CN106483929A - A kind of fault of numerical control machine tool cancellation scheme differentiates indicating means - Google Patents
A kind of fault of numerical control machine tool cancellation scheme differentiates indicating means Download PDFInfo
- Publication number
- CN106483929A CN106483929A CN201610906297.7A CN201610906297A CN106483929A CN 106483929 A CN106483929 A CN 106483929A CN 201610906297 A CN201610906297 A CN 201610906297A CN 106483929 A CN106483929 A CN 106483929A
- Authority
- CN
- China
- Prior art keywords
- text
- fault
- failure
- malfunction
- target
- 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
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B19/00—Programme-control systems
- G05B19/02—Programme-control systems electric
- G05B19/18—Numerical control [NC], i.e. automatically operating machines, in particular machine tools, e.g. in a manufacturing environment, so as to execute positioning, movement or co-ordinated operations by means of programme data in numerical form
- G05B19/406—Numerical control [NC], i.e. automatically operating machines, in particular machine tools, e.g. in a manufacturing environment, so as to execute positioning, movement or co-ordinated operations by means of programme data in numerical form characterised by monitoring or safety
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B2219/00—Program-control systems
- G05B2219/30—Nc systems
- G05B2219/32—Operator till task planning
- G05B2219/32224—Identify parameters with highest probability of failure
Landscapes
- Engineering & Computer Science (AREA)
- Human Computer Interaction (AREA)
- Manufacturing & Machinery (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides a kind of fault of numerical control machine tool cancellation scheme differentiates indicating means，Which is contrasted by splitting when coupling extracts each malfunction descriptor included in target string and improves by way of circulate coupling and extracts accuracy and comprehensive，And by the use of each malfunction descriptor for extracting as the follow-up identification basis to fault category，While also eliminating the contrast process that irrelevant fault category describes text，Reduce non-faulting state description word and irrelevant fault category describes impact and the interference that text is caused for fault category identification，Improve the identification accuracy and recognition efficiency that text is described to target faults classification，And using the corresponding relation on fault category，Text is described according to target faults classification and the Failure elimination scheme text of corresponding failure classification is extracted in order to indicate，Conveniently to Digit Control Machine Tool user or the corresponding Failure elimination suggestion of maintenance personal，Lifted to the Digit Control Machine Tool convenience that in addition field maintenance or long-distance service are assisted.
Description
Technical field
The invention belongs to mechanical fault diagnosis and computer intelligence technology of identification field, more particularly to a kind of Digit Control Machine Tool event
Barrier cancellation scheme differentiates indicating means.
Background technology
Digit Control Machine Tool is the letter of numerically-controlled machine tool (Computer numerical control machine tools)
Claim, be a kind of automated machine tool equipped with program control system.The control system logically can process with control coding or
Other symbolic instruction regulated procedures, and decoded, represented with the numeral of code, numerical control dress is input into by information carrier
Put.Various control signals are sent by numerical control device through calculation process, control the action of lathe, shape by drawing and chi
Very little, automatically part is processed.Digit Control Machine Tool preferably solves the part processing of complexity, precision, small lot, multi items
Problem, is a kind of flexible, dynamical automated machine tool, represents the developing direction of modern machine control technology, is a kind of
Typical electromechanical integrated product.
During the use of Digit Control Machine Tool, user is most concerned with how shifting to an earlier date when equipment breaks down sign signal
Diagnosed and how to be keeped in repair etc. in time upon a fault.But, today's numerical control lathe China application with
Safeguard that present situation also allows of no optimist, be mainly manifested in following aspects：
(1) substantial amounts of Digit Control Machine Tool, various in style, and the workload of plant maintenance and maintenance is big；
(2) Digit Control Machine Tool integrates mechanical, electrical, liquid, gas, with high content of technology, once break down, Digit Control Machine Tool user with
It is difficult to solve its all problems by means of the technical force of itself.
(3) Heuristics of maintenance personal is largely dependent upon to the diagnosis of Digit Control Machine Tool, and these Heuristicses
Typically exist only in the brain of maintenance personal, be unfavorable for the shared of diagnostic knowledge, while it also becomes restriction maintenance maintenance water
The flat bottleneck for improving further.
The maintenance of Digit Control Machine Tool at present is still by the way of traditional, once that is, Digit Control Machine Tool breaks down, it is necessary to rely on
In the engineer's on-site maintenance having wide experience, or the integrated producer of Digit Control Machine Tool appoints personnel to provide propping up for technical maintenance
Hold.So, the after-sale service personnel of one side Digit Control Machine Tool manufacturer have to run around all the time wears him out, and on the other hand, also extend fault and repair
The multiple stand-by period, delay the processing of Digit Control Machine Tool.
Therefore how to realize the suggestion for accurately identifying and providing Failure elimination to fault of numerical control machine tool, so as to as soon as possible by
Digit Control Machine Tool user and maintenance personal eliminate fault of numerical control machine tool, and becoming this area has problem to be solved.
Content of the invention
For the above-mentioned problems in the prior art, the invention provides a kind of fault of numerical control machine tool cancellation scheme differentiates
Indicating means, solves numerically controlled machine remote diagnostic software and can quick and precisely recognize fault, in order to corresponding in matching database
Write complete Failure elimination suggestion text, draw the Failure elimination scheme of Digit Control Machine Tool.
For achieving the above object, following technological means present invention employs：
A kind of fault of numerical control machine tool cancellation scheme differentiates indicating means, pre-builds basicvocabulary database and Failure elimination
Scheme database；Be stored with the basicvocabulary database variant malfunction descriptor set in advance, each
Malfunction descriptor is in order to state a kind of malfunction of Digit Control Machine Tool；It is stored with the Failure elimination scheme database
The each corresponding fault category of the variant fault category of Digit Control Machine Tool describes text and Failure elimination scheme text, each fault
Classification describes text in order to describe failure cause or/and the phenomenon of the failure information of its corresponding failure classification, each Failure elimination side
Case text is in order to state the Failure elimination scheme of its corresponding failure classification；Then, executed to fault of numerical control machine tool as follows
The differentiation of cancellation scheme is indicated：
A) phenomenon of the failure is provided typing operation is described, the phenomenon of the failure of typing is described text-string as Fault Identification
Target string；
B) each malfunction descriptor stored in basicvocabulary database is transferred, and it is right to carry out with target string respectively
Than coupling, so as to extract each malfunction descriptor for obtaining matching with target string, combination constitutes target keyword
Set；
C) transfer each fault category stored in Failure elimination scheme database and text is described, ask for each fault respectively
Classification describes the Similarity value of text and target critical set of words, will be maximum for Similarity value and meet presetting key Word similarity
The fault category of matching condition describes text and is judged to that target faults classification describes text；
D) fault category corresponding to text is described according to the target faults classification, from Failure elimination scheme database
The Failure elimination scheme text of corresponding failure classification is transferred, in order to carry out the instruction of fault of numerical control machine tool cancellation scheme.
