CN114897074A - Method and device for determining running state of equipment, equipment and storage medium - Google Patents
Method and device for determining running state of equipment, equipment and storage medium Download PDFInfo
- Publication number
- CN114897074A CN114897074A CN202210524527.9A CN202210524527A CN114897074A CN 114897074 A CN114897074 A CN 114897074A CN 202210524527 A CN202210524527 A CN 202210524527A CN 114897074 A CN114897074 A CN 114897074A
- Authority
- CN
- China
- Prior art keywords
- equipment operation
- operation data
- data
- abnormal
- classification model
- 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
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- 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
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Abstract
The disclosure relates to a method and a device for determining an operating state of equipment, the equipment and a storage medium, wherein the method comprises the following steps: receiving equipment operation data in a preset time period, and judging whether a multi-dimensional data classification model for classifying the equipment operation data exists or not: when the multi-dimensional data classification model exists, inputting the equipment operation data into the multi-dimensional data classification model, and outputting a classification result of whether the equipment operation data is abnormal or not so as to determine an equipment operation state according to the classification result; and when the multi-dimensional data classification model does not exist, performing dimension reduction on the equipment operation data, performing clustering processing on the dimension-reduced equipment operation data, identifying abnormal data in the equipment operation data according to a clustering result, determining the equipment operation state according to the identification result, and quickly judging whether the equipment operation data is abnormal or not by using the multi-dimensional data classification model or a dimension reduction clustering method.
Description
Technical Field
The present disclosure relates to the field of data processing technologies, and in particular, to a method and an apparatus for determining an operating state of a device, and a storage medium.
Background
The current equipment running state data is usually a high-dimensional state vector, and the running state of each equipment, the bad running condition of the early warning equipment or the auxiliary equipment running optimization scheme can not be directly judged, so that an operator can not respond in time, and therefore a machine learning method is needed to mine data containing information to determine the equipment running state.
Disclosure of Invention
In order to solve the technical problem or at least partially solve the technical problem, embodiments of the present disclosure provide a device operating state determining method and apparatus, a device, and a storage medium.
In a first aspect, an embodiment of the present disclosure provides a method for determining an operating state of a device, where the method includes:
receiving equipment operation data in a preset time period, and judging whether a multi-dimensional data classification model for classifying the equipment operation data exists, wherein the equipment operation data is multi-dimensional data at the same moment:
when the multi-dimensional data classification model exists, inputting the equipment operation data into the multi-dimensional data classification model, and outputting a classification result of whether the equipment operation data is abnormal or not so as to determine an equipment operation state according to the classification result;
and when the multi-dimensional data classification model does not exist, performing dimension reduction on the equipment operation data, performing clustering processing on the dimension-reduced equipment operation data, and identifying abnormal data in the equipment operation data according to a clustering result so as to determine the equipment operation state according to the identification result.
In one possible embodiment, the multidimensional data classification model is constructed by the following steps:
receiving historical equipment operation data;
the method comprises the steps that an upper sampling method and/or a lower sampling method are/is adopted, so that the number of samples of abnormal equipment operation data samples and non-abnormal equipment operation data samples in historical equipment operation data is balanced, wherein the abnormal equipment operation data samples are equipment operation data when equipment operates abnormally;
and constructing a multi-dimensional data classification model by using the equipment operation data samples with balanced sample number.
In a possible embodiment, the using an upsampling method and/or a downsampling method to balance the number of samples of abnormal device operation data samples and non-abnormal device operation data samples in the historical device operation data includes:
randomly sampling abnormal equipment operation data samples in historical equipment operation data by adopting the sampling method; and/or
And a down-sampling method is adopted to randomly delete non-abnormal equipment operation data samples in the historical equipment operation data, so that the number balance of the abnormal equipment operation data samples and the non-abnormal equipment operation data samples is ensured.
In a possible embodiment, the constructing a multidimensional data classification model by using the sample number balanced device operation data samples includes:
and for each dimension of data in each equipment operation data sample after the sample number is balanced, taking a corresponding threshold condition as a weak classifier, integrating a plurality of weak classifiers into a strong classifier whether the equipment operation state is abnormal or not, and taking the strong classifier as a multi-dimensional data classification model.
