CN109948466A - A kind of identification exchanges the method and system of super UHV transmission line audible noise abnormal data - Google Patents

Abstract

The present invention provides a kind of method and system of the super UHV transmission line audible noise abnormal data of identification exchange.Described method and system first calculates at determining i environment sensitive target and the Euclidean distance data of conducting wire, and generates Euclidean distance data matrix based on the Euclidean distance data；Again by constructing data matrix to be clustered at equivalent continuous A sound level data matrix and environment sensitive target and the splicing of the Euclidean distance data matrix of conducting wire；Then cluster data Matrix C is treated using DBSCAN algorithm_i2After carrying out Density Clustering, determine that the matrix that contained element is most in cluster result is attributed to valid data, and the remainder in cluster result is included into abnormal data.The method and system that the identification exchanges super UHV transmission line audible noise abnormal data can be applied to the real-time processing of extensive audible noise data, and using automatic method carry out processing reduce the new subjective error of manual operation bring, described method and system accuracy is high, comprehensive system.

Description

A kind of identification exchange the method for super UHV transmission line audible noise abnormal data with System

Technical field

The present invention relates to power grid acoustic environment fields, and exchange super extra-high voltage transmission line more particularly, to a kind of identification The method and system of road audible noise abnormal data.

Background technique

With economic sustained and rapid development, power load also increases substantially therewith.When power generation energy resource resource and electricity consumption are negative When lotus distributed pole is unbalanced, the power transmission mode of " high voltage ", " remote " is generally used to configure and clean to adapt to energy source optimization The needs of energy fast development.

" high voltage " is embodied in the strong smart grid for being dedicated to building using extra-high voltage grid as bulk transmission grid, and extra-high It presses in transmission system, transmission line of electricity corona effect is obvious.Corona effect mainly includes corona loss, radio interference and audible makes an uproar Sound.Wherein audible noise problem has become focus of attention in recent years, the reason is that, with the raising of voltage class, it is audible Numerically amplification is more significant compared to other corona effects for noise, and neighbouring resident is being felt in the growth in this amplitude Impression on official is also more intuitive.

" remote " is then embodied in extra-high voltage project transmission distance more than several hundred kilometers, or even the UHV transmission having Engineering circuit path is up to 3300 kilometers.The increase of route distance inevitably closes on populated area, thereupon Be exactly that significant increase is presented in the Population that is influenced by transmission line of electricity audible noise.

The test of general audible noise carries out in outdoor environment, and it is dry to usually there will be a variety of noises around testing location Disturb source, such as the motor-driven vehicle going sound of surrounding pavement, animal cry, mankind's activity sound etc..It is used when UHV Transmission Engineering When the surface field strength range of split conductor is between 18-24kV/cm, since conducting wire bloom is not strong, what is generated at this time is audible Noise figure is smaller, under external interference, hardly results in effective measurement data.

At present for the distinguishing validity of transmission line of alternation current audible noise, lacks systematic embodiment, mostly use Artificial scalping method, 8kHz component method, the methods such as sound level meter probe group are handled.But all there is unavoidable lack in above method It falls into.Such as artificial scalping method is difficult to apply in large-scale data, and can introduce new subjective error；8kHz component method has Effect degree is still not clear at present；Then excessively coarse to the processing of result using sound level meter probe group, the above method is difficult accurately to be had Abnormal data in the convection current transmission line of electricity audible noise of effect is differentiated, and then obtains the actual audible noise water of transmission line of electricity It is flat.

Therefore, how to identify that the valid data for exchanging super UHV transmission line audible noise and abnormal data just become one The problem of a urgent need to resolve.

Summary of the invention

The valid data of super UHV transmission line audible noise and different are exchanged in order to solve to lack in the prior art identification The problem of technical solution of regular data, the present invention provide a kind of identification exchange super UHV transmission line audible noise abnormal data Method, which comprises

According to the coordinate data of coordinate data and conducting wire at the environment sensitive target in i group environment sensitive target data, Calculating determines at i environment sensitive target and the Euclidean distance data of conducting wire, and is based at the i environment sensitive target and leads At the Euclidean distance data build environment sensitive target of line and the Euclidean distance data matrix B of conducting wire_i1；

Equivalent continuous A sound level data in i group environment sensitive target data are formed into equivalent continuous A sound level data matrix A_i1, and pass through equivalent continuous A sound level data matrix A_i1With at environment sensitive target and the Euclidean distance data matrix B of conducting wire_i1's Splicing, constructs data matrix C to be clustered_i2；

Treat cluster data Matrix C_i2Carry out Density Clustering；

The matrix for determining that contained element is most in cluster result is attributed in valid data, and by the remainder in cluster result Divide and is included into abnormal data.

