CN110110803A - A kind of robot failure diagnosis method, device and equipment - Google Patents
A kind of robot failure diagnosis method, device and equipment Download PDFInfo
- Publication number
- CN110110803A CN110110803A CN201910402385.7A CN201910402385A CN110110803A CN 110110803 A CN110110803 A CN 110110803A CN 201910402385 A CN201910402385 A CN 201910402385A CN 110110803 A CN110110803 A CN 110110803A
- Authority
- CN
- China
- Prior art keywords
- robot
- sensitive features
- collection
- fault diagnosis
- primitive character
- 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
- 238000003745 diagnosis Methods 0.000 title claims abstract description 91
- 238000000034 method Methods 0.000 title claims abstract description 91
- 239000011159 matrix material Substances 0.000 claims abstract description 23
- 238000003860 storage Methods 0.000 claims abstract description 16
- 238000012216 screening Methods 0.000 claims abstract description 15
- 239000012141 concentrate Substances 0.000 claims abstract description 14
- 230000035945 sensitivity Effects 0.000 claims abstract description 12
- 238000004590 computer program Methods 0.000 claims description 12
- 230000001133 acceleration Effects 0.000 claims description 9
- 238000004458 analytical method Methods 0.000 claims description 7
- 238000011156 evaluation Methods 0.000 claims description 4
- 230000000694 effects Effects 0.000 abstract description 8
- 238000009826 distribution Methods 0.000 abstract description 3
- 230000033001 locomotion Effects 0.000 description 16
- 230000008569 process Effects 0.000 description 13
- 238000002405 diagnostic procedure Methods 0.000 description 7
- 238000005516 engineering process Methods 0.000 description 7
- 238000004519 manufacturing process Methods 0.000 description 4
- 238000012423 maintenance Methods 0.000 description 3
- 238000001514 detection method Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000007257 malfunction Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- DMJXRYSGXCLCFP-LBPRGKRZSA-N (3s)-n-tert-butyl-1,2,3,4-tetrahydroisoquinoline-3-carboxamide Chemical compound C1=CC=C2CN[C@H](C(=O)NC(C)(C)C)CC2=C1 DMJXRYSGXCLCFP-LBPRGKRZSA-N 0.000 description 1
- 201000004569 Blindness Diseases 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000005856 abnormality Effects 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 239000003638 chemical reducing agent Substances 0.000 description 1
- 238000000151 deposition Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000005461 lubrication Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000005498 polishing Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000011309 routine diagnosis Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 238000005507 spraying Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000005303 weighing Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/211—Selection of the most significant subset of features
- G06F18/2113—Selection of the most significant subset of features by ranking or filtering the set of features, e.g. using a measure of variance or of feature cross-correlation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Manipulator (AREA)
Abstract
This application discloses a kind of robot failure diagnosis methods, the operation data of robot can be acquired using as primitive character collection, screening sensitive features collection is concentrated from primitive character, and determine that sensitive features concentrate the feature weight of sensitive features, and then sensitive features collection is clustered using the clustering method based on inverse covariance matrix according to feature weight, the fault diagnosis result of robot is finally determined according to cluster result.It can be seen that, this method realizes cluster using the clustering method based on inverse covariance matrix, since the clustering method can carry out fault diagnosis by different manifestations of the discovery robot under identical operating status, improve the reliability of diagnosis, furthermore this method considers the difference of sensitivity of each sensitive features in fault diagnosis, for the corresponding feature weight of each sensitive weight distribution, the accuracy of diagnosis is improved.Present invention also provides a kind of robot failure diagnosis device, equipment and computer readable storage medium, effect is corresponded to the above method.
Description
Technical field
This application involves fault diagnosis field, in particular to a kind of robot failure diagnosis method, device, equipment and calculating
Machine readable storage medium storing program for executing.
Background technique
Robot is a kind of machine of semi-autonomous or full utonomous working, collection Modern Manufacturing Technology, new material technology and letter
Breath control technology is integrated, and is the mainstream product of intelligence manufacture.
