CN108196986A - Unit exception detection method, device, computer equipment and storage medium - Google Patents
Unit exception detection method, device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN108196986A CN108196986A CN201711477775.8A CN201711477775A CN108196986A CN 108196986 A CN108196986 A CN 108196986A CN 201711477775 A CN201711477775 A CN 201711477775A CN 108196986 A CN108196986 A CN 108196986A
- Authority
- CN
- China
- Prior art keywords
- frequency domain
- monitoring signals
- abnormality detection
- different levels
- frequency
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
- G06F11/16—Error detection or correction of the data by redundancy in hardware
- G06F11/1608—Error detection by comparing the output signals of redundant hardware
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/22—Detection or location of defective computer hardware by testing during standby operation or during idle time, e.g. start-up testing
- G06F11/26—Functional testing
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Quality & Reliability (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Hardware Design (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
The present invention proposes a kind of unit exception detection method, device, computer equipment and storage medium, wherein, method includes:Obtain the time domain monitoring signals being monitored to equipment;Frequency-domain transform is carried out to time domain monitoring signals, obtains frequency domain monitoring signals;According to dimensionless Processing Algorithm, the amplitude of frequency domain components at different levels in frequency domain monitoring signals is adjusted;By in frequency domain monitoring signals, in the abnormality detection model that the frequency domain components at different levels input after range-adjusting is trained in advance, abnormality detection result is obtained;Wherein, abnormality detection model has learnt to obtain the correspondence using between the frequency domain components at different levels after the adjustment of dimensionless Processing Algorithm and abnormality detection result.This method can be realized carries out real-time, automatic abnormality detection to bulk device, promotes efficiency and the accuracy of result detection.
Description
Technical field
The present invention relates to technical field of information processing more particularly to a kind of unit exception detection method, device, computer to set
Standby and storage medium.
Background technology
With the continuous development and innovation of computer technology, technology of Internet of things obtains large development in many fields.Mesh
The number of devices of preceding access Internet of Things is more huge, how to realize real-time management and monitors the state of each equipment, and detect and set
It is standby exception whether occur with very profound significance.
In the prior art, for the sensor that output signal in equipment is periodic function, by manually to existing exception
Reason is summarized, and then according to the spectrogram of sensor output signal, judges whether the sensor is abnormal.This mode
Under, whether it is abnormal by artificial judgment sensor, efficiency and accuracy are relatively low, and workload is larger, are not suitable for magnanimity
The detection of equipment.
Invention content
The present invention is directed to solve at least some of the technical problems in related technologies.
For this purpose, first purpose of the present invention is to propose a kind of unit exception detection method, to realize to bulk device
Real-time, automatic abnormality detection is carried out, promotes efficiency and the accuracy of result detection, it is existing by manually sentencing for solving
Whether link sensor is abnormal, and efficiency and accuracy are relatively low, and workload is larger, is not suitable for the skill of the detection of bulk device
Art problem.
Second object of the present invention is to propose a kind of unit exception detection device.
Third object of the present invention is to propose a kind of computer equipment.
Fourth object of the present invention is to propose a kind of computer readable storage medium.
The 5th purpose of the present invention is to propose a kind of computer program product.
In order to achieve the above object, first aspect present invention embodiment proposes a kind of unit exception detection method, including:
Obtain the time domain monitoring signals being monitored to equipment;
Frequency-domain transform is carried out to the time domain monitoring signals, obtains frequency domain monitoring signals;
According to dimensionless Processing Algorithm, the amplitude of frequency domain components at different levels in the frequency domain monitoring signals is adjusted;
By the abnormality detection mould that in the frequency domain monitoring signals, the frequency domain components at different levels input after range-adjusting is trained in advance
In type, abnormality detection result is obtained;Wherein, the abnormality detection model has learnt to obtain using the dimensionless Processing Algorithm tune
The correspondence between frequency domain components at different levels and abnormality detection result after whole.
The equipment of the embodiment of the present invention method for detecting abnormality, by will be to the time domain for the single dimension that equipment is monitored
Monitoring signals are transformed to the frequency domain monitoring signals of multidimensional, and then can obtain more amplitude-frequency characteristic, then using dimensionless at
Adjustment method adjusts the amplitude of frequency domain components at different levels in frequency domain monitoring signals, can be by the width of the corresponding frequency domain components of abnormal data
Value tag is amplified, and finally using abnormality detection model trained in advance, frequency domain monitoring signals is detected, are detected
As a result, it is possible to achieve real-time, automatic abnormality detection is carried out to bulk device.Further, since abnormality detection model has learnt
It obtains using the correspondence between the frequency domain components at different levels after the adjustment of dimensionless Processing Algorithm and abnormality detection result, so as to root
According to abnormality detection model, frequency domain monitoring signals are detected, efficiency and the accuracy of result detection can be promoted, solved existing
Have in technology and whether be abnormal by artificial judgment sensor, efficiency and accuracy are relatively low, and workload is larger, are not suitable for
The technical issues of detection of bulk device.
In order to achieve the above object, second aspect of the present invention embodiment proposes a kind of unit exception detection device, including:
Acquisition module, for obtaining the time domain monitoring signals being monitored to equipment;
Conversion module for carrying out frequency-domain transform to the time domain monitoring signals, obtains frequency domain monitoring signals;
Module is adjusted, for according to dimensionless Processing Algorithm, adjusting frequency domain components at different levels in the frequency domain monitoring signals
Amplitude;
Detection module, for by the frequency domain monitoring signals, the frequency domain components at different levels input after range-adjusting to be instructed in advance
In experienced abnormality detection model, abnormality detection result is obtained;Wherein, the abnormality detection model has learnt to obtain using the nothing
The correspondence between frequency domain components at different levels and abnormality detection result after the adjustment of dimension Processing Algorithm.