Above-mentioned fault of numerical control machine tool cancellation scheme is differentiated in indicating means, as preferred concrete scheme, the step
B) it is specially：
B1) order counts symbol i=1, makes alternative keywords vectorTarget string is assigned to character string vector tg,
And transfer in basicvocabulary database store each malfunction descriptor be assigned to basicvocabulary vector t0So that transfer base
Each malfunction descriptor in plinth lexical data base is used as basicvocabulary vector t0In an element；
B2) basicvocabulary vector t is taken0In i-th element t0(i) and current character string vector tgCarry out contrast coupling；If
Element t0I () is included in current character string vector tgIn, then execution step b3)；Otherwise, direct execution step b4)；
B3) by element t0I malfunction descriptor that () represents is added to alternative keywords vector tsIn, and from current
Character string vector tgMiddle rejecting element t0I the malfunction descriptor represented by (), forms new character string vector tg, then hold
Row step b4)；
B4) judge current character string vector tgWhether the number of characters that is guaranteed replacement has been less than presetting number of characters lower threshold
k；If so, then direct execution step b7)；Otherwise, execution step b5)；
B5) judge whether the value of current count symbol i has reached basicvocabulary vector t0Included in number of elements the upper limit
Value；If so, then direct execution step b7)；Otherwise, execution step b6)；
B6) order counts the value of symbol i from Jia 1, is then back to step b2)；
B7) current alternative keywords vector t is calculatedsResolution K with target stringt：
Kt=Ks/Ktg；
Wherein, KsRepresent current alternative keywords vector tsWith the number of identical characters in target string, KtgRepresent
Character number in target string；Then judge the resolution KtValue whether be more than presetting resolution threshold k0；If
It is, then execution step b9)；Otherwise, execution step b8)；
B8) basicvocabulary vector t is adjusted according to presetting order regulation rule0Putting in order for middle element, makes and counting
Symbol i=1, makes alternative keywords vectorAgain target string is assigned to character string vector tg, it is then back to execute
Step b2)；
B9) by current alternative keywords vector tsIn each malfunction descriptor as with target string phase
The malfunction descriptor that joins, by current alternative keywords vector tsAs target critical set of words t.
In above-mentioned fault of numerical control machine tool cancellation scheme differentiation indicating means, preferably, step b7) in,
Resolution threshold k0Value be 0.5～0.6.
In above-mentioned fault of numerical control machine tool cancellation scheme differentiation indicating means, preferably, step b8) in,
Presetting order regulation rule is using the regulation rule of random reversed order.
Above-mentioned fault of numerical control machine tool cancellation scheme is differentiated in indicating means, as preferred concrete scheme, the step
C) it is specially：
C1) transfer each fault category stored in Failure elimination scheme database to describe text and be assigned to failure-description square
Battle array T so that each fault category in Failure elimination scheme database describes text as in the failure-description matrix T
Individual element；
C2) all elements in traversal failure-description matrix T, will not include step B in failure-description matrix T) gained target
In keyword set, the element of any malfunction descriptor is rejected, so as to obtain brief failure-description matrix T'；
C3) calculate the fault category in brief failure-description matrix T' represented by each element respectively and describe text and target
The Similarity value of keyword set；
C4) text is described fault category that is maximum for Similarity value and meeting presetting keyword similarity mode condition
It is judged to that target faults classification describes text.
In above-mentioned fault of numerical control machine tool cancellation scheme differentiation indicating means, specifically, step c3) in, for
Any one fault category describes text dk, itself and step B) gained target critical set of words t Similarity value SβCalculation
For：
Wherein, β represents regulation value, and Q represents that fault category describes text dkWith comprehensive degree of target critical set of words t, C
Represent that fault category describes text dkWith the multiplicity of target critical set of words t, and：
Wherein, NtRepresent the quantity of the malfunction descriptor included in target critical set of words t, NkRepresent fault category
Description text dkIn the quantity of malfunction descriptor that includes, NsRepresent that fault category describes text dkWith target critical word set
Close the quantity of the same fault state description word that t has；
qiRepresent the j-th malfunction descriptor t included in target critical set of words tjCorresponding weighted value, wjkRepresent
Fault category describes text dkWith the j-th malfunction descriptor t included in target critical set of words tjBetween pass contact
Number, and be expressed as：
NdRepresent that the fault category included in brief failure-description matrix T' describes the quantity of text, df (ti) represent brief
Include j-th malfunction descriptor t in target critical set of words t in failure-description matrix T'jFault category description text
This quantity, C (tj,dk) represent the malfunction descriptor tjText d is described in fault categorykIn single text multiplicity,
Computing formula is：
tf(tj,dk) represent j-th malfunction descriptor t in target critical set of words tjText is described in fault category
dkThe number of times of middle appearance,Represent the malfunction descriptor tjInclude in brief failure-description matrix T'
All fault categories describe the number of times occurred in text.
Above-mentioned fault of numerical control machine tool cancellation scheme is differentiated in indicating means, as preferred concrete scheme, described default
Fixed keyword similarity mode condition adopts following one of which：
1., similarity maximum principle：Directly take the maximum fault category of Similarity value and describe text and be judged to target faults
Classification describes text；
2., similarity threshold principle：Text is described for the maximum fault category of Similarity value, if its Similarity value is more than
Or equal to presetting similarity threshold, then it is judged to that target faults classification describes text；
3., the difference limen value principle of similarity minimax：Text is described for the maximum fault category of Similarity value, if its
The similarity difference described between text with the fault category of the Similarity value minimum for calculating gained is more than or equal to presetting
Similarity difference threshold, then describe text by the maximum fault category of the Similarity value and be judged to target faults classification description text
This；
4., spend comprehensively and multiplicity threshold value principle：Text is described for the maximum fault category of Similarity value, if itself and mesh
The value of comprehensive degree of mark keyword set is more than or equal to presetting comprehensive degree threshold value, and its weight with target critical set of words
The value of multiplicity is more than or equal to presetting multiplicity threshold value, then the maximum fault category of the Similarity value is described text judgement
Text is described for target faults classification.
