Transformer substation missing data preprocessing method

Abstract

The invention discloses a transformer substation missing data preprocessing method. The transformer substation missing data preprocessing method comprises the following steps: obtaining transformer substation switch cabinet data with missing data; calculating the total variation quantity of the substation switch cabinet data, and comparing the total variation quantity of the substation switch cabinet data with a preset first standard threshold value; performing piecewise linear interpolation method filling processing on the substation switch cabinet data of which the total variation quantity is higher than a first standard threshold value; calculating the missing degree of the transformer substation switch cabinet data of which the total variation quantity is lower than a first standard threshold value, and comparing the missing degree of the transformer substation switch cabinet data with a preset second standard threshold value; and performing spline interpolation method filling processing on the substation switch cabinet data of which the missing degree is higher than the second standard threshold value, and performing Newton interpolation method filling processing on the substation switch cabinet data of which the missing degree is lower than the second standard threshold value, so that the problem that a single interpolation method is adopted when data missing is processed in the prior art is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of substation missing data preprocessing, and in particular to a substation missing data preprocessing method. Background Art

[0002] With the development of my country's economy and society, the demand for electricity has increased, the scale of the power grid has gradually expanded, and the level of informatization has become higher and higher, which has put forward higher requirements for the intelligence of the power grid. Substation switchgear is an electrical equipment and an important equipment in the power system, which is used for opening and closing, control and protection in the process of power generation, transmission, distribution and power conversion.

[0003] The main functions of substation switchgear include controlling current, protecting circuits and disconnecting circuits. If the relevant parameters of the substation switchgear are missing, the switchgear may operate beyond the rated value, increasing the risk of equipment overload and failure, and the protection devices (such as overcurrent protection, short-circuit protection, etc.) cannot be set correctly, resulting in failure to cut off the faulty circuit in time, increasing the possibility of accidents and greatly enhancing safety hazards. Therefore, it is very important to complete and comprehensively fill in the missing parameters of the substation switchgear to ensure the integrity of relevant data and parameters.

[0004] Many existing technologies use a single interpolation method when dealing with missing data. However, a single interpolation method may have multiple impacts when dealing with missing data, reducing the flexibility and accuracy of data processing, and leading to information loss and uncertainty. Summary of the invention

[0005] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a substation missing data preprocessing method, which adopts multiple interpolation methods according to the data characteristics of the switch cabinet missing data to solve the problem of the prior art using a single interpolation method when processing missing data.

[0006] To achieve the above object, the present invention provides the following technical solutions: A method for preprocessing missing data of a substation comprises the following steps: in a preferred embodiment, obtaining substation switch cabinet data with missing data; calculating the total variation of the substation switch cabinet data, and comparing the total variation of the substation switch cabinet data with a preset first standard threshold; performing piecewise linear interpolation filling processing on the substation switch cabinet data with a total variation higher than the first standard threshold; calculating the missing degree of the substation switch cabinet data with a total variation lower than the first standard threshold, and comparing the missing degree of the substation switch cabinet data with a preset second standard threshold; performing spline interpolation filling processing on the substation switch cabinet data with a missing degree higher than the second standard threshold, and performing Newton interpolation filling processing on the substation switch cabinet data with a missing degree lower than the second standard threshold.

[0007] In a preferred embodiment, the calculation of the total variation of the substation switch cabinet data includes the following steps: placing the acquired substation switch cabinet data into a one-dimensional switch cabinet data set according to a time series; calculating the total variation and segmented total variation of the substation switch cabinet data; and performing comprehensive processing on the obtained total variation and segmented total variation to obtain the total variation.

[0008] In a preferred embodiment, the substation switch cabinet data whose total variation is higher than the first standard threshold is filled by piecewise linear interpolation, and the processing method is as follows: traverse the one-dimensional switch cabinet data set, record the position numbers of all missing data in the one-dimensional switch cabinet data set; take the data of the previous position number of the missing data as the front data group, and take the data of the next position number of the missing data as the back data group; substitute the front data group and the back data group into the linear interpolation formula to estimate the missing values, and update the missing values ​​at the missing data of the corresponding position number.

[0009] In a preferred embodiment, the missing degree of the substation switch cabinet data whose total variation is lower than the first standard threshold is calculated by the following calculation method: the substation switch cabinet data whose total variation is lower than the first standard threshold is placed in the first sub-switch cabinet data set; a preset unit time interval is calculated, and the missing degree of the substation switch cabinet data in each time interval is calculated; and the missing degree of the substation switch cabinet data in all time intervals is compared with the second standard threshold respectively.