The equipment of the embodiment of the present invention abnormal detector, by will be to the time domain for the single dimension that equipment is monitored
Monitoring signals are transformed to the frequency domain monitoring signals of multidimensional, and then can obtain more amplitude-frequency characteristic, then using dimensionless at
Adjustment method adjusts the amplitude of frequency domain components at different levels in frequency domain monitoring signals, can be by the width of the corresponding frequency domain components of abnormal data
Value tag is amplified, and finally using abnormality detection model trained in advance, frequency domain monitoring signals is detected, are detected
As a result, it is possible to achieve real-time, automatic abnormality detection is carried out to bulk device.Further, since abnormality detection model has learnt
It obtains using the correspondence between the frequency domain components at different levels after the adjustment of dimensionless Processing Algorithm and abnormality detection result, so as to root
According to abnormality detection model, frequency domain monitoring signals are detected, efficiency and the accuracy of result detection can be promoted, solved existing
Have in technology and whether be abnormal by artificial judgment sensor, efficiency and accuracy are relatively low, and workload is larger, are not suitable for
The technical issues of detection of bulk device.
In order to achieve the above object, third aspect present invention embodiment proposes a kind of computer equipment, including:Memory, place
The computer program managed device and storage on a memory and can run on a processor, when the processor performs described program,
Realize the unit exception detection method as described in first aspect present invention embodiment.
To achieve these goals, fourth aspect present invention embodiment proposes a kind of computer readable storage medium,
On be stored with computer program, which is characterized in that realized when the program is executed by processor such as first aspect present invention embodiment
The unit exception detection method.
In order to achieve the above object, fifth aspect present invention embodiment proposes a kind of computer program product, when the calculating
When instruction in machine program product is performed by processor, the unit exception detection as described in first aspect present invention embodiment is performed
Method.
The additional aspect of the present invention and advantage will be set forth in part in the description, and will partly become from the following description
It obtains significantly or is recognized by the practice of the present invention.
Description of the drawings
Above-mentioned and/or additional aspect and advantage of the invention will become from the following description of the accompanying drawings of embodiments
Significantly and it is readily appreciated that, wherein:
The flow diagram of the first unit exception detection method that Fig. 1 is provided by the embodiment of the present invention;
Fig. 2 is time domain monitoring signals Fourier expansion schematic diagram in the embodiment of the present invention;
The flow diagram of second of unit exception detection method that Fig. 3 is provided by the embodiment of the present invention;
Fig. 4 is a kind of structure diagram of unit exception detection device provided in an embodiment of the present invention;
Fig. 5 is the structure diagram of another unit exception detection device provided in an embodiment of the present invention;
Fig. 6 shows the block diagram suitable for being used for the exemplary computer device for realizing embodiment of the present invention.
Specific embodiment
The embodiment of the present invention is described below in detail, the example of the embodiment is shown in the drawings, wherein from beginning to end
Same or similar label represents same or similar element or the element with same or like function.Below with reference to attached
The embodiment of figure description is exemplary, it is intended to for explaining the present invention, and is not considered as limiting the invention.
Whether for being abnormal by artificial judgment sensor in the prior art, efficiency and accuracy are relatively low, and work
It measures larger, the technical issues of not being suitable for the detection of bulk device, in the embodiment of the present invention, by will be monitored to equipment
To the time domain monitoring signals of single dimension be transformed to the frequency domain monitoring signals of multidimensional, and then more amplitude-frequency characteristic can be obtained,
Then using dimensionless Processing Algorithm, the amplitude of frequency domain components at different levels in frequency domain monitoring signals is adjusted, it can be by abnormal data pair
The amplitude Characteristics for the frequency domain components answered are amplified, finally using abnormality detection model trained in advance, to frequency domain monitoring signals
It is detected, obtains testing result, can realize and real-time, automatic abnormality detection is carried out to bulk device.It is further, since different
Normal detection model has learnt to obtain using between the frequency domain components at different levels after the adjustment of dimensionless Processing Algorithm and abnormality detection result
Correspondence, so as to according to abnormality detection model, be detected to frequency domain monitoring signals, the efficiency of result detection can be promoted
And accuracy.
Below with reference to the accompanying drawings the equipment of the embodiment of the present invention method for detecting abnormality, device, computer equipment and storage are described
Medium.
The flow diagram of the first unit exception detection method that Fig. 1 is provided by the embodiment of the present invention.
As shown in Figure 1, detection method includes the following steps for the unit exception:
Step 101, the time domain monitoring signals being monitored to equipment are obtained.
The equipment of the embodiment of the present invention method for detecting abnormality, can be used for output signal in detection device is periodic function
Whether sensor is abnormal.When being carried out abnormality detection to this kind of sensor, the output signal of such sensor can be obtained,
Time domain monitoring signals are denoted as in the embodiment of the present invention, optionally, it is f (x) to mark the time domain monitoring signals.
Step 102, frequency-domain transform is carried out to time domain monitoring signals, obtains frequency domain monitoring signals.
In the embodiment of the present invention, since the time domain monitoring signals of acquisition are periodic function, time domain can be monitored and believed
Number carry out Fourier expansion, obtain frequency domain monitoring signals, the frequency domain monitoring signals be infinite multiple sine waves combination.
Specifically, time domain monitoring signals f (x) is subjected to Fourier expansion, obtained frequency domain monitoring signals are:
Wherein, k is frequency domain componentsSeries, AkFor amplitude.
As a kind of example, referring to Fig. 2, Fig. 2 is that time domain monitoring signals Fourier expansion is shown in the embodiment of the present invention
It is intended to.Waveform 1 represents that time domain monitoring signals, waveform 2,3,4 etc. are represented after time domain monitoring signals are carried out Fourier expansion,
Frequency domain components at different levels in obtained frequency domain monitoring signals.
Further, since f (x) is periodic function, intrinsic frequency is more apparent, and intrinsic frequency is in the relatively low frequency of frequency
Section.Therefore, in the embodiment of the present invention, in order to reduce the workload of system, treatment effeciency is promoted, it can be to frequency domain monitoring signals
Frequency domain components at different levels are screened, and retain the frequency domain components of default series, wherein, it is wrapped in the frequency component of the default series of reservation
Frequency domain components containing intrinsic frequency, for example, the default series of label is M.Further, since the high fdrequency component in frequency domain components is
Therefore noise, in the embodiment of the present invention, can carry out denoising, to filter out frequency to the frequency domain components at different levels of frequency domain monitoring signals
Rate is higher than the frequency domain components of predetermined threshold value, so as to promote the accuracy of testing result.Wherein, predetermined threshold value can be according to equipment
Application scenarios are configured.
Step 103, according to dimensionless Processing Algorithm, the amplitude of frequency domain components at different levels in frequency domain monitoring signals is adjusted.