In above-mentioned fault of numerical control machine tool cancellation scheme differentiation indicating means, preferably, the similarity threshold
In principle, presetting similarity threshold value is 0.7～0.8.
Compared to prior art, the present invention has the advantages that：
1st, the fault of numerical control machine tool cancellation scheme of the present invention is differentiated in indicating means, is extracted in target string in coupling and is wrapped
During each malfunction descriptor for containing, the inventive method employs the mode of circulation contrast coupling, mutual of sentence and sentence
Join and being mutually matched for word and word is converted into, meet the principle of maximum similarity, it is possible to increase the standard that malfunction descriptor is extracted
True property and comprehensive.
2nd, the fault of numerical control machine tool cancellation scheme of the present invention is differentiated in indicating means, be make use of and is mated from target string
The combined target critical set of words for constituting of each malfunction descriptor for obtaining is extracted as the follow-up identification to fault category
Basis, reduces primary fault phenomenon and describes non-faulting state description word present in text-string for fault category identification
The impact for causing and interference, while also eliminate irrelevant event using each malfunction descriptor in target critical set of words
Barrier classification describes the interference that text recognize to fault category, improve target faults classification is described the identification accuracy of text with
Recognition efficiency.
3rd, the fault of numerical control machine tool cancellation scheme of the present invention differentiates indicating means, using depositing in Failure elimination scheme database
The fault category of storage describes text and corresponding relation of the Failure elimination scheme text on fault category, can be according to target faults
Classification describes text and extracts the Failure elimination scheme text of corresponding failure classification in order to indicate, convenient to Digit Control Machine Tool user
Or the suggestion of maintenance personal's corresponding Failure elimination, lifted to the Digit Control Machine Tool facility that in addition field maintenance or long-distance service are assisted
Property.
Description of the drawings
Fig. 1 differentiates the FB(flow block) of indicating means for fault of numerical control machine tool cancellation scheme of the present invention.
Fig. 2 differentiates step B in indicating means for fault of numerical control machine tool cancellation scheme of the present invention) one kind be embodied as flow process
Figure.
Fig. 3 differentiates step C in indicating means for fault of numerical control machine tool cancellation scheme of the present invention) one kind be embodied as flow process
Figure.
Specific embodiment
The present invention proposes a kind of fault of numerical control machine tool cancellation scheme and differentiates indicating means, and the method is particularly suitable for row
Except the remote diagnosis system software of machine failure, in order to carry out the differentiation of fault type for the different faults situation of Digit Control Machine Tool
And Failure elimination scheme is indicated, to assist lathe user to exclude machine failure as soon as possible, shorten and cause to set because of equipment fault
The standby time that shuts down, reduce loss, simultaneously for also can be in order to reduce after-sale technical support for CNC Machine Tool Maintenance producer
Difficulty and maintenance cost, lifted the Digit Control Machine Tool competitiveness of product in market.
The fault of numerical control machine tool cancellation scheme of the present invention differentiates indicating means, need to pre-build basicvocabulary database and
Failure elimination scheme database；Be stored with the basicvocabulary database variant malfunction description set in advance
Word, each malfunction descriptor is in order to state a kind of malfunction of Digit Control Machine Tool；In the Failure elimination scheme database
In the variant fault category of the Digit Control Machine Tool each corresponding fault category that is stored with text and Failure elimination scheme text are described,
Each fault category describes text in order to describe failure cause or/and the phenomenon of the failure information of its corresponding failure classification, each event
Barrier cancellation scheme text is in order to state the Failure elimination scheme of its corresponding failure classification；Then, then can be according to as shown in Figure 1
Steps flow chart executes the differentiation to fault of numerical control machine tool cancellation scheme and indicates：
A) phenomenon of the failure is provided typing operation is described, the phenomenon of the failure of typing is described text-string as Fault Identification
Target string.
Phenomenon of the failure provided in the step describes typing operation, can also may be used in the form of words input is provided
To be converted to the typing form of Word message using offer voice typing, as long as typing and energy can be provided in prior art
Enough it is converted to phenomenon of the failure and describes the phenomenon of the failure of text-string and describes typing operation format.
B) each malfunction descriptor stored in basicvocabulary database is transferred, and it is right to carry out with target string respectively
Than coupling, so as to extract each malfunction descriptor for obtaining matching with target string, combination constitutes target keyword
Set.
The idiographic flow of the step is as shown in Figure 2：
B1) order counts symbol i=1, makes alternative keywords vectorTarget string is assigned to character string vector tg,
And transfer in basicvocabulary database store each malfunction descriptor be assigned to basicvocabulary vector t0So that transfer base
Each malfunction descriptor in plinth lexical data base is used as basicvocabulary vector t0In an element；
B2) basicvocabulary vector t is taken0In i-th element t0(i) and current character string vector tgCarry out contrast coupling；If
Element t0I () is included in current character string vector tgIn, then execution step b3)；Otherwise, direct execution step b4)；
B3) by element t0I malfunction descriptor that () represents is added to alternative keywords vector tsIn, and from current
Character string vector tgMiddle rejecting element t0I the malfunction descriptor represented by (), forms new character string vector tg, then hold
Row step b4)；
B4) judge current character string vector tgWhether the number of characters that is guaranteed replacement has been less than presetting number of characters lower threshold
k；If so, then direct execution step b7)；Otherwise, execution step b5)；
B5) judge whether the value of current count symbol i has reached basicvocabulary vector t0Included in number of elements the upper limit
Value；If so, then direct execution step b7)；Otherwise, execution step b6)；
B6) order counts the value of symbol i from Jia 1, is then back to step b2)；
B7) current alternative keywords vector t is calculatedsResolution K with target stringt：
Kt=Ks/Ktg；
Wherein, KsRepresent current alternative keywords vector tsWith the number of identical characters in target string, KtgRepresent
Character number in target string；Then judge the resolution KtValue whether be more than presetting resolution threshold k0；If
It is, then execution step b9)；Otherwise, execution step b8)；Here, resolution threshold k0Can value be preferably 0.5～0.6；
B8) basicvocabulary vector t is adjusted according to presetting order regulation rule0Putting in order for middle element, makes and counting
Symbol i=1, makes alternative keywords vectorAgain target string is assigned to character string vector tg, it is then back to execute
Step b2)；Here, presetting order regulation rule adopts the regulation rule of random reversed order, it is of course also possible to use
Other presetting order regulation rules；
B9) by current alternative keywords vector tsIn each malfunction descriptor as with target string phase
The malfunction descriptor that joins, by current alternative keywords vector tsAs target critical set of words t.