In a possible implementation, the performing the dimension reduction processing on the device operation data includes:
and selecting data with preset quantity dimensionality from the equipment operation data according to the importance degree of the data with different dimensionalities in the equipment operation data, wherein the preset quantity dimensionality is smaller than the original quantity dimensionality.
In a possible implementation manner, the selecting data of a preset number of dimensions from the device operation data according to the importance degree of data of different dimensions in the device operation data includes:
operating data X on the equipment n×m Each row of (a) is zero-averaged to obtain
Computing a covariance matrix
Calculating an eigenvalue lambda of the covariance matrix C and a corresponding eigenvector;
taking the first k larger eigenvalues lambda 1 To lambda k Corresponding feature vector v 1 To v k Form a matrix
Reduced matrix
Is formed by splicing m pieces of k-dimensional data according to columns.
In a possible implementation manner, the clustering process on the device operation data after the dimensionality reduction includes:
for data of each dimension in the device operation data after the dimension reduction, forming clusters corresponding to the dimension data and used for representing different data states according to data values of the dimension at different times in a preset time period;
and calculating the total distance value of the equipment operation data at each moment according to the distance between the data of each dimensionality at each moment and the corresponding cluster center, and clustering the equipment operation data subjected to dimensionality reduction.
In a second aspect, an embodiment of the present disclosure provides an apparatus for determining an operating state of a device, including:
the judging module is used for receiving equipment operation data in a preset time period and judging whether a multi-dimensional data classification model for classifying the equipment operation data exists or not, wherein the equipment operation data are multi-dimensional data at the same moment:
the output module is used for inputting the equipment operation data into the multi-dimensional data classification model when the multi-dimensional data classification model exists, outputting a classification result of whether the equipment operation data is abnormal or not, and determining an equipment operation state according to the classification result;
and the clustering module is used for performing dimensionality reduction on the equipment operation data when the multi-dimensional data classification model does not exist, performing clustering processing on the dimensionality reduced equipment operation data, identifying abnormal data in the equipment operation data according to a clustering result, and determining the equipment operation state according to the identification result.
In a third aspect, an embodiment of the present disclosure provides an electronic device, including a processor, a communication interface, a memory, and a communication bus, where the processor, the communication interface, and the memory complete communication with each other through the communication bus;
a memory for storing a computer program;
and the processor is used for realizing the equipment running state determining method when executing the program stored in the memory.
In a fourth aspect, an embodiment of the present disclosure provides a computer-readable storage medium, on which a computer program is stored, and the computer program, when executed by a processor, implements the device operating state determining method described above.
Compared with the prior art, the technical scheme provided by the embodiment of the disclosure at least has part or all of the following advantages:
the method for determining the operating state of the device according to the embodiment of the present disclosure receives device operating data in a preset time period, and determines whether a multidimensional data classification model for classifying the device operating data exists, where the device operating data is multidimensional data at the same time: when the multi-dimensional data classification model exists, inputting the equipment operation data into the multi-dimensional data classification model, and outputting a classification result of whether the equipment operation data is abnormal or not so as to determine an equipment operation state according to the classification result; and when the multi-dimensional data classification model does not exist, performing dimension reduction on the equipment operation data, performing clustering processing on the dimension-reduced equipment operation data, identifying abnormal data in the equipment operation data according to a clustering result, determining the equipment operation state according to the identification result, and quickly judging whether the equipment operation data is abnormal or not by using the multi-dimensional data classification model or a dimension reduction clustering method.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and together with the description, serve to explain the principles of the disclosure.
In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the related art will be briefly described below, and it is obvious for those skilled in the art to obtain other drawings without inventive exercise.
Fig. 1 schematically illustrates a schematic flow diagram of a device operational status determination method according to an embodiment of the present disclosure;
FIG. 2 schematically illustrates a schematic flow chart of a method of determining an operational status of a device according to an embodiment of the present disclosure;
fig. 3 schematically shows a block diagram of a device operation state determination apparatus according to an embodiment of the present disclosure; and
fig. 4 schematically shows a block diagram of an electronic device according to an embodiment of the present disclosure.
