CN104239742A - Transformer far-field noise prediction method and system - Google Patents

Abstract

The invention discloses a transformer far-field noise prediction method and system. The method comprises the following steps: 1) equivalently regarding a transformer as a hexahedron, establishing a four-point source prediction model and a five-point source prediction model of transformer far-field noise according to near-field noise values, and setting model switching distance according to external dimension of the hexahedron; 2) detecting horizontal distance between a current prediction point and a transformer case; 3) comparing the horizontal distance with the model switching distance; if the horizontal distance is shorter than or equal to the model switching distance, calculating the noise value of the current prediction point according to the established four-point source prediction model; if the horizontal distance is longer than or equal to the model switching distance, calculating the noise value of the current prediction point according to the established five-point source prediction model. The system comprises a prediction model establishment module, a distance detection module and a noise prediction module. The transformer far-field noise prediction method and system have the advantages that the realization method is simple, the prediction speed is fast and the prediction results are accurate.

Description

A kind of transformer far-field noise Forecasting Methodology and system

Technical field

The present invention relates to transformer technology field, particularly relate to a kind of transformer far-field noise Forecasting Methodology and system.

Background technology

Along with the quickening of socioeconomic development, urbanization process and improving constantly of electrical network grade, the impact that the noise on human produced during power transmission and transforming equipment runs are lived is increasing.The noise produced during substation operation and the environmental issue caused happens occasionally, therefore also more and more receive the concern of people, and in transformer station, one of topmost noise source is transformer.The manufacturer of transformer can provide transformer noise controlling value usually in technical protocol, and is generally only near field noise value, and be generally the noise figure apart from transformer case 2m place, the far field sound power value for transformer then cannot provide.Therefore transformer far-field noise is predicted for the Noise measarement of newly-built transformer station and is all had great importance in the noise abatement of transporting the transformer station that exceeds standard.Although numerous research has been carried out in the far-field noise prediction both at home and abroad for transformer, and up to the present, is also commonly recognized without any a kind of transformer far-field noise Forecasting Methodology.

Summary of the invention

The technical problem to be solved in the present invention is just: the technical matters existed for prior art, the invention provides that a kind of implementation method is simple, predetermined speed fast and predict the outcome transformer far-field noise Forecasting Methodology and system accurately.

For solving the problems of the technologies described above, the technical scheme that the present invention proposes is:

A kind of transformer far-field noise Forecasting Methodology, concrete implementation step is:

1) forecast model is set up: transformer is equivalent to a hexahedron, set up four point source forecast models and the five point source forecast models of transformer far-field noise according to the near field noise value of transformer, and models switching distance is set according to described hexahedral physical dimension; The noise that described four point source forecast models produce at future position as four point sound sources for four faces far-field noise be expressed as in described hexahedron, the noise that described five point source forecast models produce at future position as five point sound sources for five faces far-field noise be expressed as in described hexahedron;

2) distance detects: detect the horizontal range between current predictive point and transformer case;

3) noise prediction: the size of more described horizontal range and described models switching distance, if horizontal range is less than or equal to models switching distance, calculates the far-field noise value at current predictive point place according to described four point source forecast models; If horizontal range is greater than models switching distance, calculate the far-field noise value at current predictive point place according to described five point source forecast models.

Preferably, the expression formula of described four point source forecast models is:

$L_{A\left(r\right)}=L_A-201g\frac{r}{2}+M_4---\left(1\right)$

The expression formula of described five point source forecast models is:

$L_{A\left(r\right)}=L_A-201g\frac{r}{2}+M_5---\left(2\right)$

Wherein, r is the horizontal range between future position and transformer case, L _{a (r)}for transformer is in the far-field noise value of r position, L _afor the near field noise value at distance transformer 2m place, M ₄be the superimposed noise amount that four point sound sources produce, M ₅it is the superimposed noise amount that five point sound sources produce.

