CN105866636A - Transformer substation positioning method based on time difference positioning - Google Patents

Abstract

The invention provides a transformer substation positioning method based on time difference positioning, which comprises the following steps: determining an antenna arrangement area, determining an inscribed circle in the antenna arrangement area, arranging a first antenna and a second antenna at two ends of the diameter of the inscribed circle, arranging a third antenna at the center of the inscribed circle to form an antenna array, wherein the antenna array can rotate around the center of the circle, calculating a first time difference from a local discharge source to the first antenna and the third antenna and a second time difference from the local discharge source to the second antenna and the third antenna according to local discharge pulses received by the first antenna, the second antenna and the third antenna, and further performing planar two-dimensional positioning according to the first time difference and the second time difference to obtain a primary positioning point; and rotating the antenna array to enable a connecting line of the preliminary positioning point and the third antenna to be perpendicular to a connecting line of the first antenna and the second antenna, and accurately positioning again according to the first time difference and the second time difference to obtain a target positioning point. The invention has accurate positioning and easy popularization.

Description

Transformer station based on positioning using TDOA localization method

Technical field

The present invention relates to aerial array method for arranging, in particular it relates to a kind of transformer station based on positioning using TDOA location Method.

Background technology

Partial Discharge Detection is as the important means of insulation diagnosis, because it can reflect that power system is electric timely and effectively The insulation status of equipment, is just popularized by power department.

It is substantially the concrete power equipment for transformer station at present both at home and abroad and carries out partial discharge monitoring and monitoring skill The research of art, the Cleaning Principle of main application and method include pulse current method, superfrequency method (UHF), ultrasonic Ripple method, chemical method, optical method etc., wherein superfrequency method and supercritical ultrasonics technology are practical feasible methods.

Emphasis and the difficult point of Partial Discharge Detection is to use the scheme of low cost at present, is implementing Partial Discharge Detection During realize the location of Partial Discharge Sources and the assessment of degree of discharge, thus realize commenting of Electric Power Equipment Insulation state Estimate.

The current partial discharge monitoring for substation equipment and location are set mainly for GIS, transformator, capacitive The concrete single equipment such as standby is carried out, to each self installation of equipment a set of monitoring system to be monitored.And appointing in transformer station What high-tension electricity equipment all it may happen that partial discharges fault, wants an electrical equipment at full station is implemented monitoring, It is accomplished by installing and overlaps monitoring device more, then it is carried out comprehensively.Such monitoring system architecture, required expense is high, The service efficiency of monitoring system is the lowest, and the biggest to the maintenance workload of numerous on-Line Monitor Device itself.

Along with the economic and development of society, the requirement to power supply reliability is more and more higher, develops a kind of low cost, height Reliability, it is capable of full station high voltage electric equipment is carried out the novel on-line monitoring dress of real-time state monitoring, location Put and be highly desirable to.In terms of the result of Literature Consult, the unit such as Shanghai Communications University proposes several transformer station and entirely stands The method of monitoring, one is to use fixed installation formula or vehicular UHF-antenna array, based on antenna array receiver The time difference of partial-discharge ultrahigh-frequency signal carries out the three-dimensional of Partial Discharge Sources or the location of deflection.Its ultimate principle Being to receive time difference of homology partial discharge pulse according to antenna to position, at present time delay is estimated by existing a lot of scholars Algorithm and local breakdown location algorithm have made extensive and intensive studies, and have many methods, then for how The research carrying out antenna layout is few, but this has vital effect for the precision ensureing location.

Summary of the invention

For defect of the prior art, it is an object of the invention to provide a kind of transformer station based on positioning using TDOA location Method.

Transformer station based on the positioning using TDOA localization method that the present invention provides, comprises the steps:

Step 1: determine antenna layout area, determines inscribed circle in described antenna layout area, and by first antenna and Second antenna is arranged in the two ends of a diameter of described inscribed circle, and third antenna is arranged in the center of circle of inscribed circle, forms sky Linear array, described antenna array can rotate around the center of circle, and first antenna, the second antenna, third antenna are always positioned at described interior Meet round one diametrically；

Step 2: calculate Partial Discharge Sources according to the partial discharge pulse that first antenna, the second antenna, third antenna receive During to the very first time difference of first antenna and third antenna and Partial Discharge Sources to the second of the second antenna and third antenna Between poor, and then according to the very first time difference and the second time difference carry out planar location, obtain Primary Location point；

Step 3: rotary antenna battle array so that the line of Primary Location point and third antenna be perpendicular to first antenna, second day The line of line, carries out very first time difference and the calculating of the second time difference again, according to very first time difference with the second time difference again Secondary it is accurately positioned, obtains target anchor point.

