阅读说明：本技术 基于指数型非线性回归的配电变压器档位识别方法及系统 (Distribution transformer gear identification method and system based on exponential nonlinear regression ) 是由 覃日升 于辉 姜訸 段锐敏 马红升 邢超 奚鑫泽 张建 于 2021-08-19 设计创作，主要内容包括：本申请属于电力系统调度领域,提供基于指数型非线性回归的配电变压器档位识别方法及系统,包括：获取配变电压器低压侧的日电压曲线,根据日电压曲线得到第一备用日电压；查找第一备用日电压中异常的电压量,采用牛顿插值法对异常的电压量进行修正,得到第二备用日电压；采用指数型非线性回归模型对第二备用日电压进行归一化处理,得到最终日电压；计算最终日电压中所有电压量的均值,以及,根据配变档位选择原则对所述电压量的均值进行归属判别,得到所述配变电压器的档位。上述基于指数型非线性回归的配电变压器档位识别方法可以有效且准确地识别配电变压器档位。(The application belongs to the field of power system scheduling, and provides a distribution transformer gear identification method and a distribution transformer gear identification system based on exponential nonlinear regression, which comprise the following steps: acquiring a daily voltage curve of the low-voltage side of the distribution transformer, and acquiring a first standby daily voltage according to the daily voltage curve; searching abnormal voltage quantity in the first standby daily voltage, and correcting the abnormal voltage quantity by adopting a Newton interpolation method to obtain a second standby daily voltage; normalizing the second standby daily voltage by using an exponential nonlinear regression model to obtain a final daily voltage; and calculating the average value of all voltage quantities in the final daily voltage, and performing attribution judgment on the average value of the voltage quantities according to a distribution transformer gear selection principle to obtain the gear of the distribution transformer. The distribution transformer gear identification method based on the exponential nonlinear regression can effectively and accurately identify the distribution transformer gear.)

1. The distribution transformer gear identification method based on exponential nonlinear regression is characterized by comprising the following steps of:

acquiring a daily voltage curve of the low-voltage side of the distribution transformer, and acquiring a first standby daily voltage according to the daily voltage curve, wherein the first standby daily voltage comprises a voltage quantity of the daily voltage curve and acquisition time corresponding to the voltage quantity;

searching abnormal voltage quantity in the first standby daily voltage, and correcting the abnormal voltage quantity by adopting a Newton interpolation method to obtain a second standby daily voltage, wherein the abnormal voltage quantity comprises missing, sudden drop, sudden increase or negative voltage quantity;

normalizing the second standby daily voltage by using an exponential nonlinear regression model to obtain a final daily voltage;

and calculating the average value of all voltage quantities in the final daily voltage, and performing attribution judgment on the average value of the voltage quantities according to a distribution transformer gear selection principle to obtain the gear of the distribution transformer.

2. The distribution transformer gear identification method based on exponential nonlinear regression as recited in claim 1, wherein the process of obtaining a daily voltage curve of a low voltage side of a distribution transformer and obtaining a first standby daily voltage according to the daily voltage curve comprises:

acquiring a daily voltage curve of the low-voltage side of the distribution transformer;

counting abnormal voltage quantities in the daily voltage curve, and if the proportion of the abnormal voltage quantity collection number of the daily voltage curve to the total voltage quantity collection number is greater than or equal to a preset threshold value, rejecting the daily voltage curve;

and re-acquiring a daily voltage curve of the low-voltage side of the distribution transformer until the proportion of the abnormal voltage quantity acquisition number of the daily voltage curve to the total voltage quantity acquisition number is smaller than a preset threshold value, and obtaining a first standby daily voltage.

3. The distribution transformer gear identification method based on exponential nonlinear regression is characterized in that the preset threshold value is 20% if the daily voltage curve comprises 24 collection time points and corresponding voltage quantities, and 10% if the collection time points of the daily voltage curve comprises 96 collection time points and corresponding voltage quantities.

