the fault characteristics of the similar axial plunger pumps as an evaluation factor;
determining the product of the evaluation factor and the original classification sensitivity of the axial plunger pump fault characteristic sample as the classification sensitivity of the axial plunger pump fault characteristic sample after correction; wherein the classification sensitivity represents the ratio of the average distance between sample points in the fault characteristic samples of different types of axial plunger pumps to the average distance between sample points in the fault characteristic samples of the same type of axial plunger pumps;
Sorting the features in descending order based on the classification sensitivity of the axial plunger pump fault feature samples after correction;
and performing dimension reduction, recognition and sequencing on the axial plunger pump fault characteristic samples to obtain classification results of the axial plunger pump fault characteristic samples.
2. The method of claim 1, wherein said calculating distances between like axial plunger pump failure signatures in said axial plunger pump failure signature sample comprises: calculating the distance between the fault characteristics of the similar axial plunger pumps according to a first formula, wherein the first formula is as follows:
，
wherein,distance between fault features of similar axial plunger pumps; />The number of samples contained for the axial plunger pump fault feature samples; />Is the characteristic obtained after normalization;。
3. the method of claim 1, wherein determining the ratio of the median value and the maximum value of the distances between the similar axial plunger pump failure features as an evaluation factor comprises: determining an evaluation factor according to a second formula, wherein the second formula is:
，
wherein U is the evaluation factor;a median value of distances between fault features of the similar axial plunger pumps; / >Is the maximum value of the distance between the fault characteristics of the similar axial plunger pumps.
4. The method of claim 1, wherein determining the product of the evaluation factor and the original classification sensitivity of the axial plunger pump failure feature sample as the corrected classification sensitivity of the axial plunger pump failure feature sample comprises: determining the corrected classification sensitivity according to a third formula, wherein the third formula is:
，
wherein,the corrected classification sensitivity; />The original classification sensitivity is adopted; u is the evaluation factor;a median value of distances between fault features of the similar axial plunger pumps; />Is the maximum value of the distance between the fault characteristics of the similar axial plunger pumps.
5. The method of claim 1, wherein after determining the product of the evaluation factor and the original classification sensitivity as the corrected classification sensitivity, the method further comprises:
constructing probability distribution according to the similarity between sample points of the fault characteristic sample of the axial plunger pump in a high-dimensional space;
constructing probability distribution according to the similarity between sample points of the fault characteristic sample of the axial plunger pump in a low-dimensional space;
Constructing a mapping function of similarity construction probability distribution between sample points in a high-dimensional space to similarity construction probability distribution between sample points in a low-dimensional space; wherein, the mapping function is inversely related to the effect of classification sensitivity correction, and the smaller the value of the mapping function is, the better the effect of classification sensitivity correction is.
6. The method of claim 5, wherein constructing a mapping function of a similarity construct probability distribution between sample points in a high dimensional space to a similarity construct probability distribution between sample points in a low dimensional space, the constructed mapping function being:
，
wherein C is a constructed mapping function;a probability distribution for similarity between sample points in a high dimensional space for the axial plunger pump fault signature samples; />A probability distribution for similarity between sample points in a low dimensional space for the axial plunger pump fault signature samples; />Similarity between sample point i and sample point j in the high-dimensional space;is the similarity between sample point i and sample point j in the low dimensional space.
7. The method according to claim 1, wherein the method further comprises:
determining evaluation indexes of classification sensitivity correction according to the ratio of the average distance to the maximum distance among the clusters of the fault characteristic samples of the various axial plunger pumps; the clusters of the fault characteristic samples of the axial plunger pumps are obtained by clustering the fault characteristic samples of the same type of axial plunger pumps after the fault characteristics are ordered according to the classification sensitivity.
