CN109685348A - A kind of power equipment typical mode of operation determines method - Google Patents
A kind of power equipment typical mode of operation determines method Download PDFInfo
- Publication number
- CN109685348A CN109685348A CN201811536027.7A CN201811536027A CN109685348A CN 109685348 A CN109685348 A CN 109685348A CN 201811536027 A CN201811536027 A CN 201811536027A CN 109685348 A CN109685348 A CN 109685348A
- Authority
- CN
- China
- Prior art keywords
- vector
- state vector
- state
- power equipment
- typicalness
- 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
- 238000000034 method Methods 0.000 title claims abstract description 69
- 239000013598 vector Substances 0.000 claims abstract description 233
- 238000011156 evaluation Methods 0.000 claims abstract description 25
- 238000011161 development Methods 0.000 claims abstract description 16
- 239000000284 extract Substances 0.000 claims description 8
- 238000000605 extraction Methods 0.000 claims description 4
- 238000005070 sampling Methods 0.000 claims description 3
- 241000208340 Araliaceae Species 0.000 claims description 2
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 claims description 2
- 235000003140 Panax quinquefolius Nutrition 0.000 claims description 2
- 235000008434 ginseng Nutrition 0.000 claims description 2
- 230000008569 process Effects 0.000 description 5
- 230000000694 effects Effects 0.000 description 3
- 238000007689 inspection Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000004804 winding Methods 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000004888 barrier function Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 238000004587 chromatography analysis Methods 0.000 description 1
- 230000002354 daily effect Effects 0.000 description 1
- 238000006731 degradation reaction Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 238000003064 k means clustering Methods 0.000 description 1
- 238000012886 linear function Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000002203 pretreatment Methods 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/06—Electricity, gas or water supply
Abstract
The embodiment of the present application discloses a kind of power equipment typical mode of operation and determines method, comprising: according to the preset state vector space, the state vector of the corresponding preset state vector space is extracted from the operation data of similar power equipment；Parameter of the state vector extracted in each dimension is normalized；According to default classification number, to treated, state vector is clustered, and determines cluster result；For every one kind in cluster result, typicalness vector of the nearest state vector of distance-like center vector as current class is extracted, the corresponding operational mode of typicalness vector is the typical mode of operation of power equipment；Solves not representative enough the technical problem of transformer typicalness that existing method is determined.The embodiment of the present application also provides a kind of power equipment present mode of operation evaluation method and a kind of state prediction of the development trend methods of power equipment.
Description
Technical field
This application involves equipment O&M technical field more particularly to a kind of power equipment typical mode of operation determine method,
A kind of state prediction of the development trend method of power equipment present mode of operation evaluation method and a kind of power equipment.
Background technique
Core equipment of the large-scale power transformer as electric system, on-going commitment the transformation and transmitting function of electric energy,
Its healthy and stable operation is to guarantee one of the deciding factor of power system security.
As gradually the concern and attention by each field, Utilities Electric Co. set up the big of oneself to big data technology one after another
Data platform, the data barrier between the equipment of big physical extent are also constantly weakened.The development of big data is so that transformer allusion quotation
The extraction of type operational mode is more representative, helps to formulate reasonable power equipment O&M strategy.
In the existing appraisal procedure to running state of transformer, the transformer typicalness provided is that subjectivity is set
It is fixed, some intermediate state with general character during state evolution are had ignored, cause the transformer typicalness determined inadequate
It is representative.
Summary of the invention
The embodiment of the present application provides a kind of power equipment typical mode of operation and determines method, and it is true to solve existing method
Not representative enough the technical problem of the transformer typicalness made.The embodiment of the present application also provides a kind of power equipments to work as
A kind of state prediction of the development trend method of preceding operational mode evaluation method and power equipment.
In view of this, the application first aspect, which provides a kind of power equipment typical mode of operation, determines method, comprising:
According to the preset state vector space, the corresponding preset shape is extracted from the operation data of similar power equipment
The state vector of state vector space；
Parameter of the state vector extracted in each dimension is normalized；
According to default classification number, to treated, the state vector is clustered, and determines cluster result；
For every one kind in cluster result, allusion quotation of the nearest state vector of distance-like center vector as current class is extracted
Type state vector, the corresponding operational mode of the typicalness vector are the typical mode of operation of the power equipment；
Wherein, parameter of the class center vector in each dimension be, where the class center vector in class each state to
Measure the median of the parameter of corresponding dimension.
Preferably, the default classification number of the basis, to treated, the state vector is clustered, and determines cluster result
It specifically includes:
The default classification number of S1, basis, to treated, the state vector is clustered, and obtains cluster result；
S2, according to Calinski-Harabasz criterion, calculate the corresponding cluster score of the cluster result；
If S3, the cluster score are greater than target fractional, the cluster result is determined, otherwise, adjust the default classification
Number, returns to the S1.
