CN109784777A - Grid equipment state evaluating method based on timing information segment cloud measuring similarity - Google Patents
Grid equipment state evaluating method based on timing information segment cloud measuring similarity Download PDFInfo
- Publication number
- CN109784777A CN109784777A CN201910149813.XA CN201910149813A CN109784777A CN 109784777 A CN109784777 A CN 109784777A CN 201910149813 A CN201910149813 A CN 201910149813A CN 109784777 A CN109784777 A CN 109784777A
- Authority
- CN
- China
- Prior art keywords
- information segment
- grid equipment
- similarity
- typicalness
- timing information
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of grid equipment state evaluating methods based on timing information segment cloud measuring similarity.The on-line monitoring time series data that grid equipment status assessment relies on is there are monitoring data amount is big, and poor information is shared between monitoring parameters, the monitoring data obtained in the presence of a harsh environment the problems such as there are noise and deviations.Appraisal procedure proposed by the present invention utilizes expectation, entropy, online monitoring data is described in super entropy, using weighting cloud measuring similarity and probability output method, convergence analysis has been carried out to each parameter for describing grid equipment state, it is excessive to solve data volume existing for powernet monitoring time series data itself, information island, the problem of the defects of data deviation and existing grid equipment state evaluating method can not timely and accurately judge equipment state, the real-time status assessment of grid equipment can be pointedly used for, to further ensure that the safety of electric system operates normally.
Description
Technical field
The invention belongs to grid equipment status assessment technical fields, and in particular to be based on timing information segment cloud similarity degree
The grid equipment state evaluating method of amount.
Background technique
With flourishing for artificial intelligence, intellectual technology is combined with electric system to be had become for a kind of trend.By intelligence
Energy algorithm is applied in grid equipment status assessment, can be carried out going deep into excavation to grid equipment operation data, be known that power grid is set
Standby history run state and current operating conditions, instruct the formulation of grid equipment Strategies of Maintenance, evade grid equipment to realize
Failure occurs, and extends grid equipment service life, ensures the purpose of power network safety operation.
Currently, mainly passing through the offline preventative examination of analysis to the status assessment of grid equipment in technical field of power systems
Test data and online monitoring data two ways.Although the equipment operating data that offline preventive trial obtains can be more accurate
The state of ground consersion unit, but since data volume is few, and the test period is longer, substantially increases the difficulty of data mining, and
Lose the real-time of status monitoring.The real-time device operation data that on-line monitoring system obtains can preferably overcome offline pre-
The shortcomings that anti-property test data, but there is also monitoring data amount is big, there are multiple on-line monitoring systems to cause letter for same equipment
Cease island phenomenon, the monitoring data obtained in the presence of a harsh environment the problems such as there are noise and deviations.Therefore, overcome on-line monitoring number
Ensure that succeeding state assessment result is accurate, reliable precondition according to itself having a defect that.
Existing grid equipment state evaluating method is broadly divided into Weight of Expert marking, machine learning method and time series
Analysis method.Wherein, Time series analysis method can measure the similarity in on-line monitoring system between time series, obtain
Potential information in temporal sequence, to carry out status assessment to grid equipment.But existing Time Series Analysis Method tends not to
Sufficient status information is extracted from the monitoring data of highly redundant and is lacked, and the information of the multiple quantity of states of grid equipment is merged, from
And cause not accurate enough to grid equipment condition evaluation results.
Can not only it overcome in conjunction with the above deficiency it is really necessary to propose a kind of grid equipment state evaluating method that method is new
On-line monitoring system time series data amount is big, information island and the problems such as there are noise and deviations, moreover it is possible to run and join from grid equipment
Effective status information is extracted in number, merges multiple state parameters and accurate evaluation is carried out to grid equipment state.
Summary of the invention
The present invention provides the grid equipment state evaluating methods based on timing information segment cloud measuring similarity, to solve
The defects of certainly monitoring excessive data volume existing for time series data itself, information island, data deviation on-line and existing power grid are set
The problem of standby state evaluating method can not timely and accurately judge equipment state.