Preferably, described step 1 comprises the steps:

Step 101: determine that described antenna layout area is roof；

Step 102: determine inscribed circle in described antenna layout area, and first antenna and the second antenna are arranged in institute State the two ends of a diameter of inscribed circle, third antenna is arranged in the center of circle of inscribed circle, form antenna array, described antenna array Can rotate around the center of circle, and first antenna, the second antenna, third antenna are always positioned at the one of described inscribed circle diametrically.

Preferably, step 201: the partial discharge pulse received according to 3 antennas calculates Partial Discharge Sources to first antenna And Partial Discharge Sources poor with the very first time of third antenna, to the second antenna and the second time difference of third antenna, is designated as t₂₁、 t₃₁；

Step 202: the coordinate of Partial Discharge Sources is designated as (x₀,y₀), the coordinate of i-th antenna is designated as (x_i,y_i), according to Little optical path difference principle, lists equation A as follows:

$\left\{\begin{array}{c}\sqrt{{(x_2-x_0)}^2+{(y_2-y_0)}^2}-\sqrt{{(x_1-x_0)}^2+{(y_1-y_0)}^2}=c{t}_{21}\\ \sqrt{{(x_3-x_0)}^2+{(y_3-y_0)}^2}-\sqrt{{(x_1-x_0)}^2+{(y_1-y_0)}^2}={\mathrm{ct}}_{31}\end{array}\right.$

Step 203: solving equation A, obtains Primary Location point.

Preferably, step 301: rotary antenna battle array so that Primary Location point is perpendicular to first day with the line of third antenna Line, the line of the second antenna, and by third antenna, the line formed along Primary Location point and the described center of circle is to Primary Location Point direction is moved to antenna layout area edge；

Step 302: repeat step 202 and 203 and be accurately positioned, obtain target anchor point.

Compared with prior art, the present invention has a following beneficial effect:

1, present invention is generally directed to transformer station uses triantennary to position the two dimensional surface of Partial Discharge Sources, in theory Provable, the result that this method obtains is the arrangement of position error lower bound minimum；

2, the present invention arranges a little positioning equation groups based on the shortest optical path principle, utilizes the Cramer-Lao in parameter estimation Lower bound principle, provides the variance inferior boundary computing formula of position error, according to position error lower bound computing formula, determines The method for optimally arranging of sensor；

3, the present invention uses the antenna arrangement that the two dimensional surface of Partial Discharge Sources is positioned by four antennas to transformer station Also there is directive significance, because now antenna amount has redundancy, according to the result of the present invention, select wherein to position 3 antennas of error minimum position, it will obtain more accurate result.

Accompanying drawing explanation

By the detailed description non-limiting example made with reference to the following drawings of reading, the further feature of the present invention, Purpose and advantage will become more apparent upon:

Fig. 1 is positioning principle figure based on UHF antenna array in the present invention；

Fig. 2 is the angle schematic diagram of UHF antenna in the present invention；

Fig. 3 is the first arrangement location scattergram in the present invention；

Fig. 4 is the second arrangement location scattergram in the present invention；

Fig. 5 is the third arrangement location scattergram in the present invention；

Fig. 6 is the third arrangement location scattergram in the present invention；

The emulation signal based on real noise of Fig. 7 present invention；

Fig. 8 is emulation signal based on real noise in prior art.

Detailed description of the invention

Below in conjunction with specific embodiment, the present invention is described in detail.Following example will assist in those skilled in the art Member is further appreciated by the present invention, but limits the present invention the most in any form.It should be pointed out that, the common skill to this area For art personnel, without departing from the inventive concept of the premise, it is also possible to make some deformation and improvement.These broadly fall into Protection scope of the present invention.

In the present embodiment, transformer station based on the positioning using TDOA localization method that the present invention provides, use UHF antenna array, Transformer station partial discharge signals is monitored and positions, as it is shown in figure 1, utilize UHF signal to arrive from Partial Discharge Sources The time difference information of each antenna, positions based on the shortest optical path difference principle.