Preferably, the state vector described to treated cluster and is specifically included:
By K-means method, to treated, the state vector is clustered.
Preferably, after the described pair of state vector extracted is normalized further include:
It is pre- whether the judgement interval that treated between the state vector corresponding sampling time and current time is greater than
If the time cycle；
If so, deleting the state vector.
Preferably, after the described pair of state vector extracted is normalized further include:
Whether parameter in judgement treated the state vector in each dimension and the difference of corresponding predetermined normal level
Greater than the X% of predetermined normal level；Wherein, the X% is the first default percentage；
If so, deleting the state vector.
Preferably, after the described pair of state vector extracted is normalized further include:
It is all the same with the presence or absence of two parameters in each corresponding dimension in judgement treated the state vector
State vector；
If so, deleting one in two state vectors.
Preferably, specific packet is normalized in parameter of the described pair of state vector extracted in each dimension
It includes:
According to formula
Parameter of the state vector extracted in each dimension is normalized；
Wherein, xi={ xi1,…,xij,…,ximIt is the i-th bar state vector, xijIt is the i-th bar state vector in j dimension
Parameter.
Preferably, it is described extract the nearest state vector of distance-like center vector as current class typicalness vector it
Afterwards further include:
If the state vector sample number of class is less than the total number of samples of Y% where the typicalness vector, by the typical case
State vector is labeled as to be evaluated；Wherein, the Y% is the second default percentage.
The application second aspect provides a kind of power equipment present mode of operation evaluation method, the typical case of equipment to be evaluated
Operational mode is that a kind of application such as described in any item power equipment typical mode of operation of above-mentioned first aspect determine that method determines
, comprising:
Obtain the state vector of equipment to be evaluated；
It calculates between the state vector that gets each typicalness vector corresponding with the equipment to be evaluated
First distance；
Extract the minimum range in the first distance；
If the minimum range extracted is less than or equal to the corresponding target typicalness vector institute of the minimum range
In the maximum radius of class, the present mode of operation of the equipment to be evaluated is included into the corresponding allusion quotation of the target typicalness vector
Type operational mode, the evaluation of the corresponding typical mode of operation of the evaluation of the equipment present mode of operation to be evaluated；
Wherein, the maximum radius is where the target typicalness vector in class, and each state vector sample is into class
The maximum value of the distance of Heart vector.
The application third aspect provides a kind of state prediction of the development trend method of power equipment, the allusion quotation of equipment to be predicted
Type operational mode is that a kind of application such as described in any item power equipment typical mode of operation of above-mentioned first aspect determine that method is true
Fixed, comprising:
From past first moment to current time, multiple state vectors of equipment to be predicted are extracted with preset interval；
According to least square method, first direction vector and the calculating of the multiple state vector corresponding states variation are calculated
Second direction vector and second distance of the multiple state vector to each typicalness vector；
Calculate the angle of the first direction vector Yu the second direction vector；
The product for calculating the second distance Yu the angle obtains the predicted value for corresponding to each typicalness vector；
The minimum value in the predicted value is extracted, and determines the corresponding typical case of the corresponding typicalness vector of the minimum value
Operational mode is the state that the equipment to be predicted is intended to development.
As can be seen from the above technical solutions, the embodiment of the present application has the advantage that
In the embodiment of the present application, provides a kind of power equipment typical mode of operation and determine method, comprising: according to preset
The state vector space extracts the state vector of the corresponding preset state vector space from the operation data of similar power equipment；
Parameter of the state vector extracted in each dimension is normalized；According to default classification number, to treated shape
State vector is clustered, and determines cluster result；For every one kind in cluster result, the nearest shape of distance-like center vector is extracted
Typicalness vector of the state vector as current class, the corresponding operational mode of typicalness vector are the typical operation of power equipment
Mode；Wherein, parameter of the class center vector in each dimension is that each state vector corresponds to dimension in class where class center vector
The median of parameter.
Method provided by the embodiments of the present application passes through when determining such as typical mode of operation of transformer power equipment
The source of an allusion is determined by each parameter state vectorization of power equipment, then by cluster and the means calculated at a distance from class center vector
Type state vector, and then determine typical mode of operation.Method compared to existing subjective setting typicalness is more objective,
The typical mode of operation determined is more representative closer to truth, also, the means clustered have also contemplated state and drill
The intermediate state of general character during change, makes the typical mode of operation determined more have researching value.
Detailed description of the invention
Fig. 1 is the flow chart that a kind of power equipment typical mode of operation that the application one embodiment provides determines；
Fig. 2 is the process that a kind of power equipment typical mode of operation that second embodiment of the application provides determines method
Figure；
Fig. 3 is a kind of process for power equipment present mode of operation evaluation method that the application third embodiment provides
Figure；
Fig. 4 is a kind of process of the state prediction of the development trend method for power equipment that the 4th embodiment of the application provides
Figure.