Wherein, dimensionless Processing Algorithm for causing the amplitude of the frequency domain components at different levels after adjustment, becomes scalar.
In the embodiment of the present invention, according to dimensionless Processing Algorithm, the width of frequency domain components at different levels in frequency domain monitoring signals is adjusted
After value, the frequency domain components amplitude Characteristics of intrinsic frequency can be weakened, while it is special to amplify the corresponding frequency domain components amplitude of abnormal data
Sign.For example, dimensionless Processing Algorithm can be normalization algorithm, alternatively, dimensionless Processing Algorithm can weaken to be any other
The frequency domain components amplitude Characteristics of intrinsic frequency, while amplify the algorithm of the corresponding frequency domain components amplitude Characteristics of abnormal data, this hair
Bright embodiment is not restricted this.
In the embodiment of the present invention, after frequency domain monitoring signals are obtained, can frequency domain prison be adjusted according to dimensionless Processing Algorithm
Survey the amplitude of frequency domain components at different levels in signal, the frequency domain monitoring signals after being adjusted.Optionally, frequency domain monitors after label adjustment
Signal is f'(x).
As a kind of possible realization method, normalization algorithm may be used, adjust frequency domains at different levels in frequency domain monitoring signals
The amplitude of component so that the frequency domain components amplitude of the intrinsic frequency after adjustment is weakened, while the corresponding frequency domain point of abnormal data
Amount amplitude Characteristics are amplified.
Optionally, when dimensionless Processing Algorithm is normalization algorithm, it is assumed that sensor no exceptions in a device
When, frequency domain monitoring signals are:
Wherein, due to f1(x) it is periodic function, intrinsic frequency is more apparent, and since high fdrequency component is noise,
Intrinsic frequency is a in the relatively low frequency range of frequency, the frequency domain components amplitude for marking intrinsic frequencyf, then af>>ak(k≠f)。
Using normalization algorithm, respectively to f1(x) every level-one frequency domain components in are normalized so that per level-one frequency domain
The coefficient of component is 1, and records and cause used normalized parameter a when the coefficient per level-one frequency domain components is 1k。
It should be noted that normalized parameter ak, k=1,2 ... ..., that is to say, that have correspondence per level-one frequency domain components
Normalized parameter.
After normalization, frequency domain monitoring signals are:
It is found that for intrinsic frequency, before normalization, the frequency domain components amplitude and intrinsic frequency of extrinsic frequency
The ratio between frequency domain components amplitude beDue to af>>ak(k ≠ f), therefore, the frequency domain components amplitude Characteristics of intrinsic frequency are more bright
It is aobvious, and after normalizing, the ratio between the frequency domain components amplitude of extrinsic frequency and the frequency domain components amplitude of intrinsic frequency become 1 (Gu
There is the amplitude of frequency for 1), the frequency domain components amplitude Characteristics of intrinsic frequency are weakened.
And sensor in a device is when being abnormal, it is assumed that frequency domain monitoring signals are:
Wherein, bk≈ak,For abnormal data.
Using normalization algorithm, respectively to f2(x) every level-one frequency domain components divided by corresponding normalized parameter a ink, obtain
Frequency domain monitoring signals after to normalization are:
Wherein,For intrinsic frequency, before normalization, the corresponding frequency domain components amplitude of abnormal data with
The ratio between frequency domain components amplitude of intrinsic frequency isAfter normalization, the corresponding frequency domain components amplitude of abnormal data and intrinsic frequency
The ratio between frequency domain components amplitude of rate becomes(amplitude of intrinsic frequency is 1).Due to af>>ak(k ≠ f), so havingTherefore, after normalization, the ratio between the frequency domain components amplitude of abnormal data and the frequency domain components amplitude of intrinsic frequency are opposite
Increase before adjustment, i.e. the corresponding frequency domain components amplitude Characteristics of abnormal data are amplified.
Step 104, by frequency domain monitoring signals, the exception of the frequency domain components at different levels input training in advance after range-adjusting is examined
It surveys in model, obtains abnormality detection result.
Wherein, abnormality detection model learnt to obtain using the frequency domain components at different levels after the adjustment of dimensionless Processing Algorithm with it is different
Correspondence between normal testing result.
It is carried by step 103 it is found that when the sensor in equipment is abnormal, in the time domain monitoring signals of acquisition abnormal
Data according to dimensionless Processing Algorithm, are adjusted in frequency domain monitoring signals after the amplitude of frequency domain components at different levels, abnormal data is corresponding
Frequency domain components amplitude Characteristics are amplified, meanwhile, the frequency domain components amplitude Characteristics of intrinsic frequency are also weakened, and therefore, frequency domain is supervised
It surveys in signal, in the abnormality detection model trained in advance of the frequency domain components at different levels input after range-adjusting, abnormal inspection can be obtained
Survey result.Since abnormality detection model has learnt to obtain the frequency domain components at different levels and exception using after the adjustment of dimensionless Processing Algorithm
Correspondence between testing result, so as to according to abnormality detection model, be detected to frequency domain monitoring signals, knot can be promoted
The efficiency of fruit detection and accuracy.
The unit exception detection method of the present embodiment, by will be monitored to the time domain for the single dimension that equipment is monitored
Signal is transformed to the frequency domain monitoring signals of multidimensional, and then can obtain more amplitude-frequency characteristic, is then handled and calculated using dimensionless
Method adjusts the amplitude of frequency domain components at different levels in frequency domain monitoring signals, can be special by the amplitude of the corresponding frequency domain components of abnormal data
Sign is amplified, and finally using abnormality detection model trained in advance, frequency domain monitoring signals are detected, and obtains detection knot
Fruit can be realized and carry out real-time, automatic abnormality detection to bulk device.Further, since abnormality detection model has learnt
The correspondence between frequency domain components at different levels and abnormality detection result to after being adjusted using dimensionless Processing Algorithm, so as to basis
Abnormality detection model is detected frequency domain monitoring signals, can promote efficiency and the accuracy of result detection.
In the embodiment of the present invention, before detection, abnormality detection model can be trained in advance, it is right with reference to Fig. 3
The above process is described in detail.
The flow diagram of second of unit exception detection method that Fig. 3 is provided by the embodiment of the present invention.