4. The distribution transformer gear identification method based on the exponential nonlinear regression as recited in claim 1, wherein the step of finding the abnormal voltage amount in the first backup daily voltage and correcting the abnormal voltage amount by using a newton interpolation method to obtain the second backup daily voltage comprises:

searching acquisition time corresponding to the abnormal voltage quantity to obtain an abnormal time point;

calculating an updated voltage quantity corresponding to the abnormal time point according to a Newton difference polynomial model;

and correcting the original voltage quantity corresponding to the abnormal time point by using the updated voltage quantity.

5. The distribution transformer gear identification method based on exponential nonlinear regression of claim 4, characterized in that said Newton difference polynomial model is:

in the formula, x_iAcquisition time point for abnormality, f (x)_i) For the updated voltage quantity corresponding to the abnormal acquisition time point, the interpolation approximation function is as follows:

the truncation error is:

R_n(x)＝f[x_n,x_n-1,…,x₁+x_i](x_i-x₁)…(x_i-x_n)。

6. the distribution transformer gear identification method based on exponential nonlinear regression of claim 1, characterized in that the exponential nonlinear regression model is:

y'＝β₀'+β₁x'；

in the formula, x and x 'are both acquisition time, y is the voltage quantity before normalization, y' is the voltage quantity after normalization, beta₀' and beta₁As a model parameter, β₀Is' ln beta₀。

7. An exponential nonlinear regression based distribution transformer gear identification system, wherein the exponential nonlinear regression based distribution transformer gear identification system is configured to perform the exponential nonlinear regression based distribution transformer gear identification method of any one of claims 1-6, comprising:

the voltage quantity acquisition module is used for acquiring a daily voltage curve of the low-voltage side of the distribution transformer and acquiring a first standby daily voltage according to the daily voltage curve, wherein the first standby daily voltage comprises a voltage quantity of the daily voltage curve and acquisition time corresponding to the voltage quantity;

the interpolation module is used for searching abnormal voltage quantities in the first standby daily voltage and correcting the abnormal voltage quantities by adopting a Newton interpolation method to obtain a second standby daily voltage, wherein the abnormal voltage quantities comprise missing, suddenly dropping, sudden increasing or negative voltage quantities;

the regression module is used for carrying out normalization processing on the second standby daily voltage by adopting an exponential nonlinear regression model to obtain a final daily voltage;

and the attribution judging module is used for calculating the mean value of all voltage quantities in the final daily voltage and judging the attribution of the mean value of the voltage quantities according to a distribution transformer gear selecting principle to obtain the gear of the distribution transformer.

8. The system of claim 7, wherein the voltage quantity acquisition module comprises:

the acquisition unit is used for acquiring a daily voltage curve of the low-voltage side of the distribution transformer;

the rejecting unit is used for counting abnormal voltage quantities in the daily voltage curve, and rejecting the daily voltage curve if the proportion of the abnormal voltage quantity collection number of the daily voltage curve to the total voltage quantity collection number is greater than or equal to a preset threshold value;

and the determining unit is used for re-acquiring the daily voltage curve of the low-voltage side of the distribution transformer until the proportion of the abnormal voltage quantity acquisition number of the daily voltage curve to the total voltage quantity acquisition number is smaller than a preset threshold value, so as to obtain a first standby daily voltage.

9. The system of claim 8, wherein the preset threshold of the voltage level obtaining module is set to 20% if the daily voltage curve includes 24 sampling time points and corresponding voltage levels, and is set to 10% if the sampling time points of the daily voltage curve include 96 sampling time points and corresponding voltage levels.

10. The exponential-based nonlinear regression distribution transformer range identification system of claim 7, wherein the difference module comprises:

the searching unit is used for searching the acquisition time corresponding to the abnormal voltage quantity to obtain an abnormal time point;

the updating calculation unit is used for calculating the updating voltage quantity corresponding to the abnormal time point according to a Newton difference polynomial model;

and the correcting unit is used for correcting the original voltage quantity corresponding to the abnormal time point by using the updated voltage quantity.