In order to achieve the above objectives, the grid equipment state of the present invention based on timing information segment cloud measuring similarity is commented
Estimate method, comprising the following steps:
Step 1, the sample for collecting equipment to be assessed, each sample include multiple for describing the key of grid equipment state
Parametric data, each key parameters data of grid equipment form the measurement timing information piece that sequencing changes at any time
Section, grid equipment is current or the time series data to be assessed of historic state for describing for multiple measurement timing information segments compositions
Collect A, wherein key parameters data refer to the operating parameter for effectively reflecting power equipment operating status;
Step 2 collects multiple key parameters data under grid equipment typicalness, every under grid equipment typicalness
A key parameters data form a standard time sequence information segment, and multiple standard time sequence information segments form standard time series number
The standard time sequence information collection U under corresponding typicalness is formed according to collection B, standard time series data set B, with standard time sequence information collection
U is element, integrates the standard time series information bank V under grid equipment typicalness;
Step 3, the characteristic information for extracting every measurement timing information segment in time series data collection A to be assessed, with mentioning
The characteristic information for the measurement timing information segment got constructs time series eigenmatrix Q to be assessed;
The characteristic information of the timing information segment of each element in step 4, extraction standard time serial message library V, building
Typicalness time series feature set P;Using typicalness time series feature set P as element, grid equipment typicalness is integrated
Under typical temporal aspect library T;
Step 5 is calculated using cloud method for measuring similarity in temporal aspect matrix Q to be assessed and typical temporal aspect library T
The similitude of each corresponding element, the similarity of grid equipment measurement sample and each typicalness sample as to be assessed;
The similarity of step 6, the grid equipment state and each typicalness to be assessed that are obtained according to step 5 calculates to be evaluated
Estimate the probability that grid equipment is under the jurisdiction of each typicalness.
Further, in step 1, the time series data to be assessed integrates the size of A as d*n, and wherein d sets for power grid
The key parameter number of standby state, n are the length for measuring timing information segment;Timing in step 2, under the typicalness
Information collection U={ B1,B2,…,Bi,…,BN, wherein 1≤i≤N, N are the sample number of every quasi-representative state, and B is typicalness
Time series data collection, the size of B is d*n;Standard time series information bank V={ U1,U2,…,Uj,…US, wherein 1≤
J≤S, S are grid equipment typicalness type.
Further, in step 3, every measurement in time series data collection A to be assessed is extracted using timing characterizing method
The characteristic information of timing information segment;In step 4, using each in timing characterizing method extraction standard time serial message library V
The characteristic information of the timing information segment of element, the calculation formula of timing characterizing method are as follows:
Ym=diff (Xm), Zm=diff (Ym);
Exm=mean (Xm),Eym=mean (Ym),Ezm=mean (Zm);
Cm=(Exm,Eym,Ezm,Enxm,Enym,Enzm,Hexm,Heym,Hezm);
Wherein, XmFor a timing information segment of input, YmFor XmFirst difference as a result, ZmFor XmSecond order difference knot
Fruit;ExmFor XmExpectation, EymFor YmExpectation, EzmFor ZmExpectation;EnxmFor XmEntropy, EnymFor YmEntropy, EnzmFor Zm's
Entropy, HexmFor XmSuper entropy, HeymFor YmSuper entropy, HezmFor ZmSuper entropy, diff be Difference Calculation function, mean be mean value meter
Function is calculated, var is that variance calculates function;CmFor the characteristic information of timing information segment, timing information segment includes measurement timing
Information segment and standard time sequence information segment, include 9 characterization factors, and 1≤m≤d, d are the key parameter of grid equipment state
Number.
Further, in step 3, time series eigenmatrix Q size to be assessed is d*9, expression formula are as follows:
In step 4, typicalness time series feature set P={ L1,L2,…,Li,…,LN, wherein LiWhen for typicalness
Between sequence signature matrix, 1≤i≤N, LiSize be d*9;Typical temporal aspect library T={ P under grid equipment typicalness1,
P2,…,Pi,…,PS, PiFor the typicalness time series feature set under different key parameters, 1≤i≤S.
Further, in step 5, similarity between the timing information segment measured and standard time sequence information segment
Calculate step are as follows:
Step 5.1, by the characteristic information C of every timing information segmentmDisassemble into 3 independent characteristic variable Cxm、CymWith
Czm, it is as follows to disassemble mode:
Step 5.2 calculates separately the same key parameters in measurement timing information segment and standard time sequence information segment
Independent characteristic variable between similarity, with CxamAnd CxbmFor, its calculation formula is:
Uminxm=min (Exam+3Enxam,Exbm+3Enxbm), Umaxxm=max (Exam+3Enxam,Exbm+3Enxbm),
Lminxm=min (Exam-3Enxam,Exbm-3Enxbm), Lmaxxm=max (Exam-3Enxam,Exbm-3Enxbm),
Wherein, CxamFor the independent characteristic variable for measuring timing information segment, CxbmFor the independence of standard time sequence information segment
Characteristic variable;UmaxxmFor measurement timing information segment and standard time sequence information segment independent characteristic variable upper bound maximum value,
UminxmFor the upper bound minimum value of measurement timing information segment and the independent characteristic variable of standard time sequence information segment;LmaxxmFor,
LminxmFor the lower bound maximum value of measurement timing information segment and the independent characteristic variable of standard time sequence information segment, timing is measured
The lower bound minimum value of the independent characteristic variable of information segment and standard time sequence information segment;s1And s2It is measurement timing information piece
Section corresponds to the overlapping point of cloud model with the independent characteristic variable of standard time sequence information segment;μ indicates that overlapping point belongs to two independences
The confidence level of characteristic variable;Dist represents the phase of the independent characteristic variable of measurement timing information segment and standard time sequence information segment
Like degree;
CyamAnd CybmBetween similarity and CzamAnd CzbmSimilarity and CxamAnd CxbmSimilarity calculation process
It is identical;
The independent characteristic of step 5.3, the measurement timing information segment and standard time sequence information segment that are obtained according to step 5.2
The similarity of variable calculates the cloud similarity Cloud of measurement timing information segment and standard time sequence information segment:
Wherein, CamFor the characteristic information for measuring timing information segment, CbmFor the characteristic information of standard time sequence information segment;
The weight coefficient of step 5.4, each key parameters of setting;By way of weighting recombination, obtain the measurement sample of grid equipment with
The similarity M of typicalness sample:
Wherein, wmRepresent the weight coefficient of the key parameters of grid equipment.
Further, in step 5.4, according to the significance level of each parameter for describing grid equipment state, setting is each
The weight coefficient of key parameters.
Further, step 6 the following steps are included:
Step 6.1 summarizes similarity between grid equipment to be assessed measurement sample and each typicalness sample, obtains S*
N number of similarity, and sort according to the sequence of the numerical value of similarity from small to large;
Step 6.2, the similarity set omega for setting fixed size, and the preceding length (Ω) that will sort in step 6.1 is a
Element of the similarity data as set omega, length (Ω) represent the similarity set sizes of setting;
Step 6.3 filters out the similarity for being subordinate to all kinds of typicalness in similarity set omega, constructs each typicalness phase
Like degree subset Ωj, and calculate the probability P (S that grid equipment to be assessed is under the jurisdiction of each typicalnessi), calculation formula is as follows:
Wherein, SjRepresent the S kind difference typicalness of grid equipment, length (Ωj) represent typicalness similarity subset
Size.
Further, it in step 1, is used to describe the multiple of grid equipment state from what different detections or monitoring system were collected
Key parameters data.
Compared with prior art, the present invention at least has technical effect beneficial below, grid equipment proposed by the present invention
With expectation, entropy, 3 kinds of super entropy, online monitoring data is described in totally 9 characteristic quantities to state evaluating method, can effectively gram
Noise existing for data source itself and offset issue are taken.Time series data, first difference data, second order difference to device parameter
Data carry out feature extraction, can accurately know distribution and the change information of time series data, solve grid equipment data
Information caused by amount is big, redundancy is high buries problem;Using weighting cloud measuring similarity and probability output method, not only to being used for
Each parameter of description grid equipment state has carried out convergence analysis, moreover it is possible to be in each typical shape to grid equipment in a probabilistic manner
The situation of state is assessed, and solves the problems, such as status assessment inaccuracy caused by grid equipment parameter information isolated island.
Detailed description of the invention
Fig. 1 is that typicalness characteristic information library of the invention constructs flow chart;
Fig. 2 is specific implementation flow chart of the invention.
Specific embodiment
The following describes the present invention in detail with reference to the accompanying drawings and specific embodiments.
In the description of the present invention, it is to be understood that, term " center ", " longitudinal direction ", " transverse direction ", "upper", "lower",
The orientation or positional relationship of the instructions such as "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outside" is
It is based on the orientation or positional relationship shown in the drawings, is merely for convenience of description of the present invention and simplification of the description, rather than instruction or dark
Show that signified device or element must have a particular orientation, be constructed and operated in a specific orientation, therefore should not be understood as pair
Limitation of the invention.In addition, term " first ", " second " are used for description purposes only, it is not understood to indicate or imply opposite
Importance or the quantity for implicitly indicating indicated technical characteristic.Define " first " as a result, the feature of " second " can be bright
Show or implicitly include one or more of the features.In the description of the present invention, unless otherwise indicated, " multiple " contain
Justice is two or more.In the description of the present invention, it should be noted that unless otherwise clearly defined and limited, art
Language " installation ", " connected ", " connection " shall be understood in a broad sense, for example, it may be being fixedly connected, may be a detachable connection, or
It is integrally connected;It can be mechanical connection, be also possible to be electrically connected;It can be directly connected, it can also be by between intermediary
It connects connected, can be the connection inside two elements.For the ordinary skill in the art, can be understood with concrete condition
The concrete meaning of above-mentioned term in the present invention.
Due to grid equipment huge number, the present embodiment is by taking more representative grid equipment-transformer as an example, to this
The technical solution of invention is described.In the following, in conjunction with attached drawing to the grid equipment based on timing information segment cloud measuring similarity
State evaluating method is described in detail.
One, data collection and construction feature information bank
Key parameters data of 300 groups of transformer state data as emulation are collected, the key parameters data of collection include
Oil colours modal data, winding resistance value, iron core temperature etc. are obtained by different detections or monitoring system.Wherein, belong to normal condition,
Data under 4 kinds of attention state, abnormality and severe conditions typicalness have 75 groups.Every group of data include 4 kinds for retouching
State the parameter of transformer state.Each parameter has 50 sampled points, constitutes the timing information segment that length is 50.From every kind of transformation
25 groups are extracted in device typicalness data at random, constitutes the timing information collection to be assessed comprising 100 groups of transformer state data.It is surplus
200 groups of remaining transformer state data sets are as the standard time series information bank V under transformer typicalness.
According to the significance level of 4 kinds of different parameters for describing transformer state, corresponding weight, the present embodiment are set
The weighted value of middle setting is respectively w1=1, w2=2, w3=2, w4=3.According to the standard time sequence information under transformer typicalness
The size of library V, sets the size of similarity set omega, and the size of similarity set omega is the standard time sequence information under typicalness
The 15%-20% of library V, the size for the similarity set omega being arranged in the present embodiment are 30.
As shown in Figure 1, the standard time series information bank V under transformer typicalness is turned with timing characterizing method
Turn to the typical temporal aspect library T under grid equipment typicalness:
It include 4 standard time sequences information collection U, i.e. V={ U in standard time sequence information bank V under transformer typicalness1,U2,
U3,U4}.Each standard time sequence information collection U includes 50 standard time series data set B, i.e. U={ B1,B2,…,B50}.Standard
The expression of time series data collection B are as follows:
With the characteristic variable of the time series of every row in following formulas Extraction standard time series data set B, allusion quotation is obtained
Type state temporal aspect matrix L.
The expression of typicalness temporal aspect matrix L are as follows:
The typical temporal aspect library T under typicalness temporal aspect collection P and grid equipment typicalness is constructed, wherein P=
{L1,L2,…,L50, T={ P1,P2,P3,P4}。
Two, similarity measurement and probability output
1) 1 group of transformer state data to be assessed is converted temporal aspect matrix Q to be assessed by formula (1):
Wherein, subscript a is used to distinguish with the subscript b in typicalness temporal aspect matrix L.
2) formula (2)-(3) calculate between temporal aspect matrix Q to be assessed and typicalness temporal aspect matrix L
Cloud similarity obtains 200 groups of similarity vectors.Every group of similarity vector expression formula are as follows:
Wherein, Uminxm=min (Exam+3Enxam,Exbm+3Enxbm),
Umaxxm=max (Exam+3Enxam,Exbm+3Enxbm),
Lminxm=min (Exam-3Enxam,Exbm-3Enxbm), Lmaxxm=max (Exam-3Enxam,Exbm-3Enxbm),
3) according to the transformer state parameters weighting value of setting, formula (4) adds similarity vector interior element
Weight group obtains the similarity M of grid equipment state to be assessed and typicalness.
M=Cloud (Ca1,Cb1)+2·Cloud(Ca2,Cb2)+2·Cloud(Ca3,Cb3)+3·Cloud(Ca4,Cb4) (4)
4) 200 groups of similarity M are sorted from small to large, obtains M1≤M2≤…≤M199≤M200.According to the similarity of setting
Set sizes take preceding 30 groups of data to be fitted into similarity set omega, expression are as follows:
Ω={ M1,M2,...,M30}
5) similarity for being subordinate to all kinds of typicalness in similarity set omega is screened, 4 typicalness similarity are constructed
Collect Ω1~Ω4, and calculate the probability P (S that grid equipment to be assessed is under the jurisdiction of each typicalnessi):
Three, interpretation of result and verifying
100 groups of transformer state data to be assessed are passed through above step to calculate, judge each transformation to be assessed
The state of device, and be compared with time of day.Table 1 provides the sample number that transformer to be assessed is under each typicalness, comments
Estimate correct sample number, assessment accuracy and the correct output probability mean value of assessment.Table 2 provides appraisal procedure of the present invention and shows
There is the comparison of appraisal procedure.
Grid equipment condition evaluation results of the table 1 based on timing information segment cloud similarity measurement
Transformer state | Sample number | Sentence positive number | Accuracy | Sentence positive average probability |
Normal condition | 25 | 22 | 88% | 80% |
Attention state | 25 | 19 | 76% | 77% |
Abnormality | 25 | 22 | 88% | 73% |
Severe conditions | 25 | 25 | 100% | 83% |
The appraisal procedure of the present invention of table 2 and existing appraisal procedure compare
Equipment evaluation method | Assess accuracy | Operation time/s |
Neural network | 78% | 30 |
Dynamic time warping | 65% | 10 |
The method of the present invention | 88% | 2 |
Simulation result shows that the grid equipment state evaluating method based on timing information segment cloud measuring similarity can have
Effect ground judges equipment status according to the timing information of grid equipment key parameter, to the judgment accuracy of each state
Also 75% or more, while the method for the present invention exports in the form of grid equipment is subordinate to each state probability as a result, back work
Personnel further identify and judge.
Existing appraisal procedure such as neural network and dynamic time warping are compared, the method for the present invention utilizes cloud characteristic manner,
The feature in timing information segment is extracted, carries out status assessment using characteristic variable, the existing number of data source itself can be overcome
The defects of according to excessive, data information redundancy, doping noise is measured, and greatly reduce the operation time of method, ensure that equipment state is commented
The real-time estimated.Using cloud measuring similarity and weighted sum, the spacing of each state parameter time series data can be accurately calculated
From and similarity between grid equipment state to be assessed and typicalness, improve correctness, the reliability of assessment.
Using appraisal procedure proposed by the present invention, solves data volume mistake existing for powernet monitoring time series data itself
Greatly, the defects of information island, data deviation and existing grid equipment state evaluating method can not be timely and accurately to equipment shape
The problem of state is judged.To realize accurate, the rapid evaluation of grid equipment state, mentioned for the safe and reliable operation of power grid
For supporting.
The above content is merely illustrative of the invention's technical idea, and this does not limit the scope of protection of the present invention, all to press
According to technical idea proposed by the present invention, any changes made on the basis of the technical scheme each falls within claims of the present invention
Protection scope within.
Claims (8)
1. a kind of grid equipment state evaluating method based on timing information segment cloud measuring similarity, which is characterized in that including
Following steps:
Step 1, the sample for collecting equipment to be assessed, each sample include multiple for describing the key parameters of grid equipment state
Data, each key parameters data of grid equipment form the measurement timing information segment that sequencing changes at any time,
Grid equipment is current or the time series data collection to be assessed of historic state for describing for multiple measurement timing information segments compositions
A;
Step 2 collects multiple key parameters data under grid equipment typicalness, each pass under grid equipment typicalness
Key parametric data forms a standard time sequence information segment, and multiple standard time sequence information segments form standard time series data set
B, standard time series data set B form the standard time sequence information collection U under corresponding typicalness, are with standard time sequence information collection U
Element integrates the standard time series information bank V under grid equipment typicalness;
Step 3, the characteristic information for extracting every measurement timing information segment in time series data collection A to be assessed, with extracting
The characteristic information of measurement timing information segment construct time series eigenmatrix Q to be assessed;
The characteristic information of the timing information segment of each element in step 4, extraction standard time serial message library V, building are typical
State for time sequence signature collection P;Using typicalness time series feature set P as element, integrate under grid equipment typicalness
Typical temporal aspect library T;
Step 5, calculated using cloud method for measuring similarity it is each right in temporal aspect matrix Q to be assessed and typical case's temporal aspect library T
Answer the similitude of element, the similarity of grid equipment measurement sample and each typicalness sample as to be assessed;
The similarity of step 6, the grid equipment state and each typicalness to be assessed that are obtained according to step 5, calculates electricity to be assessed
Net equipment is under the jurisdiction of the probability of each typicalness.
2. a kind of grid equipment status assessment side based on timing information segment cloud measuring similarity according to claim 1
Method, which is characterized in that in step 1, the time series data to be assessed integrates the size of A as d*n, and wherein d is grid equipment
The key parameter number of state, n are the length for measuring timing information segment;Timing letter in step 2, under the typicalness
Breath collection U={ B1,B2,…,Bi,…,BN, wherein 1≤i≤N, N are the sample number of every quasi-representative state, and B is typicalness
Time series data collection, the size of B are d*n;Standard time series information bank V={ U1,U2,…,Uj,…US, wherein 1≤j
≤ S, S are grid equipment typicalness type.
3. a kind of grid equipment status assessment side based on timing information segment cloud measuring similarity according to claim 1
Method, which is characterized in that in step 3, every measurement timing in time series data collection A to be assessed is extracted using timing characterizing method
The characteristic information of information segment;In step 4, using each element in timing characterizing method extraction standard time serial message library V
Timing information segment characteristic information, the calculation formula of timing characterizing method are as follows:
Ym=diff (Xm), Zm=diff (Ym);
Exm=mean (Xm),Eym=mean (Ym),Ezm=mean (Zm);
Cm=(Exm,Eym,Ezm,Enxm,Enym,Enzm,Hexm,Heym,Hezm);
Wherein, XmFor a timing information segment of input, YmFor XmFirst difference as a result, ZmFor XmSecond order difference result;
ExmFor XmExpectation, EymFor YmExpectation, EzmFor ZmExpectation;EnxmFor XmEntropy, EnymFor YmEntropy, EnzmFor ZmEntropy,
HexmFor XmSuper entropy, HeymFor YmSuper entropy, HezmFor ZmSuper entropy, diff be Difference Calculation function, mean is mean value computation
Function, var are that variance calculates function;CmFor the characteristic information of timing information segment, timing information segment includes measurement timing letter
Segment and standard time sequence information segment are ceased, includes 9 characterization factors, 1≤m≤d, d are the key parameter of grid equipment state
Number.
4. a kind of grid equipment status assessment side based on timing information segment cloud measuring similarity according to claim 3
Method, which is characterized in that in step 3, time series eigenmatrix Q size to be assessed is d*9, expression formula are as follows:
In step 4, typicalness time series feature set P={ L1,L2,…,Li,…,LN, wherein LiFor typicalness time sequence
Column eigenmatrix, 1≤i≤N, LiSize be d*9;Typical temporal aspect library T={ P under grid equipment typicalness1,
P2,…,Pi,…,PS, PiFor the typicalness time series feature set under different key parameters, 1≤i≤S.
5. a kind of grid equipment status assessment side based on timing information segment cloud measuring similarity according to claim 3
Method, which is characterized in that in step 5, the meter of similarity between the timing information segment measured and standard time sequence information segment
Calculate step are as follows:
Step 5.1, by the characteristic information C of every timing information segmentmDisassemble into 3 independent characteristic variable Cxm、CymAnd Czm,
Dismantling mode is as follows:
Step 5.2, calculate separately measurement timing information segment and standard time sequence information segment in the same key parameters it is only
Similarity between vertical characteristic variable, with CxamAnd CxbmFor, its calculation formula is:
Uminxm=min (Exam+3Enxam,Exbm+3Enxbm), Umaxxm=max (Exam+3Enxam,Exbm+3Enxbm),
Lminxm=min (Exam-3Enxam,Exbm-3Enxbm), Lmaxxm=max (Exam-3Enxam,Exbm-3Enxbm),
Wherein, CxamFor the independent characteristic variable for measuring timing information segment, CxbmFor the independent characteristic of standard time sequence information segment
Variable;UmaxxmFor measurement timing information segment and standard time sequence information segment independent characteristic variable upper bound maximum value,
UminxmFor the upper bound minimum value of measurement timing information segment and the independent characteristic variable of standard time sequence information segment;LmaxxmFor,
LminxmFor the lower bound maximum value of measurement timing information segment and the independent characteristic variable of standard time sequence information segment, timing is measured
The lower bound minimum value of the independent characteristic variable of information segment and standard time sequence information segment;s1And s2It is measurement timing information piece
Section corresponds to the overlapping point of cloud model with the independent characteristic variable of standard time sequence information segment;μ indicates that overlapping point belongs to two independences
The confidence level of characteristic variable;Dist represents the phase of the independent characteristic variable of measurement timing information segment and standard time sequence information segment
Like degree;
CyamAnd CybmBetween similarity and CzamAnd CzbmSimilarity and CxamAnd CxbmSimilarity calculation process it is identical;
The independent characteristic variable of step 5.3, the measurement timing information segment and standard time sequence information segment that are obtained according to step 5.2
Similarity, calculate measurement timing information segment and standard time sequence information segment cloud similarity Cloud:
Wherein, CamFor the characteristic information for measuring timing information segment, CbmFor the characteristic information of standard time sequence information segment;Step
5.4, the weight coefficient of each key parameters is set;By way of weighting recombination, measurement sample and the typical case of grid equipment are obtained
The similarity M of state sample:
Wherein, wmRepresent the weight coefficient of the key parameters of grid equipment.
6. a kind of grid equipment status assessment side based on timing information segment cloud measuring similarity according to claim 5
Method, which is characterized in that in step 5.4, according to the significance level of each parameter for describing grid equipment state, each key is set
The weight coefficient of parameter.
7. a kind of grid equipment status assessment side based on timing information segment cloud measuring similarity according to claim 1
Method, which is characterized in that step 6 the following steps are included:
Step 6.1 summarizes similarity between grid equipment to be assessed measurement sample and each typicalness sample, obtains S*N
Similarity, and sort according to the sequence of the numerical value of similarity from small to large;
Step 6.2, the similarity set omega for setting fixed size, and a similar of length (Ω) before sorting in step 6.1
Degree represents the similarity set sizes of setting according to the element as set omega, length (Ω);
Step 6.3 filters out the similarity for being subordinate to all kinds of typicalness in similarity set omega, constructs each typicalness similarity
Subset Ωj, and calculate the probability P (S that grid equipment to be assessed is under the jurisdiction of each typicalnessi), calculation formula is as follows:
Wherein, SjRepresent the S kind difference typicalness of grid equipment, length (Ωj) represent the big of typicalness similarity subset
It is small.
8. a kind of grid equipment status assessment side based on timing information segment cloud measuring similarity according to claim 1
Method, which is characterized in that in step 1, from the multiple passes for being used to describe grid equipment state that different detections or monitoring system are collected
Key parametric data.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910149813.XA CN109784777B (en) | 2019-02-28 | 2019-02-28 | Power grid equipment state evaluation method based on time sequence information fragment cloud similarity measurement |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910149813.XA CN109784777B (en) | 2019-02-28 | 2019-02-28 | Power grid equipment state evaluation method based on time sequence information fragment cloud similarity measurement |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109784777A true CN109784777A (en) | 2019-05-21 |
CN109784777B CN109784777B (en) | 2021-03-02 |
Family
ID=66486565
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910149813.XA Active CN109784777B (en) | 2019-02-28 | 2019-02-28 | Power grid equipment state evaluation method based on time sequence information fragment cloud similarity measurement |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109784777B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111639141A (en) * | 2020-05-26 | 2020-09-08 | 李云祥 | Data testing method and device and computer terminal |
CN112685473A (en) * | 2020-12-29 | 2021-04-20 | 山东大学 | Network abnormal flow detection method and system based on time sequence analysis technology |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103487250A (en) * | 2013-10-08 | 2014-01-01 | 中国矿业大学(北京) | Coal mining equipment predictive maintenance method based on two-dimensional projection |
CN103778575A (en) * | 2014-03-04 | 2014-05-07 | 国网浙江宁波市鄞州区供电公司 | Transformer state evaluation method and system |
CN105259495A (en) * | 2015-07-03 | 2016-01-20 | 四川大学 | High-voltage circuit breaker operation mechanism state evaluation method based on opening-closing coil current characteristic quantity optimization |
JP2016189062A (en) * | 2015-03-30 | 2016-11-04 | 有限責任監査法人トーマツ | Abnormality detection device, abnormality detection method and network abnormality detection system |
CN106597574A (en) * | 2016-12-30 | 2017-04-26 | 重庆邮电大学 | Weather temperature prediction method and device based on time-varying cloud model |
CN108491358A (en) * | 2018-03-21 | 2018-09-04 | 广东电网有限责任公司电力科学研究院 | A kind of bushing shell for transformer state evaluating method and device |
-
2019
- 2019-02-28 CN CN201910149813.XA patent/CN109784777B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103487250A (en) * | 2013-10-08 | 2014-01-01 | 中国矿业大学(北京) | Coal mining equipment predictive maintenance method based on two-dimensional projection |
CN103778575A (en) * | 2014-03-04 | 2014-05-07 | 国网浙江宁波市鄞州区供电公司 | Transformer state evaluation method and system |
JP2016189062A (en) * | 2015-03-30 | 2016-11-04 | 有限責任監査法人トーマツ | Abnormality detection device, abnormality detection method and network abnormality detection system |