As shown in figure 3, before step 104, which can also include the following steps:
Step 201, the time domain historical signal monitored in device history operational process is obtained.
In the embodiment of the present invention, in system monitoring equipment running process, it is the period that can preserve output signal in equipment
The signal that the sensor of function is exported, so as in training abnormality detection model, obtain in device history operational process
The time domain historical signal monitored.
Step 202, frequency-domain transform is carried out to time domain historical signal, obtains frequency domain historical signal.
In the embodiment of the present invention, Fourier expansion can be carried out to time domain historical signal, obtain frequency domain historical signal,
The frequency domain historical signal is the combination of infinite multiple sine waves.
It is understood that the time domain historical signal of single dimension to be transformed to the frequency domain historical signal of multidimensional, can obtain
More amplitude-frequency characteristic when being carried out abnormality detection thereby using abnormality detection model, can improve the accuracy of testing result.
Step 203, according to dimensionless Processing Algorithm, the amplitude of frequency domain components at different levels in frequency domain historical signal is adjusted.
In the embodiment of the present invention, according to dimensionless Processing Algorithm, the width of frequency domain components at different levels in frequency domain historical signal is adjusted
After value, the frequency domain components amplitude Characteristics of intrinsic frequency can be weakened, while it is special to amplify the corresponding frequency domain components amplitude of abnormal data
Sign.For example, dimensionless Processing Algorithm can be normalization algorithm, alternatively, dimensionless Processing Algorithm can weaken to be any other
The frequency domain components amplitude Characteristics of intrinsic frequency, while amplify the algorithm of the corresponding frequency domain components amplitude Characteristics of abnormal data, this hair
Bright embodiment is not restricted this.
In the embodiment of the present invention, after frequency domain historical signal is obtained, it can adjust frequency domain according to dimensionless Processing Algorithm and go through
The amplitude of frequency domain components at different levels, the frequency domain historical signal after being adjusted in history signal.
As a kind of possible realization method, normalization algorithm may be used, adjust frequency domains at different levels in frequency domain historical signal
The amplitude of component so that the frequency domain components amplitude Characteristics of the intrinsic frequency after adjustment are weakened, while the corresponding frequency of abnormal data
Domain component amplitude feature is amplified.
Step 204, according to the frequency domain components at different levels after range-adjusting in frequency domain historical signal and the corresponding history of equipment
Operating status is trained abnormality detection model.
Wherein, history run state includes normal condition and abnormality.
Specifically, according to the frequency domain components at different levels after the corresponding historical signal range-adjusting of normal condition, generation for pair
The positive sample that abnormality detection model is trained is trained abnormality detection model using the positive sample.
In the embodiment of the present invention, by according to the frequency domain components at different levels after range-adjusting in frequency domain historical signal, Yi Jishe
Standby corresponding history run state, is trained abnormality detection model, so as to which abnormality detection model can learn to be used
The correspondence between frequency domain components at different levels and abnormality detection result after the adjustment of dimensionless Processing Algorithm, so as to using according to different
Normal detection model when being detected to frequency domain monitoring signals, can promote efficiency and the accuracy of result detection.
As a kind of possible realization method, the mode of regression forecasting may be used, to using long short-term memory (Long
Short Term Memory, LSTM) the abnormality detection model of neural network is trained so that using LSTM neural networks
Abnormality detection model learning is obtained using between the frequency domain components at different levels after the adjustment of dimensionless Processing Algorithm and abnormality detection result
Correspondence.
The unit exception detection method of the present embodiment, by the frequency domain that the time domain historical signal of single dimension is transformed to multidimensional
Historical signal can obtain more amplitude-frequency characteristic, when being carried out abnormality detection thereby using abnormality detection model, can improve inspection
Survey the accuracy of result.According to dimensionless Processing Algorithm, the amplitude of frequency domain components at different levels in frequency domain historical signal is adjusted, can be incited somebody to action
The amplitude Characteristics of the corresponding frequency domain components of abnormal data are amplified, finally according to each after range-adjusting in frequency domain historical signal
Grade frequency domain components and the corresponding history run state of equipment, are trained abnormality detection model, so as to abnormality detection model
It can learn to obtain pair using between the frequency domain components at different levels after dimensionless Processing Algorithm adjustment and abnormality detection result
It should be related to, so as to which using according to abnormality detection model, when being detected to frequency domain monitoring signals, the effect of result detection can be promoted
Rate and accuracy.
In order to realize above-described embodiment, the present invention also proposes a kind of unit exception detection device.
Fig. 4 is a kind of structure diagram of unit exception detection device provided in an embodiment of the present invention.
As shown in figure 4, the unit exception detection device 400 includes:Acquisition module 410, conversion module 420, adjustment module
430 and detection module 440.Wherein,
Acquisition module 410, for obtaining the time domain monitoring signals being monitored to equipment.
Conversion module 420 for carrying out frequency-domain transform to time domain monitoring signals, obtains frequency domain monitoring signals.
As a kind of possible realization method, conversion module 420, specifically for time domain monitoring signals are carried out with Fourier's grade
Number expansion, obtains frequency domain monitoring signalsWherein, k is frequency domain componentsGrade
Number, AkFor amplitude.
Module 430 is adjusted, for according to dimensionless Processing Algorithm, adjusting the width of frequency domain components at different levels in frequency domain monitoring signals
Value.
As a kind of possible realization method, module 430 is adjusted, specifically for using normalization algorithm, adjustment frequency domain prison
Survey the amplitude of frequency domain components at different levels in signal.
Detection module 440, for by frequency domain monitoring signals, the frequency domain components at different levels input after range-adjusting to be trained in advance
Abnormality detection model in, obtain abnormality detection result;Wherein, abnormality detection model has been learnt to obtain and calculated using dimensionless processing
The correspondence between frequency domain components at different levels and abnormality detection result after method adjustment.
Further, in a kind of possible realization method of the embodiment of the present invention, referring to Fig. 5, embodiment shown in Fig. 4
On the basis of, which can also include:Screen denoising module 450 and training module 460.
Screen denoising module 450, for carrying out frequency-domain transform to time domain monitoring signals, obtain frequency domain monitoring signals it
Afterwards, the frequency domain components at different levels of frequency domain monitoring signals are screened, retains the frequency domain components of default series;And/or frequency domain is supervised
The frequency domain components at different levels for surveying signal carry out denoising, to filter out the frequency domain components that frequency is higher than predetermined threshold value.