Detailed Description
To make the objects, technical solutions and advantages of the embodiments of the present disclosure more clear, the technical solutions of the embodiments of the present disclosure will be described clearly and completely with reference to the drawings in the embodiments of the present disclosure, and it is obvious that the described embodiments are some embodiments of the present disclosure, but not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments disclosed herein without making any creative effort, shall fall within the protection scope of the present disclosure.
Referring to fig. 1, an embodiment of the present disclosure provides a device operating state determining method, including:
s1, receiving equipment operation data in a preset time period, and judging whether a multi-dimensional data classification model for classifying the equipment operation data exists, wherein the equipment operation data are multi-dimensional data at the same moment:
if yes, go to step S2;
if not, go to step S3;
s2, inputting the equipment operation data into the multi-dimensional data classification model, and outputting the classification result of whether the equipment operation data is abnormal or not so as to determine the equipment operation state according to the classification result;
and S3, performing dimensionality reduction on the equipment operation data, performing clustering processing on the dimensionality reduced equipment operation data, and identifying abnormal data in the equipment operation data according to a clustering result to determine the equipment operation state according to the identification result.
In an actual application scene, the equipment operation data is multidimensional data at the same moment, and is obtained through the following steps:
the multi-class working parameters recorded in the running process of the equipment form multi-dimensional working condition point data, which comprises discrete data and continuous data:
discrete data:
switching conditions: including the water pump switch condition, the fan switch condition and the like, and are expressed in the forms of 0 and 1
Continuous data:
temperature: comprises the inlet water temperature of the chilled water, the outlet water temperature of the chilled water, the inlet water temperature of the cooling water, the outlet water temperature of the cooling water and the like;
flow rate: including chilled water flow, etc.;
rotating speed: the method comprises the steps of cooling water pump rotating speed, chilled water pump rotating speed and cooling tower fan rotating speed;
environment: including indoor and outdoor ambient temperatures, weather conditions, etc.;
power consumption: including the total operating power of the system, the total operating efficiency of the system, etc.
The related data jointly form multi-dimensional operating point data at a certain moment, and due to numerous related working parameters, the data has high dimensionality and can not directly identify the outlier condition, namely abnormal data.
Referring to fig. 2, in step S1, the multidimensional data classification model is constructed by the following steps:
s21, receiving historical equipment operation data;
s22, an up-sampling method and/or a down-sampling method are/is adopted to balance the number of samples of abnormal equipment operation data samples and non-abnormal equipment operation data samples in historical equipment operation data, wherein the abnormal equipment operation data samples are equipment operation data when equipment operates abnormally;
and S23, constructing a multi-dimensional data classification model by using the equipment operation data samples with balanced sample number.
In this embodiment, in step S22, the balancing the number of samples of the abnormal device operation data sample and the non-abnormal device operation data sample in the historical device operation data by using the up-sampling method and/or the down-sampling method includes:
randomly sampling abnormal equipment operation data samples in historical equipment operation data by adopting an upper sampling method; and/or
And a down-sampling method is adopted to randomly delete non-abnormal equipment operation data samples in the historical equipment operation data, so that the number balance of the abnormal equipment operation data samples and the non-abnormal equipment operation data samples is ensured.
In practical application, because the abnormal state is less in the operation of the actual equipment, the labeled outlier data is less, and because a biased classifier is trained due to unbalanced data labels during supervised learning, the labels are balanced before training, namely a sampling method is adopted, a few types of samples are randomly sampled, namely the outlier data is sampled, and a SMOTE algorithm or an ADASYN algorithm is used for generating new few types of samples; and a downsampling method is adopted, and part of majority samples are randomly deleted, so that the number of majority samples and minority samples is balanced.
In this embodiment, in step S23, the constructing a multidimensional data classification model by using the sample number balanced device operation data samples includes:
and for each dimension of data in each equipment operation data sample after the sample number is balanced, a corresponding threshold condition is used as a weak classifier, a plurality of weak classifiers are integrated into a strong classifier whether the equipment operation state is abnormal or not, and the strong classifier is used as a multi-dimensional data classification model.