Further, the method is according to the number of coordinates at the environment sensitive target in i group environment sensitive target data According to the coordinate data with conducting wire, calculates at determining i environment sensitive target and the Euclidean distance data of conducting wire further include before adopting Collect the i group environment sensitive target data on the super UHV transmission line periphery of exchange to be identified, wherein the environment sensitive number of targets According to including equivalent continuous A sound level data at environment sensitive target, the coordinate of coordinate data and conducting wire at environment sensitive target Data.

Further, the method most matrix of contained element in determining cluster result is attributed in valid data, and It further includes the valid data and abnormal data for exporting identification that remainder in cluster result, which is included into after abnormal data,.

Further, it the coordinate data at the environment sensitive target according in i group environment sensitive target data and leads The coordinate data of line, calculating determines at i environment sensitive target and the Euclidean distance data of conducting wire, its calculation formula is:

In formula: d_ij--- at environment sensitive target and the Euclidean distance of conducting wire, m；

x_ik--- the coordinate data at environment sensitive target, m；

x_jk--- the coordinate data of conducting wire, m.

Further, described to treat cluster data Matrix C_i2Density Clustering is carried out to refer to using DBSCAN algorithm to be clustered Data matrix C_i2Carry out Density Clustering comprising:

The value of MinPts is determined according to the principle of MinPts >=dim+1, wherein dim indicates the dimension of data to be clustered；

Determine radius of neighbourhood Eps, its calculation formula is:

In formula: A_max--- matrix A_i1In greatest member, m；

A_min--- matrix A_i1In least member, m；

B_max--- matrix B_i1In greatest member, m；

B_min--- matrix B_i1In least member, m；

M --- the line number of data matrix to be clustered；

N --- the columns of data matrix to be clustered；

All kernel objects are determined according to determining MinPts and Eps, and the kernel object p meets | N_Eps(p)|≥ MinPts, wherein N_Eps(p) be object p field, its calculation formula is:

N_Eps(p)={ q ∈ C_i2|dist(p,q)≤Eps}

In formula, C_i2For object matrix to be clustered, q is Matrix C_i2In any object in addition to object p, dist (p, q) For the Euclidean distance of object p and q；

To each kernel object p, finds and clustering cluster is generated by the reachable object of its density, and be attributed to matrix D_jIn, And generator matrix set D={ D₁, D₂……D_j……D_n, in formula, 1≤j≤n, j, n are natural number；

By Matrix C_i2In be not attributed to the object of any matrix and be included into matrix G.

Further, the most matrix of contained element is attributed in valid data in the determining cluster result, and will cluster As a result the remainder in is included into abnormal data

Determine set of matrices D={ D₁, D₂……D_j……D_nIn the matrix D most containing element_max, it is attributed in matrix E, then The set of element in the matrix E is the set that identification exchanges super UHV transmission line audible noise valid data, institute State the expression formula of matrix E are as follows:

In formula, e is matrix D_maxIn element, Card (D_max) it is matrix D_maxThe quantity of contained element,To remove matrix D in set of matrices D_maxThe quantity of element contained by other outer matrixes；

By set of matrices D={ D₁, D₂……D_j……D_nIn except remove matrix D_maxOther outer matrixes are attributed in matrix G, Then the set of the element in the matrix G is the set that identification exchanges super UHV transmission line audible noise abnormal data.