Preferably, the expression formula of described four point source forecast models is:

$L_{A\left(r\right)}=L_A-201g\frac{r}{2}+101g4---\left(3\right)$

The expression formula of described five point source forecast models is:

$L_{A\left(r\right)}=L_A-201g\frac{r}{2}+101g5---\left(4\right)$

Wherein, r is the horizontal range between future position and transformer case, L _{a (r)}for transformer is in the far-field noise value of r position, L _afor the near field noise value at distance transformer 2m place.

Preferably, described hexahedron is cube; In described four point source forecast models, four faces are: remove lower surface in described hexahedral six surfaces, away from four surfaces of all the other beyond the surface of future position side; In described five point source forecast models, five faces are: remove all the other five surfaces beyond lower surface in described hexahedral six surfaces.

Preferably, described step 1) in the models switching distance that arranges according to described hexahedral physical dimension be the twice of maximal value in described hexahedral physical dimension.

A kind of transformer far-field noise prognoses system, comprising:

Forecast model sets up module, for transformer being equivalent to a hexahedron, set up four point source forecast models and the five point source forecast models of transformer far-field noise according to the near field noise value of transformer, and models switching distance is set according to described hexahedral physical dimension; The noise that described four point source forecast models produce at future position as four point sound sources for four faces far-field noise be expressed as in described hexahedron, the noise that described five point source forecast models produce at future position as five point sound sources for five faces far-field noise be expressed as in described hexahedron;

Distance detection module, detects the horizontal range between current predictive point and transformer case;

Noise prediction module, for the size of more described horizontal range and described models switching distance, if horizontal range is less than or equal to models switching distance, calculates the far-field noise value at current predictive point places according to described four point source forecast models; If horizontal range is greater than models switching distance, calculate the far-field noise value at current predictive point place according to described five point source forecast models.

Preferably, described forecast model is set up the expression formula that module sets up described four point source forecast models and is:

$L_{A\left(r\right)}=L_A-201g\frac{r}{2}+M_4---\left(1\right)$

Described forecast model sets up the expression formula that module sets up described five point source forecast models:

$L_{A\left(r\right)}=L_A-201g\frac{r}{2}+M_5---\left(2\right)$

Wherein, r is the horizontal range between future position and transformer case, L _{a (r)}for transformer is in the far-field noise value of r position, L _afor the near field noise value at distance transformer 2m place, M ₄be the superimposed noise amount that four point sound sources produce, M ₅it is the superimposed noise amount that five point sound sources produce.

Preferably, described forecast model is set up the expression formula that module sets up described four point source forecast models and is:

$L_{A\left(r\right)}=L_A-201g\frac{r}{2}+101g4---\left(3\right)$

Described forecast model sets up the expression formula that module sets up described five point source forecast models:

$L_{A\left(r\right)}=L_A-201g\frac{r}{2}+101g5---\left(4\right)$

Wherein, L _{a (r)}for transformer is in the far-field noise value of r position, L _afor the near field noise value at distance transformer 2m place, r is the horizontal range between future position and transformer case.

Preferably, described forecast model sets up the described hexahedron of module is cube; Described forecast model sets up four faces in the described four point source forecast models of module: remove lower surface in described hexahedral six surfaces, away from four surfaces of all the other beyond the surface of future position side; Described forecast model sets up five faces in the described five point source forecast models of module: remove all the other five surfaces beyond lower surface in described hexahedral six surfaces.

Preferably, described models switching distance arranges the twice that the models switching distance arranged according to described hexahedral physical dimension in module is maximal value in described hexahedral physical dimension.

Compared with prior art, the invention has the advantages that:

1) the present invention sets up four point source forecast models and the five point source forecast models of far-field noise according near field noise value, different forecast models is adopted to calculate the far-field noise value of transformer according to horizontal range again, just far-field noise prediction fast and effectively can be realized without the need to the parameter detecting of complexity and computation process, take into full account the different forecast models at different prediction distance place, implementation method is simple and predict the outcome accurately simultaneously.