At this stage, the integrated level of robot system and complexity are also higher and higher, the operation process of most robot manipulating tasks
Parameter setting, O&M control also uniquely rely on the execution of skilled worker scene, and simple dependence experience and process knowledge have carried out O&M
Be unable to satisfy the demand of current complication system, thus be widely present blindness regular inspection, periodical repair cause maintenance cost improve, low efficiency
The problems such as lower, influence production efficiency and quality.Generally, there are equipment operating efficiencies for the application process of industrial robot at this stage
It is low, process knowledge intelligent decision level is low, operation troubles rate is high, maintenance response not in time, the problems such as maintenance efficiency low cost is high.
Fault diagnosis technology is developed so far, and substantially experienced three phases, respectively the early stage Artificial Diagnosis stage, with automatic
Routine diagnosis stage, intelligent diagnostics stage based on detection.By to the complex equipments fault diagnosis such as domestic and international industrial robot
Current situation is researched and analysed, and can sum up method for diagnosing faults mainly has method for diagnosing faults, base based on signal processing
In the method for diagnosing faults and Knowledge based engineering diagnostic method of analytic modell analytical model.However, the fault diagnosis real-time of these schemes and
Accuracy is lower, is unable to satisfy actual use demand.
Summary of the invention
The purpose of the application is to provide a kind of robot failure diagnosis method, device, equipment and computer-readable storage medium
Matter, the diagnosis real-time and accuracy to solve the problems, such as traditional robot failure diagnosis scheme are lower.Concrete scheme is such as
Under:
In a first aspect, this application provides a kind of robot failure diagnosis methods, comprising:
The operation data of robot is acquired using as primitive character collection;
Screening sensitive features collection is concentrated from the primitive character, and determines that the sensitive features concentrate the feature of sensitive features
Weight;
According to the feature weight, the sensitive features collection is gathered using the clustering method based on inverse covariance matrix
Class obtains cluster result;
According to the cluster result, the fault diagnosis result of the robot is determined.
Optionally, described to concentrate screening sensitive features collection from the primitive character, and it is quick to determine that the sensitive features are concentrated
Feel the feature weight of feature, comprising:
Screening sensitive features collection is concentrated from the primitive character using uncompensation distance appraisal procedure;
According to Distance evaluation standard, determine that the sensitive features concentrate sensitivity of the sensitive features in fault diagnosis,
Using as feature weight.
Optionally, described according to the cluster result, determine the fault diagnosis result of the robot, comprising:
According to the cluster result, outlier and the corresponding operating status of the outlier are determined;
According to the outlier and the operating status, the fault type of the robot is determined.
Optionally, the operation data of the acquisition robot is using as primitive character collection, comprising:
The feedback signal for acquiring motor encoder in robot, using as primitive character collection.
Optionally, the feedback signal includes following any one or more: location of instruction signal, feedback position signal,
Command speed signal, feedback speed signal, command acceleration signal, feedback acceleration signal, instruction torque signals, feedback moment
Signal, error signal.
Optionally, described according to the cluster result, determine the fault diagnosis result of the robot, comprising:
According to the axis feedback moment of the multiple axis of the robot, the axis weight of each axis is determined;
According to the cluster result of multiple axis and corresponding axis weight, fault diagnosis result is determined.
Second aspect, this application provides a kind of robot failure diagnosis devices, comprising:
Primitive character collection determining module: for acquiring the operation data of robot using as primitive character collection;
Sensitive features collection determining module: for concentrating screening sensitive features collection from the primitive character, and determination is described quick
Feel the feature weight of sensitive features in feature set;
Cluster module: it is used for according to the feature weight, using the clustering method based on inverse covariance matrix to described quick
Sense feature set is clustered, and cluster result is obtained;
Fault diagnosis module: for determining the fault diagnosis result of the robot according to the cluster result.
Optionally, the fault diagnosis module includes:
Cluster cell: for determining outlier and the corresponding operating status of the outlier according to the cluster result;
Accident analysis unit: for determining the failure classes of the robot according to the outlier and the operating status
Type.
The third aspect, this application provides a kind of robot failure diagnosis equipment, comprising:
Memory: for storing computer program;
Processor: for executing the computer program to realize a kind of robot failure diagnosis method as described above
Step.
Fourth aspect, this application provides a kind of computer readable storage medium, the computer readable storage medium is used
In storage computer program, for realizing a kind of robot fault as described above when the computer program is executed by processor
The step of diagnostic method.