In practical application, a supervised learning method can be used for constructing a two-class classifier as a multi-dimensional data classification model, whether working point data is an outlier or not can be distinguished, an extreme gradient lifting tree can be used, a strong classifier is formed by integrating a plurality of weak classifiers, the outlier classification can be better completed under the conditions of limited sample quantity and lack of parameter adjusting background knowledge, the classifier is obtained by using data training after certain degree of equalization, and the method can be further applied to classifying unmarked data.
In this embodiment, in step S3, the performing dimension reduction processing on the device operation data includes:
selecting data with preset number dimensionality from the equipment operation data according to the importance degree of the data with different dimensionality in the equipment operation data, wherein the preset number dimensionality is smaller than the original number dimensionality,
in an embodiment, the selecting data of a preset number of dimensions from the device operation data according to the importance degree of data of different dimensions in the device operation data includes:
operating data X on the equipment n×m Each row of (a) is zero-averaged to obtain
Wherein, X n×m The system is formed by splicing m pieces of n-dimensional data according to columns, each piece of data is one type of state information of the working point, if a certain piece of information is the switching condition (0 or 1) of various water pumps and fans on the working point, a certain piece of information is the temperature of chilled water and the like of the working point related to the working point, for each working point, m types of state information is recorded, n dimensions of each type of information are recorded, and if the recording dimension of a certain type of state is smaller than n, the m types of state information are supplemented by 0;
computing a covariance matrix
Calculating an eigenvalue lambda of the covariance matrix C and a corresponding eigenvector;
taking the first k larger eigenvalues lambda 1 To lambda k Corresponding feature vector v 1 To v k Form a matrix
Reduced matrix
Is formed by splicing m pieces of k-dimensional data according to columns.
In another embodiment, the selecting data of a preset number of dimensions from the device operation data according to the importance degree of data of different dimensions in the device operation data includes:
under the condition of knowing the importance of each data point of the working condition, manually selecting by adopting a characteristic selection method;
when experience is poor or the workload of feature selection is large, different source data from different proportion mixtures is identified, high-order redundancy among components is removed through an independent component analysis algorithm, and dimension reduction is achieved, wherein the different source data comprise various types of data from different machines.
In one embodiment, in step S3, the clustering the device operation data after the dimension reduction includes:
for data of each dimension in the device operation data after the dimension reduction, forming clusters corresponding to the dimension data and used for representing different data states according to data values of the dimension at different times in a preset time period;
and calculating the total distance value of the equipment operation data at each moment according to the distance between the data of each dimensionality at each moment and the corresponding cluster center, and clustering the equipment operation data subjected to dimensionality reduction.
In another embodiment, in step S3, the clustering process performed on the device operation data after the dimension reduction includes:
by adopting a K-Means clustering algorithm, under the condition that the data category number is judged to be 2 in advance, the distance between each data point and a central point is calculated, the data is rapidly clustered, and normal working point data of a large cluster and outlier working point data of a small cluster are gathered, so that the operating condition of the working point is judged, wherein the central point is obtained by the following steps: and presetting a classification number k group, randomly selecting k objects as initial clustering centers, calculating the distance between each data point and the current k clustering centers in the clustering process, classifying the data point into the category to which the clustering center closest to the data point belongs, recalculating a new central point for the category after each new clustering is performed, and continuously circulating the clustering process until all the data points are clustered.
In another embodiment, in step S3, the clustering the dimensionality reduced device operation data includes:
and forming the most possible class condition by calculating the probability that each data point belongs to a certain class through expectation maximization clustering in a Gaussian mixture model, and obtaining the working condition points which cannot be clustered with the large class working condition points, namely the outlier working condition points.
It is worth to be noted that when the multidimensional data classification model is used for classifying the equipment operation data, the classification result of the multidimensional data classification model can be verified by performing dimension reduction and clustering on the equipment operation data.