2) the present invention affects the noise effect that less surface produces carry out Simplified prediction model by ignoring in transformer, when horizontal range is less than or equal to models switching distance, adopt four point source forecast models to calculate far-field noise value, noise is expressed as the noise that in hexahedron, four faces produce at future position as four point sound sources; When horizontal range is greater than models switching distance, adopt five point source forecast model calculating noise values, noise is expressed as the noise that in hexahedron, five faces produce at future position as five point sound sources, can effectively reduces prediction complexity, thus improve predetermined speed and precision of prediction.

Accompanying drawing explanation

Fig. 1 is the present embodiment transformer far-field noise Forecasting Methodology realization flow schematic diagram.

Fig. 2 is transformer far-field noise Forecasting Methodology principle schematic in the present embodiment.

Fig. 3 is Four-point Forecast model prediction principle schematic in the present embodiment.

Fig. 4 is five point prediction model prediction principle schematic in the present embodiment.

Marginal data:

1, transformer equivalent; 11, upper surface; 12, lower surface; 13, front surface; 14, rear surface; 15, left-hand face; 16, right lateral surface.

Embodiment

Below in conjunction with Figure of description and concrete preferred embodiment, the invention will be further described, but protection domain not thereby limiting the invention.

As shown in Figure 1, the present embodiment transformer far-field noise Forecasting Methodology, concrete implementation step is:

1) forecast model is set up: transformer is equivalent to a hexahedron, set up four point source forecast models and the five point source forecast models of transformer far-field noise according to the near field noise value of transformer, and models switching distance is set according to hexahedral physical dimension; The noise that four point source forecast models produce at future position as four point sound sources for four faces far-field noise be expressed as in hexahedron, the noise that five point source forecast models produce at future position as five point sound sources for five faces far-field noise be expressed as in hexahedron;

2) distance detects: detect the horizontal range between current predictive point and transformer case;

3) noise prediction: comparison level distance and the size of models switching distance, if horizontal range is less than or equal to models switching distance, calculate the far-field noise value at current predictive point place according to four point source forecast models; If horizontal range is greater than models switching distance, calculate the far-field noise value at current predictive point place according to five point source forecast models.

Transformer is equivalent to a hexahedron by the present embodiment, and hexahedron is cube, and using each the point sound source as generation far-field noise in hexahedron, the far-field noise of transformer is the vibration radiation noise that each point sound source produces at future position place, and the horizontal range r between future position and transformer case is greater than near field noise distance r ₁time noise be the far-field noise of transformer, r in the present embodiment ₁get 2m.The noise effect that each face of transformer produces at future position place as point sound source is different, the noise effect that the such as lower surface (bottom) of usual transformer produces as point sound source is less, when horizontal range r is less, the noise effect produced as point sound source away from the surface of future position side is also less, thus the far-field noise forecast model that the noise effect foundation can ignoring the less surface of impact according to horizontal range r simplifies, reduce prediction complexity, thus improve predetermined speed and precision of prediction.

In the present embodiment, transformer far-field noise Forecasting Methodology principle as shown in Figure 2,3, 4, future position is positioned on the surface level of distance ground 1.5m, transformer is rear surface 14 (in figure with center origin illustrate each surface) away from the surface of future position side, is less than or equal to models switching distance r for horizontal range r ₀far-field noise prediction, lower surface 12, rear surface 14 (as shown in shaded surface in figure) at future position place the noise effect that produces less, therefore ignore lower surface 12 in transformer equivalent 1, four point source forecast models are set up as the noise effect that point sound source produces in rear surface 14, and all the other four surfaces (upper surface 11, front surface 13, left-hand face 15 and right lateral surface 16) only in transformer except lower surface 12, rear surface 14 are as the noise figure of point sound source at the noise effect computational prediction point place that future position produces, and obtain four point source forecast models; Models switching distance r is greater than for horizontal range r ₀far-field noise prediction, lower surface 12 (as shown in shaded surface in figure) at future position place the noise effect that produces less, therefore the lower surface 12 ignored in transformer equivalent 1 sets up five point source forecast models as the noise effect that point sound source produces, and the noise figure at the noise effect computational prediction point place that five faces of all the other in transformer except lower surface 12 (upper surface 11, front surface 13, rear surface 14, left-hand face 15 and right lateral surface 16) is produced as point sound source, obtain five point source forecast models.