It is designated as d from the distance of Partial Discharge Sources to No. n-th antenna_n, UHF signal is from Partial Discharge Sources to No. n-th antenna institute The time needed is designated as t_n, the light velocity is designated as c, then can be obtained by relational expression (1).

d_n=ct_n (1)

X time yet with UHF signal is unknown, so t can not be directly obtained_n.By the most different The UHF signal waveform information that it line locking receives, such that it is able to the time of advent of the UHF signal obtained between different antennae Difference, utilizes this time difference, can be in the hope of the position of Partial Discharge Sources.

UHF signal arrival time difference between i-th antenna and jth antenna is designated as t_ij, range difference is designated as d_ij, i-th The distance of individual antenna and Partial Discharge Sources is designated as d_i, then can be obtained by equation group (2).

d_ij=d_i-d_j=ct_ij (2)

In theory, in two dimensional surface positions, it is only necessary to two equations in (2) just can putting down in the hope of Partial Discharge Sources Position, face；In three-dimensional fix, three equations in needs (2) can be in the hope of the plan-position of Partial Discharge Sources. In the present invention, the coordinate of Partial Discharge Sources is designated as (x₀,y₀), the coordinate of i-th antenna is designated as (x_i,y_i), then can Obtain relational expression (3).

$d_i=\sqrt{{(x_0-x_i)}^2+{(y_0-y_i)}^2}---\left(3\right)$

Non-linear due to equation group (2), is usually the optimal solution asking equation group (2) under the meaning of least square, generally This equation group is solved with the optimized algorithm such as Newton method and grid data service.

Due to environment noise, signal attenuation, there is the impact of the factors such as error, time delay estimated value t in waveform comparison algorithm_ij There is also error.Current invention assumes that the error that each time delay is estimated meets the same distribution of Gauss of zero-mean, divide on this basis Analysis position error.Consider the situation of three antenna two-dimensional localization, utilize the Cramer-Lao lower bound principle in parameter estimation, Provide the variance inferior boundary of position error, and this lower bound is analyzed.

The when of using three antennas to position, two equations in equation group (2) are used to can be carried out location, will It is designated as equation group (4).

$\left\{\begin{array}{c}\sqrt{{(x_2-x_0)}^2+{(y_2-y_0)}^2}-\sqrt{{(x_1-x_0)}^2+{(y_1-y_0)}^2}={\mathrm{ct}}_{21}\\ \sqrt{{(x_3-x_0)}^2+{(y_3-y_0)}^2}-\sqrt{{(x_1-x_0)}^2+{(y_1-y_0)}^2}={\mathrm{ct}}_{31}\end{array}\right.---\left(4\right)$

Due to the impact of error, (4) can be written as equation group (5) further.

$\left\{\begin{array}{c}\sqrt{{(x_2-x_0)}^2+{(y_2-y_0)}^2}-\sqrt{{(x_1-x_0)}^2+{(y_1-y_0)}^2}={\mathrm{ct}}_{21}+{\mathrm{c\ϵ}}_1\\ \sqrt{{(x_3-x_0)}^2+{(y_3-y_0)}^2}-\sqrt{{(x_1-x_0)}^2+{(y_1-y_0)}^2}={\mathrm{ct}}_{31}+{\mathrm{c\ϵ}}_2\end{array}\right.---\left(5\right)$

Wherein, ε₁,ε₂～N (0, σ²), i.e. ε₁,ε₂Obey average be zero variance be σ²Gauss distribution, ε₁、ε₂It is respectively t₂₁、 t₃₁Estimation difference.So t₂₁、t₃₁Joint probability density function f (T, θ) be represented by (6).

$f(T,\mathrm{\θ})=\frac{1}{2\mathrm{\π}{\left|Q\right|}^{\frac{1}{2}}}\exp (-\frac{1}{2}{\left(T-T_T\right)}^T{Q}^{-1}(T-T_T\left)\right)---\left(6\right)$

Wherein, Q and T is that the time delay that waveform comparison algorithm obtains is estimated, T_TFor in theory True time delay, It is hidden in T_TExpression Among formula.Take the logarithm probability density function, (7) can be obtained

$ln\left(f\right(T,\mathrm{\θ}\left)\right)=ln\left(\frac{1}{2\mathrm{\π}{\left|Q\right|}^{\frac{1}{2}}}\right)-\frac{1}{2}{(T-T_T)}^T{Q}^{-1}(T-T_T)---\left(7\right)$

Fisher information matrix I (θ) can be calculated accordingly, as shown in (8) formula.