Training module 460, for obtaining the time domain historical signal monitored in device history operational process;To time domain history
Signal carries out frequency-domain transform, obtains frequency domain historical signal;According to dimensionless Processing Algorithm, frequencies at different levels in frequency domain historical signal are adjusted
The amplitude of domain component;According to the frequency domain components at different levels after range-adjusting in frequency domain historical signal and the corresponding history fortune of equipment
Row state is trained abnormality detection model;History run state includes normal condition and abnormality.
It should be noted that the aforementioned explanation to unit exception detection method embodiment is also applied for the embodiment
Unit exception detection device 400, details are not described herein again.
The unit exception detection device of the present embodiment, by will be monitored to the time domain for the single dimension that equipment is monitored
Signal is transformed to the frequency domain monitoring signals of multidimensional, and then can obtain more amplitude-frequency characteristic, is then handled and calculated using dimensionless
Method adjusts the amplitude of frequency domain components at different levels in frequency domain monitoring signals, can be special by the amplitude of the corresponding frequency domain components of abnormal data
Sign is amplified, and finally using abnormality detection model trained in advance, frequency domain monitoring signals are detected, and obtains detection knot
Fruit can be realized and carry out real-time, automatic abnormality detection to bulk device.Further, since abnormality detection model has learnt
The correspondence between frequency domain components at different levels and abnormality detection result to after being adjusted using dimensionless Processing Algorithm, so as to basis
Abnormality detection model is detected frequency domain monitoring signals, can promote efficiency and the accuracy of result detection.
In order to realize above-described embodiment, the present invention also proposes a kind of computer equipment, including:Memory and processor,
In, the processor is run and the executable program generation by reading the executable program code stored in the memory
The corresponding program of code, for performing unit exception detection method as in the foregoing embodiment.
In order to clearly illustrate the concrete structure of aforementioned computer equipment, Fig. 6 shows to be used for realizing implementation of the present invention
The block diagram of the exemplary computer device 12 of mode.The computer equipment 12 that Fig. 6 is shown is only an example, should not be to this hair
The function and use scope of bright embodiment bring any restrictions.
As shown in fig. 6, computer equipment 12 is showed in the form of general purpose computing device.The component of computer equipment 12 can
To include but not limited to:One or more processor or processing unit 16, system storage 28 connect different system component
The bus 18 of (including system storage 28 and processing unit 16).
Bus 18 represents one or more in a few class bus structures, including memory bus or Memory Controller,
Peripheral bus, graphics acceleration port, processor or the local bus using the arbitrary bus structures in a variety of bus structures.It lifts
For example, these architectures include but not limited to industry standard architecture (Industry Standard
Architecture, abbreviation ISA) bus, microchannel architecture (Micro Channel Architecture, abbreviation MAC)
Bus, enhanced isa bus, video electronics standard (Vedio Electronic Standard Association, abbreviation
VESA) local bus and peripheral component interconnection (Peripheral Component Interconnect, abbreviation PCI) bus.
Computer equipment 12 typically comprises a variety of computer system readable media.These media can be it is any can be by
The usable medium that computer equipment 12 accesses, including volatile and non-volatile medium, moveable and immovable medium.
System storage 28 can include the computer system readable media of form of volatile memory, such as arbitrary access
Memory (Random Access Memory, abbreviation RAM) 30 and/or cache memory 32.Computer equipment 12 can be with
Further comprise other removable/nonremovable, volatile/non-volatile computer system storage mediums.Only as an example,
Storage system 34 can be used for reading and writing immovable, non-volatile magnetic media, and (Fig. 6 do not show, commonly referred to as " hard drive
Device ").Although being not shown in Fig. 6, can provide to drive the disk for moving non-volatile magnetic disk (such as " floppy disk ") read-write
Dynamic device and the disc drives to moving anonvolatile optical disk (such as CD-ROM, DVD-ROM or other optical mediums) read-write
Device.In these cases, each driver can be connected by one or more data media interfaces with bus 18.Memory
28 can include at least one program product, which has one group of (for example, at least one) program module, these programs
Module is configured to perform the function of various embodiments of the present invention.
Program/utility 40 with one group of (at least one) program module 42 can be stored in such as memory 28
In, such program module 42 include but not limited to operating system, one or more application program, other program modules and
Program data may include the realization of network environment in each or certain combination in these examples.Program module 42 is usual
Perform the function and/or method in embodiment described in the invention.
Computer equipment 12 can also be with one or more external equipments 14 (such as keyboard, sensing equipment, display 24
Deng) communication, the equipment interacted with the computer equipment 12 communication can be also enabled a user to one or more and/or with making
The computer equipment 12 any equipment (such as network interface card, the modulatedemodulate that can communicate with one or more of the other computing device
Adjust device etc.) communication.This communication can be carried out by input/output (I/O) interface 22.Also, computer equipment 12 may be used also
To pass through network adapter 20 and one or more network (such as LAN, wide area network and/or public network, such as because of spy
Net) communication.As shown in the figure, network adapter 20 is communicated by bus 18 with other modules of computer equipment 12.It should be understood that
Although not shown in the drawings, can combine computer equipment 12 uses other hardware and/or software module, including but not limited to:It is micro-
Code, device driver, redundant processing unit, external disk drive array, redundant array of independent disks (Redundant
Array of Independent Disks, abbreviation RAID) system, tape drive and data backup storage system etc..
Processing unit 16 is stored in program in system storage 28 by operation, so as to perform various functions application and
Above equipment method for detecting abnormality is realized in data processing.
In order to achieve the above object, the present invention also proposes a kind of computer program product, when the instruction in computer program product
When being performed by processor, unit exception detection method as in the foregoing embodiment is performed.
In order to realize above-described embodiment, the present invention also proposes a kind of non-transitorycomputer readable storage medium, deposits thereon
Computer program is contained, unit exception as in the foregoing embodiment can be realized when the computer program is executed by processor
Detection method.