8. The method of claim 7, wherein determining the evaluation index for classification sensitivity correction based on the ratio of the average distance to the maximum distance between clusters of axial plunger pump failure feature samples of each type comprises: determining an evaluation index of the classification sensitivity correction according to a fourth formula, wherein the fourth formula is as follows:
，
wherein,the higher the evaluation index is, the better the effect of the classification sensitivity correction is represented; />The average distance among clusters of the fault characteristic samples of the axial plunger pumps is the average distance among clusters of the fault characteristic samples of the axial plunger pumps; />And clustering maximum distances among fault characteristic samples of the axial plunger pumps.
9. A classification device for axial plunger pump fault signature samples, wherein the device is applied to axial plunger pump fault signature samples, the axial plunger pump fault signature samples comprise multiple types of axial plunger pump fault signatures, and different types of axial plunger pump fault signature faults generate different positions, the device comprising: a calculation module and a determination module; wherein,
the calculation module is used for calculating the distance between similar axial plunger pump fault characteristics in the axial plunger pump fault characteristic sample;
The determining module is used for determining the ratio of the median value to the maximum value of the distances between the fault characteristics of the similar axial plunger pumps as an evaluation factor;
the determining module is used for determining the product of the evaluation factor and the original classification sensitivity of the axial plunger pump fault characteristic sample as the classification sensitivity of the axial plunger pump fault characteristic sample after correction; wherein the classification sensitivity represents the ratio of the average distance between sample points in the fault characteristic samples of different types of axial plunger pumps to the average distance between sample points in the fault characteristic samples of the same type of axial plunger pumps;
the classifying device of the axial plunger pump fault characteristic samples is also used for sorting the characteristics in a descending order based on the classification sensitivity of the axial plunger pump fault characteristic samples after correction; and performing dimension reduction, recognition and sequencing on the axial plunger pump fault characteristic samples to obtain classification results of the axial plunger pump fault characteristic samples.
10. A sorting device for axial plunger pump failure feature samples, comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the steps of the method according to any one of claims 1-8 when executing the program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311420841.3A CN117150376B (en) | 2023-10-30 | 2023-10-30 | Classification method, device and equipment for fault characteristic samples of axial plunger pump |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311420841.3A CN117150376B (en) | 2023-10-30 | 2023-10-30 | Classification method, device and equipment for fault characteristic samples of axial plunger pump |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117150376A CN117150376A (en) | 2023-12-01 |
| CN117150376B true CN117150376B (en) | 2024-02-20 |
Family
ID=88903011
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311420841.3A Active CN117150376B (en) | 2023-10-30 | 2023-10-30 | Classification method, device and equipment for fault characteristic samples of axial plunger pump |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117150376B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106092574A (en) * | 2016-05-30 | 2016-11-09 | 西安工业大学 | The Method for Bearing Fault Diagnosis selected with sensitive features is decomposed based on improving EMD |
| CN110991544A (en) * | 2019-12-10 | 2020-04-10 | 上海交通大学 | Axial plunger pump cavitation level identification method based on PICA-VMD and Hilbert marginal spectrum |
| CN112836581A (en) * | 2021-01-05 | 2021-05-25 | 北京航空航天大学 | Sensitive fault feature extraction method and device based on correlation analysis |
| CN113435314A (en) * | 2021-06-25 | 2021-09-24 | 陈珅 | Rolling bearing acoustic signal early fault sensitivity characteristic screening method and system |