Specific embodiment
In order to make those skilled in the art more fully understand application scheme, below in conjunction in the embodiment of the present application
Attached drawing, the technical scheme in the embodiment of the application is clearly and completely described, it is clear that described embodiment is only this
Apply for a part of the embodiment, instead of all the embodiments.Based on the embodiment in the application, those of ordinary skill in the art exist
Every other embodiment obtained under the premise of creative work is not made, shall fall in the protection scope of this application.
The embodiment of the present application provides a kind of power equipment typical mode of operation and determines method, excavates by clustering method
The typical mode of operation of power equipment, it can be found that some intermediate state with general character during state evolution, thus more
Aggregation behaviour that is objective and comprehensively describing power equipment.
The embodiment of the present application also provides a kind of power equipment present mode of operation evaluation methods, are set based on above-mentioned electric power
Standby typical mode of operation determines method, can determine which kind of typical mode of operation is equipment to be evaluated be in, to set to be evaluated
Standby state in which is evaluated.
The embodiment of the present application also provides a kind of state prediction of the development trend methods of power equipment, equally based on above-mentioned
Power equipment typical mode of operation determines method, and can predict equipment to be predicted will develop toward which kind of typical mode of operation, from
And guidance can be provided to subsequent O&M strategy.
Under request in person participation Fig. 1, Fig. 1 is that a kind of power equipment typical mode of operation that the application one embodiment provides is true
Fixed flow chart, this method comprises:
Step 101, according to the preset state vector space, extract that corresponding this is pre- from the operation data of similar power equipment
If the state vector space state vector.
Power equipment usually have it is numerous can be with the parameter of indexing.Such as transformer comprising oil chromatography (H2 content,
C2H4 content, CO content, ratioRatioRatioRatio), electrical test (absorptance, winding
Dielectric loss, winding D.C. resistance, core clamping earth current, partial discharge etc.), oil test (micro- water etc.), operating condition (operation temperature
Degree, overvoltage, overload, short-circuit conditions etc.), the parameters of the various aspects such as fortune inspection information (make an inspection tour situation, all kinds of maintenance situations etc.).
By way of state vector, various parameters can be merged together, a kind of parameter correspondence is made to be incorporated into state
In one dimension of vector.The preset state vector space may be considered the frame of the state vector set, according to
This frame can be convenient and directly mention from the operation data (including daily monitoring data and fortune inspection data) of similar power equipment
Take out the state vector for meeting this frame.
Parameter of the state vector extracted in each dimension is normalized in step 102.
Since numerical value has a long way to go between the parameter on different dimensions, but the ratio of the variable quantity of parameter is identical.Cause
This, can be normalized the parameter in each dimension of state vector, so that state vector, in cluster, each parameter is to poly-
The influence degree of class result corresponds to percentage shared by its numerical value and nonumeric size itself, the quality of cluster result are protected
Card.
Specifically, can be normalized according to the following formula.
Wherein, xi={ xi1,…,xij,…,ximIndicate a state vector, xijIndicate the i-th bar state vector in j dimension
On parameter.
The default classification number of step 103, basis, to treated, state vector is clustered, and determines cluster result.
The algorithm clustered has very much, such as K-means clustering procedure, mean shift clustering procedure, Agglomerative Hierarchical Clustering method
Deng these clustering methods may be applicable to technical solution provided by the present application.
Default classification number is the quantity of preset class, can be limited cluster result.But in view of different
The default corresponding cluster result of number of classifying is different, optimal for cluster result of which kind of the determining default classification under several, can be by more
The secondary default classification number of adjustment is realized, is specifically illustrated in subsequent embodiment.
Step 104, for every one kind in cluster result, extract the nearest state vector of distance-like center vector as working as
The typicalness vector of preceding class.
It should be noted that it includes to belong to this in every one kind that state vector, which will be aggregated into multiple classes, after the completion of cluster
The state vector of class, class center vector are a virtual vectors having in every one kind, remaining state vector as sample
There are same frame, but the middle position of its parameter for corresponding to dimension for each state vector where it in class in the parameter in each dimension
Number.For example, having x in certain one kind1、x2、x3Three state vectors, the then parameter that the first of such class center vector is tieed up are
x11、x21、x31Median, two-dimensional parameter be x12、x22、x32Median (xijIndicate the i-th bar state vector in j dimension
On parameter) ... in this way, formed a class center vector, take away from nearest state vector be such typicalness to
Amount.
Typicalness vector contains the parameters of power equipment, and corresponding operational mode is the typical case of power equipment
Operational mode.
Wherein, parameter of the class center vector in each dimension be, where the class center vector in class each state to
Measure the median of the parameter of corresponding dimension.
A kind of power equipment present mode of operation evaluation method provided in this embodiment is set in determining such as transformer electric power
When standby typical mode of operation, by by each parameter state vectorization of power equipment, then by cluster and calculate and class center
The means of the distance of vector determine typicalness vector, and then determine typical mode of operation.Compared to existing subjective setting
The method of typicalness is more objective, and the typical mode of operation determined is more representative closer to truth, and
And the means of cluster have also contemplated the intermediate state of general character during state evolution, there is the typical mode of operation determined more
Researching value.
The above are a kind of power equipment typical mode of operation provided the application one embodiment to determine specifically
It is bright, Fig. 2 is referred to below, and Fig. 2 is a kind of power equipment typical mode of operation determination side that second embodiment of the application provides
The flow chart of method, this method comprises:
Step 201, according to the preset state vector space, extracted from the operation data of similar power equipment it is corresponding described in
The state vector of the preset state vector space.
The step is identical as the step 101 in above-mentioned one embodiment.
Parameter of the state vector extracted in each dimension is normalized in step 202.
The step is identical as the step 102 in above-mentioned one embodiment.
Whether the interval between step 203, the judgement state vector corresponding sampling time extracted and current time is big
In preset period of time, if so, the bar state vector is deleted, if it is not, not dealing with.
If the state vector and current time interval in view of extracting are too long, the ginseng of the operating status of state vector reflection
Examining value will substantially reduce, and therefore, the state vector for being more than preset period of time apart from current time of this part can be deleted
It removes.For example, preset period of time can be set as 15 days according to 15 days tour vacuum phases of regulation, if a then newest tour
Before being recorded as 20 days, then it is assumed that the data invalid deletes its corresponding state vector.
Parameter in step 204, judgement treated state vector in each dimension and the difference of corresponding predetermined normal level
Whether the X% of predetermined normal level is greater than, if so, the bar state vector is deleted, if it is not, not dealing with.
Further, can exist in state vector it is some have obvious degradation trend, the confidence level of the partial status vector
It is lower, it is in particular in parameter, the parameter in each dimension has big difference with predetermined normal level, and difference is greater than default normal
The X%'s of value, reference value is very low, can delete.Wherein, X% is the first default percentage, such as 20%.
It is homogeneous with the presence or absence of two parameters in each corresponding dimension in step 205, judgement treated state vector
Same state vector, if so, one in two bar state vectors is deleted, if it is not, not dealing with.
To have duplicate state vector sample when avoiding cluster, duplicate two bar states vector if it exists then can be with
Delete one in the two.
The default classification number of step 206, basis, to treated, the state vector is clustered, and obtains cluster result.
Step 207, according to Calinski-Harabasz criterion, calculate the corresponding cluster score of cluster result.
It should be noted that Calinski-Harabasz criterion is the algorithm that marking evaluation can be carried out to cluster result,
Cluster score can be calculated by the following formula out.
Wherein, SSB is inter-class variance, and SSW is variance within clusters,For complexity, VRCk is cluster score.VRCk is got over
Greatly, Clustering Effect is better.
If step 208, calculated cluster score are greater than target fractional, cluster result is determined, otherwise, adjust default classification
Number, return step 206.
If calculated cluster score is greater than target fractional, it is believed that cluster result has reached requirement, determines the cluster
As a result, the numerical value of the default classification number of readjustment influences cluster result, directly otherwise it is assumed that the setting of default classification number is unreasonable
Meet target fractional to cluster score.
Step 209, for every one kind in cluster result, extract the nearest state vector of distance-like center vector as working as
The typicalness vector of preceding class.
The step is identical as the step 104 in above-mentioned one embodiment.
If the state vector sample number of class is less than total sample of Y% where step 210, the typicalness vector extracted
Number, by the typicalness vector labeled as to be evaluated.
After the completion of cluster, it is understood that there may be the seldom class of state vector sample number, the typicalness extracted in these classes
Vector is not necessarily effective, it is possible to belong to invalid noise, therefore can first be marked as to be evaluated, facilitate and subsequent ask industry
Expert to the typicalness vector carry out deeper into analysis.
A kind of power equipment present mode of operation evaluation method provided in this embodiment is set in determining such as transformer electric power
It, will be invalid by by each parameter state vectorization of power equipment, and by three kinds of pretreatments when standby typical mode of operation
State vector is rejected, and clustering result quality is improved；And when being clustered by clustering algorithm, also use cluster score evaluation cluster
Effect adjusts default classification number when cluster score is not up to standard, Clustering Effect is made to reach requirement；Finally, calculating shape in every one kind
State vector determines typicalness vector at a distance from class center vector, and then determines typical mode of operation.Compared to existing
The method of subjectivity setting typicalness is more objective, and the typical mode of operation determined has more generation closer to truth
Table, also, the means clustered have also contemplated the intermediate state of general character during state evolution, make the typical operation mould determined
Formula more has researching value.
The above are a kind of power equipment typical mode of operation provided second embodiment of the application to determine the detailed of method
It describes in detail bright, refers to Fig. 3 below, Fig. 3 is that a kind of power equipment present mode of operation that the application third embodiment provides is commented
The flow chart of valence method.
It is understood that any reality provided using above-mentioned the application one embodiment or second embodiment
Power equipment typical mode of operation under existing mode determines method, can determine a kind of typical mode of operation of power equipment,
For example, each typical mode of operation of transformer can be determined.In order to which typical mode of operation is used, typical can run
Mode carries out preparatory evaluation.
Specifically, can ask industry specialists is that various typical mode of operation carry out marking evaluation, naturally it is also possible to take and beat
Other evaluation methods other than point.The score that each typical mode of operation obtains can characterize the excellent of the typical mode of operation
It is bad.After being evaluated in advance typical mode of operation, can use its to by by the state vector of certain equipment and typicalness to
Amount is matched, so that it is determined that it evaluates the present mode of operation of power equipment, the specific method is as follows:
Step 301, the state vector for obtaining equipment to be evaluated.
The state vector of equipment to be evaluated should have the identical state vector space with typicalness vector, i.e., in each dimension
It is corresponded in the parameter of degree.It is also possible to be normalized as needed.
Between state vector each typicalness vector corresponding with equipment to be evaluated that step 302, calculating are got
First distance.
Equipment to be evaluated has corresponding multiple typicalness vectors to require to calculate for each typicalness vector
First distance between remaining state vector.
Minimum range in step 303, extraction first distance.
It extracts in multiple first distances apart from shortest minimum range.
If step 304, the minimum range extracted are less than or equal to the corresponding target typicalness vector institute of minimum range
In the maximum radius of class, the present mode of operation of equipment to be evaluated is included into the corresponding typical operation of the target typicalness vector
Mode, the evaluation of the corresponding typical mode of operation of the evaluation of equipment present mode of operation to be evaluated.
It should be noted that being obtained in cluster result in the application one embodiment and second embodiment, every one kind
All it is corresponding with a maximum radius, in a maximum radius i.e. class, the distance of each state vector sample to class center vector is most
Big value.If minimum range is less than or equal to such maximum radius, the state vector of equipment to be evaluated belongs to such, can incite somebody to action
Itself and such typicalness vector association, corresponding, the present mode of operation of equipment to be evaluated can be included into target typical case
The corresponding typical mode of operation of state vector, and the evaluation of equipment present mode of operation to be evaluated then corresponds to the typical mode of operation
Evaluation.The evaluation of typical mode of operation can be obtained by preparatory expert opinion, can specifically be referred to and be described above.
In the present embodiment, a kind of power equipment present mode of operation evaluation method is provided, utilizes above-mentioned the application
Power equipment typical mode of operation determination side under any implementation that one embodiment or second embodiment provide
Method determines the typical mode of operation of the power equipment type of equipment to be evaluated, recycle the state vector of equipment to be evaluated with
Typicalness vector is matched, and may thereby determine that the present mode of operation of equipment to be evaluated is close or it is typical which kind of belongs to
Operational mode provides guidance meaning further, it is possible to evaluate the superiority and inferiority of its present mode of operation for the subsequent O&M of equipment to be evaluated
See.
The above are a kind of the detailed of the power equipment present mode of operation evaluation method provided the application third embodiment
It describes in detail bright, refers to Fig. 4 below, Fig. 4 is a kind of state development trend for power equipment that the 4th embodiment of the application provides
The flow chart of prediction technique.
It is similar with above-mentioned third embodiment, it is equally to a kind of power equipment provided by the present application in the present embodiment
Typical mode of operation determines the further utilization for the typical mode of operation that method is determined, can obtain further by calculating
The degree of association of the corresponding each typical mode of operation of the state of some particular device, so as to predict which it would tend to
Kind typical mode of operation development.
Specifically method includes:
Step 401, from past first moment to current time, multiple states of equipment to be predicted are extracted with preset interval
Vector.
For example, extracting and conforming to from the corresponding operation data of equipment to be predicted every day to today before 10 days
The state vector asked, the requirement of state vector are identical as the requirement in any of the above-described a embodiment.
Step 402, according to least square method, calculate multiple state vector corresponding states variations first direction vector and
Calculate multiple state vectors to each typicalness vector second direction vector and second distance.
Specifically, linear function fitting can be carried out using least square method, calculate in recent a period of time (i.e. from mistake
The first moment to current time gone), the state vector of equipment to be predicted according to the state change of time sequencing first direction to
Measure dx, meanwhile, calculate multiple state vectors to each typicalness vector second direction vector dyp=yp-xt.Between dx and dyp
Angle:
Step 403, the angle for calculating first direction vector and second direction vector.