In the description of this specification, reference term " one embodiment ", " example ", " is specifically shown " some embodiments "
The description of example " or " some examples " etc. means specific features, structure, material or the spy for combining the embodiment or example description
Point is contained at least one embodiment of the present invention or example.In the present specification, schematic expression of the above terms are not
It must be directed to identical embodiment or example.Moreover, particular features, structures, materials, or characteristics described can be in office
It is combined in an appropriate manner in one or more embodiments or example.In addition, without conflicting with each other, the skill of this field
Art personnel can tie the different embodiments or examples described in this specification and the feature of different embodiments or examples
It closes and combines.
In addition, term " first ", " second " are only used for description purpose, and it is not intended that instruction or hint relative importance
Or the implicit quantity for indicating indicated technical characteristic.Define " first " as a result, the feature of " second " can be expressed or
Implicitly include at least one this feature.In the description of the present invention, " multiple " are meant that at least two, such as two, three
It is a etc., unless otherwise specifically defined.
Any process described otherwise above or method description are construed as in flow chart or herein, represent to include
Module, segment or the portion of the code of the executable instruction of one or more the step of being used to implement custom logic function or process
Point, and the range of the preferred embodiment of the present invention includes other realization, wherein can not press shown or discuss suitable
Sequence, including according to involved function by it is basic simultaneously in the way of or in the opposite order, to perform function, this should be of the invention
Embodiment person of ordinary skill in the field understood.
Expression or logic and/or step described otherwise above herein in flow charts, for example, being considered use
In the order list for the executable instruction for realizing logic function, may be embodied in any computer-readable medium, for
Instruction execution system, device or equipment (such as computer based system, including the system of processor or other can be held from instruction
The system of row system, device or equipment instruction fetch and execute instruction) it uses or combines these instruction execution systems, device or set
It is standby and use.For the purpose of this specification, " computer-readable medium " can any can be included, store, communicate, propagate or pass
Defeated program is for instruction execution system, device or equipment or the dress used with reference to these instruction execution systems, device or equipment
It puts.The more specific example (non-exhaustive list) of computer-readable medium is including following:Electricity with one or more wiring
Connecting portion (electronic device), portable computer diskette box (magnetic device), random access memory (RAM), read-only memory
(ROM), erasable edit read-only storage (EPROM or flash memory), fiber device and portable optic disk is read-only deposits
Reservoir (CDROM).In addition, computer-readable medium can even is that the paper that can print described program on it or other are suitable
Medium, because can be for example by carrying out optical scanner to paper or other media, then into edlin, interpretation or when necessary with it
His suitable method is handled electronically to obtain described program, is then stored in computer storage.
It should be appreciated that each section of the present invention can be realized with hardware, software, firmware or combination thereof.Above-mentioned
In embodiment, software that multiple steps or method can in memory and by suitable instruction execution system be performed with storage
Or firmware is realized.Such as, if realized with hardware in another embodiment, following skill well known in the art can be used
Any one of art or their combination are realized:With for data-signal realize logic function logic gates from
Logic circuit is dissipated, the application-specific integrated circuit with suitable combinational logic gate circuit, programmable gate array (PGA), scene can compile
Journey gate array (FPGA) etc..
Those skilled in the art are appreciated that realize all or part of step that above-described embodiment method carries
Suddenly it is that relevant hardware can be instructed to complete by program, the program can be stored in a kind of computer-readable storage medium
In matter, the program when being executed, one or a combination set of the step of including embodiment of the method.
In addition, each functional unit in each embodiment of the present invention can be integrated in a processing module, it can also
That each unit is individually physically present, can also two or more units be integrated in a module.Above-mentioned integrated mould
The form that hardware had both may be used in block is realized, can also be realized in the form of software function module.The integrated module is such as
Fruit is realized in the form of software function module and is independent product sale or in use, can also be stored in a computer
In read/write memory medium.
Storage medium mentioned above can be read-only memory, disk or CD etc..Although it has been shown and retouches above
The embodiment of the present invention is stated, it is to be understood that above-described embodiment is exemplary, it is impossible to be interpreted as the limit to the present invention
System, those of ordinary skill in the art can be changed above-described embodiment, change, replace and become within the scope of the invention
Type.
Claims (10)
1. a kind of unit exception detection method, which is characterized in that method includes the following steps:
Obtain the time domain monitoring signals being monitored to equipment;
Frequency-domain transform is carried out to the time domain monitoring signals, obtains frequency domain monitoring signals;
According to dimensionless Processing Algorithm, the amplitude of frequency domain components at different levels in the frequency domain monitoring signals is adjusted;
By the abnormality detection model that in the frequency domain monitoring signals, the frequency domain components at different levels input after range-adjusting is trained in advance
In, obtain abnormality detection result;Wherein, the abnormality detection model has been learnt to obtain and adjusted using the dimensionless Processing Algorithm
Correspondence between rear frequency domain components at different levels and abnormality detection result.
2. unit exception detection method according to claim 1, which is characterized in that it is described according to dimensionless Processing Algorithm,
The amplitude of frequency domain components at different levels in the frequency domain monitoring signals is adjusted, including:
Using normalization algorithm, the amplitude of frequency domain components at different levels in the frequency domain monitoring signals is adjusted.
3. unit exception detection method according to claim 1, which is characterized in that it is described to the time domain monitoring signals into
Row frequency-domain transform after obtaining frequency domain monitoring signals, further includes:
The frequency domain components at different levels of the frequency domain monitoring signals are screened, retain the frequency domain components of default series;
And/or denoising is carried out to the frequency domain components at different levels of the frequency domain monitoring signals, to filter out frequency higher than predetermined threshold value
Frequency domain components.
4. unit exception detection method according to claim 1, which is characterized in that described by the frequency domain monitoring signals
In, in the abnormality detection model that the frequency domain components at different levels input after range-adjusting is trained in advance, before obtaining abnormality detection result,
It further includes:
Obtain the time domain historical signal monitored in device history operational process;
Frequency-domain transform is carried out to the time domain historical signal, obtains frequency domain historical signal;
According to the dimensionless Processing Algorithm, the amplitude of frequency domain components at different levels in the frequency domain historical signal is adjusted;
According to the frequency domain components at different levels after range-adjusting in the frequency domain historical signal and the corresponding history run of the equipment
State is trained the abnormality detection model;The history run state includes normal condition and abnormality.