In the present embodiment, the expression formula of four point source forecast models is:

$L_{A\left(r\right)}=L_A-201g\frac{r}{2}+M_4---\left(1\right)$

The expression formula of five point source forecast models is:

$L_{A\left(r\right)}=L_A-201g\frac{r}{2}+M_5---\left(2\right)$

Wherein, r is horizontal range between future position and transformer case and r>r ₁, L _{a (r)}for transformer is in the far-field noise value of r position, L _afor the near field noise value at distance transformer 2m place, M ₄be the superimposed noise amount that four point sound sources produce, M ₅it is the superimposed noise amount that five point sound sources produce.

In the present embodiment, the derivation principle of four point source forecast models, five point source forecast models is as follows:

Because the expression formula of point sound source spherical radiation is such as formula shown in (1-1);

$L_{\mathrm{WA}}=L_{A\left(r\right)}^{\′}+101g\frac{{4\mathrm{\πr}}^2}{K}---(1-1)$

Wherein, L _wAfor transformer A weighted acoustical power (dB (A)), a weighted sound pressure level (dB (A)) during for being r position apart from transformer case distance, K is directivity coefficient and is 2 for semi-free space (outdoor substation) K value.

K value is that 2 up-to-date styles (1-1) can be expressed as:

$L_{\mathrm{WA}}=L_{A\left(r\right)}^{\′}+101g2{\mathrm{\πr}}^2=L_{A\left(r\right)}^{\′}+201\mathrm{gr}+8---(1-2)$

Can be obtained by formula (1-2) computing formula be:

$L_{A\left(r\right)}^{\′}=L_{\mathrm{WA}}-201\mathrm{gr}-8---(1-3)$

Relational expression then between near field noise value and far-field noise value is:

$L_{A\left(r2\right)}=L_{A\left(r1\right)}-201g\frac{r_2}{r_1}---(1-4)$

Wherein, L _{a (r1)}, L _{a (r2)}be respectively transformer at r ₁, r ₂position near field noise value, far-field noise value, and L _{a (r1)}=L _a, then according to the near field noise value L of transformer _aand the predictor formula that formula (1-4) can obtain far-field noise is:

$L=L_A-201g\frac{r}{r_1}---(1-5)$

Wherein L is the initial prediction obtained by near field noise value.

Because N number of identical sound pressure level sound source can superpose generation superimposed noise, therefore in the present embodiment, the forecast model of far-field noise considers the noise recruitment of superimposed noise on the basis of near field noise prediction (formula 1-5), and the superimposed noise amount produced in four point sound sources, five point sound sources is respectively M ₄, M ₅, to set up more accurate far-field noise forecast model, improve the precision of prediction of far-field noise.

Near field noise distance r in the present embodiment ₁get 2m, the initial prediction L obtained by near field noise value ₀be specially:

$L_0=L_A-201g\frac{r}{2}---(1-6)$

The superimposed noise amount that point sound source produces is specially 10lgN, i.e. M ₄=10lg4, M ₅=10lg5, then the expression formula obtaining four point source forecast models is specially:

$L_{A\left(r\right)}=L_A-201g\frac{r}{2}+101g4---\left(3\right)$

The expression formula of five point source forecast models is specially:

$L_{A\left(r\right)}=L_A-201g\frac{r}{2}+101g5---\left(4\right)$

In the present embodiment, models switching distance r ₀arrange according to hexahedral physical dimension in advance, be specially the twice 2l of maximal value in hexahedral physical dimension _max, l _maxfor maximal value in hexahedral physical dimension (length, width and height), be namely less than 2l _maxfuture position place can ignore the noise effect of lower surface 12 and rear surface 14, be greater than 2l _maxfuture position place then can ignore the noise effect of rear surface 14.When horizontal range r is less than or equal to 2l _maxtime, i.e. r ₁< r≤2l _max, adopt four point source noise model expression formulas (3) to calculate the noise figure at current predictive point place; When horizontal range r is greater than models switching distance r ₀time, i.e. r > 2l _max, adopt five point source noise model expression formulas (4) to calculate the noise figure at current predictive point place.

The present embodiment sets up four point source forecast models and the five point source forecast models of far-field noise according near field noise value, different forecast models is adopted to calculate the far-field noise value of transformer according to horizontal range again, just fast prediction can be realized without the need to the parameter detecting of complexity and computation process, take into full account the different forecast models at different prediction distance place, implementation method is simple and predict the outcome accurately simultaneously.

Be predicted as example with 1 the 50000kVA main-transformer far-field noise configured in certain 110kV transformer station to be below further described the present invention.

Step 1: forecast model is set up

The actual measurement of Aiwa 6270 acoustic meter is adopted to be 63.1dB (A), i.e. near field noise value L apart from the noise figure of transformer case 2m distant place _a=63.1dB (A), according near field noise value L _aset up four point source forecast models and five point source forecast models, wherein four point source forecast models are: $L_{A\left(r\right)}=63.1-201g\frac{r}{2}+101g4=L_A-201g\frac{r}{2}+6,$ Five point source forecast models are: $L_{A\left(r\right)}=63.1-201g\frac{r}{2}+101g5=L_A-201g\frac{r}{2}+7,$

In the present embodiment, the physical dimension of transformer is about 6m × 6m × 3m, then maximal value l in hexahedral physical dimension _maxfor 6m, then models switching distance r is set ₀=2l _max=12m.

Step 2: distance detects

Detect the horizontal range r between current predictive point and transformer case.

Step 3: noise prediction

Comparison level distance r and models switching distance r ₀size, when horizontal range r is less than or equal to 12m, adopt four point-source models prediction r positions noise figure L _{a (r)}; When horizontal range r is greater than 12m, adopt the noise figure L of five point-source model prediction r positions _{a (r)}.

The predicted value obtained by said method and measured value compare, comparative result is as shown in table 1, wherein predicted value and the difference of measured value when each horizontal range are all within 1dB (A), show to adopt this enforcement transformer far-field noise Forecasting Methodology can dope the far-field noise of transformer preferably.

Table 1: predicted value compares with measured value.

Horizontal range	2m	5m	10m	15m	20m	25m	30m
								Test value	63.1	60.4	55	52.4	50.2	48.7	46.8
Four point-source model predicted values	/	61.2	55.4	/	/	/	/
								Five point-source model predicted values	/	/	/	52.6	50.1	48.2	46.6

The present invention also provides a kind of transformer far-field noise prognoses system, comprising:

Forecast model sets up module, for transformer being equivalent to a hexahedron, set up four point source forecast models and the five point source forecast models of transformer far-field noise according to the near field noise value of transformer, and models switching distance is set according to hexahedral physical dimension; The noise that four point source forecast models produce at future position as four point sound sources for four faces far-field noise be expressed as in hexahedron, the noise that five point source forecast models produce at future position as five point sound sources for five faces far-field noise be expressed as in hexahedron;

Distance detection module, detects the horizontal range between current predictive point and transformer case;

Noise prediction module, for the size of comparison level distance with models switching distance, if horizontal range is less than or equal to models switching distance, calculates the far-field noise value at current predictive point place according to the four point source forecast models set up; If horizontal range is greater than models switching distance, calculate the far-field noise value at current predictive point place according to the five point source forecast models set up.