It can be according to formula:
Calculate the angle of first direction vector Yu second direction vector.Obviously, each corresponding typicalness vector is equal
There is calculated corresponding angle.
Step 404, the product for calculating second distance and angle, obtain the predicted value for corresponding to each typicalness vector.
Deviate the degree of a certain typical mode of operation using the product of second distance and angle as equipment to be predicted.
Minimum value in step 405, extraction predicted value, and determine the corresponding allusion quotation of the corresponding typicalness vector of the minimum value
Type operational mode is the state that equipment to be predicted is intended to development.
Predicted value is the smallest, it is believed that it is minimum that equipment to be predicted deviates the corresponding typicalness vector of the predicted value, most
It is possible that being intended to the corresponding typical mode of operation development of the typicalness vector.
In the present embodiment, a kind of state prediction of the development trend method of power equipment is provided, based on provided by the present application
A kind of power equipment typical mode of operation determines method, realizes and would tend to which kind of typical fortune to the state of certain power equipment
The prediction of row mode development.Further, according to prediction result, corresponding O&M strategy can be formulated in advance, so that prevention is each
The generation of kind accident, improves power supply reliability.
The description of the present application and term " first " in above-mentioned attached drawing, " second ", " third ", " the 4th " etc. are (if deposited
) it is to be used to distinguish similar objects, without being used to describe a particular order or precedence order.It should be understood that use in this way
Data are interchangeable under appropriate circumstances, so that embodiments herein described herein for example can be in addition to illustrating herein
Or the sequence other than those of description is implemented.In addition, term " includes " and " having " and their any deformation, it is intended that
Cover it is non-exclusive include, for example, containing the process, method, system, product or equipment of a series of steps or units need not limit
In step or unit those of is clearly listed, but may include be not clearly listed or for these process, methods, produce
The other step or units of product or equipment inherently.
It should be appreciated that in this application, " at least one (item) " refers to one or more, and " multiple " refer to two or two
More than a."and/or" indicates may exist three kinds of relationships, for example, " A and/or B " for describing the incidence relation of affiliated partner
It can indicate: only exist A, only exist B and exist simultaneously tri- kinds of situations of A and B, wherein A, B can be odd number or plural number.Word
Symbol "/" typicallys represent the relationship that forward-backward correlation object is a kind of "or"." at least one of following (a) " or its similar expression, refers to
Any combination in these, any combination including individual event (a) or complex item (a).At least one of for example, in a, b or c
(a) can indicate: a, b, c, " a and b ", " a and c ", " b and c ", or " a and b and c ", and wherein a, b, c can be individually, can also
To be multiple.
The above, above embodiments are only to illustrate the technical solution of the application, rather than its limitations；Although referring to before
Embodiment is stated the application is described in detail, those skilled in the art should understand that: it still can be to preceding
Technical solution documented by each embodiment is stated to modify or equivalent replacement of some of the technical features；And these
It modifies or replaces, the spirit and scope of each embodiment technical solution of the application that it does not separate the essence of the corresponding technical solution.
Claims (10)
1. a kind of power equipment typical mode of operation determines method characterized by comprising
According to the preset state vector space, extracted from the operation data of similar power equipment the corresponding preset state to
The state vector of quantity space；
Parameter of the state vector extracted in each dimension is normalized；
According to default classification number, to treated, the state vector is clustered, and determines cluster result；
For every one kind in cluster result, typical shape of the nearest state vector of distance-like center vector as current class is extracted
State vector, the corresponding operational mode of the typicalness vector are the typical mode of operation of the power equipment；
Wherein, parameter of the class center vector in each dimension is each state vector pair in class where the class center vector
Answer the median of the parameter of dimension.
2. a kind of power equipment typical mode of operation according to claim 1 determines method, which is characterized in that the basis
Default classification number, to treated, the state vector is clustered, and determines that cluster result specifically includes:
The default classification number of S1, basis, to treated, the state vector is clustered, and obtains cluster result；
S2, according to Calinski-Harabasz criterion, calculate the corresponding cluster score of the cluster result；
If S3, the cluster score are greater than target fractional, the cluster result is determined, otherwise, adjust the default classification number, return
Return the S1.
3. a kind of power equipment typical mode of operation according to claim 2 determines method, which is characterized in that at described pair
The state vector after reason carries out cluster and specifically includes:
By K-means method, to treated, the state vector is clustered.
4. a kind of power equipment typical mode of operation according to claim 1 determines method, which is characterized in that described pair mentions
After the state vector taken out is normalized further include:
When whether the judgement interval that treated between the state vector corresponding sampling time and current time is greater than default
Between the period；
If so, deleting the state vector.