A kind of robot failure diagnosis method provided herein can acquire the operation data of robot using as original
Beginning feature set concentrates screening sensitive features collection from primitive character, and determines that sensitive features concentrate the feature weight of sensitive features, into
And sensitive features collection is clustered using the clustering method based on inverse covariance matrix according to feature weight, finally according to cluster
As a result the fault diagnosis result of robot is determined.Gather as it can be seen that this method is realized using the clustering method based on inverse covariance matrix
Class is mentioned since robot motion quickly can be divided into several states by the clustering method based on inverse covariance matrix
The speed of diagnosis is risen, it is each that furthermore this method, which considers the difference of sensitivity of each sensitive features in fault diagnosis,
The corresponding feature weight of a sensitivity weight distribution, improves the accuracy of fault diagnosis.
In addition, present invention also provides a kind of robot failure diagnosis device, equipment and computer readable storage medium,
Effect corresponds to the above method, and which is not described herein again.
Detailed description of the invention
It, below will be to embodiment or existing for the clearer technical solution for illustrating the embodiment of the present application or the prior art
Attached drawing needed in technical description is briefly described, it should be apparent that, the accompanying drawings in the following description is only this Shen
Some embodiments please for those of ordinary skill in the art without creative efforts, can be with root
Other attached drawings are obtained according to these attached drawings.
Fig. 1 is a kind of implementation flow chart of robot failure diagnosis method embodiment one provided herein;
Fig. 2 is a kind of implementation flow chart of robot failure diagnosis method embodiment two provided herein;
Fig. 3 is a kind of functional block diagram of robot failure diagnosis Installation practice provided herein;
Fig. 4 is a kind of structural schematic diagram of robot failure diagnosis apparatus embodiments provided herein.
Specific embodiment
The core of the application is to provide a kind of robot failure diagnosis method, device, equipment and computer-readable storage medium
Matter realizes the real-time and accuracy of hoisting machine people's failure diagnostic process.
In order to make those skilled in the art more fully understand application scheme, with reference to the accompanying drawings and detailed description
The application is described in further detail.Obviously, described embodiments are only a part of embodiments of the present application, rather than
Whole embodiments.Based on the embodiment in the application, those of ordinary skill in the art are not making creative work premise
Under every other embodiment obtained, shall fall in the protection scope of this application.
A kind of robot failure diagnosis method embodiment one provided by the present application is introduced below, referring to Fig. 1, is implemented
Example one includes:
Step S101: the operation data of robot is acquired using as primitive character collection;
Specifically, the operation data of above-mentioned robot may include: the location of instruction, feedback position, command speed, feedback speed
Degree, command acceleration, feedback acceleration, instruction torque, feedback moment, error etc..It, can be according to suitable in practical application scene
When frequency the operation data of robot is acquired, obtain above-mentioned primitive character collection, can specifically acquire single or multiple
Movement executes the operation data in the period.When above-mentioned robot lacks corresponding collector or sensor, it can pass through
The feedback signal of the motor encoder of read machine people realizes above-mentioned data acquisition.It is noted that working as above-mentioned machine
When the artificial multi-axis robot of device, can the operation data respectively to each axis of robot be acquired.
Step S102: screening sensitive features collection is concentrated from the primitive character, and it is sensitive to determine that the sensitive features are concentrated
The feature weight of feature;
Different features has different significance levels, and some features are the namely closely related with failure of sensitivity,
But other features are quite different.Therefore, in order to improve fault diagnosis accuracy and avoid dimension disaster, can filter out can more be provided
The sensitive features of robot fault relevant information abandon or weaken uncorrelated or redundancy feature.Therefore, the present embodiment is according to original
Each feature filters out sensitive features from primitive character concentration for the difference of the significance level of fault diagnosis in beginning feature set
Collection, wherein sensitive features collection includes one or more sensitive features.As a kind of specific embodiment, compensation can be chosen
Apart from assessment technology (CDET) the Lai Shixian above process.
Although having chosen sensitive features from primitive character concentration by uncompensation distance assessment technology, selected sensitivity is special
Sign has different sensibility in fault diagnosis.Therefore, in order to obtain more reliable diagnostic result, the present embodiment is each quick
Sense feature imparts corresponding feature weight.Specifically, the present embodiment is the number that each sensitive features is distributed in one [0,1]
Word, for indicating the sensitivity of the sensitive features.In Euclidean space, characteristic weighing is carried out to the corresponding axis of sensitive features
Extension, shrinks axis corresponding with the unrelated feature of failure.