$I\left(\mathrm{\θ}\right)=-E\left({\▿}_{\mathrm{\θ}}\right({\▿}_{\mathrm{\θ}}\left(\ln \[f\left(T,\mathrm{\θ}\right)\]\right)\left)\right)---\left(8\right)$

Wherein, E (.) represents expectation,Represent and seek gradient,

According to chain rule, (8), through calculating, can turn to (9).

I (θ)=HQ^-1H^-1 (9)

$H=\frac{1}{c}\left[\begin{array}{cc}\frac{\∂d_{21}}{\∂x_0}& \frac{\∂d_{31}}{\∂x_0}\\ \frac{\∂d_{21}}{\∂y_0}& \frac{\∂d_{31}}{\∂y_0}\end{array}\right]=\frac{1}{c}\left[\begin{array}{cc}\frac{x_0-x_2}{d_2}-\frac{x_0-x_1}{d_1}& \frac{x_0-x_3}{d_3}-\frac{x_0-x_1}{d_1}\\ \frac{y_0-y_2}{d_2}-\frac{y_0-y_1}{d_1}& \frac{y_0-y_3}{d_3}-\frac{y_0-y_1}{d_1}\end{array}\right]$

According to Cramér-Rao lower bound principle, the lower bound σ of position error² _CRBAs shown in formula (10).

${\mathrm{\σ}}_{CRB}^2=tr\left(I{\left(\mathrm{\θ}\right)}^{-1}\right)---\left(10\right)$

Wherein, tr (.) is matrix trace, i.e. Matrix diagonal entry sum.

Comprehensively (9) formula and (10) formula, can obtain σ² _CRBExpression formula, as shown in formula (11).

${\mathrm{\σ}}_{CRB}^2=c^2{\mathrm{\σ}}^2\frac{{(\frac{x_0-x_2}{d_2}-\frac{x_0-x_1}{d_1})}^2+{(\frac{x_0-x_3}{d_3}-\frac{x_0-x_1}{d_1})}^2+{(\frac{y_0-y_2}{d_2}-\frac{y_0-y_1}{d_1})}^2+{(\frac{y_0-y_3}{d_3}-\frac{y_0-y_1}{d_1})}^2}{{\left(\right(\frac{x_0-x_2}{d_2}-\frac{x_0-x_1}{d_1}\left)\right(\frac{y_0-y_3}{d_3}-\frac{y_0-y_1}{d_1})-(\frac{x_0-x_3}{d_3}-\frac{x_0-x_1}{d_1}\left)\right(\frac{y_0-y_2}{d_2}-\frac{y_0-y_1}{d_1}\left)\right)}^2}---\left(11\right)$

From formula (11) it can be seen that the lower bound σ of position error² _CRBBy the position of Partial Discharge Sources, the position of transducer arrangements And the size of time delay error together decides on.

In the present invention, transformer station based on positioning using TDOA localization method is exactly under certain restrictive condition so that position error Minimum a kind of sensor arrangement.The most also can be considered that a kind of unbiased is estimated owing to using optimized algorithm to obtain positioning result Meter, so finding a kind of sensor arrangement method so that the error lower bound that formula (11) is stated is minimum, also can make location Error becomes minimum generally.

Assuming that sensor can only be arranged in certain limit, outside this scope, there is a Partial Discharge Sources, as in figure 2 it is shown, Then observe under which kind of is arranged, can minimum to the position error of this Partial Discharge Sources.

In figure, P point represents Partial Discharge Sources position, and dash area is the region that sensor can be arranged, β is P point The maximum angle that can be formed with the line of any two points in dash area, S₁、S₂、S₃Be respectively first antenna, the Two antennas, third antenna, θ₁, θ₂, θ₃Be respectively the deflection of first antenna, the deflection of the second antenna, the 3rd day The deflection of line, then formula (11) can be converted into formula (12).