According to another aspect of the present invention, it is different to provide a kind of super UHV transmission line audible noise of identification exchange by the present invention The system of regular data, the system comprises:

First matrix unit is used for according to the number of coordinates at the environment sensitive target in i group environment sensitive target data According to the coordinate data with conducting wire, calculating determines at i environment sensitive target and the Euclidean distance data of conducting wire, and is based on the i At a environment sensitive target and at the Euclidean distance data build environment sensitive target of conducting wire and the Euclidean distance data square of conducting wire Battle array B_i1；

Second matrix unit is used for the equivalent continuous A sound level data composition in i group environment sensitive target data is equivalent Continuous A sound level data matrix A_i1, and pass through equivalent continuous A sound level data matrix A_i1With at the environment sensitive target and Europe of conducting wire Formula distance data matrix B_i1Splicing, construct data matrix C to be clustered_i2；

Density Clustering unit is used to treat cluster data Matrix C_i2Carry out Density Clustering；

As a result determination unit, the matrix for being used to determine that contained element to be most in cluster result are attributed in valid data, and Remainder in cluster result is included into abnormal data.

Further, the system also includes data acquisition unit, it is used to acquire the super UHV transmission of exchange to be identified The i group environment sensitive target data on route periphery, wherein the environment sensitive target data include at environment sensitive target etc. Continuous A sound level data are imitated, the coordinate data of coordinate data and conducting wire at environment sensitive target.

Further, the system also includes result output units, are used to export the valid data and abnormal number of identification According to.

Further, first matrix unit according to the environment sensitive target in i group environment sensitive target data at The coordinate data of coordinate data and conducting wire, calculating determines at i environment sensitive target and the Euclidean distance data of conducting wire, calculates Formula are as follows:

In formula: d_ij--- at environment sensitive target and the Euclidean distance of conducting wire, m；

x_ik--- the coordinate data at environment sensitive target, m；

x_jk--- the coordinate data of conducting wire, m.

Further, the Density Clustering unit treats cluster data Matrix C using DBSCAN algorithm_i2It is poly- to carry out density Class comprising:

First parameters unit is used to determine the value of MinPts according to the principle of MinPts >=dim+1, wherein dim is indicated The dimension of data to be clustered；

Second parameters unit is used to determine radius of neighbourhood Eps, its calculation formula is:

In formula: A_max--- matrix A_i1In greatest member, m；

A_min--- matrix A_i1In least member, m；

B_max--- matrix B_i1In greatest member, m；

B_min--- matrix B_i1In least member, m；

M --- the line number of data matrix to be clustered；

N --- the columns of data matrix to be clustered；

Kernel object unit is used to determine all kernel objects, the core pair according to determining MinPts and Eps As p meets | N_Eps(p) | >=MinPts, wherein N_Eps(p) be object p field, its calculation formula is:

N_Eps(p)={ q ∈ C_i2|dist(p,q)≤Eps}

In formula, C_i2For object matrix to be clustered, q is Matrix C_i2In any object in addition to object p, dist (p, q) For the Euclidean distance of object p and q；

Set of matrices unit is used to find each kernel object p the object generation cluster reachable by its density Cluster is attributed to matrix D_jIn, and generator matrix set D={ D₁, D₂……D_j……D_n, in formula, 1≤j≤n, j, n are nature Number；

Third matrix unit is used for Matrix C_i2In be not attributed to the object of any matrix and be included into matrix G.

Further, the result determination unit includes:

Valid data determination unit is used to determine set of matrices D={ D₁, D₂……D_j……D_nIn containing element it is most Matrix D_max, it is attributed in matrix E, then the set of the element in the matrix E is to identify that the super UHV transmission line of exchange can Listen the set of noise valid data, the expression formula of the matrix E are as follows:

In formula, e is matrix D_maxIn element, Card (D_max) it is matrix D_maxThe quantity of contained element,To remove matrix D in set of matrices D_maxThe quantity of element contained by other outer matrixes；

Abnormal data determination unit is used for set of matrices D={ D₁, D₂……D_j……D_nIn remove matrix D_maxOuter Other matrixes are attributed in matrix G, then the set of the element in the matrix G is to identify that the super UHV transmission line of exchange is audible The set of noise abnormal data.