Technical Field

The application relates to the field of power system scheduling, in particular to a distribution transformer gear identification method and system based on exponential nonlinear regression.

Background

The distribution network refers to a network which plays a role in electric energy distribution in an electric power network, and generally refers to a network in which a low-voltage side of a distribution transformer in an electric power system directly supplies power to users. In recent years, with the increasing of the number of users, the fluctuation of the power load is faster and faster, the voltage fluctuation of the low-voltage side of the distribution transformer is more and more frequent, and the voltage is lower due to the overload of the distribution transformer when the users in part of areas use the power in the peak period, so that the problem of low voltage occurs.

The gear identification of the distribution transformer refers to checking the distribution transformer and judging the actual voltage gear attribution of the distribution transformer. At present, most distribution transformers installed in a power distribution network are no-load voltage regulating devices, the adopted identification mode mainly comprises manual offline verification and online identification based on transformer windings, the manual offline verification refers to that the existing portable distribution transformer gear checking device is manually carried for field checking, the online identification based on the transformer windings refers to that the resistance value of the windings is detected through a resistance value detection circuit, and the resistance value data is analyzed through a controller to obtain gear data.

However, the manual checking mode needs to firstly make a work ticket, close the distribution transformer and then carry the checking device to climb onto the transformer for detection and checking, so that the flow is busy, and time and labor are consumed; the mode based on the transformer winding is easy to interfere the normal operation of the transformer, even the transformer can be damaged, and due to frequent voltage fluctuation, the attribution of the distribution transformer gear can hardly be really judged by manual off-line verification at a certain time point or original data identified by the transformer winding.

Disclosure of Invention

The application provides a distribution transformer gear identification method and system based on exponential nonlinear regression, so as to provide a more effective and accurate distribution transformer gear online identification method and system.

The application provides a distribution transformer gear identification method based on exponential nonlinear regression in a first aspect, and the distribution transformer gear identification method based on exponential nonlinear regression comprises the following steps:

acquiring a daily voltage curve of the low-voltage side of the distribution transformer, and acquiring a first standby daily voltage according to the daily voltage curve, wherein the first standby daily voltage comprises a voltage quantity of the daily voltage curve and acquisition time corresponding to the voltage quantity;

searching abnormal voltage quantity in the first standby daily voltage, and correcting the abnormal voltage quantity by adopting a Newton interpolation method to obtain a second standby daily voltage, wherein the abnormal voltage quantity comprises missing, sudden drop, sudden increase or negative voltage quantity;

normalizing the second standby daily voltage by using an exponential nonlinear regression model to obtain a final daily voltage;

and calculating the average value of all voltage quantities in the final daily voltage, and performing attribution judgment on the average value of the voltage quantities according to a distribution transformer gear selection principle to obtain the gear of the distribution transformer.

Optionally, the obtaining a daily voltage curve of the low-voltage side of the distribution transformer and obtaining a first standby daily voltage according to the daily voltage curve includes:

acquiring a daily voltage curve of the low-voltage side of the distribution transformer;

counting abnormal voltage quantities in the daily voltage curve, and if the proportion of the abnormal voltage quantity collection number of the daily voltage curve to the total voltage quantity collection number is greater than or equal to a preset threshold value, rejecting the daily voltage curve;

and re-acquiring a daily voltage curve of the low-voltage side of the distribution transformer until the proportion of the abnormal voltage quantity acquisition number of the daily voltage curve to the total voltage quantity acquisition number is smaller than a preset threshold value, and obtaining a first standby daily voltage.

Optionally, if the daily voltage curve includes 24 acquisition time points and corresponding voltage quantities, the preset threshold is 20%, and if the acquisition time points of the daily voltage curve includes 96 acquisition time points and corresponding voltage quantities, the preset threshold is 10%.