Step S103: according to the feature weight, using the clustering method based on inverse covariance matrix to described sensitive special
Collection is clustered, and cluster result is obtained;
As described above, the present embodiment realizes cluster using the clustering method (abbreviation TICC) based on inverse covariance matrix, it should
Clustering method is clustered according to internal relations between each feature, and cluster result has interpretation, that is to say, that cluster result
Several states with actual physical meaning such as the acceleration of corresponding robot, deceleration, steering, even running, substantially static, therefore
Analysis on Fault Diagnosis can be realized according to cluster.It should be noted that when the artificial multi-axis robot of machine, in cluster process
It needs respectively to cluster the sensitive features matrix of each axis.
Step S104: according to the cluster result, the fault diagnosis result of the robot is determined.
Specifically, the present embodiment notes abnormalities class according to the cluster result, so that it is determined that the failure of the robot is examined
Disconnected result.As a kind of specific mode, outlier can be determined whether by above-mentioned cluster process, and is peeled off determining to exist
When point, fault diagnosis result is determined in conjunction with expertise, wherein so-called outlier just refers in eigenmatrix far from square
The extremum of battle array mean level.
The present embodiment provides a kind of robot failure diagnosis method, can acquire the operation data of robot using as original
Beginning feature set concentrates screening sensitive features collection from primitive character, and determines that sensitive features concentrate the feature weight of sensitive features, into
And sensitive features collection is clustered using the clustering method based on inverse covariance matrix according to feature weight, finally according to cluster
As a result the fault diagnosis result of robot is determined.Gather as it can be seen that this method is realized using the clustering method based on inverse covariance matrix
Class, due to the clustering method based on inverse covariance matrix it can be found that robot run when hidden state, pass through discovery machine
The different manifestations of people in the same state carry out fault diagnosis, improve the reliability of diagnosis, and furthermore this method is in view of each
The difference of sensitivity of a sensitive features in fault diagnosis mentions for the corresponding feature weight of each sensitive weight distribution
The accuracy of fault diagnosis is risen.
Start that a kind of robot failure diagnosis method embodiment two provided by the present application, embodiment diyl is discussed in detail below
It is realized in above-described embodiment one, and has carried out expansion to a certain extent on the basis of example 1.
Specifically, referring to fig. 2, embodiment two includes:
Step S201: the feedback signal of motor encoder in acquisition robot, using as primitive character collection;
It can specifically acquire and execute the feedback signal of the motor encoder in the period, the feedback signal packet in individual part
It includes following any one or more: location of instruction signal, feedback position signal, command speed signal, feedback speed signal, instruction
Acceleration signal, feedback acceleration signal, instruction torque signals, feedback moment signal, error signal.It is noted that
In collection process can the feedback signal respectively to each axis of robot be acquired, obtain the primitive character collection of each axis.
Specifically, in the present embodiment, it can be assumed that the primitive character collection of C condition are as follows:
{qm,c,j, m=1,2 ..., Mc;C=1,2 ..., C;J=1,2 ..., J } (1)
Wherein qm,c,jIt is j-th of characteristic value of m-th of sample under the conditions of c-th, McIt is the total sample number of c-th of condition
Amount, J is the feature total quantity of each sample.Therefore, total sample number amount is Mc× C, feature total quantity are Mc×C×J。
Step S202: according to the axis feedback moment of the multiple axis of the robot, the axis weight of each axis is determined;
It may polish, carry in robot real work, loading and unloading, the various motions such as spraying, difference movement is to machine
The usage degree of six axis of device people is different, for example, polishing movement in 4,5,6 axis movement it is more, so should more pay close attention to this three
The movement of a axis;And the usage degree of 1,2 axis is higher in carrying movement, is also easier to break down, so should more pay close attention to this
Two axis.Therefore, the present embodiment considers the different work condition states of multi-axis robot each axis when running different movements, is each
A axis assigns corresponding axis weight, in addition, can dynamically adjust axis weight in real time according to operating condition during actual diagnosis.Tool
Body, axis weight can be calculated according to the mean value of each axis feedback moment.It is noted that this implementation in cluster process
Example not clusters the operation data of several axis together, but clusters respectively to the operation data of each axis, above-mentioned
Axis weight is only used as the reference conditions of consequent malfunction diagnostic analysis, does not influence cluster result.Therefore, the present embodiment does not limit step
S202's executes sequence, as long as guarantee before step S208, and step S202 provided in this embodiment executes sequence
As a preferred embodiment, can be to avoid the treatment process of the feature set to non-targeted axis.