Can see, when coupling extracts each malfunction descriptor included in target string, the inventive method
The mode of circulation contrast coupling is employed, being mutually matched for sentence and sentence is converted into being mutually matched for word and word, satisfaction maximum phase
Principle like degree, it is possible to increase the accuracy and comprehensive that malfunction descriptor is extracted.
C) transfer each fault category stored in Failure elimination scheme database and text is described, ask for each fault respectively
Classification describes the Similarity value of text and target critical set of words, will be maximum for Similarity value and meet presetting key Word similarity
The fault category of matching condition describes text and is judged to that target faults classification describes text.
The idiographic flow of the step is as shown in Figure 3：
C1) transfer each fault category stored in Failure elimination scheme database to describe text and be assigned to failure-description square
Battle array T so that each fault category in Failure elimination scheme database describes text as in the failure-description matrix T
Individual element.
C2) all elements in traversal failure-description matrix T, will not include step B in failure-description matrix T) gained target
In keyword set, the element of any malfunction descriptor is rejected, so as to obtain brief failure-description matrix T'.
Reject the fault category not comprising any malfunction descriptor and text element is described, then it is unnecessary to reduce
Similarity computing, reduces Failure elimination scheme and differentiates the data processing amount for indicating；Meanwhile, text element is described not to fault category
Differentiation comprising malfunction descriptor, it is also possible to which which is zero with the Similarity value of target critical set of words to be considered as a kind of determination
Simple and easy method.
C3) calculate the fault category in brief failure-description matrix T' represented by each element respectively and describe text and target
The Similarity value of keyword set.
The matching of text and target critical set of words is described herein for measurement fault category, introduces three signs
Join the index of similarity：
A. spend comprehensively：The comprehensive of terminology match is weighed, reflects that keyword is closed to searched targets field in the document that is retrieved
The level of coverage of keyword；
B. multiplicity：Weigh the frequency that aiming field keyword repeats in the document that is retrieved；
C. similarity：Consider the important indicator of degree comprehensively and the two reflection text retrieval matching effects of multiplicity,
Weigh aiming field and the similitude being retrieved between document.
Text d is described for any one fault categoryk, itself and step B) gained target critical set of words t similarity
Value SβCalculation be：
Wherein, β represents regulation value, and Q represents that fault category describes text dkWith comprehensive degree of target critical set of words t, C
Represent that fault category describes text dkWith the multiplicity of target critical set of words t, and：
In the calculating of degree Q comprehensively, NtRepresent the quantity of the malfunction descriptor included in target critical set of words t, Nk
Represent that fault category describes text dkIn the quantity of malfunction descriptor that includes, NsRepresent that fault category describes text dkWith
The quantity of the same fault state description word that target critical set of words t has；
In the calculating of multiplicity C, qiRepresent the j-th malfunction descriptor t included in target critical set of words tjRight
The weighted value that answers, the setting of the weighted value can consider that corresponding failure state description word is true for the significance level of search target
Fixed；wjkRepresent that fault category describes text dkWith the j-th malfunction descriptor t included in target critical set of words tjBetween
Incidence coefficient, can be calculated using TFIDF weighted strategy, be expressed as：
NdRepresent that the fault category included in brief failure-description matrix T' describes the quantity of text, df (ti) represent brief
Include j-th malfunction descriptor t in target critical set of words t in failure-description matrix T'jFault category description text
This quantity, C (tj,dk) represent the malfunction descriptor tjText d is described in fault categorykIn single text multiplicity,
Computing formula is：
tf(tj,dk) represent j-th malfunction descriptor t in target critical set of words tjText is described in fault category
dkThe number of times of middle appearance,Represent the malfunction descriptor tjInclude in brief failure-description matrix T'
All fault categories describe the number of times occurred in text.
C4) text is described fault category that is maximum for Similarity value and meeting presetting keyword similarity mode condition
It is judged to that target faults classification describes text.
In the step, need first to find out the fault category maximum with the Similarity value of target critical set of words and text described,
Further judge that the fault category describes whether text meets presetting keyword similarity mode condition.It is being embodied as
When, presetting keyword similarity mode condition adopts following one of which：
1., similarity maximum principle：Directly take the maximum fault category of Similarity value and describe text and be judged to target faults
Classification describes text.The determinating treatment course that target faults classification can be caused to describe text using the maximum principle of similarity is more
Simply.
2., similarity threshold principle：Text is described for the maximum fault category of Similarity value, if its Similarity value is more than
Or equal to presetting similarity threshold, then it is judged to that target faults classification describes text.Here, presetting similarity threshold
Value can value be preferably 0.7～0.8.
3., the difference limen value principle of similarity minimax：Text is described for the maximum fault category of Similarity value, if its
The similarity difference described between text with the fault category of the Similarity value minimum for calculating gained is more than or equal to presetting
Similarity difference threshold, then describe text by the maximum fault category of the Similarity value and be judged to target faults classification description text
This.
4., spend comprehensively and multiplicity threshold value principle：Text is described for the maximum fault category of Similarity value, if itself and mesh
The value of comprehensive degree of mark keyword set is more than or equal to presetting comprehensive degree threshold value, and its weight with target critical set of words
The value of multiplicity is more than or equal to presetting multiplicity threshold value, then the maximum fault category of the Similarity value is described text judgement
Text is described for target faults classification.
Former using similarity threshold principle, the difference limen value principle of similarity minimax and comprehensive degree and multiplicity threshold value
Then, the fault category maximum because of Similarity value can be further avoided to describe the Similarity value of text and target critical set of words
It is not high enough to and causes the situation that draws fault cancellation scheme, it is to avoid indicates the Failure elimination scheme of mistake to Digit Control Machine Tool
Maintenance cause mislead.
Can see, make use of step B in this step) in from target string coupling extract each fault shape for obtaining
The combined target critical set of words for constituting of state descriptor reduces primary fault as the follow-up identification basis to fault category
Phenomenon describes impact and the interference that non-faulting state description word present in text-string is caused for fault category identification, with
When also using each malfunction descriptor in target critical set of words eliminate irrelevant fault category describe text to therefore
The interference of barrier classification identification, improves the identification accuracy and recognition efficiency for describing text to target faults classification.
D) fault category corresponding to text is described according to the target faults classification, from Failure elimination scheme database
The Failure elimination scheme text of corresponding failure classification is transferred, in order to carry out the instruction of fault of numerical control machine tool cancellation scheme.