In the present embodiment, forecast model is set up the expression formula that module sets up four point source forecast models and is:

$L_{A\left(r\right)}=L_A-201g\frac{r}{2}+M_4---\left(1\right)$

Forecast model sets up the expression formula that module sets up five point source forecast models:

$L_{A\left(r\right)}=L_A-201g\frac{r}{2}+M_5---\left(2\right)$

Wherein, r is the horizontal range between future position and transformer case, L _{a (r)}for transformer is in the far-field noise value of r position, L _afor the near field noise value at distance transformer 2m place, M ₄be the superimposed noise amount that four point sound sources produce, M ₅it is the superimposed noise amount that five point sound sources produce.

In the present embodiment, forecast model is set up the expression formula that module sets up four point source forecast models and is specially:

$L_{A\left(r\right)}=L_A-201g\frac{r}{2}+101g4---\left(3\right)$

Forecast model is set up the expression formula that module sets up five point source forecast models and is specially:

$L_{A\left(r\right)}=L_A-201g\frac{r}{2}+101g5---\left(4\right)$

In the present embodiment, the hexahedron that forecast model sets up module is cube; Forecast model sets up four faces in four point source forecast models of module: remove lower surface in hexahedral six surfaces, away from four surfaces of all the other beyond the surface of future position side; Forecast model sets up five faces in five point source forecast models of module: remove all the other five surfaces beyond lower surface in hexahedral six surfaces.

In the present embodiment, models switching distance arranges the twice that the models switching distance arranged according to hexahedral physical dimension in module is maximal value in hexahedral physical dimension.

Above-mentioned just preferred embodiment of the present invention, not does any pro forma restriction to the present invention.Although the present invention discloses as above with preferred embodiment, but and be not used to limit the present invention.Any those of ordinary skill in the art, when not departing from technical solution of the present invention scope, can utilize the technology contents of above-mentioned announcement to make many possible variations and modification to technical solution of the present invention, or being revised as the Equivalent embodiments of equivalent variations.Therefore, every content not departing from technical solution of the present invention, according to the technology of the present invention essence to any simple modification made for any of the above embodiments, equivalent variations and modification, all should drop in the scope of technical solution of the present invention protection.

Claims (10)

1. a transformer far-field noise Forecasting Methodology, is characterized in that concrete implementation step is:

1) forecast model is set up: transformer is equivalent to a hexahedron, set up four point source forecast models and the five point source forecast models of transformer far-field noise according to the near field noise value of transformer, and models switching distance is set according to described hexahedral physical dimension; The noise that described four point source forecast models produce at future position as four point sound sources for four faces far-field noise be expressed as in described hexahedron, the noise that described five point source forecast models produce at future position as five point sound sources for five faces far-field noise be expressed as in described hexahedron;

2) distance detects: detect the horizontal range between current predictive point and transformer case;

3) noise prediction: the size of more described horizontal range and described models switching distance, if horizontal range is less than or equal to models switching distance, calculates the far-field noise value at current predictive point place according to described four point source forecast models; If horizontal range is greater than models switching distance, calculate the far-field noise value at current predictive point place according to described five point source forecast models.

2. transformer far-field noise Forecasting Methodology according to claim 1, is characterized in that, the expression formula of described four point source forecast models is:

$L_{A\left(r\right)}=L_A-20\lg \frac{r}{2}+M_4---\left(1\right)$

The expression formula of described five point source forecast models is:

$L_{A\left(r\right)}=L_A-20\lg \frac{r}{2}+M_5---\left(2\right)$

Wherein, r is the horizontal range between future position and transformer case, L _{a (r)}for transformer is in the far-field noise value of r position, L _afor the near field noise value at distance transformer 2m place, M ₄be the superimposed noise amount that four point sound sources produce, M ₅it is the superimposed noise amount that five point sound sources produce.