| WO2023115486A1 (en) * | 2021-12-23 | 2023-06-29 | 烟台杰瑞石油服务集团股份有限公司 | Fault early-warning method and apparatus for plunger pump device and plunger pump device system |
| CN116702054A (en) * | 2023-05-16 | 2023-09-05 | 中国人民解放军战略支援部队航天工程大学 | Early fault diagnosis method, equipment and storage medium for DC-DC converter |
-
2023
- 2023-10-30 CN CN202311420841.3A patent/CN117150376B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106092574A (en) * | 2016-05-30 | 2016-11-09 | 西安工业大学 | The Method for Bearing Fault Diagnosis selected with sensitive features is decomposed based on improving EMD |
| CN110991544A (en) * | 2019-12-10 | 2020-04-10 | 上海交通大学 | Axial plunger pump cavitation level identification method based on PICA-VMD and Hilbert marginal spectrum |
| CN112836581A (en) * | 2021-01-05 | 2021-05-25 | 北京航空航天大学 | Sensitive fault feature extraction method and device based on correlation analysis |
| CN113435314A (en) * | 2021-06-25 | 2021-09-24 | 陈珅 | Rolling bearing acoustic signal early fault sensitivity characteristic screening method and system |
| WO2023115486A1 (en) * | 2021-12-23 | 2023-06-29 | 烟台杰瑞石油服务集团股份有限公司 | Fault early-warning method and apparatus for plunger pump device and plunger pump device system |
| CN116702054A (en) * | 2023-05-16 | 2023-09-05 | 中国人民解放军战略支援部队航天工程大学 | Early fault diagnosis method, equipment and storage medium for DC-DC converter |
Non-Patent Citations (4)
| Title |
|---|
| A novel RSG-based intelligent bearing fault diagnosis method for motors in high-noise industrial environment;Pin Lyu et al.;《AdvancedEngineeringInformatics》;20220226;1-16 * |
| 串联式轴向柱塞泵转位角对合流流量脉动的影响;徐楠 等;《液压与气动》;20230831;第47卷(第8期);8-16 * |
| 基于FMI的轴向柱塞泵分布式联合仿真与动态优化;郭志敏 等;《工程设计学报》;20230831;第30卷(第4期);495-502 * |
| 基于振动信号分析的转辙机故障诊断方法研究;宋迪;《中国优秀硕士学位论文全文数据库》;20230715;1-108 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117150376A (en) | 2023-12-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Jia et al. | A novel ranking-based clustering approach for hyperspectral band selection | |
| Lange et al. | Stability-based validation of clustering solutions | |
| Bhise et al. | Breast cancer detection using machine learning techniques | |
| HUE027904T2 (en) | Method and system for determining whether a drug will be effective on a patient with a disease | |
| CN109817339B (en) | Patient grouping method and device based on big data | |
| CN111553127A (en) | Multi-label text data feature selection method and device | |
| CN111507470A (en) | Abnormal account identification method and device | |
| Chiang et al. | A combination of rough-based feature selection and RBF neural network for classification using gene expression data | |
| Amelio et al. | A genetic algorithm for color image segmentation | |
| Vengatesan et al. | The performance analysis of microarray data using occurrence clustering | |
| CN117150376B (en) | Classification method, device and equipment for fault characteristic samples of axial plunger pump | |
| CN112287036A (en) | Outlier detection method based on spectral clustering | |
| Chi et al. | A novel local human visual perceptual texture description with key feature selection for texture classification | |
| Abouabdallah et al. | Does clustering of DNA barcodes agree with botanical classification directly at high taxonomic levels? Trees in French Guiana as a case study | |
| Handaga et al. | Similarity approach on fuzzy soft set based numerical data classification | |
| Moon et al. | A predictive modeling for detecting fraudulent automobile insurance claims | |
| KR102266950B1 (en) | Method of under-sampling based ensemble for data imbalance problem | |
| Watanabe et al. | SACMiner: A new classification method based on statistical association rules to mine medical images | |
| CN107992886B (en) | Method for predicting failure trend of gas turbine compressor equipment | |
| CN113111935A (en) | Same transaction subject judgment method based on transaction data real-time clustering in bulk commodity electronic commerce market | |
| Ren et al. | Multivariate functional data clustering using adaptive density peak detection | |
| Porwal et al. | Outlier detection by consistent data selection method | |
| Wang et al. | Cosine kernel based density peaks clustering algorithm | |
| Masmoudi et al. | A binarization strategy for modelling mixed data in multigroup classification | |
| Shahbaba et al. | Efficient unimodality test in clustering by signature testing |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |