CN104239970B - A kind of conductor galloping method for prewarning risk based on Adaboost - Google Patents

A kind of conductor galloping method for prewarning risk based on Adaboost Download PDF

Info

Publication number
CN104239970B
CN104239970B CN201410448630.5A CN201410448630A CN104239970B CN 104239970 B CN104239970 B CN 104239970B CN 201410448630 A CN201410448630 A CN 201410448630A CN 104239970 B CN104239970 B CN 104239970B
Authority
CN
China
Prior art keywords
mrow
msub
mtd
conductor galloping
transmission line
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.)
Active
Application number
CN201410448630.5A
Other languages
Chinese (zh)
Other versions
CN104239970A (en
Inventor
梁允
熊小伏
周宁
翁世杰
王建
苑司坤
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Chongqing University
State Grid Corp of China SGCC
Henan Electric Power Research Institute
Original Assignee
Chongqing University
State Grid Corp of China SGCC
Henan Electric Power Research Institute
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Chongqing University, State Grid Corp of China SGCC, Henan Electric Power Research Institute filed Critical Chongqing University
Priority to CN201410448630.5A priority Critical patent/CN104239970B/en
Publication of CN104239970A publication Critical patent/CN104239970A/en
Application granted granted Critical
Publication of CN104239970B publication Critical patent/CN104239970B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06QDATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00Administration; Management
    • G06Q10/04Forecasting or optimisation, e.g. linear programming, "travelling salesman problem" or "cutting stock problem"

Abstract

The present invention provides a kind of conductor galloping method for prewarning risk based on Adaboost, comprises the following steps:The internal cause of conductor galloping is classified, and the Meteorological Characteristics factor under accident is waved to power transmission line history by classification results and is counted;According to the transmission line information being predicted, selection and corresponding one kind in conductor galloping internal cause classification results, the Meteorological Characteristics factor information waved with history in such under emergency conditions records composing training sample set, grader is formed with Adaboost Ensemble Learning Algorithms, again using the forecast data of conductor galloping Meteorological Characteristics factor as input, transmission line galloping early warning result is obtained by grader;According to early warning result, judgement obtains conductor galloping Risk-warning grade.The present invention has considered the internal cause and external cause for influenceing conductor galloping, makes full use of power transmission line history to wave information and Weather Forecast Information, more meets reality;And algorithm generalization ability used is strong, easily coding, early warning result precision is high.

Description

A kind of conductor galloping method for prewarning risk based on Adaboost
Technical field
The invention belongs to the failure risk early warning technology field of the overhead transmission line of power system, specifically a kind of base In the conductor galloping method for prewarning risk of Adaboost algorithm.
Background technology
Conductor galloping be guide line wind-force and (or) caused low frequency in the presence of asymmetric icing, significantly Self-excited vibration, it is a kind of Aerodynamic Instability phenomenon.Conductor galloping mostly occurs in the winter time, and its energy is very big, and the duration It is long, mechanical failure and electric fault are easily caused to transmission line of electricity, gently then causes alternate flashover, damage wires, ground wire and gold utensil etc., It is heavy then cause stranded, broken string, even fall severe accident, the serious threat such as tower and the safe and stable operation of transmission line of electricity.
Operation and observation and statistics show that China is one of country of conductor galloping disaster most serious.With China The development of power network scale and it is boisterous frequently occur, the occurrence frequency and the extent of injury of conductor galloping accident have substantially Increase, and conductor galloping region is not limited in a small number of scopes, over to the most area of national grid yet.Cause This, the research to conductor galloping and its precautionary measures has important theory significance and engineering practical value.
In the last few years, domestic and foreign scholars were to conductor galloping excitation mechanism, computer sim- ulation and power transmission line Anti-Oscillation Measures etc. Many-sided research has been carried out, many important achievements has been achieved and is applied to Practical Project.Unfortunately, due to power transmission line and air-flow Geometrical non-linearity caused by coupling and significantly the moving of power transmission line caused by interaction etc., makes conductor galloping problem Become sufficiently complex, there is no unified, pervasive conductor galloping excitation theory so far.
Existing power transmission line Anti-Oscillation Measures, which sum up, can be divided into three major types:First, considering from meteorological condition, avoid easily In the icing region that formation is waved and line alignment;Second, improve the anti-dance of line system with electric angle, design from mechanical Kinetic force;Third, taking various anti-dancing devices, suppress the generation waved, for built circuit, even more unique feasible is done Method.It will be appreciated, however, that current conductor galloping defensive measure still has following deficiency:
1) consider to save the economic design requirement of line corridor and the factors such as cheap property of constructing so that transmit electricity part Circuit can not be avoided waving area completely;
2) in actual applications, power transmission line quality enhancement techniques and it is anti-dance design still not full and accurate enough and specification, economy with And operability is poor, while also lack practical experience;
3) anti-dance device is to develop to obtain based on different conductor galloping mechanism, causes several anti-dances that application is more at present to fill Put, all with its obvious design feature and application limitation, there is also very big difference for anti-dance effect.
Obviously, wanting uniform conductor galloping completely also needs the effort of a very long time, main there is an urgent need to one kind at present Dynamic property is stronger, the wider array of power transmission line Anti-galloping aid decision-making method of application, and disaster is waved mitigate that transmission system is subjected to. Conductor galloping on-Line Monitor Device and method are flourished in recent years, and conductor galloping on-line monitoring system into Work(operates to the meteorological data that research conductor galloping have accumulated preciousness;In addition, weather forecast in recent years become more meticulous degree with It is more and more closer with power network cooperation and the degree of accuracy all has a distinct increment, make to realize conductor galloping risk using weather information Early warning is a kind of feasible, science approach.
The content of the invention
For the deficiency of the existing measure of above-mentioned analysis, the invention provides a kind of conductor galloping based on Adaboost Method for prewarning risk, it can realize to structural parameters of the forecast information of conductor galloping Meteorological Characteristics factor and transmission line of electricity etc. The calculating processing of related data, and export the conductor galloping disaster alarm analysis result of region.
In order to solve the technical problem, the technical solution adopted by the present invention is:
A kind of conductor galloping method for prewarning risk based on Adaboost, comprises the following steps;
(1), the internal cause of conductor galloping is classified, and waves the gas under accident to power transmission line history by classification results As characteristic factor is counted;
Wherein internal cause classification results include:1. the type of circuit, is divided into single conductor and split conductor;2. the section of circuit Product, is divided into large, medium and small section circuit;3. line span, it is divided into large, medium and small span circuit;Meteorological Characteristics factor includes:① Wind speed;2. wind direction is to the angle of wire axial direction;3. temperature;4. relative humidity;
(2), according to the transmission line information that is predicted, selection with corresponding one in conductor galloping internal cause classification results Class, and Meteorological Characteristics factor information under emergency conditions is waved with history in such and records composing training sample set, determined with individual layer Plan tree forms strong classifier as Weak Classifier, using Adaboost Ensemble Learning Algorithms, then pre- by being obtained from meteorological department Count off obtains transmission line galloping early warning output result and confidence level margin values, pre- count off according to as input by grader According to Meteorological Characteristics include:1. wind speed;2. wind direction is to the angle of wire axial direction;3. temperature;4. relative humidity;
(3), the early warning output result according to step (2), according to the confidence level of conductor galloping prediction result Margin values, judgement obtain conductor galloping Risk-warning grade;
The step of wherein described (2), is specific as follows:
2.1), input:Training sample set X, wherein sample class label should be included;N is number of training;T is training time Number, namely Weak Classifier number;Wherein Weak Classifier sorting algorithm is C, and sample class label is yi={ -1,1 }, wherein 1 table Show that power transmission line is waved, -1 expression power transmission line is not waved;
2.2), initialize:Sample weights are distributed w1(i)=1/N, i=1,2 ..., N;
2.3) t=1,2, is worked as ..., T:
1. it is distributed w according to the sample weights of the t timest(i) sampling put back to is carried out from original sample collection X, generation is new Sample set Xt
2. in XtUpper training Weak Classifier Ct(X), and C is usedt(X) original sample collection X is classified;
3. calculate Weak Classifier Ct(X) classification error rate;
In formula (1), work as Ct(xi)≠yiWhen, I () is 1, and remaining is then 0;
4. calculate Weak Classifier Ct(X) coefficient;
5. update weights distribution;
In formula (3),It is normalization factor so that
2.4), final classification device:
In formula (4), function sgn () is sign function, and its specific mathematic(al) representation is
Then conductor galloping early warning output includes grader prediction result y and confidence level margin (x, y):
In formula, x is the forecast data of conductor galloping Meteorological Characteristics factor;Y ∈ { -1 ,+1 }, it is grader prediction result: 1 is predicts that the power transmission line will be waved, and -1 is to predict that the power transmission line will not be waved;Margin ∈ [- 1 ,+1], margin Closer+1 represents to predict that the confidence level that the circuit is waved is higher, and margin closer -1 represents to predict that the circuit does not occur The confidence level waved is higher, and margin represents that the confidence level of prediction result is relatively low closer to 0.
Because meteorological condition is the most important extraneous factor that influence conductor galloping excites, when internal cause is relatively constant, thing The change of thing will be determined by external cause.In order to make up that existing conductor galloping defensive measure application is narrow, initiative is poor and to gas Image information lacks the deficiencies of utilizing, and the present invention proposes a kind of conductor galloping method for prewarning risk from operation angle, and the present invention will Conductor galloping early warning problem is attributed to the classification forecasting problem under supervised learning, and power transmission line is gone through by conductor galloping internal cause Meteorological Characteristics factor information in the case of history is waved carries out disaggregatedly statistical disposition, passes through Adaboost Ensemble Learning Algorithms Strong classifier is established, COMPREHENSIVE CALCULATING handles the related datas such as the location parameter of Weather Forecast Information and transmission line of electricity, where providing Region transmission line of electricity waves risk class, realizes the conductor galloping disaster alarm of science.Early warning result can be operation of power networks Dispatcher carry out Anti-Oscillation Measures provide science decision support, carry out in advance targetedly wind resistance, except ice-melt etc. is transmitted electricity Line Anti-Oscillation Measures, avoid strick precaution deficiency from causing accident and excessively take precautions against waste of resource, transmission line galloping failure can be reduced Rate, ensure the safe and stable operation of transmission line of electricity.
It is regional present invention is generally applicable to power system conductor galloping early warning, particularly conductor galloping Frequent Accidents, Compared with prior art, the invention has the advantages that:
1) present invention has considered the internal cause and external cause for influenceing conductor galloping, makes full use of power transmission line history to wave letter Breath, more meets reality;
2) present invention use Adaboost Ensemble Learning Algorithms, with generalization ability (i.e. from sample data learning to Rule can be applied to the ability of new data) strong, the advantages that easily encoding, early warning result reliability is high.
Brief description of the drawings
Fig. 1 is the conductor galloping Risk-warning flow chart based on Adaboost;
Fig. 2 is decision-making pile (Weak Classifier) flow chart for splitting criterion based on Gini.
Embodiment
Because the physical model of existing conductor galloping is not accurate enough, and the Some Parameters in model are difficult on actual track To be obtained by measuring so that it is relatively low using the practicality and accuracy of physical model progress conductor galloping early warning, and now Machine Learning Theory is just to we provide good method for early warning.Machine learning is obtained accurate based on observation before Prediction, it provide it is a kind of obtain the rule that can not still be obtained by principle analysis at present from observation data, and then utilize The method of these law forecasting Future Datas.
The present invention proposes a kind of conductor galloping method for prewarning risk based on Adaboost algorithm, described Adaboost algorithm is adaptive enhancing (Adaptive boosting, abbreviation Adaboost) algorithm in Ensemble Learning Algorithms, Its basic thought is to utilize substantial amounts of classification capacity in general Weak Classifier, is superimposed, gathered by certain method, forms one The stronger final classification device (strong classifier) of individual classification capacity.The present invention can be realized to conductor galloping Meteorological Characteristics factor The calculating processing of the related datas such as the structural parameters of forecast information and transmission line of electricity, and export the conductor galloping calamity of region Evil early warning analysis result.
With reference to embodiment and accompanying drawing, the inventive method is done and further clearly and completely described, but the reality of the present invention The mode of applying is not limited to this.
As shown in figure 1, the present invention comprises the following steps:
(1) (conductor galloping internal cause refers to influence what conductor galloping excited the internal cause for, exciting influence conductor galloping Internal factor) --- line construction is classified with parameter as shown in table 1, altogether 18 kinds of combinations, such as:Single conductor, small bore, Small span circuit combines to be a kind of.
The conductor galloping internal cause categorised statistical form of table 1
And Meteorological Characteristics factor information this external cause is waved to power transmission line history and carries out statistics classification, conductor galloping is meteorological Characteristic factor refers to the external meteorological factor that influence conductor galloping excites, and conductor galloping Meteorological Characteristics factor mainly includes wind Speed, wind direction are to wire axis angle, temperature and relative humidity.Significantly, since the change of height can have shadow to wind speed Ring, and it is usually to be defaulted as the high wind speed of liftoff 10m that obtained wind speed and weather forecast wind speed are observed in weather station, therefore should be by Following formula is unified to be converted to the wind speed v at conductor heightl
In formula, vqFor liftoff 10m eminences wind speed, H is conductor height (m);μ is ground roughness exponent, according to marine, township The ground roughness exponent in village, city and the class of big city downtown 4 is respectively 0.12,0.15,0.22 and 0.30.
(2), according to the actual conditions of tested transmission line of electricity, corresponding one kind in above-mentioned 18 kinds of combinations is selected, with step (1) the power transmission line history described in waves the Meteorological Characteristics factor information record composing training sample set under accident, with this implementation The following Adaboost Ensemble Learning Algorithms of example form strong classification learning device, and the Meteorological Characteristics factor provided according to meteorological department Forecast data carry out conductor galloping early warning.It is specific as follows:
2.1), input:Conductor galloping training sample set X={ (x1,y1),(x2,y2),…,(xN,yN)};Wherein, xiFor The Meteorological Characteristics factor vector of i-th of conductor galloping sample;yi={ -1,1 } represent the category label of i-th of sample:- 1 table Show that power transmission line is not waved, 1 expression power transmission line is waved;N is number of training;T is frequency of training (namely Weak Classifier Number).Weak Classifier sorting algorithm C uses existing method individual layer decision tree, can also use what SVMs etc. can be realized Existing method.
2.2), initialize:Sample weights are distributed w1(i)=1/N, i=1,2 ..., N.
2.3) t=1,2, is worked as ..., T:
1. it is distributed w according to the sample weights of the t timest(i) sampling put back to is carried out from original sample collection X, generation is new Sample set Xt
2. in XtUpper training Weak Classifier Ct(X), and C is usedt(X) original sample collection X is classified;
3. calculate Weak Classifier Ct(X) classification error rate;
In formula (2), work as Ct(xi)≠yiWhen, I () is 1, and remaining is then 0.
4. calculate Weak Classifier Ct(X) coefficient;
5. update weights distribution;
In formula (4),It is normalization factor so that
2.4), final classification device:
In formula (5), function sgn () is sign function, and its specific mathematic(al) representation is
Further, the conventional individual layer decision tree of step (2) the Weak Classifier selection, the decision tree are based only upon single input Feature simultaneously does decision-making using threshold value division methods, i.e. an only node, and because this sets only once fission process, with Stub is similar to, therefore it is also referred to as decision-making pile.Decision-making pile construction the most critical issue be how judgment threshold division result Quality, to select optimal partition point.The present embodiment assesses the good and bad degree of segmentation rule using Gini impurity levels index, right In the data set S for including c classification, it is defined as follows:
In formula, pjRepresent the ratio shared by classification j sample in set S.If split regular rule is divided into S by S1And S2 Two subsets, then the regular Gini assessed values be designated as:
In formula, n1For subset S1Number of samples, n2For subset S2Number of samples, n be set S number of samples.
For a Numeric Attributes, the two categorised decision stakes segmentation thought based on Gini indexes is:Travel through it is all can After the dividing method of energy, selection makes assessed value gini (S, rule) reach the regular as the optimal dividing at this node of minimum. Its flow as shown in Figure 2, is described as:
1) sample value of logarithm value type attribute is ranked up, it is assumed that the result after sequence is (x1,y1), (x2,y2) ..., (xn,yn);yi(i=1,2 ..., n) it is category label.
2) due to segmentation only occur in two data points between, so generally taking midpoint (xi+xi+1)/2 are used as cut-point, so Take different cut-points successively from small to large afterwards, and the gini values of each segmentation rule are calculated by formula (7) and formula (8).
3) point for making gini values minimum is taken as optimal partition point, and constructs decision-making pile by threshold value of the cut-point.
Further, the conductor galloping early warning output described in step (2) includes grader prediction result y and confidence level margin(x,y):
In formula, x is the forecast data of conductor galloping Meteorological Characteristics factor;Y ∈ { -1 ,+1 }, it is grader prediction result: 1 will wave for prediction power transmission line, and -1 will not wave for prediction power transmission line;Margin ∈ [- 1 ,+1], larger positive side Boundary is (i.e. nearer from 1) to be represented to predict that the circuit waves negative edge with a high credibility, larger expression (i.e. nearer from -1) in advance Survey the circuit and do not wave with a high credibility, less border (i.e. nearer from 0) then represents that the confidence level of prediction result is relatively low.
(3), the early warning output result according to step (2), judgement obtain conductor galloping Risk-warning grade.
According to the margin values of conductor galloping prediction result, the Risk-warning of conductor galloping can be judged shown according to the form below Grade.
The conductor galloping warning grade table of table 2
Above-described embodiment is the preferable embodiment of the present invention, but embodiments of the present invention are not by the embodiment Limitation, other any Spirit Essences without departing from the present invention with made under principle change, modification, replacement, combine, simplification, Equivalent substitute mode is should be, is included within protection scope of the present invention.

Claims (5)

1. a kind of conductor galloping method for prewarning risk based on Adaboost, it is characterised in that comprise the following steps;
(1), the internal cause of conductor galloping is classified, and the meteorology waved by classification results to power transmission line history under accident is special Sign factor is counted;
Wherein internal cause classification results include:1. the type of circuit, is divided into single conductor and split conductor;2. the sectional area of circuit, It is divided into large, medium and small section circuit;3. line span, it is divided into large, medium and small span circuit;
Meteorological Characteristics factor includes:1. wind speed;2. wind direction is to the angle of wire axial direction;3. temperature;4. relative humidity;
(2), according to the transmission line information being predicted, selection and corresponding one kind in conductor galloping internal cause classification results, and The Meteorological Characteristics factor information waved with history in such under emergency conditions records composing training sample set, is made with individual layer decision tree For Weak Classifier, strong classifier, then the forecast data that will be obtained from meteorological department are formed using Adaboost Ensemble Learning Algorithms As input, transmission line galloping early warning output result and confidence level margin values, the gas of forecast data are obtained by grader As feature includes:1. wind speed;2. wind direction is to the angle of wire axial direction;3. temperature;4. relative humidity;
(3), the early warning output result according to step (2), according to the confidence level margin values of conductor galloping prediction result, Judgement obtains conductor galloping Risk-warning grade;
Described step (2) is specific as follows:
2.1), input:Training sample set X, wherein sample class label should be included;N is number of training;T is frequency of training, That is Weak Classifier number;Wherein Weak Classifier sorting algorithm is C, and sample class label is yi={ -1,1 }, wherein 1 represents transmission of electricity Line is waved, and -1 expression power transmission line is not waved;
2.2), initialize:Sample weights are distributed w1(i)=1/N, i=1,2 ..., N;
2.3) t=1,2, is worked as ..., T:
1. it is distributed w according to the sample weights of the t timest(i) sampling put back to is carried out from original sample collection X, generates new sample Set Xt
2. in XtUpper training Weak Classifier Ct(X), and C is usedt(X) original sample collection X is classified;
3. calculate Weak Classifier Ct(X) classification error rate;
<mrow> <msub> <mi>&amp;epsiv;</mi> <mi>t</mi> </msub> <mo>=</mo> <munderover> <mo>&amp;Sigma;</mo> <mrow> <mi>i</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>N</mi> </munderover> <msub> <mi>w</mi> <mi>t</mi> </msub> <mrow> <mo>(</mo> <mi>i</mi> <mo>)</mo> </mrow> <mi>I</mi> <mrow> <mo>(</mo> <msub> <mi>C</mi> <mi>t</mi> </msub> <mo>(</mo> <msub> <mi>x</mi> <mi>i</mi> </msub> <mo>)</mo> <mo>)</mo> </mrow> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>1</mn> <mo>)</mo> </mrow> </mrow>
In formula (1), work as Ct(xi)≠yiWhen, I () is 1, and remaining is then 0;
4. calculate Weak Classifier Ct(X) coefficient;
<mrow> <msub> <mi>a</mi> <mi>t</mi> </msub> <mo>=</mo> <mfrac> <mn>1</mn> <mn>2</mn> </mfrac> <mi>l</mi> <mi>n</mi> <mrow> <mo>(</mo> <mfrac> <mrow> <mn>1</mn> <mo>-</mo> <msub> <mi>&amp;epsiv;</mi> <mi>t</mi> </msub> </mrow> <msub> <mi>&amp;epsiv;</mi> <mi>t</mi> </msub> </mfrac> <mo>)</mo> </mrow> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>2</mn> <mo>)</mo> </mrow> </mrow>
5. update weights distribution;
<mrow> <msub> <mi>w</mi> <mrow> <mi>t</mi> <mo>+</mo> <mn>1</mn> </mrow> </msub> <mrow> <mo>(</mo> <mi>i</mi> <mo>)</mo> </mrow> <mo>=</mo> <mfrac> <mrow> <msub> <mi>w</mi> <mi>t</mi> </msub> <mrow> <mo>(</mo> <mi>i</mi> <mo>)</mo> </mrow> </mrow> <msub> <mi>Z</mi> <mi>t</mi> </msub> </mfrac> <mo>&amp;times;</mo> <mfenced open = "{" close = ""> <mtable> <mtr> <mtd> <mrow> <msup> <mi>e</mi> <mrow> <mo>-</mo> <msub> <mi>a</mi> <mi>t</mi> </msub> </mrow> </msup> <mo>,</mo> </mrow> </mtd> <mtd> <mrow> <msub> <mi>C</mi> <mi>t</mi> </msub> <mrow> <mo>(</mo> <msub> <mi>x</mi> <mi>i</mi> </msub> <mo>)</mo> </mrow> <mo>=</mo> <msub> <mi>y</mi> <mi>i</mi> </msub> </mrow> </mtd> </mtr> <mtr> <mtd> <mrow> <msup> <mi>e</mi> <msub> <mi>a</mi> <mi>t</mi> </msub> </msup> <mo>,</mo> </mrow> </mtd> <mtd> <mrow> <msub> <mi>C</mi> <mi>t</mi> </msub> <mrow> <mo>(</mo> <msub> <mi>x</mi> <mi>i</mi> </msub> <mo>)</mo> </mrow> <mo>&amp;NotEqual;</mo> <msub> <mi>y</mi> <mi>i</mi> </msub> </mrow> </mtd> </mtr> </mtable> </mfenced> <mo>=</mo> <mfrac> <mrow> <msub> <mi>w</mi> <mi>t</mi> </msub> <mrow> <mo>(</mo> <mi>i</mi> <mo>)</mo> </mrow> </mrow> <msub> <mi>Z</mi> <mi>t</mi> </msub> </mfrac> <mo>&amp;CenterDot;</mo> <msup> <mi>e</mi> <mrow> <mo>-</mo> <msub> <mi>a</mi> <mi>t</mi> </msub> <msub> <mi>y</mi> <mi>i</mi> </msub> <msub> <mi>C</mi> <mi>t</mi> </msub> <mrow> <mo>(</mo> <msub> <mi>x</mi> <mi>i</mi> </msub> <mo>)</mo> </mrow> </mrow> </msup> <mo>,</mo> <mi>i</mi> <mo>=</mo> <mn>1</mn> <mo>,</mo> <mn>2</mn> <mo>,</mo> <mn>...</mn> <mo>,</mo> <mi>N</mi> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>3</mn> <mo>)</mo> </mrow> </mrow>
In formula (3),It is normalization factor so that
2.4), final classification device:
<mrow> <mi>C</mi> <mrow> <mo>(</mo> <mi>X</mi> <mo>)</mo> </mrow> <mo>=</mo> <mi>sgn</mi> <mo>&amp;lsqb;</mo> <munderover> <mo>&amp;Sigma;</mo> <mrow> <mi>t</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>T</mi> </munderover> <msub> <mi>a</mi> <mi>t</mi> </msub> <msub> <mi>C</mi> <mi>t</mi> </msub> <mrow> <mo>(</mo> <mi>X</mi> <mo>)</mo> </mrow> <mo>&amp;rsqb;</mo> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>4</mn> <mo>)</mo> </mrow> </mrow>
In formula (4), function sgn () is sign function, and its specific mathematic(al) representation is
<mrow> <mi>sgn</mi> <mrow> <mo>(</mo> <mi>x</mi> <mo>)</mo> </mrow> <mo>=</mo> <mo>{</mo> <mtable> <mtr> <mtd> <mrow> <mn>1</mn> <mo>,</mo> </mrow> </mtd> <mtd> <mrow> <mi>x</mi> <mo>&amp;GreaterEqual;</mo> <mn>0</mn> </mrow> </mtd> </mtr> <mtr> <mtd> <mrow> <mo>-</mo> <mn>1</mn> <mo>,</mo> </mrow> </mtd> <mtd> <mrow> <mi>x</mi> <mo>&lt;</mo> <mn>0</mn> </mrow> </mtd> </mtr> </mtable> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>5</mn> <mo>)</mo> </mrow> </mrow>
Then conductor galloping early warning output includes grader prediction result y and confidence level margin (x, y):
<mrow> <mi>y</mi> <mo>=</mo> <mi>C</mi> <mrow> <mo>(</mo> <mi>x</mi> <mo>)</mo> </mrow> <mo>=</mo> <mi>sgn</mi> <mo>&amp;lsqb;</mo> <munderover> <mo>&amp;Sigma;</mo> <mrow> <mi>t</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>T</mi> </munderover> <msub> <mi>a</mi> <mi>t</mi> </msub> <msub> <mi>C</mi> <mi>t</mi> </msub> <mrow> <mo>(</mo> <mi>x</mi> <mo>)</mo> </mrow> <mo>&amp;rsqb;</mo> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>6</mn> <mo>)</mo> </mrow> </mrow>
<mrow> <mi>m</mi> <mi>arg</mi> <mi>i</mi> <mi>n</mi> <mrow> <mo>(</mo> <mi>x</mi> <mo>,</mo> <mi>y</mi> <mo>)</mo> </mrow> <mo>=</mo> <mfrac> <mrow> <mi>y</mi> <munder> <mo>&amp;Sigma;</mo> <mi>t</mi> </munder> <msub> <mi>a</mi> <mi>t</mi> </msub> <msub> <mi>C</mi> <mi>t</mi> </msub> <mrow> <mo>(</mo> <mi>x</mi> <mo>)</mo> </mrow> </mrow> <mrow> <munder> <mo>&amp;Sigma;</mo> <mi>t</mi> </munder> <mrow> <mo>|</mo> <msub> <mi>a</mi> <mi>t</mi> </msub> <mo>|</mo> </mrow> </mrow> </mfrac> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>7</mn> <mo>)</mo> </mrow> </mrow>
In formula, x is the forecast data of conductor galloping Meteorological Characteristics factor;Y ∈ { -1 ,+1 }, it is grader prediction result:1 is Predict that the power transmission line will be waved, -1 is to predict that the power transmission line will not be waved;Margin ∈ [- 1 ,+1], margin is got over Represent to predict that the confidence level that the circuit is waved is higher close to+1, margin closer -1 represents to predict that the circuit is not waved Dynamic confidence level is higher, and margin represents that the confidence level of prediction result is relatively low closer to 0.
CN201410448630.5A 2014-09-04 2014-09-04 A kind of conductor galloping method for prewarning risk based on Adaboost Active CN104239970B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201410448630.5A CN104239970B (en) 2014-09-04 2014-09-04 A kind of conductor galloping method for prewarning risk based on Adaboost

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201410448630.5A CN104239970B (en) 2014-09-04 2014-09-04 A kind of conductor galloping method for prewarning risk based on Adaboost
PCT/CN2014/092740 WO2016033883A1 (en) 2014-09-04 2014-12-02 Power transmission line gallop risk early-warning method based on adaboost

Publications (2)

Publication Number Publication Date
CN104239970A CN104239970A (en) 2014-12-24
CN104239970B true CN104239970B (en) 2017-11-28

Family

ID=52227992

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201410448630.5A Active CN104239970B (en) 2014-09-04 2014-09-04 A kind of conductor galloping method for prewarning risk based on Adaboost

Country Status (2)

Country Link
CN (1) CN104239970B (en)
WO (1) WO2016033883A1 (en)

Families Citing this family (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105974219B (en) * 2016-04-11 2021-01-15 中国电力科学研究院 Classification and identification method for energy-saving electrical appliance load types
CN106779274B (en) * 2016-04-20 2020-07-07 海南电力技术研究院 Typhoon risk early warning method and system for power equipment
CN106129944B (en) * 2016-06-24 2017-11-28 国网河南省电力公司电力科学研究院 A kind of horizontally disposed transmission line of electricity anti-dance approach of extra-high voltage
CN106338708B (en) * 2016-08-30 2020-04-24 中国电力科学研究院 Electric energy metering error analysis method combining deep learning and recurrent neural network
CN106503751A (en) * 2016-11-10 2017-03-15 国网河南省电力公司电力科学研究院 A kind of power transmission line Louis dance potential prediction method based on SVM classifier
CN106529837B (en) * 2016-12-13 2020-09-01 国网新疆电力公司电力科学研究院 Correlation method of wind data and transmission line body operation data
CN107045638B (en) * 2016-12-30 2021-03-30 中国民航管理干部学院 Flight safety event analysis method based on situational awareness model
CN107436162B (en) * 2017-07-31 2019-07-12 国网湖南省电力有限公司 A kind of power network line waves Occurrence forecast method and system
CN107491839B (en) * 2017-08-17 2020-09-01 国网湖南省电力有限公司 Power grid galloping forecasting method and system based on historical galloping characteristics
CN107818339A (en) * 2017-10-18 2018-03-20 桂林电子科技大学 Method for distinguishing is known in a kind of mankind's activity
CN107526083B (en) * 2017-10-18 2019-05-31 国网新疆电力公司电力科学研究院 A kind of strong convection wind scale prediction technique based on weather radar data
CN108898765A (en) * 2018-05-11 2018-11-27 国网湖北省电力有限公司检修公司 A kind of overhead transmission line external force damage alarm method based on analysis of vibration signal
CN108831115B (en) * 2018-06-22 2020-11-06 国网湖南省电力有限公司 Adaboost-based power transmission line rainstorm disaster risk early warning method
CN109492756A (en) * 2018-11-19 2019-03-19 中国气象局公共气象服务中心 More element conductor galloping method for early warning and relevant apparatus based on deep learning
CN112269907A (en) * 2020-11-02 2021-01-26 山东万里红信息技术有限公司 Processing method of health big data of Internet of things

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102095449A (en) * 2010-10-28 2011-06-15 华南理工大学 Method for alarming dancing of overhead transmission circuit
CN103245379A (en) * 2013-04-07 2013-08-14 上海申瑞继保电气有限公司 Method for monitoring real-time status of high voltage transmission line
CN103559557A (en) * 2013-11-01 2014-02-05 国家电网公司 Gallop warning method and system based on electric transmission line of electrical power system
CN203595550U (en) * 2013-10-09 2014-05-14 云南电网公司玉溪供电局 Power transmission line aeolian vibration safety early warning system

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB2209911B (en) * 1987-09-17 1991-08-21 Joseph Luis Cruz Bite indicator
CN203502996U (en) * 2013-11-01 2014-03-26 国家电网公司 System for alarming dancing based on power transmission line of electric power system

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102095449A (en) * 2010-10-28 2011-06-15 华南理工大学 Method for alarming dancing of overhead transmission circuit
CN103245379A (en) * 2013-04-07 2013-08-14 上海申瑞继保电气有限公司 Method for monitoring real-time status of high voltage transmission line
CN203595550U (en) * 2013-10-09 2014-05-14 云南电网公司玉溪供电局 Power transmission line aeolian vibration safety early warning system
CN103559557A (en) * 2013-11-01 2014-02-05 国家电网公司 Gallop warning method and system based on electric transmission line of electrical power system

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
ADABOOST+: An Ensemble Learning Approach for Estimating Weather-Related Outages in Distribution Systems;Padmavathy Kankanala etc.;《IEEE Transactions on Power Systems》;20140131;第29卷(第1期);正文第359页左栏第1节,第360页左栏第1节,第362页第4节,第363页第5节 *

Also Published As

Publication number Publication date
WO2016033883A1 (en) 2016-03-10
CN104239970A (en) 2014-12-24

Similar Documents

Publication Publication Date Title
CN104239970B (en) A kind of conductor galloping method for prewarning risk based on Adaboost
Yilmaz et al. A statistical approach to estimate the wind speed distribution: the case of Gelibolu region
CN106650767B (en) Flood forecasting method based on cluster analysis and real-time correction
CN105701596A (en) Method for lean distribution network emergency maintenance and management system based on big data technology
Wang et al. Early warning method for transmission line galloping based on SVM and AdaBoost bi-level classifiers
CN102789447B (en) Based on the icing of grey multiple linear regression and the analytical approach of meteorological relation
Yuan et al. Predicting traffic accidents through heterogeneous urban data: A case study
CN103488869A (en) Wind power generation short-term load forecast method of least squares support vector machine
CN103530527A (en) Wind power probability forecasting method based on numerical weather forecasting ensemble forecasting results
CN106503751A (en) A kind of power transmission line Louis dance potential prediction method based on SVM classifier
CN103278326A (en) Method for diagnosing faults of wind generating set gear case
Gong et al. Special issue on meteorological disaster risk analysis and assessment: on basis of grey systems theory
CN103902837A (en) Method for wind speed prediction based on experience Copula function
CN104318503A (en) System and method for rainfall forecasting according to typhoons
CN102147839A (en) Method for forecasting photovoltaic power generation quantity
CN105912857A (en) Selection and configuration method of distribution equipment state monitoring sensors
CN105184067A (en) Isolator pollution flashover state fuzzy evaluation method
CN104835073A (en) Unmanned aerial vehicle control system operation performance evaluating method based on intuitionistic fuzzy entropy weight
CN108831115B (en) Adaboost-based power transmission line rainstorm disaster risk early warning method
CN106682776A (en) Fine forecasting and early warning method and system for dancing of overhead transmission line
CN106127331A (en) Civil aviaton based on rough set theory radio interference Forecasting Methodology
CN105303268A (en) Wind power generation output power prediction method based on similarity theory
CN104036330A (en) Rainfall classification prediction method based on MapReduce
Heckenbergerová et al. Estimation of wind direction distribution with genetic algorithms
CN103473476A (en) Wind energy resource calculation method based on wind measurement data of several intra-area wind measuring towers

Legal Events

Date Code Title Description
PB01 Publication
C06 Publication
SE01 Entry into force of request for substantive examination
C10 Entry into substantive examination
CB02 Change of applicant information

Address after: 450052 Songshan, Zhengzhou, Henan District No. 27 South Road, No. 85

Applicant after: Electric Power Research Institute, State Grid Henan Electric Power Company

Applicant after: Chongqing University

Applicant after: State Grid Corporation of China

Address before: 100031 Xicheng District West Chang'an Avenue, No. 86, Beijing

Applicant before: State Grid Corporation of China

Applicant before: Electric Power Research Institute, State Grid Henan Electric Power Company

Applicant before: Chongqing University

CB02 Change of applicant information
GR01 Patent grant
GR01 Patent grant