CN105259495A (en) * | 2015-07-03 | 2016-01-20 | 四川大学 | High-voltage circuit breaker operation mechanism state evaluation method based on opening-closing coil current characteristic quantity optimization |
CN106597574A (en) * | 2016-12-30 | 2017-04-26 | 重庆邮电大学 | Weather temperature prediction method and device based on time-varying cloud model |
CN108491358A (en) * | 2018-03-21 | 2018-09-04 | 广东电网有限责任公司电力科学研究院 | A kind of bushing shell for transformer state evaluating method and device |
Non-Patent Citations (2)
Title |
---|
冯燕军: "《基于云模型的电力变压器风险评估及检修决策研究》", 《中国优秀硕士学位论文全文数据库 工程科技II辑》 * |
邓炳杰: "《面向不同寿命阶段的机电设备健康状态评价研究》", 《中国优秀硕士学位论文全文数据库 经济与管理科学辑》 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111639141A (en) * | 2020-05-26 | 2020-09-08 | 李云祥 | Data testing method and device and computer terminal |
CN111639141B (en) * | 2020-05-26 | 2023-08-15 | 湖北华中电力科技开发有限责任公司 | Data testing method and device and computer terminal |
CN112685473A (en) * | 2020-12-29 | 2021-04-20 | 山东大学 | Network abnormal flow detection method and system based on time sequence analysis technology |
Also Published As
Publication number | Publication date |
---|---|
CN109784777B (en) | 2021-03-02 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106777984B (en) | A method of photovoltaic array Working state analysis and fault diagnosis are realized based on density clustering algorithm | |
CN103617568B (en) | Setting method for abnormal data determination threshold in steady-state power quality early-warning mechanism | |
CN102944416B (en) | Multi-sensor signal fusion technology-based fault diagnosis method for wind turbine blades | |
CN109856515A (en) | A kind of direct current cables state of insulation judgment method and system | |
CN105095963A (en) | Method for accurately diagnosing and predicting fault of wind tunnel equipment | |
CN103810328A (en) | Transformer maintenance decision method based on hybrid model | |
CN109766952A (en) | Photovoltaic array fault detection method based on Partial Least Squares and extreme learning machine | |
CN109376801A (en) | Blade of wind-driven generator icing diagnostic method based on integrated deep neural network | |
CN103103570B (en) | Based on the aluminium cell condition diagnostic method of pivot similarity measure | |
CN108304931A (en) | A kind of Condition-based Maintenance of Substation Equipment method for diagnosing faults | |
CN105022021A (en) | State discrimination method for gateway electrical energy metering device based on the multiple agents | |
CN103886518A (en) | Early warning method for voltage sag based on electric energy quality data mining at monitoring point | |
CN109784777A (en) | Grid equipment state evaluating method based on timing information segment cloud measuring similarity | |
CN108562821A (en) | A kind of method and system determining Single-phase Earth-fault Selection in Distribution Systems based on Softmax | |
CN106768933A (en) | A kind of wind power generation unit blade method for diagnosing faults based on genetic algorithm | |
CN111612326A (en) | Comprehensive evaluation method for power supply reliability of distribution transformer | |
CN108287327A (en) | Metering automation terminal fault diagnostic method based on Bayes's classification | |
CN112288293A (en) | Comprehensive evaluation method for electric energy quality of large charging station | |
CN110378549A (en) | A kind of transmission tower bird pest grade appraisal procedure based on FAHP- entropy assessment | |
CN115684776A (en) | Method, device and processor for determining route loss branch of high-voltage transmission line | |
CN114169424A (en) | Discharge capacity prediction method based on k nearest neighbor regression algorithm and electricity utilization data | |
CN110443481B (en) | Power distribution automation terminal state evaluation system and method based on hybrid K-nearest neighbor algorithm | |
CN110096723B (en) | High-voltage switch cabinet insulation state analysis method based on operation and maintenance detection big data | |
CN112926686B (en) | BRB and LSTM model-based power consumption anomaly detection method and device for big power data | |
CN106644436B (en) | A kind of assessment method of breaker mechanic property |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
TR01 | Transfer of patent right | ||
TR01 | Transfer of patent right |
Effective date of registration: 20210830 Address after: No.69 Feitian Avenue, Airport Economic Development Zone, Jiangning District, Nanjing City, Jiangsu Province Patentee after: Nanjing Shoufeng Smart Power Research Institute Co.,Ltd. Address before: 710049 No. 28 West Xianning Road, Shaanxi, Xi'an Patentee before: XI'AN JIAOTONG University |