Optionally, the step of searching for an abnormal voltage magnitude in the first standby daily voltage, and correcting the abnormal voltage magnitude by using a newton interpolation method to obtain a second standby daily voltage includes:

searching acquisition time corresponding to the abnormal voltage quantity to obtain an abnormal time point;

calculating an updated voltage quantity corresponding to the abnormal time point according to a Newton difference polynomial model;

and correcting the original voltage quantity corresponding to the abnormal time point by using the updated voltage quantity.

Optionally, the newton difference polynomial model is:

in the formula, x_iAcquisition time point for abnormality, f (x)_i) For the updated voltage quantity corresponding to the abnormal acquisition time point, the interpolation approximation function is as follows:

the truncation error is:

R_n(x)＝f[x_n,x_n-1,…,x₁+x_i](x_i-x₁)…(x_i-x_n)。

optionally, the exponential nonlinear regression model is:

in the formula, x and x 'are both acquisition time, y is the voltage quantity before normalization, y' is the voltage quantity after normalization, beta₀' and beta₁As a model parameter, β₀Is' ln beta₀。

The second aspect of the present application provides a distribution transformer gear identification system based on exponential nonlinear regression, where the distribution transformer gear identification system based on exponential nonlinear regression is used to execute a distribution transformer gear identification method based on exponential nonlinear regression provided by the first aspect of the present application, and the method includes:

the voltage quantity acquisition module is used for acquiring a daily voltage curve of the low-voltage side of the distribution transformer and acquiring a first standby daily voltage according to the daily voltage curve, wherein the first standby daily voltage comprises a voltage quantity of the daily voltage curve and acquisition time corresponding to the voltage quantity;

the interpolation module is used for searching abnormal voltage quantities in the first standby daily voltage and correcting the abnormal voltage quantities by adopting a Newton interpolation method to obtain a second standby daily voltage, wherein the abnormal voltage quantities comprise missing, suddenly dropping, sudden increasing or negative voltage quantities;

the regression module is used for carrying out normalization processing on the second standby daily voltage by adopting an exponential nonlinear regression model to obtain a final daily voltage;

and the attribution judging module is used for calculating the mean value of all voltage quantities in the final daily voltage and judging the attribution of the mean value of the voltage quantities according to a distribution transformer gear selecting principle to obtain the gear of the distribution transformer.

Optionally, the voltage quantity obtaining module includes:

the acquisition unit is used for acquiring a daily voltage curve of the low-voltage side of the distribution transformer;

the rejecting unit is used for counting abnormal voltage quantities in the daily voltage curve, and rejecting the daily voltage curve if the proportion of the abnormal voltage quantity collection number of the daily voltage curve to the total voltage quantity collection number is greater than or equal to a preset threshold value;

and the determining unit is used for re-acquiring the daily voltage curve of the low-voltage side of the distribution transformer until the proportion of the abnormal voltage quantity acquisition number of the daily voltage curve to the total voltage quantity acquisition number is smaller than a preset threshold value, so as to obtain a first standby daily voltage.

Optionally, if the daily voltage curve includes 24 collection time points and corresponding voltage quantities, the preset threshold of the voltage quantity obtaining module is set to 20%, and if the collection time points of the daily voltage curve includes 96 collection time points and corresponding voltage quantities, the preset threshold of the voltage quantity obtaining module is set to 10%.

Optionally, the difference module includes:

the searching unit is used for searching the acquisition time corresponding to the abnormal voltage quantity to obtain an abnormal time point;

the updating calculation unit is used for calculating the updating voltage quantity corresponding to the abnormal time point according to a Newton difference polynomial model;

and the correcting unit is used for correcting the original voltage quantity corresponding to the abnormal time point by using the updated voltage quantity.