5. unit exception detection method according to claim 4, which is characterized in that it is described to the abnormality detection model into
Row training, including:
By the way of regression forecasting, to being trained using the abnormality detection model of LSTM neural networks.
6. unit exception detection method according to claim 1, which is characterized in that it is described to the time domain monitoring signals into
Row frequency-domain transform obtains frequency domain monitoring signals, including:
Fourier expansion is carried out to time domain monitoring signals, obtains frequency domain monitoring signalsIts
In, k is frequency domain componentsSeries, AkFor amplitude.
7. a kind of unit exception detection device, which is characterized in that device includes:
Acquisition module, for obtaining the time domain monitoring signals being monitored to equipment;
Conversion module for carrying out frequency-domain transform to the time domain monitoring signals, obtains frequency domain monitoring signals;
Module is adjusted, for according to dimensionless Processing Algorithm, adjusting the amplitude of frequency domain components at different levels in the frequency domain monitoring signals;
Detection module, for by the frequency domain monitoring signals, the frequency domain components at different levels after range-adjusting to input training in advance
In abnormality detection model, abnormality detection result is obtained;Wherein, the abnormality detection model has learnt to obtain using the dimensionless
The correspondence between frequency domain components at different levels and abnormality detection result after Processing Algorithm adjustment.
8. a kind of computer equipment, which is characterized in that including:Memory, processor and storage on a memory and can handled
The computer program run on device when the processor performs described program, is realized as described in any one of claim 1-6
Unit exception detection method.
9. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the program is held by processor
The unit exception detection method as described in any in claim 1-6 is realized during row.
10. a kind of computer program product when the instruction in the computer program product is performed by processor, is performed as weighed
Profit requires any unit exception detection method in 1-6.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711477775.8A CN108196986B (en) | 2017-12-29 | 2017-12-29 | Equipment abnormality detection method and device, computer equipment and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711477775.8A CN108196986B (en) | 2017-12-29 | 2017-12-29 | Equipment abnormality detection method and device, computer equipment and storage medium |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108196986A true CN108196986A (en) | 2018-06-22 |
CN108196986B CN108196986B (en) | 2021-03-30 |
Family
ID=62586522
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711477775.8A Active CN108196986B (en) | 2017-12-29 | 2017-12-29 | Equipment abnormality detection method and device, computer equipment and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108196986B (en) |
Cited By (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110414603A (en) * | 2019-07-29 | 2019-11-05 | 中国工商银行股份有限公司 | For detecting the method, apparatus, computer system and medium of mobile device |
CN110542474A (en) * | 2019-09-04 | 2019-12-06 | 中国科学院上海高等研究院 | Method, system, medium, and apparatus for detecting vibration signal of device |
CN110763493A (en) * | 2018-07-27 | 2020-02-07 | 珠海格力电器股份有限公司 | Method and device for determining fault type |
CN111124794A (en) * | 2018-11-01 | 2020-05-08 | 百度在线网络技术(北京)有限公司 | Monitoring method and device of data processing system and computer equipment |
CN111855195A (en) * | 2019-04-29 | 2020-10-30 | 富士通株式会社 | Abnormality detection method for gearbox and information processing apparatus |
CN113362850A (en) * | 2020-03-03 | 2021-09-07 | 杭州海康威视数字技术股份有限公司 | Detection method and device of audio signal acquisition device and storage medium |
CN113516023A (en) * | 2021-04-23 | 2021-10-19 | 广东电网有限责任公司计量中心 | Equipment vibration abnormality diagnosis method and system |
CN113794680A (en) * | 2021-08-04 | 2021-12-14 | 清华大学 | Malicious traffic detection method and device under high-bandwidth scene based on frequency domain analysis |
CN114089033A (en) * | 2022-01-24 | 2022-02-25 | 天津安力信通讯科技有限公司 | Abnormal signal detection method and system based on spectrum analysis |
CN114417940A (en) * | 2022-03-25 | 2022-04-29 | 阿里巴巴(中国)有限公司 | Equipment for detecting data center, method and device for obtaining equipment detection model |
CN115993807A (en) * | 2023-03-23 | 2023-04-21 | 日照鲁光电子科技有限公司 | Production monitoring optimization control method and system for silicon carbide |
CN116992254A (en) * | 2023-09-25 | 2023-11-03 | 北京博华信智科技股份有限公司 | Reconstruction method, device and equipment for shell vibration signal of variable frequency motor and storage medium |
CN117289066A (en) * | 2023-11-22 | 2023-12-26 | 南通至正电子有限公司 | Voltage stability monitoring method and system |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101170843A (en) * | 2007-11-30 | 2008-04-30 | 清华大学 | Speaker online pure voice failure diagnosis method |
US20100204813A1 (en) * | 2007-02-01 | 2010-08-12 | Museami, Inc. | Music transcription |
CN103034170A (en) * | 2012-11-27 | 2013-04-10 | 华中科技大学 | Numerical control machine tool machining performance prediction method based on intervals |
CN103175687A (en) * | 2013-03-07 | 2013-06-26 | 温州大学 | Fault location method for sliding-tooth reducer |
CN104502103A (en) * | 2014-12-07 | 2015-04-08 | 北京工业大学 | Bearing fault diagnosis method based on fuzzy support vector machine |
CN106124989A (en) * | 2016-06-29 | 2016-11-16 | 华北电力科学研究院有限责任公司 | Turbine generators machines under rotor winding faults diagnostic method based on diagnostic cast and device |
JP2017194371A (en) * | 2016-04-21 | 2017-10-26 | 株式会社トクヤマ | Method for diagnosing abnormality of diagnosis object in rotational drive device and abnormality diagnosis device used therefor |
-
2017
- 2017-12-29 CN CN201711477775.8A patent/CN108196986B/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20100204813A1 (en) * | 2007-02-01 | 2010-08-12 | Museami, Inc. | Music transcription |
CN101170843A (en) * | 2007-11-30 | 2008-04-30 | 清华大学 | Speaker online pure voice failure diagnosis method |