The identification that technical solution of the present invention provides exchange the method for super UHV transmission line audible noise abnormal data with Coordinate data of the system according to coordinate data and conducting wire at the environment sensitive target in the environment sensitive target data of acquisition, meter It calculates and determines at i environment sensitive target and the Euclidean distance data of conducting wire, and based at the environment sensitive target and conducting wire At Euclidean distance data build environment sensitive target and the Euclidean distance data matrix of conducting wire；It again will be in environment sensitive target data Equivalent continuous A sound level data form equivalent continuous A sound level data matrix, and pass through equivalent continuous A sound level data matrix and ring At the sensitive target of border and the splicing of the Euclidean distance data matrix of conducting wire, data matrix to be clustered is constructed；Then DBSCAN is used Algorithm treats cluster data Matrix C_i2After carrying out Density Clustering, determine that the matrix that contained element is most in cluster result is attributed to effectively Data, and the remainder in cluster result is included into abnormal data.The super UHV transmission line of the identification exchange is audible to make an uproar The method and system of sound abnormal data can be applied to the real-time processing of extensive audible noise data, and use automation side Method, which carries out processing, reduces the new subjective error of manual operation bring, and described method and system accuracy is high, and comprehensive system is right In instructing transmission tower to design in exchanging super UHV transmission line construction, transmission pressure selection and power transformation building are arranged, It is of great significance in terms of reducing engineering integrated environment estimated risk.

Detailed description of the invention

By reference to the following drawings, exemplary embodiments of the present invention can be more fully understood by:

Fig. 1 is to exchange super UHV transmission line audible noise abnormal data according to the identification of the preferred embodiment for the present invention Method flow chart；

Fig. 2 is the signal according to the coordinate of coordinate and conducting wire at the environment sensitive target of the preferred embodiment for the present invention Figure；

Fig. 3 is to exchange super UHV transmission line audible noise abnormal data according to the identification of the preferred embodiment for the present invention System structural schematic diagram.

Specific embodiment

Exemplary embodiments of the present invention are introduced referring now to the drawings, however, the present invention can use many different shapes Formula is implemented, and is not limited to the embodiment described herein, and to provide these embodiments be at large and fully disclose The present invention, and the scope of the present invention is sufficiently conveyed to person of ordinary skill in the field.Show for what is be illustrated in the accompanying drawings Term in example property embodiment is not limitation of the invention.In the accompanying drawings, identical cells/elements use identical attached Icon note.

Unless otherwise indicated, term (including scientific and technical terminology) used herein has person of ordinary skill in the field It is common to understand meaning.Further it will be understood that with the term that usually used dictionary limits, should be understood as and its The context of related fields has consistent meaning, and is not construed as Utopian or too formal meaning.

Fig. 1 is to exchange super UHV transmission line audible noise abnormal data according to the identification of the preferred embodiment for the present invention Method flow chart.As shown in Figure 1, identification described in this preferred embodiment exchanges super UHV transmission line audible noise The method 100 of abnormal data is since step 101.

In step 101, the i group environment sensitive target data on the super UHV transmission line periphery of exchange to be identified is acquired, In, the environment sensitive target data includes the equivalent continuous A sound level data at environment sensitive target, at environment sensitive target The coordinate data of coordinate data and conducting wire.

In step 102, according to the coordinate data and conducting wire at the environment sensitive target in i group environment sensitive target data Coordinate data, calculating determines at i environment sensitive target and the Euclidean distance data of conducting wire, and is based on the i environment sensitive At target and at the Euclidean distance data build environment sensitive target of conducting wire and the Euclidean distance data matrix B of conducting wire_i1, wherein Data matrix B_i1Expression formula are as follows:

In step 103, the equivalent continuous A sound level data in i group environment sensitive target data are formed into equivalent continuous A sound level Data matrix A_i1, and pass through equivalent continuous A sound level data matrix A_i1With at environment sensitive target and the Euclidean distance data of conducting wire Matrix B_i1Splicing, construct data matrix C to be clustered_i2, wherein data matrix A_i1And C_i2Expression formula be respectively as follows:

In step 104, cluster data Matrix C is treated_i2Carry out Density Clustering.

In step 105, the matrix for determining that contained element is most in cluster result is attributed in valid data, and by cluster result In remainder be included into abnormal data.

In step 106, the valid data and abnormal data of identification are exported.