${\mathrm{\σ}}_{CRB}^2=c^2{\mathrm{\σ}}^2\frac{{({\mathrm{cos\θ}}_2-{\mathrm{cos\θ}}_1)}^2+{({\mathrm{cos\θ}}_3-{\mathrm{cos\θ}}_1)}^2+{({\mathrm{sin\θ}}_2-{\mathrm{sin\θ}}_1)}^2+{({\mathrm{sin\θ}}_3-{\mathrm{sin\θ}}_1)}^2}{{\left(\right({\mathrm{cos\θ}}_2-{\mathrm{cos\θ}}_1\left)\right({\mathrm{sin\θ}}_3-{\mathrm{sin\θ}}_1)-({\mathrm{cos\θ}}_3-{\mathrm{cos\θ}}_1\left)\right({\mathrm{sin\θ}}_2-{\mathrm{sin\θ}}_1\left)\right)}^2}---\left(12\right)$

Formula (12) abbreviation can be obtained formula (13).

${\mathrm{\σ}}_{CRB}^2=2c^2{\mathrm{\σ}}^2\frac{2-cos({\mathrm{\θ}}_1-{\mathrm{\θ}}_2)-cos({\mathrm{\θ}}_1-{\mathrm{\θ}}_3)}{{\left(sin\right({\mathrm{\θ}}_1-{\mathrm{\θ}}_2)+sin({\mathrm{\θ}}_2-{\mathrm{\θ}}_3)+\sin ({\mathrm{\θ}}_3-{\mathrm{\θ}}_1\left)\right)}^2}---\left(13\right)$

According to symmetry, when using t₂₁, t₂₃And t₂₃, t₃₁When calculate σ² _CRBCan obtain shown in formula (14) and formula (15).

${\mathrm{\σ}}_{CRB}^2=2c^2{\mathrm{\σ}}^2\frac{2-cos({\mathrm{\θ}}_1-{\mathrm{\θ}}_2)-cos({\mathrm{\θ}}_2-{\mathrm{\θ}}_3)}{{\left(sin\right({\mathrm{\θ}}_1-{\mathrm{\θ}}_2)+sin({\mathrm{\θ}}_2-{\mathrm{\θ}}_3)+\sin ({\mathrm{\θ}}_3-{\mathrm{\θ}}_1\left)\right)}^2}---\left(14\right)$

${\mathrm{\σ}}_{CRB}^2=2c^2{\mathrm{\σ}}^2\frac{2-cos({\mathrm{\θ}}_2-{\mathrm{\θ}}_3)-cos({\mathrm{\θ}}_1-{\mathrm{\θ}}_3)}{{\left(sin\right({\mathrm{\θ}}_1-{\mathrm{\θ}}_2)+sin({\mathrm{\θ}}_2-{\mathrm{\θ}}_3)+\sin ({\mathrm{\θ}}_3-{\mathrm{\θ}}_1\left)\right)}^2}---\left(15\right)$

By formula (13), (14) and (15) make an arithmetic average, it is possible to obtain a σ² _CRBOnly with respect to θ₁, θ₂, θ₃And σ² Expression formula, as shown in formula (16).

${\mathrm{\σ}}_{CRB}^2=\frac{4}{3}{c}^2{\mathrm{\σ}}^2\frac{3-cos({\mathrm{\θ}}_1-{\mathrm{\θ}}_2)-cos({\mathrm{\θ}}_1-{\mathrm{\θ}}_3)-cos({\mathrm{\θ}}_2-{\mathrm{\θ}}_3)}{{\left(sin\right({\mathrm{\θ}}_1-{\mathrm{\θ}}_2)+sin({\mathrm{\θ}}_2-{\mathrm{\θ}}_3)+sin({\mathrm{\θ}}_3-{\mathrm{\θ}}_1\left)\right)}^2}---\left(16\right)$

If θ₁≥θ₂≥θ₃, α=θ₁-θ₃, η=θ₂-θ₃, then formula (16) can be written as formula (17).

${\mathrm{\σ}}_{CRB}^2=\frac{4}{3}{c}^2{\mathrm{\σ}}^2\frac{3-\cos \mathrm{\η}-\cos (\mathrm{\α}-\mathrm{\η})-\cos \mathrm{\α}}{{(\sin \mathrm{\η}+\sin (\mathrm{\α}-\mathrm{\η})-\sin \mathrm{\α})}^2}---\left(17\right)$

η is first asked local derviation can obtain formula (18) by formula (17).