Step S203: it according to the axis weight, determines and executes the target axis in the period in this movement;
The quantity of the unlimited axis that sets the goal of the present embodiment is specifically based on actual condition and determines.
Step S204: screening sensitive features are concentrated from the primitive character of the target axis using uncompensation distance appraisal procedure
Collection;
Concentrate the process of screening sensitive features that can refer to from the primitive character of target axis based on uncompensation distance appraisal procedure
Existing scheme is no longer discussed in detail herein.
Step S205: according to Distance evaluation standard, determine that the sensitive features concentrate sensitive features in fault diagnosis
Sensitivity, using as feature weight;
After the completion of Feature Selection, several sensitive features can be obtained from primitive character.Then, as a kind of optional
Embodiment can assess each sensitive features by executing uncompensation distance appraisal procedure again, according to Distance evaluation standard
Sensitivity size of each sensitive features in failure diagnostic process is determined, using as feature weight.
It is noted that needing that operation is normalized to each sensitive features, specifically before executing cluster operation
's.It can be by the characteristic value v of t-th of characteristic parametertIt is normalized as the following formula:
Wherein, min (vt) indicate characteristic value vtMinimum value, max (vt) indicate characteristic value vtMaximum value, T is sensitive special
The total quantity of sign.
Step S206: according to the feature weight, using the clustering method based on inverse covariance matrix to described sensitive special
Collection is clustered, and cluster result is obtained;
The present embodiment selects the clustering method based on inverse covariance matrix to realize cluster, which can be based on Ma Er
Can the random field of husband accurately interpretable cluster result is found from time series data automatically, core concept is exactly not demarcate
In the case where, by being split simultaneously to data and each segmentation result being clustered, signal is divided into if reaching
The effect of dry possible state, such as mechanical arm accelerate state, mechanical arm even running state.Pass through drawing for these states
Point, the acquisition data that robot difference acts can be compared, it might even be possible to carry out the robot of the different speeds of service
Anomalies contrast expands the application range of consistency fault diagnosis, improves the convenience of fault diagnosis.
In addition, the clustering method based on inverse covariance matrix has bigger applicability, it can be used for practical plant produced
The fault detection of the robot of operation.In the production activity of factory, more robots are likely to have different program tasks, place
In different working conditions, the different exceptional value of analysis waveform will be unable to using traditional algorithm, and be based on inverse covariance square
The clustering method of battle array is not clustered to classify according to the state of run action by the way of sliding window with program, so
It can be adapted for bigger range.
Specifically the cluster process of the clustering method based on inverse covariance matrix is referred to existing scheme, and the present embodiment is not
Reinflated introduction.
Step S207: according to the cluster result, outlier and the corresponding operating status of the outlier are determined;
Specifically, matrix distance can be sought to the more of a sort eigenmatrixes of robot, if distance is more than normal value
Then it is considered outlier, determines that there is exception in this robot.
Step S208: according to the outlier and the operating status, the fault diagnosis result of the robot is determined.
In addition, as a preferred embodiment, the present embodiment can be according to the outlier and the operating status
Determine general fault type.Such as shake of the motor when low speed is run, it can be judged as outlier, fault type may be
The factors such as stator failure or rotor fault or mechanical wear, inertia variation, lubrication generate shake.If it is adding and subtracting
Occur abnormal shake during speed, then may be the failure of speed reducer.In short, according to cluster result, it can be determined that robot
Failure is the failure under which kind of operating status, such as accelerated motion, retarded motion, uniform motion, the last spy according to failure
The corresponding motion state of point, failure determines fault type in conjunction with expertise.