[0010] In a preferred embodiment, the substation switch cabinet data with a missing degree higher than the second standard threshold is filled with spline interpolation, and the processing method is as follows: the unit time intervals are connected with a polynomial; the end time series of each unit time interval is used as the spline node of the spline equation; a cubic spline function is constructed, and the spline nodes are solved based on the cubic spline function to obtain the value to be inserted.

[0011] In a preferred embodiment, the substation switch cabinet data with a missing degree lower than the second standard threshold is filled with Newton interpolation method, and the processing method is as follows: the substation switch cabinet data with a missing degree lower than the second standard threshold is placed in the second sub-switch cabinet data set; based on the second sub-switch cabinet data set, the quotient difference of the missing data is calculated; a Newton interpolation polynomial is constructed, and the calculated difference quotient is substituted into the Newton interpolation polynomial to obtain the value to be filled.

[0012] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: 1. Through the missing data preprocessing method of the substation, multiple interpolation methods are adopted according to the data characteristics of the missing data of the switch cabinet, which effectively solves the problem of using a single interpolation method when processing missing data in the prior art.

[0013] 2. The total variation is used as an indicator to measure the degree of data change and is compared with the preset first standard threshold, so that the volatility of the data can be quantified, thereby achieving a quick judgment of the impact of missing data on the smoothness of the overall data.

[0014] 3. By classifying the substation switch cabinet data according to the degree of missingness, different processing strategies can be used for data with different degrees of missingness, thereby improving the pertinence and effectiveness of data repair. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 A flow chart of a method for preprocessing missing data of a substation provided in an embodiment of the present application; DETAILED DESCRIPTION The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0016] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the statement "include..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0017] Embodiment 1, Figure 1 The present invention provides a substation missing data preprocessing method, comprising the following steps: S1 obtains substation switch cabinet data with missing data; S2 calculates the total variation of the substation switch cabinet data, and compares the total variation of the substation switch cabinet data with a preset first standard threshold; S3 performs piecewise linear interpolation filling processing on the substation switch cabinet data whose total variation is higher than the first standard threshold; S4 calculates the missing degree of substation switch cabinet data whose total variation is lower than the first standard threshold, and compares the missing degree of substation switch cabinet data with a preset second standard threshold; S5 performs spline interpolation to fill in the substation switch cabinet data whose missing degree is higher than the second standard threshold, and performs Newton interpolation to fill in the substation switch cabinet data whose missing degree is lower than the second standard threshold.

[0018] The method of calculating the total variation of the substation switch cabinet data comprises the following steps: The acquired substation switch cabinet data is put into a one-dimensional switch cabinet data set according to the time series; Calculate the total variation and section total variation of substation switch cabinet data; The total variation and segmented total variation are processed comprehensively to obtain the total variation.

[0019] The total variation is a quantity used to characterize the degree of "oscillation" of a function. It can be used to quantify the degree of change of a signal or data set. In addition, this quantification method is also applicable to non-smooth functions and has no particular preference for the smoothness of the function. The total variation is an indicator used to measure the smoothness of an image.

[0020] The one-dimensional switch cabinet dataset is defined as follows:

[0021] Where: k is the one-dimensional switchgear data set, and m is the total number of data in the data set.

[0022] The total variation is defined as follows:

[0023] Where: is the total variation, and m is the total number of data in the data set.

[0024] The piecewise total variation is defined as follows:

[0025] Where: is the total variation of the segment.

[0026] The total variation is calculated as follows:

[0027] Where: is the total segment variation, For total variation.

[0028] The specific steps of comparing the total variation of the substation switch cabinet data with a preset first standard threshold are as follows: If the total variation is greater than or equal to the preset first standard threshold, it means that the data variation caused by missing data is large and the data smoothness is weak; If the total variation is less than the preset first standard threshold, it means that the data change caused by the missing data is small and the data is relatively smooth.

[0029] As an indicator to measure the degree of data change, total variation can be used to evaluate the performance of substation switchgear. When the operating status of the equipment fluctuates, analyzing the changing trend of the variation can help determine whether the working status of the equipment is normal.

[0030] In addition, by comparing the total variation with the standard threshold, potential faults in equipment operation can be discovered in a timely manner. This early warning mechanism can greatly improve the reliability and safety of equipment and avoid economic losses and safety hazards caused by major faults.