Referring to fig. 3, an embodiment of the present disclosure provides an apparatus operating state determining device, including:
a determining module 11, configured to receive device operation data in a preset time period, and determine whether a multidimensional data classification model for classifying the device operation data already exists, where the device operation data is multidimensional data at the same time:
an output module 12, configured to, when the multidimensional data classification model exists, input the device operation data into the multidimensional data classification model, and output a classification result of whether the device operation data is abnormal, so as to determine a device operation state according to the classification result;
and the clustering module 13 is configured to, when the multidimensional data classification model does not exist, perform dimension reduction on the device operation data, perform clustering on the dimension-reduced device operation data, identify abnormal data in the device operation data according to a clustering result, and determine a device operation state according to the identification result.
The implementation process of the functions and actions of each unit in the above device is specifically described in the implementation process of the corresponding step in the above method, and is not described herein again.
For the device embodiment, since it basically corresponds to the method embodiment, reference may be made to the partial description of the method embodiment for relevant points. The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of the present invention. One of ordinary skill in the art can understand and implement it without inventive effort.
In the second embodiment, any plurality of the judging module 11, the outputting module 12 and the clustering module 13 may be combined into one module to be implemented, or any one of the modules may be split into a plurality of modules. Alternatively, at least part of the functionality of one or more of these modules may be combined with at least part of the functionality of the other modules and implemented in one module. At least one of the determining module 11, the outputting module 12 and the clustering module 13 may be at least partially implemented as a hardware circuit, such as a Field Programmable Gate Array (FPGA), a Programmable Logic Array (PLA), a system on a chip, a system on a substrate, a system on a package, an Application Specific Integrated Circuit (ASIC), or may be implemented by hardware or firmware in any other reasonable manner of integrating or packaging a circuit, or implemented by any one of three implementations of software, hardware and firmware, or an appropriate combination of any of them. Alternatively, at least one of the determination module 11, the output module 12 and the clustering module 13 may be at least partially implemented as a computer program module, which when executed may perform a corresponding function.
Referring to fig. 4, an electronic device provided by a third exemplary embodiment of the present disclosure includes a processor 1110, a communication interface 1120, a memory 1130, and a communication bus 1140, where the processor 1110, the communication interface 1120, and the memory 1130 complete communication with each other through the communication bus 1140;
a memory 1130 for storing computer programs;
the processor 1110, when executing the program stored in the memory 1130, implements a method for determining an operation state of an apparatus as follows:
receiving equipment operation data in a preset time period, and judging whether a multi-dimensional data classification model for classifying the equipment operation data exists, wherein the equipment operation data is multi-dimensional data at the same moment:
when the multi-dimensional data classification model exists, inputting the equipment operation data into the multi-dimensional data classification model, and outputting a classification result of whether the equipment operation data is abnormal or not so as to determine an equipment operation state according to the classification result;
and when the multi-dimensional data classification model does not exist, performing dimension reduction on the equipment operation data, performing clustering processing on the dimension-reduced equipment operation data, and identifying abnormal data in the equipment operation data according to a clustering result so as to determine the equipment operation state according to the identification result.
The communication bus 1140 may be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The communication bus 1140 may be divided into an address bus, a data bus, a control bus, and the like. For ease of illustration, only one thick line is shown, but this does not mean that there is only one bus or one type of bus.
The communication interface 1120 is used for communication between the electronic device and other devices.
The Memory 1130 may include a Random Access Memory (RAM) or a non-volatile Memory (non-volatile Memory), such as at least one disk Memory. Optionally, the memory 1130 may also be at least one memory device located remotely from the processor 1110.
The Processor 1110 may be a general-purpose Processor, and includes a Central Processing Unit (CPU), a Network Processor (NP), and the like; the Integrated Circuit may also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, a discrete Gate or transistor logic device, or a discrete hardware component.
A fourth exemplary embodiment of the present disclosure also provides a computer-readable storage medium. The above-mentioned computer-readable storage medium has stored thereon a computer program which, when executed by a processor, implements the method of data processing as described above.
The computer-readable storage medium may be contained in the apparatus/device described in the above embodiments; or may be present alone without being assembled into the device/apparatus. The above-mentioned computer-readable storage medium carries one or more programs which, when executed, implement the method of data processing according to an embodiment of the present disclosure.
According to embodiments of the present disclosure, the computer-readable storage medium may be a non-volatile computer-readable storage medium, which may include, for example but is not limited to: a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the present disclosure, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.