$\begin{array}{c}\frac{\∂{\mathrm{\σ}}_{CRB}^2}{\∂\mathrm{\η}}=\frac{4}{3}{c}^2{\mathrm{\σ}}^2\left(\right(\sin \mathrm{\η}-\sin \left(\mathrm{\α}-\mathrm{\η}\right)){\left(\sin \mathrm{\η}+\sin \left(\mathrm{\α}-\mathrm{\η}\right)-\sin \mathrm{\α}\right)}^2-2(3-\cos \mathrm{\η}-\cos \left(\mathrm{\α}-\mathrm{\η}\right)-\cos \mathrm{\α})\\ \left(\sin \mathrm{\η}+\sin \left(\mathrm{\α}-\mathrm{\η}\right)-\sin \mathrm{\α}\right)\left(\cos \mathrm{\η}-\cos \left(\mathrm{\α}-\mathrm{\η}\right)\right))/{\left(\sin \mathrm{\η}+\sin \left(\mathrm{\α}-\mathrm{\η}\right)-\sin \mathrm{\α}\right)}^4)\end{array}---\left(18\right)$

Observation type (18), it is judged that sign symbol can obtain formula (19),

$\frac{\&part;{\mathrm{\&sigma;}}_{CRB}^2}{\&part;\mathrm{\&eta;}}\left\{\begin{array}{c}<0,0\&le;\mathrm{\&eta;}<\frac{\mathrm{\&alpha;}}{2}\\ =0,\mathrm{\&eta;}=\frac{\mathrm{\&alpha;}}{2}\\ >0,\frac{\mathrm{\&alpha;}}{2}<\mathrm{\&eta;}\&le;\mathrm{\&alpha;}\end{array}\right.---\left(19\right)$

So, as η=α/2, σ² _CRBMinimize.Now, substitute into formula (17) and formula (20) can be obtained

${\mathrm{\&sigma;}}_{CRB}^2=2c^2{\mathrm{\&sigma;}}^2\frac{3-2cos\frac{\mathrm{\&alpha;}}{2}-cos\mathrm{\&alpha;}}{{(2sin\frac{\mathrm{\&alpha;}}{2}-sin\mathrm{\&alpha;})}^2}---\left(20\right)$

Again α is sought local derviation, formula (21) can be obtained,

$\frac{\&part;{\mathrm{\&sigma;}}_{CRB}^2}{\&part;\mathrm{\&alpha;}}=4c^2{\mathrm{\&sigma;}}^2\sin \frac{3}{4}\mathrm{\&alpha;}\frac{\sin \frac{3}{4}\mathrm{\&alpha;}+2\cos \frac{\mathrm{\&alpha;}}{2}\sin \frac{\mathrm{\&alpha;}}{4}+\cos \mathrm{\&alpha;}\sin \frac{\mathrm{\&alpha;}}{4}-6\sin \frac{\mathrm{\&alpha;}}{4}}{{(2\sin \frac{\mathrm{\&alpha;}}{2}-\sin \mathrm{\&alpha;})}^3}<0---\left(21\right)$

So, as α=β, σ² _CRBMinimize.

In sum, when the angle of Partial Discharge Sources Yu the line of two of which sensor reaches maximum, and the 3rd sensing The when that device being positioned on the angular bisector at this angle, the lower bound σ of position error² _CRBMinimize, thus have found three The optimum layout mode under sensor two dimensional surface location.

The present invention is launched simulation analysis.Designing 4 kinds of sensor arrangement, as follows, unit is rice.

$\left\{\begin{array}{c}1,S_1(-5,5),S_2(5,5),S_3(5,-5)\\ 2,S_1(-5,5),S_2(0,0),S_3(5,-5)\\ 3,S_1(-3,3),S_2(3,3),S_3(3,-3)\\ 4,S_1(-5,0),S_2(0,0),S_3(5,0)\end{array}\right.$

Assume there is a Partial Discharge Sources at coordinate (-5 ,-5) place, then for every kind of arrangement, between sensor Will there is the theoretical time delay value that one group of signal receives, give a suitable gaussian random amount to this time delay value.For often Plant arrangement, be one group with 1000 emulation, then each group all can have a location scattergram, it is also possible to according to These 1000 results calculate error.Fig. 3 to Fig. 6 gives the location scattergram of these 4 kinds of modes.

Table 1 gives the variance result table of 10 groups.

Positioning variances under table 1 Gaussian noise

Tab.1:variance of location with Gaussian noise

It can be seen that the first is little with the location difference in distribution of the second arrangement from Fig. 3 to Fig. 6, the third Distribution is more sparse than the first and the second, the 4th kind the most sparse.From the point of view of table 1, the first and the second cloth The positioning variances putting mode is substantially similar, and the third is bigger than them, and the variance of relative the 4th kind is maximum.Complete Entirely meet the notional result of chapter 4, it was demonstrated that the correctness of result.