Text is described with Failure elimination scheme text in event using the fault category stored in Failure elimination scheme database
Corresponding relation in barrier classification, can describe text according to target faults classification and extract corresponding Failure elimination scheme text
In order to indicate, conveniently to Digit Control Machine Tool user or the corresponding Failure elimination suggestion of maintenance personal, lifting is showed to Digit Control Machine Tool
The convenience that field maintenance or long-distance service are assisted.
Certainly, indicating means is differentiated using the above-mentioned fault of numerical control machine tool cancellation scheme of the present invention, there is also and cannot have
The possibility that the fault of numerical control machine tool cancellation scheme of body is indicated.
For example, in above-mentioned steps b1)～b9) processing procedure in, if adjustment basicvocabulary vector t0The row of middle element
The adjustment number of times of row order reaches presetting adjustment number of times limit value, but alternative keywords vector tsIdentification with target string
Degree KtStill fail more than presetting resolution threshold k0, then can design and jump out step B) circulation, execution step E)：
E) prompting fails to show that the phenomenon of the failure of current typing describes the fault of numerical control machine tool corresponding to text-string and disappears
Except scheme.
Now can then can be shown that the malfunction descriptor stored in basicvocabulary database is also insufficient to comprehensively, fail root
The malfunction descriptor that text-string extracts preparation enough is described according to the phenomenon of the failure of typing, it is thus possible to require supplementation with
With adjustment basicvocabulary database；Or, it is also possible to show that the phenomenon of the failure of current typing describes text-string for fault
The description of phenomenon is not accurate enough, needs to re-type phenomenon of the failure description.
Again for example, in above-mentioned steps c1)～c4) processing procedure in, if the maximum fault category description of Similarity value
Text cannot meet presetting keyword similarity mode condition, then can also design and jump out step C) circulation, in execution
The step of stating E).The fault category description of the dependent failure classification that in Failure elimination scheme database store now can then can be shown that
Text and Failure elimination scheme text comprehensive not enough, the phenomenon of the failure for failing recorded typing is described corresponding to text-string
Fault type, it is thus possible to require supplementation with and adjust Failure elimination scheme database.
But in general, the fault of numerical control machine tool cancellation scheme of the present invention differentiates indicating means, can be soft by computer
Part runs, and can carry out remote technology support by using software, instead of " flow of personnel " with " data flowing ", reduce product
The number of times of going on business of product maintenance personal, reduces after-sale technical support expense, contributes to strengthening the market competition of Digit Control Machine Tool product
Power, and Digit Control Machine Tool user or field service personnel can be helped in time to find failure cause, aid in fixing a breakdown, effectively
The equipment downtime that causes because of equipment fault of shortening, so as to be that Digit Control Machine Tool user reduces breakdown loss as far as possible.
More specific detail is carried out to the present invention below by embodiment.
Embodiment：
As a example by the present embodiment describes " knife rest breaks down, and tool-post structure has a slight fever, and operates abnormal " with phenomenon of the failure, more
Specific description fault of numerical control machine tool cancellation scheme of the present invention differentiates that indicating means differentiates in actual fault of numerical control machine tool and indicates
During application.
Before implementing fault of numerical control machine tool and differentiating and indicate, need to pre-build basicvocabulary database and Failure elimination side
Case database；Be stored with the basicvocabulary database variant malfunction descriptor set in advance, each event
Barrier state description word is in order to state a kind of malfunction of Digit Control Machine Tool；Be stored with the Failure elimination scheme database number
The each corresponding fault category of the variant fault category of control lathe describes text and Failure elimination scheme text, each failure classes
Text is not described in order to describe failure cause or/and the phenomenon of the failure information of its corresponding failure classification, each Failure elimination scheme
Text is in order to state the Failure elimination scheme of its corresponding failure classification；Then, execute as follows and fault of numerical control machine tool is disappeared
Except the differentiation of scheme is indicated：
A) phenomenon of the failure is provided typing operation is described, the phenomenon of the failure of typing is described text-string as Fault Identification
Target string.
In the present embodiment, the phenomenon of the failure of the step institute typing describes text-string for " knife rest breaks down, tool-post structure
Have a slight fever, operate abnormal ", using it as target string.
B) each malfunction descriptor stored in basicvocabulary database is transferred, and it is right to carry out with target string respectively
Than coupling, so as to extract each malfunction descriptor for obtaining matching with target string, combination constitutes target keyword
Set.The step is specially：
B1) order counts symbol i=1, makes alternative keywords vectorTarget string is assigned to character string vector tg,
And transfer in basicvocabulary database store each malfunction descriptor be assigned to basicvocabulary vector t0So that transfer base
Each malfunction descriptor in plinth lexical data base is used as basicvocabulary vector t0In an element；
B2) basicvocabulary vector t is taken0In i-th element t0(i) and current character string vector tgCarry out contrast coupling；If
Element t0I () is included in current character string vector tgIn, then execution step b3)；Otherwise, direct execution step b4)；
B3) by element t0I malfunction descriptor that () represents is added to alternative keywords vector tsIn, and from current
Character string vector tgMiddle rejecting element t0I the malfunction descriptor represented by (), forms new character string vector tg, then hold
Row step b4)；
B4) judge current character string vector tgWhether the number of characters that is guaranteed replacement has been less than presetting number of characters lower threshold
k；If so, then direct execution step b7)；Otherwise, execution step b5)；
B5) judge whether the value of current count symbol i has reached basicvocabulary vector t0Included in number of elements the upper limit
Value；If so, then direct execution step b7)；Otherwise, execution step b6)；
B6) order counts the value of symbol i from Jia 1, is then back to step b2)；
B7) current alternative keywords vector t is calculatedsResolution K with target stringt：
Kt=Ks/Ktg；
Wherein, KsRepresent current alternative keywords vector tsWith the number of identical characters in target string, KtgRepresent
Character number in target string；Then judge the resolution KtValue whether be more than presetting resolution threshold k0；If
It is, then execution step b9)；Otherwise, execution step b8)；
B8) basicvocabulary vector t is adjusted according to presetting order regulation rule0Putting in order for middle element, makes and counting
Symbol i=1, makes alternative keywords vectorAgain target string is assigned to character string vector tg, it is then back to execute
Step b2)；
B9) by current alternative keywords vector tsIn each malfunction descriptor as with target string phase
The malfunction descriptor that joins, by current alternative keywords vector tsAs target critical set of words t.
In the present embodiment, presetting resolution threshold k0For 0.6；Execute to step b7) when, from target string
Join each malfunction descriptor for extracting and be respectively " knife rest ", " fault ", " tool-post structure ", " heating ", " operating ", " not just
Often ", the alternative keywords vector t of thus combined compositionsFor knife rest, and fault, tool-post structure, heating, operating, abnormal, bag
Containing 14 characters；And target string original in the present embodiment " knife rest breaks down, and tool-post structure has a slight fever, and operating is not just
Often " include 18 characters；Therefore resolution K is calculatedt=0.778, more than presetting resolution threshold k0, cause determination
Knife rest, and fault, tool-post structure, heating, operating, abnormal as target critical set of words t.
C) transfer each fault category stored in Failure elimination scheme database and text is described, ask for each fault respectively
Classification describes the Similarity value of text and target critical set of words, will be maximum for Similarity value and meet presetting key Word similarity
The fault category of matching condition describes text and is judged to that target faults classification describes text.
In the present embodiment, presetting keyword similarity mode condition employs similarity threshold principle, and presetting
Similarity threshold value be 0.7, judge that " tool-post structure heating " fault category is corresponding former thus according to target critical set of words
Barrier classification describes text and describes text for target faults classification.
D) fault category corresponding to text is described according to the target faults classification, from Failure elimination scheme database
The Failure elimination scheme text of corresponding failure classification is transferred, in order to carry out the instruction of fault of numerical control machine tool cancellation scheme.
In the present embodiment, according to " tool-post structure heating " fault category, the fault category is transferred corresponding " based on tool-post structure
The knife rest Failure elimination scheme of heating " carries out the instruction of fault of numerical control machine tool cancellation scheme.
In sum, it can be seen that the fault of numerical control machine tool cancellation scheme of the present invention differentiates indicating means, extract in coupling
Contrasted by splitting during each malfunction descriptor included in target string and extraction is improve by way of circulating coupling
Accuracy and comprehensive, and by the use of each malfunction descriptor for extracting as the follow-up identification basis to fault category,
Process while also eliminating irrelevant fault category and describing the contrast of text, reduce non-faulting state description word and irrelevant
Fault category describes impact and the interference that text is caused for fault category identification, improves and describes text to target faults classification
Identification accuracy and recognition efficiency, and using the corresponding relation on fault category, text is described according to target faults classification and
The Failure elimination scheme text of corresponding failure classification is extracted in order to indicate, convenient corresponding to Digit Control Machine Tool user or maintenance personal
Failure elimination suggestion, lifted to the Digit Control Machine Tool convenience that in addition field maintenance or long-distance service are assisted.
Finally illustrate, above example is only unrestricted in order to technical scheme to be described, although with reference to relatively
Good embodiment has been described in detail to the present invention, it will be understood by those within the art that, can be to the skill of the present invention
Art scheme is modified or equivalent, the objective without deviating from technical solution of the present invention and scope, and which all should be covered at this
In the middle of the right of invention.
Claims (8)
1. a kind of fault of numerical control machine tool cancellation scheme differentiates indicating means, it is characterised in that pre-build basicvocabulary database
With Failure elimination scheme database；The variant malfunction set in advance that is stored with the basicvocabulary database is retouched
Predicate, each malfunction descriptor is in order to state a kind of malfunction of Digit Control Machine Tool；In the Failure elimination protocol
The each corresponding fault category of the variant fault category of Digit Control Machine Tool that is stored with storehouse describes text and Failure elimination scheme text
This, each fault category describes text in order to describe failure cause or/and the phenomenon of the failure information of its corresponding failure classification, each
Failure elimination scheme text is in order to state the Failure elimination scheme of its corresponding failure classification；Then, logarithm is executed as follows
The differentiation of control machine failure cancellation scheme is indicated：
A) phenomenon of the failure is provided typing operation is described, the phenomenon of the failure of typing is described text-string as the mesh of Fault Identification
Mark character string；
B each malfunction descriptor stored in basicvocabulary database) is transferred, carries out contrast respectively with target string
Join, so as to extract each malfunction descriptor for obtaining matching with target string, combination constitutes target critical set of words；
C) transfer each fault category stored in Failure elimination scheme database and text is described, ask for each fault category respectively
The Similarity value of description text and target critical set of words, will be maximum for Similarity value and meet presetting keyword similarity mode
The fault category of condition describes text and is judged to that target faults classification describes text；
D) fault category corresponding to text is described according to the target faults classification, transfers from Failure elimination scheme database
The Failure elimination scheme text of corresponding failure classification, in order to carry out the instruction of fault of numerical control machine tool cancellation scheme.
2. fault of numerical control machine tool cancellation scheme according to claim 1 differentiates indicating means, it is characterised in that the step
B) it is specially：
B1) order counts symbol i=1, makes alternative keywords vectorTarget string is assigned to character string vector tg, and adjust
Take each malfunction descriptor stored in basicvocabulary database and be assigned to basicvocabulary vector t0So that transfer basic word
Each the malfunction descriptor converged in database is used as basicvocabulary vector t0In an element；
B2) basicvocabulary vector t is taken0In i-th element t0(i) and current character string vector tgCarry out contrast coupling；If element
t0I () is included in current character string vector tgIn, then execution step b3)；Otherwise, direct execution step b4)；
B3) by element t0I malfunction descriptor that () represents is added to alternative keywords vector tsIn, and from current character
String vector tgMiddle rejecting element t0I the malfunction descriptor represented by (), forms new character string vector tg, then execute step
Rapid b4)；
B4) judge current character string vector tgWhether the number of characters that is guaranteed replacement has been less than presetting number of characters lower threshold k；If
It is, then direct execution step b7)；Otherwise, execution step b5)；
B5) judge whether the value of current count symbol i has reached basicvocabulary vector t0Included in number of elements higher limit；If
It is, then direct execution step b7)；Otherwise, execution step b6)；
B6) order counts the value of symbol i from Jia 1, is then back to step b2)；
B7) current alternative keywords vector t is calculatedsResolution K with target stringt：
Kt=Ks/Ktg；
Wherein, KsRepresent current alternative keywords vector tsWith the number of identical characters in target string, KtgRepresent target word
Character number in symbol string；Then judge the resolution KtValue whether be more than presetting resolution threshold k0；If so, then
Execution step b9)；Otherwise, execution step b8)；
B8) basicvocabulary vector t is adjusted according to presetting order regulation rule0Putting in order for middle element, makes and counts symbol i=
1, make alternative keywords vectorAgain target string is assigned to character string vector tg, it is then back to execution step
b2)；
B9) by current alternative keywords vector tsIn each malfunction descriptor as matching with target string
Malfunction descriptor, by current alternative keywords vector tsAs target critical set of words t.
3. fault of numerical control machine tool cancellation scheme according to claim 2 differentiates indicating means, it is characterised in that the step
B7 in), resolution threshold k0Value be 0.5～0.6.
4. fault of numerical control machine tool cancellation scheme according to claim 2 differentiates indicating means, it is characterised in that the step
B8, in), presetting order regulation rule adopts the regulation rule of random reversed order.
5. fault of numerical control machine tool cancellation scheme according to claim 1 differentiates indicating means, it is characterised in that the step
C) it is specially：
C1) transfer each fault category stored in Failure elimination scheme database to describe text and be assigned to failure-description matrix T,
So that each fault category in Failure elimination scheme database describes text as a unit in the failure-description matrix T
Element；
C2) all elements in traversal failure-description matrix T, will not include step B in failure-description matrix T) gained target critical
In set of words, the element of any malfunction descriptor is rejected, so as to obtain brief failure-description matrix T'；
C3) calculate the fault category in brief failure-description matrix T' represented by each element respectively and describe text and target critical
The Similarity value of set of words；
C4) text judgement is described fault category that is maximum for Similarity value and meeting presetting keyword similarity mode condition
Text is described for target faults classification.
6. fault of numerical control machine tool cancellation scheme according to claim 5 differentiates indicating means, it is characterised in that the step
C3, in), text d is described for any one fault categoryk, itself and step B) gained target critical set of words t Similarity value Sβ
Calculation be：
Wherein, β represents regulation value, and Q represents that fault category describes text dkWith comprehensive degree of target critical set of words t, C represents
Fault category describes text dkWith the multiplicity of target critical set of words t, and：
Wherein, NtRepresent the quantity of the malfunction descriptor included in target critical set of words t, NkRepresent fault category description
Text dkIn the quantity of malfunction descriptor that includes, NsRepresent that fault category describes text dkHave with target critical set of words t
The quantity of some same fault state description words；
qiRepresent the j-th malfunction descriptor t included in target critical set of words tjCorresponding weighted value, wjkRepresent fault
Classification describes text dkWith the j-th malfunction descriptor t included in target critical set of words tjBetween incidence coefficient, and
It is expressed as：
NdRepresent that the fault category included in brief failure-description matrix T' describes the quantity of text, df (ti) represent brief fault
Include j-th malfunction descriptor t in target critical set of words t in Description Matrix T'jFault category text is described
Quantity, C (tj,dk) represent the malfunction descriptor tjText d is described in fault categorykIn single text multiplicity, calculate
Formula is：
tf(tj,dk) represent j-th malfunction descriptor t in target critical set of words tjText d is described in fault categorykIn go out
Existing number of times,Represent the malfunction descriptor tjThe all events included in brief failure-description matrix T'
Barrier classification describes the number of times occurred in text.
7. fault of numerical control machine tool cancellation scheme according to claim 5 differentiates indicating means, it is characterised in that described default
Fixed keyword similarity mode condition adopts following one of which：
1., similarity maximum principle：Directly take the maximum fault category of Similarity value and describe text and be judged to target faults classification
Description text；
2., similarity threshold principle：Text is described for the maximum fault category of Similarity value, if its Similarity value is more than or waits
In presetting similarity threshold, then it is judged to that target faults classification describes text；
3., the difference limen value principle of similarity minimax：Text is described for the maximum fault category of Similarity value, if itself and meter
Calculate similarity difference that the minimum fault category of Similarity value of gained described between text similar more than or equal to presetting
Degree difference threshold, then describe text by the maximum fault category of the Similarity value and be judged to that target faults classification describes text；
4., spend comprehensively and multiplicity threshold value principle：Text is described for the maximum fault category of Similarity value, if which is closed with target
The value of comprehensive degree of keyword set is more than or equal to presetting comprehensive degree threshold value, and its multiplicity with target critical set of words
Value be more than or equal to presetting multiplicity threshold value, then the maximum fault category of the Similarity value is described text and is judged to mesh
Mark fault category describes text.
8. fault of numerical control machine tool cancellation scheme according to claim 5 differentiates indicating means, it is characterised in that described similar
In degree threshold value principle, presetting similarity threshold value is 0.7～0.8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610906297.7A CN106483929A (en) | 2016-10-18 | 2016-10-18 | A kind of fault of numerical control machine tool cancellation scheme differentiates indicating means |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610906297.7A CN106483929A (en) | 2016-10-18 | 2016-10-18 | A kind of fault of numerical control machine tool cancellation scheme differentiates indicating means |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106483929A true CN106483929A (en) | 2017-03-08 |
Family
ID=58270073
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610906297.7A Pending CN106483929A (en) | 2016-10-18 | 2016-10-18 | A kind of fault of numerical control machine tool cancellation scheme differentiates indicating means |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106483929A (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106528723A (en) * | 2016-10-27 | 2017-03-22 | 重庆大学 | Fault tree-based numerical control machine tool fault removal scheme judgment indication method |
| CN107291044A (en) * | 2017-06-23 | 2017-10-24 | 深圳市盛路物联通讯技术有限公司 | The intelligent control method and system of a kind of lathe running status |
| CN108416534A (en) * | 2018-03-27 | 2018-08-17 | 上海复旦后勤服务发展有限公司 | Common equipment operation management system |
| CN108470218A (en) * | 2018-03-27 | 2018-08-31 | 上海复旦后勤服务发展有限公司 | Public utility operation management system |
| CN109597392A (en) * | 2017-09-30 | 2019-04-09 | 西门子公司 | Facilitate the method, apparatus and equipment and machine readable media of fault diagnosis |
| CN109727138A (en) * | 2018-12-29 | 2019-05-07 | 航天信息股份有限公司 | Credential match method and system based on confidence level |
| CN110287301A (en) * | 2019-06-27 | 2019-09-27 | 上海华兴数字科技有限公司 | Failure consultation method, apparatus and system |
| CN110309885A (en) * | 2019-07-05 | 2019-10-08 | 黑龙江电力调度实业有限公司 | Computer room state judging method based on big data |
| CN110837247A (en) * | 2018-08-17 | 2020-02-25 | 智能云科信息科技有限公司 | Machine tool performance evaluation method, system, comprehensive system and cloud platform based on machine tool data |
| CN110969222A (en) * | 2018-09-29 | 2020-04-07 | 西门子股份公司 | Information providing method and system |
| CN111580495A (en) * | 2020-04-29 | 2020-08-25 | 北京绪水互联科技有限公司 | Remote fault processing method, device and system for medical instrument |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102252709A (en) * | 2011-04-22 | 2011-11-23 | 上海海洋大学 | Method for diagnosing faults of non-electricity measurement system |
| CN102819239A (en) * | 2011-06-08 | 2012-12-12 | 同济大学 | Intelligent fault diagnosis method of numerical control machine tool |
| CN103149878A (en) * | 2011-12-06 | 2013-06-12 | 中国科学院沈阳计算技术研究所有限公司 | Self-adaptive learning system of numerical control machine fault diagnosis system in multi-agent structure |
-
2016
- 2016-10-18 CN CN201610906297.7A patent/CN106483929A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102252709A (en) * | 2011-04-22 | 2011-11-23 | 上海海洋大学 | Method for diagnosing faults of non-electricity measurement system |
| CN102819239A (en) * | 2011-06-08 | 2012-12-12 | 同济大学 | Intelligent fault diagnosis method of numerical control machine tool |
| CN103149878A (en) * | 2011-12-06 | 2013-06-12 | 中国科学院沈阳计算技术研究所有限公司 | Self-adaptive learning system of numerical control machine fault diagnosis system in multi-agent structure |
Non-Patent Citations (5)
| Title |
|---|
| KOLODNER JL: "An introduction to case-based reasoning", 《ARTIFICIAL INTELLIGENCE REVIEW》 * |
| 李荣祖: "数控机床远程故障诊断技术研究与应用", 《中国优秀硕士学位论文全文数据库 工程科技I辑》 * |
| 王小姣: "FANUC数控机床故障智能诊断***研究", 《中国优秀硕士学位论文全文数据库》 * |
| 程德蓉: "案例推理在数控机床故障诊断学习***中的应用", 《机床与液压》 * |
| 陈永强: "基于CBR技术的数控机床故障诊断专家***", 《西昌学院院报 自然科学版》 * |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106528723A (en) * | 2016-10-27 | 2017-03-22 | 重庆大学 | Fault tree-based numerical control machine tool fault removal scheme judgment indication method |
| CN107291044A (en) * | 2017-06-23 | 2017-10-24 | 深圳市盛路物联通讯技术有限公司 | The intelligent control method and system of a kind of lathe running status |
| CN109597392A (en) * | 2017-09-30 | 2019-04-09 | 西门子公司 | Facilitate the method, apparatus and equipment and machine readable media of fault diagnosis |
| CN108416534A (en) * | 2018-03-27 | 2018-08-17 | 上海复旦后勤服务发展有限公司 | Common equipment operation management system |
| CN108470218A (en) * | 2018-03-27 | 2018-08-31 | 上海复旦后勤服务发展有限公司 | Public utility operation management system |
| CN110837247A (en) * | 2018-08-17 | 2020-02-25 | 智能云科信息科技有限公司 | Machine tool performance evaluation method, system, comprehensive system and cloud platform based on machine tool data |
| CN110969222A (en) * | 2018-09-29 | 2020-04-07 | 西门子股份公司 | Information providing method and system |
| CN109727138A (en) * | 2018-12-29 | 2019-05-07 | 航天信息股份有限公司 | Credential match method and system based on confidence level |
| CN109727138B (en) * | 2018-12-29 | 2021-03-30 | 航天信息股份有限公司 | Confidence-based certificate matching method and system |
| CN110287301A (en) * | 2019-06-27 | 2019-09-27 | 上海华兴数字科技有限公司 | Failure consultation method, apparatus and system |
| CN110309885A (en) * | 2019-07-05 | 2019-10-08 | 黑龙江电力调度实业有限公司 | Computer room state judging method based on big data |
| CN111580495A (en) * | 2020-04-29 | 2020-08-25 | 北京绪水互联科技有限公司 | Remote fault processing method, device and system for medical instrument |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106483929A (en) | A kind of fault of numerical control machine tool cancellation scheme differentiates indicating means | |
| CN104361127B (en) | The multilingual quick constructive method of question and answer interface based on domain body and template logic | |
| US20150081277A1 (en) | System and Method for Automatically Classifying Text using Discourse Analysis | |
| US20200034689A1 (en) | A method for retrieving a recommendation from a knowledge database of a ticketing system | |
| CN109948808A (en) | The banking process in substation equipment fault case library, fault diagnosis method and system | |
| CN110175324B (en) | Power grid operation instruction verification method and system based on data mining | |
| CN106528723A (en) | Fault tree-based numerical control machine tool fault removal scheme judgment indication method | |
| CN111552804B (en) | Knowledge graph construction method of power grid fault handling plan | |
| Suntwal et al. | On the importance of delexicalization for fact verification | |
| Tohidi et al. | Optimizing Persian Multi-objective Question Answering System | |
| Guo et al. | A hybrid process monitoring and fault diagnosis approach for chemical plants | |
| CN114492926A (en) | Method and system for analyzing and predicting potential safety hazard text of coal mine | |
| Bokinsky et al. | Application of natural language processing techniques to marine V-22 maintenance data for populating a CBM-oriented database | |
| Ferreira et al. | Evaluating human-machine translation with attention mechanisms for industry 4.0 environment SQL-based systems | |
| WO2022018104A1 (en) | System and method for improved spare part search for maintenance services using topic modelling | |
| Farzana et al. | Knowledge graph-enhanced neural query rewriting | |
| JP2015165348A (en) | Operation support system, operation support system control method, and operation support system control program | |
| US11295039B2 (en) | Support system and method | |
| Hasan et al. | Pattern-matching based for Arabic question answering: a challenge perspective | |
| Wan et al. | Evaluation model of power operation and maintenance based on text emotion analysis | |
| Gladkoff et al. | Predicting Perfect Quality Segments in MT Output with Fine-Tuned OpenAI LLM: Is it possible to capture editing distance patterns from historical data? | |
| CN112801537B (en) | Non-stop accident analysis method, system, medium and equipment for power generation enterprise | |
| Mandal et al. | Adaptive Voting in Multiple Classifier Systems for Word Level Language Identification. | |
| CN114461735A (en) | Industrial and mining data classification method and device and computer equipment | |
| CN113468882A (en) | Method for identifying similar spare parts |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | 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: 20170308 |