Preferably, the coordinate data and conducting wire at the environment sensitive target according in i group environment sensitive target data Coordinate data, calculate determine i environment sensitive target at and conducting wire Euclidean distance data, its calculation formula is:

In formula: d_ij--- at environment sensitive target and the Euclidean distance of conducting wire, m；

x_ik--- the coordinate data at environment sensitive target, m；

x_jk--- the coordinate data of conducting wire, m.

Fig. 2 is the signal according to the coordinate of coordinate and conducting wire at the environment sensitive target of the preferred embodiment for the present invention Figure.As shown in Fig. 2, the coordinate at the environment sensitive target refer to environment sensitive target to conducting wire extended line it is horizontal away from From the coordinate of the conducting wire is the linear distance at the guide line endpoint opposite with hanging point to environment sensitive target.

Preferably, described to treat cluster data Matrix C_i2Progress Density Clustering, which refers to, treats cluster numbers using DBSCAN algorithm According to Matrix C_i2Carry out Density Clustering comprising:

The value of MinPts is determined according to the principle of MinPts >=dim+1, wherein dim indicates the dimension of data to be clustered, In this preferred embodiment, since the dimension of cluster data is 2, therefore MinPts value is 3；

Determine radius of neighbourhood Eps, its calculation formula is:

In formula: A_max--- matrix A_i1In greatest member, m；

A_min--- matrix A_i1In least member, m；

B_max--- matrix B_i1In greatest member, m；

B_min--- matrix B_i1In least member, m；

M --- the line number of data matrix to be clustered；

N --- the columns of data matrix to be clustered；

All kernel objects are determined according to determining MinPts and Eps, and the kernel object p meets | N_Eps(p)|≥ MinPts, wherein N_Eps(p) be object p field, its calculation formula is:

N_Eps(p)={ q ∈ C_i2|dist(p,q)≤Eps}

In formula, C_i2For object matrix to be clustered, q is Matrix C_i2In any object in addition to object p, dist (p, q) For the Euclidean distance of object p and q；

To each kernel object p, finds and clustering cluster is generated by the reachable object of its density, and be attributed to matrix D_jIn, And generator matrix set D={ D₁, D₂……D_j……D_n, in formula, 1≤j≤n, j, n are natural number；

By Matrix C_i2In be not attributed to the object of any matrix and be included into matrix G.

Preferably, the most matrix of contained element is attributed in valid data in the determining cluster result, and cluster is tied Remainder in fruit is included into abnormal data

Determine set of matrices D={ D₁, D₂……D_j……D_nIn the matrix D most containing element_max, it is attributed in matrix E, then The set of element in the matrix E is the set that identification exchanges super UHV transmission line audible noise valid data, institute State the expression formula of matrix E are as follows:

In formula, e is matrix D_maxIn element, Card (D_max) it is matrix D_maxThe quantity of contained element,To remove matrix D in set of matrices D_maxThe quantity of element contained by other outer matrixes；

By set of matrices D={ D₁, D₂……D_j……D_nIn except remove matrix D_maxOther outer matrixes are attributed in matrix G, Then the set of the element in the matrix G is the set that identification exchanges super UHV transmission line audible noise abnormal data.

Fig. 3 is to exchange super UHV transmission line audible noise abnormal data according to the identification of the preferred embodiment for the present invention System structural schematic diagram.As shown in figure 3, the super UHV transmission line of the exchange of identification described in this preferred embodiment is audible The system 300 of noise abnormal data includes:

Data acquisition unit 301 is used to acquire the i group environment sensitive on the super UHV transmission line periphery of exchange to be identified Target data, wherein the environment sensitive target data includes the equivalent continuous A sound level data at environment sensitive target, environment The coordinate data of coordinate data and conducting wire at sensitive target.

First matrix unit 302 is used for according to the coordinate at the environment sensitive target in i group environment sensitive target data The coordinate data of data and conducting wire, calculating determines at i environment sensitive target and the Euclidean distance data of conducting wire, and based on described At i environment sensitive target and at the Euclidean distance data build environment sensitive target of conducting wire and the Euclidean distance data square of conducting wire Battle array B_i1；

Second matrix unit 303 is used for the equivalent continuous A sound level data composition in i group environment sensitive target data Equivalent continuous A sound level data matrix A_i1, and pass through equivalent continuous A sound level data matrix A_i1With at environment sensitive target and conducting wire Euclidean distance data matrix B_i1Splicing, construct data matrix C to be clustered_i2；

Density Clustering unit 304 is used to treat cluster data Matrix C_i2Carry out Density Clustering；

As a result determination unit 305, the matrix for being used to determine that contained element to be most in cluster result are attributed in valid data, And the remainder in cluster result is included into abnormal data.