To use real noise that conclusion is launched checking below, and measure noise data from substation field herein, folded It is added on emulation signal, then tries to achieve time delay by waveform comparison algorithm (conventional is energy supposition method), carry out discharge source Location.Wave data to be dealt with is the most as shown in Figure 7.

In figure, the signal of three kinds of colors represents the signal that three sensors receive respectively.Arrangement in above-mentioned four is entered Row Error Calculation is added up, and result is as shown in table 2

Positioning variances under table 2 real noise

Table 2:variance of location with realnoise

Because environment noise is the most unalterable, so having bigger difference between 6 prescription difference data groups and group, and Relation between arrangement is substantially the most identical with table 1, effective to real noise of this result verification notional result Property.

The squaerial array of 4 sensor compositions of the many employings of transformer station now carries out shelf depreciation location, it is considered to sensor Rectangular arrangement mode: (-5,5), (5,5), (5 ,-5), (-5 ,-5).Being the center of circle when Partial Discharge Sources is positioned at initial point, radius is 30 Time on the circle of rice, observe the curve changed with the change of discharge source deflection, as shown in Figure 8.

From figure it can be seen that error minimum at 45 °, 135 °, 225 °, about 315 ° variances minimums, 4 coordinate axess Neighbouring variance is maximum.Obviously, the discharge source in region that error is minimum and the maximum of sensor line angle are also maximum , the effectiveness of conclusion herein is so also illustrate that from a side.

Local inter-time positioning principle is set forth by the present invention, is analyzed positioning equation and has described, for Demand gives the positioning equation of 3 sensor two dimensional surface location.Former according to the Cramer-Lao lower bound in parameter estimation Reason, the variance inferior boundary of the positioning result under Gauss distribution of having derived meaning.The lower bound of position error is by the position of Partial Discharge Sources Putting, the position of transducer arrangements and the size of time delay error together decide on.Variance inferior boundary is analyzed, is passed three In the case of sensor plane positioning, give the optimal sensor placement mode minimum so that this lower bound in theory.Authorities The angle of the line of portion's discharge source and two of which sensor reaches maximum, and the angle that the 3rd sensor is positioned at this angle is put down The when of on separated time, this lower bound is minimum.

When using the sensor method for arranging to carry out transformer station based on positioning using TDOA location, comprise the steps:

Step 1: determine antenna layout area, determines inscribed circle in described antenna layout area, and by first antenna and Second antenna is arranged in the two ends of a diameter of described inscribed circle, and third antenna is arranged in the center of circle of inscribed circle, forms sky Linear array, described antenna array can rotate around the center of circle, and first antenna, the second antenna, third antenna are always positioned at described interior Meet round one diametrically；

Step 2: calculate Partial Discharge Sources according to the partial discharge pulse that first antenna, the second antenna, third antenna receive During to the very first time difference of first antenna and third antenna and Partial Discharge Sources to the second of the second antenna and third antenna Between poor, and then according to the very first time difference and the second time difference carry out planar location, obtain Primary Location point；

Step 3: rotary antenna battle array so that the line of Primary Location point and third antenna be perpendicular to first antenna, second day The line of line, carries out very first time difference and the calculating of the second time difference again, according to very first time difference with the second time difference again Secondary it is accurately positioned, obtains target anchor point.

Described step 1 comprises the steps:

Step 101: determine that described antenna layout area is roof；Roof is optimal, during because aerial position is too low, locally Discharge signal is easily hindered by other buildings, thus affects the straightline propagation of electromagnetic wave；

Step 102: determine inscribed circle in described antenna layout area, and first antenna and the second antenna are arranged in institute State the two ends of a diameter of inscribed circle, third antenna is arranged in the center of circle of inscribed circle, form antenna array, described antenna array Can rotate around the center of circle, and first antenna, the second antenna, third antenna are always positioned at the one of described inscribed circle diametrically.