The application provides a distribution transformer gear identification method and a distribution transformer gear identification system based on exponential nonlinear regression, the distribution transformer gear identification system based on exponential nonlinear regression is used for executing the steps of the distribution transformer gear identification method based on exponential nonlinear regression, a daily voltage curve of the low-voltage side of a distribution transformer is obtained, a first standby daily voltage is obtained according to the daily voltage curve, the first standby daily voltage comprises the voltage quantity of the daily voltage curve and the acquisition time corresponding to the voltage quantity, the abnormal voltage quantity in the first standby daily voltage is searched, the abnormal voltage quantity is corrected by adopting a Newton interpolation method to obtain a second standby daily voltage, the abnormal voltage quantity comprises the voltage quantity which is missing, suddenly dropping, suddenly increasing or is negative, the second standby daily voltage is normalized by adopting an exponential nonlinear regression model, and obtaining the final daily voltage, calculating the mean value of all voltage quantities in the final daily voltage, and performing attribution judgment on the mean value of the voltage quantities according to a distribution transformer gear selection principle to obtain the gear of the distribution transformer.

According to the scheme, the distribution transformer gear identification method based on the exponential nonlinear regression provided by the embodiment of the application interpolates abnormal values of daily voltage curves by acquiring the daily voltage curves of the low-voltage side of the distribution transformer, and eliminates the situation that the voltage is too high or too low by utilizing the exponential nonlinear regression, so that the voltage quantity is distributed near a certain voltage gear, the gear identification is more accurate, and the influence of voltage fluctuation is avoided.

Drawings

In order to more clearly explain the technical solution of the present application, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious to those skilled in the art that other drawings can be obtained according to the drawings without creative efforts.

Fig. 1 is a schematic flowchart of a distribution transformer gear identification method based on exponential nonlinear regression according to an embodiment of the present application.

Fig. 2 is a schematic structural diagram of a simulation node system model provided in an embodiment of the present application.

FIG. 3 is a diagram illustrating Newton's interpolation before and after comparison according to the examples of the present application.

FIG. 4 is a schematic diagram of a first set of 24-point daily voltage curve exponential non-linear regression provided by an embodiment of the present application.

FIG. 5 is a diagram of a second set of 24-point daily voltage curve exponential nonlinear regression provided by an embodiment of the present application.

FIG. 6 is a third set of 24-point daily voltage curve exponential nonlinear regression presented in the examples of this application.

Fig. 7 is a schematic structural diagram of a distribution transformer gear identification system based on exponential nonlinear regression according to an embodiment of the present application.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments.

As shown in fig. 1, a schematic flow chart of a distribution transformer gear identification method based on exponential nonlinear regression provided in the embodiment of the present application is shown, in which a daily voltage curve on a low-voltage side of a distribution transformer is obtained, an abnormal value of the daily voltage curve is interpolated, voltage quantities are regressed by using exponential nonlinear regression, conditions of too high or too low voltage are eliminated, a voltage quantity average value of the regressed daily voltage curve is finally calculated, and attribution determination is performed on the voltage quantity average value according to a distribution gear selection principle.

Referring to fig. 2, a structural schematic diagram of a simulation node system model provided in the embodiment of the present application is shown, in the embodiment of the present application, a standard IEEE14 node system simulation model is built in a PSCAD, simulation is performed by using the simulation model, loads of different user sides are continuously adjusted, and a plurality of daily voltage curves of the low voltage side of the distribution transformer are obtained, and basic parameters of the model are voltage level 10kV, frequency 50Hz, and transformer capacity 10 MW.

The distribution transformer gear identification method based on the exponential nonlinear regression comprises the steps 1 to 4.

Step 1, a daily voltage curve of the low-voltage side of the distribution transformer is obtained, and a first standby daily voltage is obtained according to the daily voltage curve.

The voltage quantity of the low-voltage side of the distribution transformer is data changing along with time, and the first standby daily voltage comprises the voltage quantity of a daily voltage curve and acquisition time corresponding to the voltage quantity. When the voltage quantity of the low-voltage side of the distribution transformer is collected, due to the fact that the measuring device is abnormal and the like, data measurement is incomplete, and a complete daily voltage curve cannot be formed, the acquired daily voltage curve needs to be preliminarily cleaned, unusable daily voltage curves are eliminated, and the main cleaning steps comprise S101 to S102.