As a result output unit 306 are used to export the valid data and abnormal data of identification.

Preferably, first matrix unit 302 according to the environment sensitive target in i group environment sensitive target data at The coordinate data of coordinate data and conducting wire, calculating determines at i environment sensitive target and the Euclidean distance data of conducting wire, calculates Formula are as follows:

In formula: d_ij--- at environment sensitive target and the Euclidean distance of conducting wire, m；

x_ik--- the coordinate data at environment sensitive target, m；

x_jk--- the coordinate data of conducting wire, m.

Preferably, the Density Clustering unit 304 treats cluster data Matrix C using DBSCAN algorithm_i2It is poly- to carry out density Class comprising:

First parameters unit 341 is used to determine the value of MinPts according to the principle of MinPts >=dim+1, wherein dim Indicate the dimension of data to be clustered；

Second parameters unit 342, is used to determine radius of neighbourhood Eps, its calculation formula is:

In formula: A_max--- matrix A_i1In greatest member, m；

A_min--- matrix A_i1In least member, m；

B_max--- matrix B_i1In greatest member, m；

B_min--- matrix B_i1In least member, m；

M --- the line number of data matrix to be clustered；

N --- the columns of data matrix to be clustered；

Kernel object unit 343 is used to determine all kernel objects, the core according to determining MinPts and Eps Heart object p meets | N_Eps(p) | >=MinPts, wherein N_Eps(p) be object p field, its calculation formula is:

N_Eps(p)={ q ∈ C_i2|dist(p,q)≤Eps}

In formula, C_i2For object matrix to be clustered, q is Matrix C_i2In any object in addition to object p, dist (p, q) For the Euclidean distance of object p and q；

Set of matrices unit 344 is used to find each kernel object p to be generated by the reachable object of its density and gather Class cluster, is attributed to matrix D_jIn, and generator matrix set D={ D₁, D₂……D_j……D_n, in formula, 1≤j≤n, j, n are nature Number；

Third matrix unit 345 is used for Matrix C_i2In be not attributed to the object of any matrix and be included into matrix G.

Preferably, the result determination unit 305 includes:

Valid data determination unit 351 is used to determine set of matrices D={ D₁, D₂……D_j……D_nIn containing element most More matrix Ds_max, it is attributed in matrix E, then the set of the element in the matrix E is and identifies to exchange super UHV transmission line The set of audible noise valid data, the expression formula of the matrix E are as follows:

In formula, e is matrix D_maxIn element, Card (D_max) it is matrix D_maxThe quantity of contained element,To remove matrix D in set of matrices D_maxThe quantity of element contained by other outer matrixes；

Abnormal data determination unit 352 is used for set of matrices D={ D₁, D₂……D_j……D_nIn remove matrix D_max Other outer matrixes are attributed in matrix G, then the set of the element in the matrix G is and identifies to exchange super UHV transmission line The set of audible noise abnormal data.

The present invention is described by reference to a small amount of embodiment.However, it is known in those skilled in the art, as Defined by subsidiary Patent right requirement, in addition to the present invention other embodiments disclosed above equally fall in it is of the invention In range.

Normally, all terms used in the claims are all solved according to them in the common meaning of technical field It releases, unless in addition clearly being defined wherein.All references " one/described/be somebody's turn to do [device, component etc.] " are all opened ground At least one example being construed in described device, component etc., unless otherwise expressly specified.Any method disclosed herein Step need not all be run with disclosed accurate sequence, unless explicitly stated otherwise.