Step 201: the partial discharge pulse received according to 3 antennas calculates Partial Discharge Sources to first antenna and the 3rd day The very first time difference of line and Partial Discharge Sources, to the second antenna and the second time difference of third antenna, are designated as t₂₁、t₃₁；

Step 202: the coordinate of Partial Discharge Sources is designated as (x₀,y₀), the coordinate of i-th antenna is designated as (x_i,y_i), according to Little optical path difference principle, lists equation A as follows:

$\left\{\begin{array}{c}\sqrt{{(x_2-x_0)}^2+{(y_2-y_0)}^2}-\sqrt{{(x_1-x_0)}^2+{(y_1-y_0)}^2}=c{t}_{21}\\ \sqrt{{(x_3-x_0)}^2+{(y_3-y_0)}^2}-\sqrt{{(x_1-x_0)}^2+{(y_1-y_0)}^2}={\mathrm{ct}}_{31}\end{array}\right.$

Step 203: solving equation A, obtains Primary Location point.

Step 301: rotary antenna battle array so that the line of Primary Location point and third antenna be perpendicular to first antenna, second The line of antenna, and by third antenna, the line formed along Primary Location point and the described center of circle moves to Primary Location point direction Move to antenna layout area edge；

Step 302: repeat step 202 and 203 and be accurately positioned, obtain target anchor point.

Above the specific embodiment of the present invention is described.It is to be appreciated that the invention is not limited in Stating particular implementation, those skilled in the art can make various deformation or amendment within the scope of the claims, This has no effect on the flesh and blood of the present invention.

Claims (4)

1. transformer station based on a positioning using TDOA localization method, it is characterised in that comprise the steps:

Step 1: determine antenna layout area, determines inscribed circle in described antenna layout area, and by first antenna and Second antenna is arranged in the two ends of a diameter of described inscribed circle, and third antenna is arranged in the center of circle of inscribed circle, forms sky Linear array, described antenna array can rotate around the center of circle, and first antenna, the second antenna, third antenna are always positioned at described interior Meet round one diametrically；

Step 2: calculate Partial Discharge Sources according to the partial discharge pulse that first antenna, the second antenna, third antenna receive During to the very first time difference of first antenna and third antenna and Partial Discharge Sources to the second of the second antenna and third antenna Between poor, and then according to the very first time difference and the second time difference carry out planar location, obtain Primary Location point；

Step 3: rotary antenna battle array so that the line of Primary Location point and third antenna be perpendicular to first antenna, second day The line of line, carries out very first time difference and the calculating of the second time difference again, according to very first time difference with the second time difference again Secondary it is accurately positioned, obtains target anchor point.

Transformer station based on positioning using TDOA the most according to claim 1 localization method, it is characterised in that described step Rapid 1 comprises the steps:

Step 101: determine that described antenna layout area is roof；

Step 102: determine inscribed circle in described antenna layout area, and first antenna and the second antenna are arranged in institute State the two ends of a diameter of inscribed circle, third antenna is arranged in the center of circle of inscribed circle, form antenna array, described antenna array Can rotate around the center of circle, and first antenna, the second antenna, third antenna are always positioned at the one of described inscribed circle diametrically.

Transformer station based on positioning using TDOA the most according to claim 2 localization method, it is characterised in that

Step 201: the partial discharge pulse received according to 3 antennas calculates Partial Discharge Sources to first antenna and the 3rd day The very first time difference of line and Partial Discharge Sources, to the second antenna and the second time difference of third antenna, are designated as t₂₁、t₃₁；

Step 202: the coordinate of Partial Discharge Sources is designated as (x₀,y₀), the coordinate of i-th antenna is designated as (x_i,y_i), according to Little optical path difference principle, lists equation A as follows:

$\left\{\begin{array}{c}\sqrt{{(x_2-x_0)}^2+{(y_2-y_0)}^2}-\sqrt{{(x_1-x_0)}^2+{(y_1-y_0)}^2}=c{t}_{21}\\ \sqrt{{(x_3-x_0)}^2+{(y_3-y_0)}^2}-\sqrt{{(x_1-x_0)}^2+{(y_1-y_0)}^2}={\mathrm{ct}}_{31}\end{array}\right.$

Step 203: solving equation A, obtains Primary Location point.

Transformer station based on positioning using TDOA the most according to claim 3 localization method, it is characterised in that

Step 301: rotary antenna battle array so that the line of Primary Location point and third antenna be perpendicular to first antenna, second The line of antenna, and by third antenna, the line formed along Primary Location point and the described center of circle moves to Primary Location point direction Move to antenna layout area edge；

Step 302: repeat step 202 and 203 and be accurately positioned, obtain target anchor point.