CN103034170A (en) * | 2012-11-27 | 2013-04-10 | 华中科技大学 | Numerical control machine tool machining performance prediction method based on intervals |
CN103175687A (en) * | 2013-03-07 | 2013-06-26 | 温州大学 | Fault location method for sliding-tooth reducer |
CN104502103A (en) * | 2014-12-07 | 2015-04-08 | 北京工业大学 | Bearing fault diagnosis method based on fuzzy support vector machine |
JP2017194371A (en) * | 2016-04-21 | 2017-10-26 | 株式会社トクヤマ | Method for diagnosing abnormality of diagnosis object in rotational drive device and abnormality diagnosis device used therefor |
CN106124989A (en) * | 2016-06-29 | 2016-11-16 | 华北电力科学研究院有限责任公司 | Turbine generators machines under rotor winding faults diagnostic method based on diagnostic cast and device |
Non-Patent Citations (1)
Title |
---|
易健雄: "基于小波分析的旋转机械故障诊断仪的研究与开发", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 * |
Cited By (19)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110763493A (en) * | 2018-07-27 | 2020-02-07 | 珠海格力电器股份有限公司 | Method and device for determining fault type |
CN111124794B (en) * | 2018-11-01 | 2023-07-07 | 百度在线网络技术(北京)有限公司 | Monitoring method and device for data processing system and computer equipment |
CN111124794A (en) * | 2018-11-01 | 2020-05-08 | 百度在线网络技术(北京)有限公司 | Monitoring method and device of data processing system and computer equipment |
CN111855195B (en) * | 2019-04-29 | 2022-08-30 | 富士通株式会社 | Abnormality detection method for gearbox and information processing apparatus |
CN111855195A (en) * | 2019-04-29 | 2020-10-30 | 富士通株式会社 | Abnormality detection method for gearbox and information processing apparatus |
CN110414603B (en) * | 2019-07-29 | 2022-02-22 | 中国工商银行股份有限公司 | Method, apparatus, computer system, and medium for detecting mobile device |
CN110414603A (en) * | 2019-07-29 | 2019-11-05 | 中国工商银行股份有限公司 | For detecting the method, apparatus, computer system and medium of mobile device |
CN110542474A (en) * | 2019-09-04 | 2019-12-06 | 中国科学院上海高等研究院 | Method, system, medium, and apparatus for detecting vibration signal of device |
CN113362850B (en) * | 2020-03-03 | 2024-05-28 | 杭州海康威视数字技术股份有限公司 | Method and device for detecting audio signal acquisition device and storage medium |
CN113362850A (en) * | 2020-03-03 | 2021-09-07 | 杭州海康威视数字技术股份有限公司 | Detection method and device of audio signal acquisition device and storage medium |
CN113516023A (en) * | 2021-04-23 | 2021-10-19 | 广东电网有限责任公司计量中心 | Equipment vibration abnormality diagnosis method and system |
CN113794680A (en) * | 2021-08-04 | 2021-12-14 | 清华大学 | Malicious traffic detection method and device under high-bandwidth scene based on frequency domain analysis |
CN114089033A (en) * | 2022-01-24 | 2022-02-25 | 天津安力信通讯科技有限公司 | Abnormal signal detection method and system based on spectrum analysis |
CN114417940A (en) * | 2022-03-25 | 2022-04-29 | 阿里巴巴(中国)有限公司 | Equipment for detecting data center, method and device for obtaining equipment detection model |
CN115993807A (en) * | 2023-03-23 | 2023-04-21 | 日照鲁光电子科技有限公司 | Production monitoring optimization control method and system for silicon carbide |
CN116992254B (en) * | 2023-09-25 | 2024-01-19 | 北京博华信智科技股份有限公司 | Reconstruction method, device and equipment for shell vibration signal of variable frequency motor and storage medium |
CN116992254A (en) * | 2023-09-25 | 2023-11-03 | 北京博华信智科技股份有限公司 | Reconstruction method, device and equipment for shell vibration signal of variable frequency motor and storage medium |
CN117289066A (en) * | 2023-11-22 | 2023-12-26 | 南通至正电子有限公司 | Voltage stability monitoring method and system |
CN117289066B (en) * | 2023-11-22 | 2024-02-13 | 南通至正电子有限公司 | Voltage stability monitoring method and system |
Also Published As
Publication number | Publication date |
---|---|
CN108196986B (en) | 2021-03-30 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108196986A (en) | Unit exception detection method, device, computer equipment and storage medium | |
CN108063698A (en) | Unit exception detection method and device, program product and storage medium | |
US10338982B2 (en) | Hybrid and hierarchical outlier detection system and method for large scale data protection | |
US11048729B2 (en) | Cluster evaluation in unsupervised learning of continuous data | |
US7899763B2 (en) | System, method and computer program product for evaluating a storage policy based on simulation | |
CN107729210A (en) | The abnormality diagnostic method and device of Distributed Services cluster | |
CN107610702A (en) | Terminal device standby wakeup method, apparatus and computer equipment | |
US10791036B2 (en) | Infrastructure costs and benefits tracking | |
US8843347B2 (en) | Method, monitoring system and computer program product for monitoring the health of a monitored system utilizing an associative memory | |
CN104346221B (en) | Server hardware device grade classification, schedule management method and device, server | |
Cipollini et al. | Unsupervised deep learning for induction motor bearings monitoring | |
AU2015414767A1 (en) | Anomaly detection in multiple correlated sensors | |
US9772896B2 (en) | Identifying intervals of unusual activity in information technology systems | |
US11747405B2 (en) | System and method for audio and vibration based power distribution equipment condition monitoring | |
JPWO2019239542A1 (en) | Abnormality detection device, abnormality detection method, and abnormality detection program | |
US11250351B2 (en) | System and method for one-class similarity machines for anomaly detection | |
Xu et al. | Automatic roller bearings fault diagnosis using DSAE in deep learning and CFS algorithm | |
US20140053028A1 (en) | Anomaly detection at the level of run time data structures | |
Song et al. | Cubic spline interpolation-based refined composite multiscale dispersion entropy and its application to bearing fault identification | |
US10345883B2 (en) | Power estimation | |
US11030995B2 (en) | Acoustic feature extractor selected according to status flag of frame of acoustic signal | |
Oncescu et al. | A self-supervised classification algorithm for sensor fault identification for robust structural health monitoring | |
Jarrett et al. | Monitoring variability and analyzing multivariate autocorrelated processes | |
CN107562608A (en) | Resource metrics importance appraisal procedure and device | |
JP2021156631A (en) | Anomaly detection device and method |
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 |