Claims (12)

1. a kind of method that identification exchanges super UHV transmission line audible noise abnormal data, which is characterized in that the method Include:

According to the coordinate data of coordinate data and conducting wire at the environment sensitive target in i group environment sensitive target data, calculate It determines at i environment sensitive target and the Euclidean distance data of conducting wire, and based at the i environment sensitive target and conducting wire At Euclidean distance data build environment sensitive target and the Euclidean distance data matrix B of conducting wire_i1；

Equivalent continuous A sound level data in i group environment sensitive target data are formed into equivalent continuous A sound level data matrix A_i1, and Pass through equivalent continuous A sound level data matrix A_i1With at environment sensitive target and the Euclidean distance data matrix B of conducting wire_i1Splicing, Construct data matrix C to be clustered_i2；

Treat cluster data Matrix C_i2Carry out Density Clustering；

The matrix for determining that contained element is most in cluster result is attributed in valid data, and the remainder in cluster result is returned Enter abnormal data.

2. the method according to claim 1, wherein the method is according in i group environment sensitive target data Environment sensitive target at coordinate data and conducting wire coordinate data, calculate and determine at i environment sensitive target and conducting wire It further include the i group environment sensitive number of targets for acquiring the super UHV transmission line periphery of exchange to be identified before Euclidean distance data According to, wherein the environment sensitive target data includes the equivalent continuous A sound level data at environment sensitive target, environment sensitive mesh The coordinate data of coordinate data and conducting wire at mark.

3. the method according to claim 1, wherein the method contained element in determining cluster result is most Matrix be attributed in valid data, and it further includes having for output identification that the remainder in cluster result, which is included into after abnormal data, Imitate data and abnormal data.

4. the method according to claim 1, wherein the environment according in i group environment sensitive target data The coordinate data of coordinate data and conducting wire at sensitive target, calculate determine i environment sensitive target at and conducting wire it is European away from From data, its calculation formula is:

In formula: d_ij--- at environment sensitive target and the Euclidean distance of conducting wire, m；

x_ik--- the coordinate data at environment sensitive target, m；

x_jk--- the coordinate data of conducting wire, m.

5. the method according to claim 1, wherein described treat cluster data Matrix C_i2Carrying out Density Clustering is Refer to and treats cluster data Matrix C using DBSCAN algorithm_i2Carry out Density Clustering comprising:

The value of MinPts is determined according to the principle of MinPts >=dim+1, wherein dim indicates the dimension of data to be clustered；

Determine radius of neighbourhood Eps, its calculation formula is:

In formula: A_max--- matrix A_i1In greatest member, m；

A_min--- matrix A_i1In least member, m；

B_max--- matrix B_i1In greatest member, m；

B_min--- matrix B_i1In least member, m；

M --- the line number of data matrix to be clustered；

N --- the columns of data matrix to be clustered；

All kernel objects are determined according to determining MinPts and Eps, and the kernel object p meets | N_Eps(p)|≥ MinPts, wherein N_Eps(p) be object p field, its calculation formula is:

N_Eps(p)={ q ∈ C_i2|dist(p,q)≤Eps}

In formula, C_i2For object matrix to be clustered, q is Matrix C_i2In any object in addition to object p, dist (p, q) is pair As the Euclidean distance of p and q；

To each kernel object p, finds and clustering cluster is generated by the reachable object of its density, and be attributed to matrix D_jIn, and generate Set of matrices D={ D₁, D₂……D_j……D_n, in formula, 1≤j≤n, j, n are natural number；

By Matrix C_i2In be not attributed to the object of any matrix and be included into matrix G.

6. according to the method described in claim 5, it is characterized in that, the most matrix of contained element in the determining cluster result It is attributed in valid data, and the remainder in cluster result is included into abnormal data and includes:

Determine set of matrices D={ D₁, D₂……D_j……D_nIn the matrix D most containing element_max, be attributed in matrix E, then it is described The set of element in matrix E is the set that identification exchanges super UHV transmission line audible noise valid data, the square The expression formula of battle array E are as follows:

In formula, e is matrix D_maxIn element, Card (D_max) it is matrix D_maxThe quantity of contained element,For Matrix D is removed in set of matrices D_maxThe quantity of element contained by other outer matrixes；

By set of matrices D={ D₁, D₂……D_j……D_nIn except remove matrix D_maxOther outer matrixes are attributed in matrix G, then described The set of element in matrix G is the set that identification exchanges super UHV transmission line audible noise abnormal data.