As it can be seen that a kind of robot failure diagnosis method provided in this embodiment, using uncompensation distance appraisal procedure and is based on
The clustering method of inverse covariance matrix realizes the failure diagnostic process of robot, at least has following advantages: independent of additional
Sensor, economical and efficient;Sensitivity based on sensitive features is that each sensitive features are assigned with corresponding feature weight;Base
Have interpretation in the cluster result that the clustering method of inverse covariance matrix obtains, is analyzed convenient for consequent malfunction;To multiaxis machine
The different axis of device people is weighted processing, because axis when executing different movements mainly is also different, according to what is executed
Movement selects different feature axis to make fault diagnosis effect more preferable;According to the outlier under different conditions, subsequent combination is supported
Expertise carries out accident analysis.
A kind of robot failure diagnosis device provided by the embodiments of the present application is introduced below, one kind described below
Robot failure diagnosis device can correspond to each other reference with a kind of above-described robot failure diagnosis method.
As shown in figure 3, the device includes:
Primitive character collection determining module 301: for acquiring the operation data of robot using as primitive character collection;
Sensitive features collection determining module 302: for concentrating screening sensitive features collection from the primitive character, and described in determination
The feature weight of sensitive features concentration sensitive features;
Cluster module 303: it is used for according to the feature weight, using the clustering method based on inverse covariance matrix to described
Sensitive features collection is clustered, and cluster result is obtained;
Fault diagnosis module 304: for determining the fault diagnosis result of the robot according to the cluster result.
As a kind of specific embodiment, the fault diagnosis module 304 includes:
Cluster cell: for determining outlier and the corresponding operating status of the outlier according to the cluster result;
Accident analysis unit: for determining the failure classes of the robot according to the outlier and the operating status
Type.
The robot failure diagnosis device of the present embodiment is for realizing robot failure diagnosis method above-mentioned, therefore the dress
The embodiment part of the visible robot failure diagnosis method hereinbefore of specific embodiment in setting, for example, primitive character collection
Determining module 301, sensitive features collection determining module 302, cluster module 303, fault diagnosis module 304, are respectively used in realization
State step S101 in robot failure diagnosis method, S102, S103, S104.So specific embodiment is referred to accordingly
Various pieces embodiment description, not reinflated introduction herein.
In addition, since the robot failure diagnosis device of the present embodiment is for realizing robot failure diagnosis side above-mentioned
Method, therefore its effect is corresponding with the effect of the above method, which is not described herein again.
In addition, present invention also provides a kind of robot failure diagnosis equipment, as shown in Figure 4, comprising:
Memory 401: for storing computer program;
Processor 402: for executing the computer program to realize a kind of robot failure diagnosis side as described above
The step of method.
Finally, the computer readable storage medium is for depositing this application provides a kind of computer readable storage medium
Computer program is stored up, for realizing a kind of robot failure diagnosis as described above when the computer program is executed by processor
The step of method.
The robot failure diagnosis equipment of the present embodiment, computer readable storage medium are former for realizing robot above-mentioned
Hinder diagnostic method, therefore the visible robot fault hereinbefore of the equipment, the specific embodiment of computer readable storage medium
The embodiment part of diagnostic method, and the effect of the two is corresponding with above method embodiment, which is not described herein again.
Each embodiment in this specification is described in a progressive manner, the highlights of each of the examples are with it is other
The difference of embodiment, same or similar part may refer to each other between each embodiment.For being filled disclosed in embodiment
For setting, since it is corresponded to the methods disclosed in the examples, so being described relatively simple, related place is referring to method part
Explanation.
The step of method described in conjunction with the examples disclosed in this document or algorithm, can directly be held with hardware, processor
The combination of capable software module or the two is implemented.Software module can be placed in random access memory (RAM), memory, read-only deposit
Reservoir (ROM), electrically programmable ROM, electrically erasable ROM, register, hard disk, moveable magnetic disc, CD-ROM or technology
In any other form of storage medium well known in field.
Above to a kind of robot failure diagnosis method provided herein, device, equipment and computer-readable storage
Medium is described in detail, and specific examples are used herein to illustrate the principle and implementation manner of the present application, with
The explanation of upper embodiment is merely used to help understand the present processes and its core concept;Meanwhile for the general of this field
Technical staff, according to the thought of the application, there will be changes in the specific implementation manner and application range, in conclusion
The contents of this specification should not be construed as limiting the present application.
Claims (10)
1. a kind of robot failure diagnosis method characterized by comprising
The operation data of robot is acquired using as primitive character collection;
Screening sensitive features collection is concentrated from the primitive character, and determines that the sensitive features concentrate the feature power of sensitive features
Weight;
According to the feature weight, the sensitive features collection is clustered using the clustering method based on inverse covariance matrix,
Obtain cluster result;
According to the cluster result, the fault diagnosis result of the robot is determined.
2. the method as described in claim 1, which is characterized in that it is described to concentrate screening sensitive features collection from the primitive character,
And determine that the sensitive features concentrate the feature weight of sensitive features, comprising:
Screening sensitive features collection is concentrated from the primitive character using uncompensation distance appraisal procedure;
According to Distance evaluation standard, determine that the sensitive features concentrate sensitivity of the sensitive features in fault diagnosis, to make
It is characterized weight.
3. the method as described in claim 1, which is characterized in that it is described according to the cluster result, determine the robot
Fault diagnosis result, comprising:
According to the cluster result, outlier and the corresponding operating status of the outlier are determined;
According to the outlier and the operating status, the fault type of the robot is determined.
4. the method as described in claim 1, which is characterized in that the operation data of the acquisition robot is using as primitive character
Collection, comprising:
The feedback signal for acquiring motor encoder in robot, using as primitive character collection.
5. method as claimed in claim 4, which is characterized in that the feedback signal includes following any one or more: being referred to
Enable position signal, feedback position signal, command speed signal, feedback speed signal, command acceleration signal, feedback acceleration letter
Number, instruction torque signals, feedback moment signal, error signal.
6. the method as described in claim 1-5 any one, which is characterized in that it is described according to the cluster result, determine institute
State the fault diagnosis result of robot, comprising:
According to the axis feedback moment of the multiple axis of the robot, the axis weight of each axis is determined;
According to the cluster result of multiple axis and corresponding axis weight, fault diagnosis result is determined.
7. a kind of robot failure diagnosis device characterized by comprising
Primitive character collection determining module: for acquiring the operation data of robot using as primitive character collection;
Sensitive features collection determining module: it for concentrating screening sensitive features collection from the primitive character, and determines described sensitive special
The feature weight of sensitive features in collection;
Cluster module: it is used for according to the feature weight, using the clustering method based on inverse covariance matrix to described sensitive special
Collection is clustered, and cluster result is obtained;
Fault diagnosis module: for determining the fault diagnosis result of the robot according to the cluster result.
8. device as claimed in claim 7, which is characterized in that the fault diagnosis module includes:
Cluster cell: for determining outlier and the corresponding operating status of the outlier according to the cluster result;
Accident analysis unit: for determining the fault type of the robot according to the outlier and the operating status.
9. a kind of robot failure diagnosis equipment characterized by comprising
Memory: for storing computer program;
Processor: a kind of robot as claimed in any one of claims 1 to 6 is realized for executing the computer program
The step of method for diagnosing faults.
10. a kind of computer readable storage medium, which is characterized in that the computer readable storage medium is for storing computer
Program, for realizing a kind of machine as claimed in any one of claims 1 to 6 when the computer program is executed by processor
The step of people's method for diagnosing faults.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910402385.7A CN110110803A (en) | 2019-05-15 | 2019-05-15 | A kind of robot failure diagnosis method, device and equipment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910402385.7A CN110110803A (en) | 2019-05-15 | 2019-05-15 | A kind of robot failure diagnosis method, device and equipment |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110110803A true CN110110803A (en) | 2019-08-09 |
Family
ID=67490162
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910402385.7A Pending CN110110803A (en) | 2019-05-15 | 2019-05-15 | A kind of robot failure diagnosis method, device and equipment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110110803A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111126822A (en) * | 2019-12-19 | 2020-05-08 | 佛山科学技术学院 | Industrial robot health assessment method and device and storage medium |