It is noted that, in this document, relational terms such as "first" and "second," and the like, may be 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. Also, 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 an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
The previous description is only for the purpose of describing particular embodiments of the present disclosure, so as to enable those skilled in the art to understand or implement the present disclosure. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the disclosure. Thus, the present disclosure is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (10)
1. A method for determining an operating state of a device, the method comprising:
receiving equipment operation data in a preset time period, and judging whether a multi-dimensional data classification model for classifying the equipment operation data exists, wherein the equipment operation data is multi-dimensional data at the same moment:
when the multi-dimensional data classification model exists, inputting the equipment operation data into the multi-dimensional data classification model, and outputting a classification result of whether the equipment operation data is abnormal or not so as to determine an equipment operation state according to the classification result;
and when the multi-dimensional data classification model does not exist, performing dimension reduction on the equipment operation data, performing clustering processing on the dimension-reduced equipment operation data, and identifying abnormal data in the equipment operation data according to a clustering result so as to determine the equipment operation state according to the identification result.
2. The method of claim 1, wherein the multidimensional data classification model is constructed by:
receiving historical equipment operation data;
the method comprises the steps that an upper sampling method and/or a lower sampling method are/is adopted, so that the number of samples of abnormal equipment operation data samples and non-abnormal equipment operation data samples in historical equipment operation data is balanced, wherein the abnormal equipment operation data samples are equipment operation data when equipment operates abnormally;
and constructing a multi-dimensional data classification model by using the equipment operation data samples with balanced sample number.
3. The method of claim 2, wherein the using an upsampling method and/or a downsampling method to balance the number of samples of abnormal equipment operation data samples and non-abnormal equipment operation data samples in historical equipment operation data comprises:
randomly sampling abnormal equipment operation data samples in historical equipment operation data by adopting an upper sampling method; and/or
And a down-sampling method is adopted to randomly delete non-abnormal equipment operation data samples in the historical equipment operation data, so that the number balance of the abnormal equipment operation data samples and the non-abnormal equipment operation data samples is ensured.
4. The method of claim 2, wherein the constructing the multi-dimensional data classification model using the sample number balanced device operation data samples comprises:
and for each dimension of data in each equipment operation data sample after the sample number is balanced, a corresponding threshold condition is used as a weak classifier, a plurality of weak classifiers are integrated into a strong classifier whether the equipment operation state is abnormal or not, and the strong classifier is used as a multi-dimensional data classification model.
5. The method of claim 1, wherein the performing the dimension reduction on the device operation data comprises:
and selecting data with preset quantity dimensionality from the equipment operation data according to the importance degree of the data with different dimensionalities in the equipment operation data, wherein the preset quantity dimensionality is smaller than the original quantity dimensionality.
6. The method according to claim 5, wherein the selecting a preset number of dimensions of data from the device operation data according to the importance degree of the data with different dimensions in the device operation data comprises:
operating data X on the equipment n×m Each row of (a) is zero-averaged to obtain
Computing a covariance matrix
Calculating an eigenvalue lambda of the covariance matrix C and a corresponding eigenvector;
taking the first k larger eigenvalues lambda 1 To lambda k Corresponding feature vector v 1 To v k Form a matrix
Reduced matrix
Is formed by splicing m pieces of k-dimensional data according to columns.
7. The method according to claim 1, wherein the clustering the reduced-dimension device operation data includes:
for data of each dimension in the device operation data after the dimension reduction, forming clusters corresponding to the dimension data and used for representing different data states according to data values of the dimension at different times in a preset time period;
and calculating the total distance value of the equipment operation data at each moment according to the distance between the data of each dimensionality at each moment and the corresponding cluster center, and clustering the equipment operation data subjected to dimensionality reduction.
8. An apparatus operating state determining device, comprising:
the judging module is used for receiving equipment operation data in a preset time period and judging whether a multi-dimensional data classification model for classifying the equipment operation data exists or not, wherein the equipment operation data are multi-dimensional data at the same moment:
the output module is used for inputting the equipment operation data into the multi-dimensional data classification model when the multi-dimensional data classification model exists, outputting a classification result of whether the equipment operation data is abnormal or not, and determining an equipment operation state according to the classification result;
and the clustering module is used for performing dimensionality reduction on the equipment operation data when the multi-dimensional data classification model does not exist, performing clustering processing on the dimensionality reduced equipment operation data, identifying abnormal data in the equipment operation data according to a clustering result, and determining the equipment operation state according to the identification result.