7. a kind of system that identification exchanges super UHV transmission line audible noise abnormal data, which is characterized in that the system Include:

First matrix unit, be used for according at the environment sensitive target in i group environment sensitive target data coordinate data and The coordinate data of conducting wire, calculating determines at i environment sensitive target and the Euclidean distance data of conducting wire, and is based on the i ring At the sensitive target of border and at the Euclidean distance data build environment sensitive target of conducting wire and the Euclidean distance data matrix B of conducting wire_i1；

Second matrix unit is used for the equivalent continuous A sound level data composition in i group environment sensitive target data is equivalent continuous A sound level data matrix A_i1, and pass through equivalent continuous A sound level data matrix A_i1With at environment sensitive target and conducting wire it is European away from From data matrix B_i1Splicing, construct data matrix C to be clustered_i2；

Density Clustering unit is used to treat cluster data Matrix C_i2Carry out Density Clustering；

As a result determination unit, the matrix for being used to determine that contained element to be most in cluster result are attributed in valid data, and will gather Remainder in class result is included into abnormal data.

8. system according to claim 7, which is characterized in that the system also includes data acquisition unit, be used to adopt Collect the i group environment sensitive target data on the super UHV transmission line periphery of exchange to be identified, wherein the environment sensitive number of targets According to including equivalent continuous A sound level data at environment sensitive target, the coordinate of coordinate data and conducting wire at environment sensitive target Data.

9. system according to claim 7, which is characterized in that the system also includes result output units, are used for defeated The valid data and abnormal data identified out.

10. system according to claim 7, which is characterized in that first matrix unit is according to i group environment sensitive target The coordinate data of the coordinate data and conducting wire at environment sensitive target in data, calculate determine i environment sensitive target at and The Euclidean distance data of conducting wire, its calculation formula is:

In formula: d_ij--- at environment sensitive target and the Euclidean distance of conducting wire, m；

x_ik--- the coordinate data at environment sensitive target, m；

x_jk--- the coordinate data of conducting wire, m.

11. system according to claim 7, which is characterized in that the Density Clustering unit is treated using DBSCAN algorithm Cluster data Matrix C_i2Carry out Density Clustering comprising:

First parameters unit is used to determine the value of MinPts according to the principle of MinPts >=dim+1, wherein dim is indicated to poly- The dimension of class data；

Second parameters unit is used to determine radius of neighbourhood Eps, its calculation formula is:

In formula: A_max--- matrix A_i1In greatest member, m；

A_min--- matrix A_i1In least member, m；

B_max--- matrix B_i1In greatest member, m；

B_min--- matrix B_i1In least member, m；

M --- the line number of data matrix to be clustered；

N --- the columns of data matrix to be clustered；

Kernel object unit is used to determine all kernel objects, the kernel object p according to determining MinPts and Eps Meet | N_Eps(p) | >=MinPts, wherein N_Eps(p) be object p field, its calculation formula is:

N_Eps(p)={ q ∈ C_i2|dist(p,q)≤Eps}

In formula, C_i2For object matrix to be clustered, q is Matrix C_i2In any object in addition to object p, dist (p, q) is pair As the Euclidean distance of p and q；

Set of matrices unit is used to find each kernel object p the object generation clustering cluster reachable by its density, returns In matrix D j, and generator matrix set D={ D₁, D₂……D_j……D_n, in formula, 1≤j≤n, j, n are natural number；

Third matrix unit is used for Matrix C_i2In be not attributed to the object of any matrix and be included into matrix G.

12. system according to claim 11, which is characterized in that the result determination unit includes:

Valid data determination unit is used to determine set of matrices D={ D₁, D₂……D_j……D_nIn the matrix most containing element D_max, it is attributed in matrix E, then the set of the element in the matrix E is and identifies to exchange super UHV transmission line audible noise The set of valid data, the expression formula of the matrix E are as follows:

In formula, e is matrix D_maxIn element, Card (D_max) it is matrix D_maxThe quantity of contained element,For Matrix D is removed in set of matrices D_maxThe quantity of element contained by other outer matrixes；

Abnormal data determination unit is used for set of matrices D={ D₁, D₂……D_j……D_nIn remove matrix D_maxOuter other Matrix is attributed in matrix G, then the set of the element in the matrix G is and identifies to exchange super UHV transmission line audible noise The set of abnormal data.