5. a kind of power equipment typical mode of operation according to claim 4 determines method, which is characterized in that described pair mentions
After the state vector taken out is normalized further include:
Whether the parameter in judgement treated the state vector in each dimension is greater than with the difference of corresponding predetermined normal level
The X% of predetermined normal level；Wherein, the X% is the first default percentage；
If so, deleting the state vector.
6. a kind of power equipment typical mode of operation according to claim 5 determines method, which is characterized in that described pair mentions
After the state vector taken out is normalized further include:
The state all the same with the presence or absence of two parameters in each corresponding dimension in judgement treated the state vector
Vector；
If so, deleting one in two state vectors.
7. a kind of power equipment typical mode of operation according to claim 1 determines method, which is characterized in that described pair mentions
Parameter of the state vector taken out in each dimension, which is normalized, to be specifically included:
According to formula
Parameter of the state vector extracted in each dimension is normalized；
Wherein, xi={ xi1,…,xij,…,ximIt is the i-th bar state vector, xijFor ginseng of the i-th bar state vector in j dimension
Number.
8. a kind of power equipment typical mode of operation according to claim 1 determines method, which is characterized in that the extraction
After the nearest state vector of distance-like center vector is as the typicalness vector of current class further include:
If the state vector sample number of class is less than the total number of samples of Y% where the typicalness vector, by the typicalness
Vector is labeled as to be evaluated；Wherein, the Y% is the second default percentage.
9. a kind of power equipment present mode of operation evaluation method, which is characterized in that the typical mode of operation of equipment to be evaluated is
Determine what method determined using a kind of power equipment typical mode of operation as claimed in any one of claims 1 to 8, comprising:
Obtain the state vector of equipment to be evaluated；
Calculate first between the state vector that gets each typicalness vector corresponding with the equipment to be evaluated
Distance；
Extract the minimum range in the first distance；
If the minimum range extracted is less than or equal to class where the corresponding target typicalness vector of the minimum range
Maximum radius, the present mode of operation of the equipment to be evaluated is included into the target typicalness vector is corresponding typical to be transported
Row mode, the evaluation of the corresponding typical mode of operation of the evaluation of the equipment present mode of operation to be evaluated；
Wherein, the maximum radius is where the target typicalness vector in class, each state vector sample to class center to
The maximum value of the distance of amount.
10. a kind of state prediction of the development trend method of power equipment, which is characterized in that the typical mode of operation of equipment to be predicted
To apply a kind of power equipment typical mode of operation as claimed in any one of claims 1 to 8 to determine what method determined, comprising:
From past first moment to current time, multiple state vectors of equipment to be predicted are extracted with preset interval；
According to least square method, calculate described in first direction vector and the calculating of the multiple state vector corresponding states variation
Second direction vector and second distance of multiple state vectors to each typicalness vector；
Calculate the angle of the first direction vector Yu the second direction vector；
The product for calculating the second distance Yu the angle obtains the predicted value for corresponding to each typicalness vector；
The minimum value in the predicted value is extracted, and determines the corresponding typical operation of the corresponding typicalness vector of the minimum value
Mode is the state that the equipment to be predicted is intended to development.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811536027.7A CN109685348A (en) | 2018-12-14 | 2018-12-14 | A kind of power equipment typical mode of operation determines method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811536027.7A CN109685348A (en) | 2018-12-14 | 2018-12-14 | A kind of power equipment typical mode of operation determines method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109685348A true CN109685348A (en) | 2019-04-26 |
Family
ID=66187837
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811536027.7A Pending CN109685348A (en) | 2018-12-14 | 2018-12-14 | A kind of power equipment typical mode of operation determines method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109685348A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113095425A (en) * | 2021-04-21 | 2021-07-09 | 广东电网有限责任公司 | Method, device, equipment and medium for automatically judging and deciding on site start-up conditions |
| CN113283504A (en) * | 2021-05-24 | 2021-08-20 | 平安国际融资租赁有限公司 | Device state detection method and device based on clustering, computer device and medium |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030161500A1 (en) * | 2002-02-22 | 2003-08-28 | Andrew Blake | System and method for probabilistic exemplar-based pattern tracking |
| US20080101705A1 (en) * | 2006-10-31 | 2008-05-01 | Motorola, Inc. | System for pattern recognition with q-metrics |