[0031] When using the Total Variation (TV) method, boundary effects are usually a significant problem, especially in image processing and signal analysis. Boundary effects refer to the edges of data where the application of TV may lead to unnatural results, such as oversmoothing or artifacts. These effects often affect the overall analysis results, so the present invention combines a variant of TV, namely, segmented TV, to reduce the boundary effects of TV.

[0032] Total variation refers to the degree of fluctuation of the data series in the entire time interval, which can reflect the variability of the signal; while the piecewise total variation divides the time series into multiple segments and calculates the total variation in each segment. By combining the total variation with the piecewise total variation, the changing characteristics of the data can be captured more finely.

[0033] Piecewise linear interpolation is a local interpolation method that approximates the value of an unknown point by connecting straight line segments between data points. The mathematical principle is based on linear interpolation. This method divides a given sample point interval into multiple different intervals and uses a linear equation for interpolation in each interval. It is assumed that within each small interval, the data change is linear, so the values ​​of other points in the interval can be calculated from known data points.

[0034] Specifically, piecewise linear interpolation is a method of estimating missing data points by drawing straight lines between known data points. This method assumes that the data changes linearly between adjacent data points, so the value of any point between them can be calculated by using two known data points.

[0035] The substation switch cabinet data whose total variation is higher than the first standard threshold value is filled by piecewise linear interpolation method, and the specific processing method is as follows: Traverse the one-dimensional switch cabinet data set and record the position sequence numbers of all missing data in the one-dimensional switch cabinet data set; The data of the position number before the missing data is taken as the front data group, and the data of the position number after the missing data is taken as the back data group; Substitute the previous data group and the next data group into the linear interpolation formula to estimate the missing values, and update the missing values ​​at the missing data of the corresponding position number.

[0036] The piecewise linear interpolation method is actually the first-order form of Newton's interpolation method, in which the piecewise linear interpolation polynomial is expressed as follows:

[0037] in, is the interpolation coefficient, as follows:

[0038] Where: is a piecewise linear interpolation polynomial, n is the total number of missing data, i is the position number of the i-th missing data, are the missing values ​​to be interpolated, is the posterior data set corresponding to the i-th missing data, is the previous data group corresponding to the i-th missing data.

[0039] The linear interpolation formula is specifically calculated as follows:

[0040] Where: are the missing values ​​to be interpolated, is the posterior data set corresponding to the i-th missing data, is the previous data group corresponding to the i-th missing data.

[0041] When performing interpolation processing using piecewise linear interpolation, the influence of outliers on the interpolation results can be reduced by analyzing the surrounding data points, thereby improving the reliability of the data.

[0042] The calculation method of the missing degree of the substation switch cabinet data whose total variation is lower than the first standard threshold is as follows: placing the substation switch cabinet data with a total variation lower than a first standard threshold value in a first sub-switch cabinet data set; Preset a unit time interval and calculate the missing degree of substation switch cabinet data in each time interval; The missingness of the substation switch cabinet data in all time intervals is compared with the second standard threshold value respectively.

[0043] The missing degree is calculated as follows:

[0044] Where: Q is the missing degree calculation formula, is the number of missing items in a unit time interval, It is the total number of substation switch cabinet data whose total variation within a unit time interval is lower than the first standard threshold.

[0045] By calculating the missing degree and processing it according to the preset threshold, different strategies can be adopted according to the missing situation of specific data sets, avoiding the "one-size-fits-all" approach. In data analysis, the transparency and interpretability of missing data processing are enhanced, making the data analysis results more credible.

[0046] The strip interpolation method is an improvement of the piecewise linear interpolation method. It is a method to fit a smooth continuous function curve through a given set of scattered data. It approximates the data in a small interval through low-degree polynomials, and uses the connection conditions of these polynomials to ensure the smoothness of the entire curve.

[0047] The basic principle of the strip interpolation method is to obtain the curve function group by solving the three-moment equations. Specifically, the strip interpolation method uses a cubic equation on each segment interval and satisfies the interpolation condition, continuity condition, first-order derivative continuity condition, and second-order derivative continuity condition to ensure the smoothness of the curve.

[0048] The substation switch cabinet data with a missing degree higher than the second standard threshold is filled by the spline interpolation method, and the processing method is as follows: Connecting the unit time intervals with a cubic polynomial as a cubic spline; The terminal time series of each unit time interval is used as the spline node of the spline equation; Construct a cubic spline function, solve the spline nodes based on the cubic spline function, and obtain the value to be inserted.