CN112180251A (en) * | 2020-08-25 | 2021-01-05 | 安徽华电宿州发电有限公司 | Motor fault diagnosis method based on non-convex and non-smooth optimization and graph model |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101509839A (en) * | 2009-03-12 | 2009-08-19 | 上海交通大学 | Cluster industrial robot failure diagnosis method based on outlier excavation |
CN106647286A (en) * | 2017-02-16 | 2017-05-10 | 吉林大学 | Early failure removal testing method for carrying industrial robot |
CN107272586A (en) * | 2016-04-08 | 2017-10-20 | 发那科株式会社 | Rote learning device, learning by rote, failure precognition apparatus and system |
CN107727395A (en) * | 2017-07-21 | 2018-02-23 | 中国矿业大学 | A kind of Method for Bearing Fault Diagnosis assessed based on full variation and uncompensation distance |
-
2019
- 2019-05-15 CN CN201910402385.7A patent/CN110110803A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101509839A (en) * | 2009-03-12 | 2009-08-19 | 上海交通大学 | Cluster industrial robot failure diagnosis method based on outlier excavation |
CN107272586A (en) * | 2016-04-08 | 2017-10-20 | 发那科株式会社 | Rote learning device, learning by rote, failure precognition apparatus and system |
CN106647286A (en) * | 2017-02-16 | 2017-05-10 | 吉林大学 | Early failure removal testing method for carrying industrial robot |
CN107727395A (en) * | 2017-07-21 | 2018-02-23 | 中国矿业大学 | A kind of Method for Bearing Fault Diagnosis assessed based on full variation and uncompensation distance |
Non-Patent Citations (3)
Title |
---|
HALLAC,D. ET AL.: "Toeplitz Inverse Covariance-based Clustering of Multivariate Time Series Data", 《WENTY-SEVENTH INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE 》 * |
刘鹏 等: "基于补偿距离评估技术与灰色关联分析的滚动轴承故障程度识别", 《机械传动》 * |
王秀青 等: "基于多传感器信息融合的机器人故障诊断", 《上海交通大学学报》 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111126822A (en) * | 2019-12-19 | 2020-05-08 | 佛山科学技术学院 | Industrial robot health assessment method and device and storage medium |
CN111126822B (en) * | 2019-12-19 | 2023-04-28 | 佛山科学技术学院 | Industrial robot health assessment method, device and storage medium |
CN112180251A (en) * | 2020-08-25 | 2021-01-05 | 安徽华电宿州发电有限公司 | Motor fault diagnosis method based on non-convex and non-smooth optimization and graph model |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105510038B (en) | Wind turbine generator fault monitoring method and device | |
US10503146B2 (en) | Control system, control device, and control method | |
CN106777622A (en) | The method and system of the electromechanical equipment on-line fault diagnosis based on artificial intelligence | |
CN110110803A (en) | A kind of robot failure diagnosis method, device and equipment | |
CN109765490A (en) | A kind of power battery fault detection method and system based on high dimensional data diagnosis | |
CN112765890B (en) | Dynamic domain adaptive network-based multi-working-condition rotating machine residual life prediction method | |
CN110018322B (en) | Rotating speed detection method and system based on deep learning | |
CN115600891B (en) | Big data analysis method and system applied to production monitoring of aqueous acrylic resin | |
CN114216640A (en) | Method, apparatus and medium for detecting fault status of industrial equipment | |
CN115683235B (en) | Industrial robot servo motor vibration fault detection method and device | |
CN116871978A (en) | Drilling tool state monitoring method based on multi-sensor fusion | |
CN105675321B (en) | A kind of equipment performance degeneration radar map determines method | |
CN114723187A (en) | Intelligent tower crane, fault prediction method of intelligent tower crane and storage medium | |
CN113211426B (en) | Robot fault diagnosis method and device, computer equipment and storage medium | |
CN114298110A (en) | Rolling bearing fault diagnosis method and system based on interpretable 1DCNN model | |
CN110774269A (en) | Robot drag teaching method and device, electronic device and storage medium | |
CN116482460A (en) | Fault diagnosis method for power grid equipment and related equipment | |
CN115712962A (en) | Industrial robot state monitoring method and system based on digital twinning | |
US11339763B2 (en) | Method for windmill farm monitoring | |
Kolar et al. | Identification of inability states of rotating machinery subsystems using industrial IoT and convolutional neural network–initial research | |
Meyer et al. | Reliable manufacturing of heavy copper wire bonds using online parameter adaptation | |
CN115685037A (en) | Fault detection method, device and medium of current sensor | |
WO2022036689A1 (en) | Method for fault diagnosis using a fault diagnosis model trained by virtual data | |
Pfaff et al. | Adaptive, predictive machine condition assessment for resilient digital solutions | |
CN111624977A (en) | Mechanical control device and method of capacitive touch screen equipment |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190809 |