S101, counting abnormal voltage quantities in the daily voltage curve, and if the proportion of the abnormal voltage quantity collection number of the daily voltage curve to the total voltage quantity collection number is larger than or equal to a preset threshold value, rejecting the daily voltage curve.

S102, a daily voltage curve of the low-voltage side of the distribution transformer is obtained again until the proportion of the abnormal voltage quantity collection number of the daily voltage curve to the total voltage quantity collection number is smaller than a preset threshold value, and a first standby daily voltage is obtained.

Further, if the daily voltage curve includes 24 collection time points and corresponding voltage quantities, the preset threshold is 20%, and if the collection time points of the daily voltage curve includes 96 collection time points and corresponding voltage quantities, the preset threshold is 10%. For example, in a plurality of daily voltage curves of the low-voltage side of the distribution transformer obtained through the simulation node system model provided in the embodiment of the present application, there are 24 collection time points included in a certain daily voltage curve, that is, 24-point daily voltage curves, and each collection time point corresponds to one voltage quantity, where the number of abnormal voltage quantities including missing, sudden drop, sudden increase, or negative voltage quantities is totally 6, and the proportion of the number of abnormal voltage quantities to the total number of voltage quantities is 25% and is greater than the preset threshold value by 10%, the daily voltage curve is rejected and the daily voltage curve is obtained again.

And 2, searching abnormal voltage quantity in the first standby daily voltage, and correcting the abnormal voltage quantity by adopting a Newton interpolation method to obtain a second standby daily voltage.

Referring to fig. 3, a diagram illustrating a comparison between before and after newton interpolation provided in the embodiments of the present application is shown. According to the embodiment of the application, the Newton interpolation method is adopted to supplement and correct missing and abnormal data, the operation times are few, and the defect that recalculation is needed when the nodes are added to the interpolation polynomial is overcome. Firstly, searching the acquisition time corresponding to the abnormal voltage quantity to obtain the abnormal time point, for example, the voltage quantity corresponding to n acquisition time points and the acquisition time points on the daily voltage curve is expressed as { (x)₁,f(x₁)),(x₂,f(x₂))…(x_n,f(x_n) In which the deletion point is (x) }_i,f(x_i) Get x as the abnormal time point)_iAnd calculating an updated voltage quantity corresponding to the abnormal time point according to the Newton interpolation polynomial model, and correcting the original voltage quantity corresponding to the abnormal time point by using the updated voltage quantity.

Wherein, the Newton difference polynomial model is:

in the formula, x_iAcquisition time point for abnormality, f (x)_i) For the updated voltage quantity corresponding to the abnormal acquisition time point, the interpolation approximation function is as follows:

the truncation error is:

R_n(x)＝f[x_n,x_n-1,…,x₁+x_i](x_i-x₁)…(x_i-x_n)。

and 3, carrying out normalization processing on the second standby daily voltage by adopting an exponential nonlinear regression model to obtain a final daily voltage.

Referring to fig. 4, a first set of 24-point daily voltage curve exponential nonlinear regression diagrams is provided in the examples of the present application.

Referring to fig. 5, a second set of 24-point daily voltage curve exponential nonlinear regression diagrams is provided in the examples of the present application.

Referring to fig. 6, a third set of 24-point daily voltage curve exponential nonlinear regression diagrams is provided in the examples of the present application.

Because the voltage fluctuates frequently along with time, the voltage difference between different acquisition time points is large, and the authenticity of the voltage cannot be accurately represented. Therefore, the voltage quantity of the daily voltage curve is subjected to regression processing by using the exponential nonlinear regression model, analysis and utilization of voltage quantity data are facilitated, the nonlinear regression straight line is obtained by using the exponential nonlinear regression calculation, the condition that the voltage is too high or too low can be eliminated by the regression straight line, the voltage quantity is distributed around a certain distribution transformer gear, and the subsequent attribution judgment is facilitated.