3. transformer far-field noise Forecasting Methodology according to claim 2, is characterized in that: the expression formula of described four point source forecast models is:

$L_{A\left(r\right)}=L_A-20\lg \frac{r}{2}+10\lg 4---\left(3\right)$

The expression formula of described five point source forecast models is:

$L_{A\left(r\right)}=L_A-20\lg \frac{r}{2}+10\lg 5---\left(4\right)$

Wherein, r is the horizontal range between future position and transformer case, L _{a (r)}for transformer is in the far-field noise value of r position, L _afor the near field noise value at distance transformer 2m place.

4. the transformer far-field noise Forecasting Methodology according to claim 1 or 2 or 3, is characterized in that, described hexahedron is cube; In described four point source forecast models, four faces are: remove lower surface in described hexahedral six surfaces, away from four surfaces of all the other beyond the surface of future position side; In described five point source forecast models, five faces are: remove all the other five surfaces beyond lower surface in described hexahedral six surfaces.

5. transformer far-field noise Forecasting Methodology according to claim 4, is characterized in that: described step 1) in the models switching distance that arranges according to the described hexahedral physical dimension twice that is maximal value in described hexahedral physical dimension.

6. a transformer far-field noise prognoses system, is characterized in that comprising:

Forecast model sets up module, for transformer being equivalent to a hexahedron, set up four point source forecast models and the five point source forecast models of transformer far-field noise according to the near field noise value of transformer, and models switching distance is set according to described hexahedral physical dimension; The noise that described four point source forecast models produce at future position as four point sound sources for four faces far-field noise be expressed as in described hexahedron, the noise that described five point source forecast models produce at future position as five point sound sources for five faces far-field noise be expressed as in described hexahedron;

Distance detection module, detects the horizontal range between current predictive point and transformer case;

Noise prediction module, for the size of more described horizontal range and described models switching distance, if horizontal range is less than or equal to models switching distance, calculates the far-field noise value at current predictive point places according to described four point source forecast models; If horizontal range is greater than models switching distance, calculate the far-field noise value at current predictive point place according to described five point source forecast models.

7. transformer far-field noise prognoses system according to claim 6, is characterized in that: described forecast model is set up the expression formula that module sets up described four point source forecast models and is:

$L_{A\left(r\right)}=L_A-20\lg \frac{r}{2}+M_4---\left(1\right)$

Described forecast model sets up the expression formula that module sets up described five point source forecast models:

$L_{A\left(r\right)}=L_A-20\lg \frac{r}{2}+M_5---\left(2\right)$

Wherein, r is the horizontal range between future position and transformer case, L _{a (r)}for transformer is in the far-field noise value of r position, L _afor the near field noise value at distance transformer 2m place, M ₄be the superimposed noise amount that four point sound sources produce, M ₅it is the superimposed noise amount that five point sound sources produce.

8. transformer far-field noise prognoses system according to claim 7, is characterized in that: described forecast model is set up the expression formula that module sets up described four point source forecast models and is:

$L_{A\left(r\right)}=L_A-20\lg \frac{r}{2}+10\lg 4---\left(3\right)$

Described forecast model sets up the expression formula that module sets up described five point source forecast models:

$L_{A\left(r\right)}=L_A-20\lg \frac{r}{2}+10\lg 5---\left(4\right)$

Wherein, L _{a (r)}for transformer is in the far-field noise value of r position, L _afor the near field noise value at distance transformer 2m place, r is the horizontal range between future position and transformer case.

9. the transformer far-field noise prognoses system according to claim 6 or 7 or 8, is characterized in that: the described hexahedron that described forecast model sets up module is cube; Described forecast model sets up four faces in the described four point source forecast models of module: remove lower surface in described hexahedral six surfaces, away from four surfaces of all the other beyond the surface of future position side; Described forecast model sets up five faces in the described five point source forecast models of module: remove all the other five surfaces beyond lower surface in described hexahedral six surfaces.

10. transformer far-field noise prognoses system according to claim 9, is characterized in that: described models switching distance arranges the twice that the models switching distance arranged according to described hexahedral physical dimension in module is maximal value in described hexahedral physical dimension.