9. An electronic device is characterized by comprising a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory are communicated with each other through the communication bus;
a memory for storing a computer program;
a processor for implementing the method of determining an operational status of a device according to any one of claims 1 to 7 when executing a program stored in the memory.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the method for determining an operational state of an apparatus according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210524527.9A CN114897074A (en) | 2022-05-13 | 2022-05-13 | Method and device for determining running state of equipment, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210524527.9A CN114897074A (en) | 2022-05-13 | 2022-05-13 | Method and device for determining running state of equipment, equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114897074A true CN114897074A (en) | 2022-08-12 |
Family
ID=82721638
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210524527.9A Pending CN114897074A (en) | 2022-05-13 | 2022-05-13 | Method and device for determining running state of equipment, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114897074A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115981970A (en) * | 2023-03-20 | 2023-04-18 | 建信金融科技有限责任公司 | Operation and maintenance data analysis method, device, equipment and medium |
| CN116680550A (en) * | 2023-05-23 | 2023-09-01 | 南京航空航天大学 | Rolling bearing fault diagnosis method based on active learning under sample imbalance |
-
2022
- 2022-05-13 CN CN202210524527.9A patent/CN114897074A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115981970A (en) * | 2023-03-20 | 2023-04-18 | 建信金融科技有限责任公司 | Operation and maintenance data analysis method, device, equipment and medium |
| CN115981970B (en) * | 2023-03-20 | 2023-05-16 | 建信金融科技有限责任公司 | Fortune dimension analysis method, device, equipment and medium |
| CN116680550A (en) * | 2023-05-23 | 2023-09-01 | 南京航空航天大学 | Rolling bearing fault diagnosis method based on active learning under sample imbalance |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114897074A (en) | Method and device for determining running state of equipment, equipment and storage medium | |
| US11461659B2 (en) | Feature set determining method and apparatus | |
| KR101879416B1 (en) | Apparatus and method for detecting abnormal financial transaction | |
| WO2020114108A1 (en) | Clustering result interpretation method and device | |
| EP1388804A2 (en) | Method and system for face detection using pattern classifier | |
| WO2021071911A1 (en) | Sensor contribution ranking | |
| US10452717B2 (en) | Technologies for node-degree based clustering of data sets | |
| CN102201062A (en) | Information processing apparatus, method and program | |
| CN108629358B (en) | Object class prediction method and device | |
| CN111260243A (en) | Risk assessment method, device, equipment and computer readable storage medium | |
| CN113918554A (en) | Commodity data cleaning method and device, equipment, medium and product thereof | |
| CN113537337A (en) | Training method, abnormality detection method, apparatus, device, and storage medium | |
| CN109902731B (en) | Performance fault detection method and device based on support vector machine | |
| WO2020114109A1 (en) | Interpretation method and apparatus for embedding result | |
| CN111291807A (en) | Fine-grained image classification method and device and storage medium | |
| CN115705279A (en) | Intelligent fault early warning method and device based on index data | |
| FI129676B (en) | Building an ensemble of anomaly detection models for analyzing measurement results | |
| CN115567371B (en) | Abnormity detection method, device, equipment and readable storage medium | |
| CN115984671A (en) | Model online updating method and device, electronic equipment and readable storage medium | |
| CN115567224A (en) | Method for detecting abnormal transaction of block chain and related product | |
| US20210133080A1 (en) | Interpretable prediction using extracted temporal and transition rules | |
| CN114064900A (en) | Power distribution automation terminal fault diagnosis method, device, equipment and storage medium | |
| CN114913008A (en) | Decision tree-based bond value analysis method, device, equipment and storage medium | |
| CN114022086A (en) | Purchasing method, device, equipment and storage medium based on BOM identification | |
| AU2021106250A4 (en) | An iot based intrusion detection system |
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 |