Wherein the exponential nonlinear regression model is as follows:

in the formula, x and x 'are both acquisition time, y is the voltage quantity before normalization corresponding to the acquisition time, y' is the voltage quantity after normalization corresponding to the acquisition time, beta₀' and beta₁As a model parameter, β₀Is' ln beta₀。

Taking the first group of 24-point daily voltage curves in fig. 4 as an example, if only peak value gear setting is adopted, an error is certainly generated, so that the regression processing is performed on each daily voltage curve to obtain a final daily voltage, and the voltage quantity after the regression processing of the first daily voltage curve has a minimum value of 420.70V and a maximum value of 424.18V; the minimum value of the voltage quantity after the regression processing of the second daily voltage curve is 411.83V, and the maximum value is 415.89V; the minimum value of the voltage quantity after the regression processing of the third daily voltage curve is 401.80V, and the maximum value is 403.81V; the minimum value of the voltage quantity after the regression processing of the fourth daily voltage curve is 390.27V, and the maximum value is 395.18V; after the regression treatment of the fifth daily voltage curve, the minimum value of the voltage quantity is 378.57V, and the maximum value is 386.26V.

And 4, calculating the average value of all voltage quantities in the final daily voltage, and performing attribution judgment on the average value of the voltage quantities according to a distribution transformer gear selection principle to obtain the gear of the distribution transformer.

Calculating the mean value of all voltage quantities in the final daily voltage, and performing attribution judgment on the mean value of the voltage quantities according to a distribution gear selection principle, wherein the specific process is as follows:

if the average value of the voltage quantities is greater than or equal to 420 and less than 430, the first gear is attributed.

If the average value of the voltage quantities is greater than or equal to 410 and less than 420, the second gear is assigned.

If the average value of the voltage quantities is greater than or equal to 400 and less than 410, the third gear is assigned.

If the average value of the voltage quantities is greater than or equal to 390 and less than 400, the fourth gear is assigned.

And if the average value of the voltage quantities is larger than or equal to 380 and smaller than 390, the fifth gear is assigned.

Taking the first group of 24-point daily voltage curves in fig. 4 as an example, the regression-processed voltage quantity average value of the first daily voltage curve is 421.68V, and the first gear is assigned; the voltage quantity average value after regression processing of the second daily voltage curve is 412.97V, and the second gear is assigned; the voltage quantity average value after regression processing of the third daily voltage curve is 402.36V, and the result is assigned as the third gear; the mean value of the voltage quantities after regression processing of the fourth daily voltage curve is 391.65V, and the fourth daily voltage curve is assigned as the fourth gear; the regression processing of the fifth daily voltage curve gave a mean value of 380.73V, which was assigned as the fifth gear.

In a second aspect of the embodiments of the present application, a distribution transformer gear identification system based on exponential nonlinear regression is provided, where the distribution transformer gear identification system based on exponential nonlinear regression is used to execute a distribution transformer gear identification method based on exponential nonlinear regression provided in the first aspect of the embodiments of the present application, and for details disclosed in the distribution transformer gear identification system based on exponential nonlinear regression provided in the second aspect of the embodiments of the present application, please refer to the distribution transformer gear identification method based on exponential nonlinear regression provided in the first aspect of the embodiments of the present application.

As shown in fig. 7, a schematic structural diagram of a distribution transformer gear identification system based on exponential nonlinear regression provided in the embodiment of the present application is shown, where the distribution transformer gear identification system based on exponential nonlinear regression includes a voltage quantity obtaining module, an interpolation module, a regression module, and an attribution distinguishing module.

The voltage quantity acquisition module is used for acquiring a daily voltage curve of the low-voltage side of the distribution transformer and acquiring a first standby daily voltage according to the daily voltage curve, wherein the first standby daily voltage comprises the voltage quantity of the daily voltage curve and acquisition time corresponding to the voltage quantity.

And the interpolation module is used for searching abnormal voltage quantities in the first standby daily voltage, and correcting the abnormal voltage quantities by adopting a Newton interpolation method to obtain a second standby daily voltage, wherein the abnormal voltage quantities comprise missing, suddenly dropping, sudden increasing or negative voltage quantities.