| US20110091113A1 (en) * | 2009-10-19 | 2011-04-21 | Canon Kabushiki Kaisha | Image processing apparatus and method, and computer-readable storage medium |
| CN105653835A (en) * | 2014-11-14 | 2016-06-08 | 北京宇航系统工程研究所 | Cluster analysis-based anomaly detection method |
| CN106709662A (en) * | 2016-12-30 | 2017-05-24 | 山东鲁能软件技术有限公司 | Electrical equipment operation condition classification method |
| CN106791758A (en) * | 2016-12-07 | 2017-05-31 | 浙江大华技术股份有限公司 | The determination methods and device of natural light mixing colour temperature in a kind of image |
| CN107895222A (en) * | 2017-10-26 | 2018-04-10 | 华北电力大学 | The bad Leakage Reactance discrimination method of transformer based on DBSCAN algorithms |
| CN108764335A (en) * | 2018-05-28 | 2018-11-06 | 南方电网科学研究院有限责任公司 | A kind of integrated energy system multi-energy requirement typical scene generation method and device |
-
2018
- 2018-12-14 CN CN201811536027.7A patent/CN109685348A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030161500A1 (en) * | 2002-02-22 | 2003-08-28 | Andrew Blake | System and method for probabilistic exemplar-based pattern tracking |
| US20080101705A1 (en) * | 2006-10-31 | 2008-05-01 | Motorola, Inc. | System for pattern recognition with q-metrics |
| US20110091113A1 (en) * | 2009-10-19 | 2011-04-21 | Canon Kabushiki Kaisha | Image processing apparatus and method, and computer-readable storage medium |
| CN105653835A (en) * | 2014-11-14 | 2016-06-08 | 北京宇航系统工程研究所 | Cluster analysis-based anomaly detection method |
| CN106791758A (en) * | 2016-12-07 | 2017-05-31 | 浙江大华技术股份有限公司 | The determination methods and device of natural light mixing colour temperature in a kind of image |
| CN106709662A (en) * | 2016-12-30 | 2017-05-24 | 山东鲁能软件技术有限公司 | Electrical equipment operation condition classification method |
| CN107895222A (en) * | 2017-10-26 | 2018-04-10 | 华北电力大学 | The bad Leakage Reactance discrimination method of transformer based on DBSCAN algorithms |
| CN108764335A (en) * | 2018-05-28 | 2018-11-06 | 南方电网科学研究院有限责任公司 | A kind of integrated energy system multi-energy requirement typical scene generation method and device |
Non-Patent Citations (1)
| Title |
|---|
| 唐胜等: "一种基于海量数据挖掘的设备状态预测算法", 《计算机科学》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113095425A (en) * | 2021-04-21 | 2021-07-09 | 广东电网有限责任公司 | Method, device, equipment and medium for automatically judging and deciding on site start-up conditions |
| CN113283504A (en) * | 2021-05-24 | 2021-08-20 | 平安国际融资租赁有限公司 | Device state detection method and device based on clustering, computer device and medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109614981B (en) | Power system intelligent fault detection method and system based on spearman level-dependent convolutional neural network | |
| CN105260595B (en) | The feature extracting method and switch breakdown diagnostic method of track switch action current curve | |
| CN110441629B (en) | Fault diagnosis method and device based on turnout switch machine operating current curve | |
| Demir et al. | Learning the topological properties of brain tumors | |
| CN107274105B (en) | Linear discriminant analysis-based multi-attribute decision tree power grid stability margin evaluation method | |
| CN109753499A (en) | A kind of O&M monitoring data administering method | |
| CN106896219B (en) | The identification of transformer sub-health state and average remaining lifetime estimation method based on Gases Dissolved in Transformer Oil data | |
| CN109685348A (en) | A kind of power equipment typical mode of operation determines method | |
| CN112200263B (en) | Self-organizing federal clustering method applied to power distribution internet of things | |
| CN104655948A (en) | Novel multistage fault diagnosis method for power transformer | |
| Ali et al. | COMSATS University Islamabad | |
| CN108197175A (en) | The treating method and apparatus of technical supervision data, storage medium, processor | |
| CN107895212A (en) | Lead-acid battery life-span prediction method based on sliding window and various visual angles Fusion Features | |
| CN110059913A (en) | A kind of quantitative estimation method counted and the power failure of future-state is planned | |
| CN109709507A (en) | Intelligent electric energy meter method for predicting reliability based on failure rate level | |
| Eke et al. | Transformer condition assessment using fuzzy C-means clustering techniques | |
| CN110852441B (en) | Fire disaster early warning method based on improved naive Bayes algorithm | |
| MansourLakouraj et al. | Event classification in active distribution grids using physics-informed graph neural network and PMU measurements | |
| De et al. | Classification of power system voltage stability conditions using Kohonen's self‐organising feature map and learning vector quantisation | |
| CN110852370A (en) | Clustering algorithm-based large-industry user segmentation method | |
| CN116911161A (en) | Data-enhanced deep learning transient voltage stability evaluation method | |
| Tang et al. | Classification for transient overvoltages in offshore wind farms based on sparse decomposition | |
| CN107451346B (en) | Power grid operation situation model construction method | |
| CN108828377A (en) | A kind of Diagnosis Method of Transformer Faults | |
| Li et al. | Improved density peak clustering for separation of multiple source partial discharge in large generators |
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 |
Application publication date: 20190426 |
|
| RJ01 | Rejection of invention patent application after publication |