And the regression module is used for carrying out normalization processing on the second standby daily voltage by adopting an exponential nonlinear regression model to obtain a final daily voltage.

And the attribution judging module is used for calculating the mean value of all voltage quantities in the final daily voltage and judging the attribution of the mean value of the voltage quantities according to a distribution transformer gear selecting principle to obtain the gear of the distribution transformer.

Further, the voltage quantity acquisition module includes:

and the acquisition unit is used for acquiring a daily voltage curve of the low-voltage side of the distribution transformer.

And the rejecting unit is used for counting abnormal voltage quantities in the daily voltage curve, and rejecting the daily voltage curve if the proportion of the abnormal voltage quantity collection number of the daily voltage curve to the total voltage quantity collection number is greater than or equal to a preset threshold value.

And the determining unit is used for re-acquiring the daily voltage curve of the low-voltage side of the distribution transformer until the proportion of the abnormal voltage quantity acquisition number of the daily voltage curve to the total voltage quantity acquisition number is smaller than a preset threshold value, so as to obtain a first standby daily voltage.

Further, if the daily voltage curve includes 24 collection time points and corresponding voltage quantities, the preset threshold of the voltage quantity obtaining module is set to 20%, and if the collection time points of the daily voltage curve includes 96 collection time points and corresponding voltage quantities, the preset threshold of the voltage quantity obtaining module is set to 10%.

Further, the difference module includes:

and the searching unit is used for searching the acquisition time corresponding to the abnormal voltage quantity to obtain an abnormal time point.

And the updating calculation unit is used for calculating the updating voltage quantity corresponding to the abnormal time point according to the Newton difference polynomial model.

And the correcting unit is used for correcting the original voltage quantity corresponding to the abnormal time point by using the updated voltage quantity.

The embodiment of the application provides a distribution transformer gear identification method and system based on exponential nonlinear regression, the distribution transformer gear identification system based on exponential nonlinear regression is used for executing the steps of the distribution transformer gear identification method based on exponential nonlinear regression, a daily voltage curve of the low-voltage side of a distribution transformer is obtained, a first standby daily voltage is obtained according to the daily voltage curve, the first standby daily voltage comprises the voltage quantity of the daily voltage curve and the acquisition time corresponding to the voltage quantity, the abnormal voltage quantity in the first standby daily voltage is searched, the abnormal voltage quantity is corrected by adopting a Newton interpolation method to obtain a second standby daily voltage, the abnormal voltage quantity comprises the voltage quantity which is missing, suddenly dropping, sudden increase or negative, the normalization processing is carried out on the second standby daily voltage by adopting an exponential nonlinear regression model, and obtaining the final daily voltage, calculating the mean value of all voltage quantities in the final daily voltage, and performing attribution judgment on the mean value of the voltage quantities according to a distribution transformer gear selection principle to obtain the gear of the distribution transformer.

The distribution transformer gear identification method based on exponential nonlinear regression provided by the embodiment of the application, through obtaining the daily voltage curve of the low-voltage side of the distribution transformer, interpolation is carried out on the abnormal value of the daily voltage curve, the condition that the voltage is too high or too low is eliminated through the exponential nonlinear regression, the voltage quantity is distributed near a certain voltage gear, the gear identification is more accurate, the influence of voltage fluctuation can not be caused, and finally, gear attribution is carried out more conveniently and rapidly by calculating the mean value of the voltage quantity and combining with the gear selection principle of the distribution transformer.

The present application has been described in detail with reference to specific embodiments and illustrative examples, but the description is not intended to limit the application. Those skilled in the art will appreciate that various equivalent substitutions, modifications or improvements may be made to the presently disclosed embodiments and implementations thereof without departing from the spirit and scope of the present disclosure, and these fall within the scope of the present disclosure. The protection scope of this application is subject to the appended claims.