[0049] In each unit time interval, the spline function is defined as follows:

[0050] Where: is the spline function, is a cubic polynomial in the time interval, is the interval of the spline equation, and z is the length of the unit time interval.

[0051] The cubic spline function is constructed as follows:

[0052] Where: is the cubic spline function within each unit time interval, are the coefficients of the interval of the spline equation, is the missing data to be inserted.

[0053] Newton interpolation is an interpolation method widely used in numerical analysis. It was proposed by British scientist Isaac Newton in the 17th century. It uses known discrete data points to construct a polynomial function so that the value of the polynomial function at a given data point is equal to the value of the original function, and the original function is approximated by this polynomial function.

[0054] The characteristic of Newton interpolation method is its inheritance, that is, when adding additional interpolation points, the previous calculation results can be used to reduce the amount of calculation. Each additional point will not cause the previous recalculation, and only the difference quotient related to the newly added point needs to be calculated. This makes Newton interpolation method have higher computational efficiency when processing large amounts of data.

[0055] The substation switch cabinet data with missing degree lower than the second standard threshold is filled by Newton interpolation method, and the processing method is as follows: The substation switch cabinet data with a missing degree lower than the second standard threshold is placed in the second sub-switch cabinet data set; Based on the second sub-switch cabinet data set, calculate the quotient difference of the missing data; Construct a Newton interpolation polynomial, and substitute the calculated difference quotient into the Newton interpolation polynomial to obtain the value to be filled.

[0056] The second sub-switch cabinet data set is in the following format:

[0057] Where: is the second sub-switch cabinet data set, u is the total number of substation switch cabinet data with a missing degree lower than the second standard threshold, is the position number of the second sub-switch cabinet data set, for The corresponding numerical value.

[0058] The calculation formula for calculating the quotient difference of missing data is as follows:

[0059] Where: is the formula for calculating the quotient difference, is the first-order quotient difference, is the second-order quotient difference, u is the total number of substation switch cabinet data with missingness lower than the second standard threshold, is the position number of the second sub-switch cabinet data set, for The corresponding numerical value.

[0060] The Newton interpolation polynomial is constructed as follows:

[0061] Where: is the Newton interpolation polynomial, is the first-order quotient difference, is the second-order quotient difference, is the value corresponding to the position number of the second sub-switch cabinet data set, is the missing data in the second sub-switch cabinet dataset.

[0062] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0063] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product.

[0064] Those of ordinary skill in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0065] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0066] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0067] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for preprocessing missing data of a substation, characterized in that: The following steps are involved: Obtain substation switchgear data with missing data; Calculating the total variation of the substation switch cabinet data, and comparing the total variation of the substation switch cabinet data with a preset first standard threshold; The substation switch cabinet data with a total variation higher than the first standard threshold value is filled by piecewise linear interpolation method; Calculate the missing degree of substation switch cabinet data whose total variation is lower than the first standard threshold, and compare the missing degree of substation switch cabinet data with a preset second standard threshold; The substation switch cabinet data with a missing degree higher than the second standard threshold are filled by the spline interpolation method, and the substation switch cabinet data with a missing degree lower than the second standard threshold are filled by the Newton interpolation method.

2. The method for preprocessing missing data of a substation according to claim 1, characterized in that: The method of calculating the total variation of the substation switch cabinet data comprises the following steps: The acquired substation switch cabinet data is put into a one-dimensional switch cabinet data set according to the time series; Calculate the total variation and section total variation of substation switch cabinet data; The total variation and segmented total variation are processed comprehensively to obtain the total variation.

3. The method for preprocessing missing data of a substation according to claim 2, characterized in that: The substation switch cabinet data whose total variation is higher than the first standard threshold value is filled by piecewise linear interpolation method, and the specific processing method is as follows: Traverse the one-dimensional switch cabinet data set and record the position sequence numbers of all missing data in the one-dimensional switch cabinet data set; The data of the position number before the missing data is taken as the front data group, and the data of the position number after the missing data is taken as the back data group; Substitute the previous data group and the next data group into the linear interpolation formula to estimate the missing values, and update the missing values ​​at the missing data of the corresponding position number.

4. The method for preprocessing missing data of a substation according to claim 3, characterized in that: The calculation method of the missing degree of the substation switch cabinet data whose total variation is lower than the first standard threshold is as follows: placing the substation switch cabinet data with a total variation lower than a first standard threshold value in a first sub-switch cabinet data set; Preset a unit time interval and calculate the missing degree of substation switch cabinet data in each time interval; The missingness of the substation switch cabinet data in all time intervals is compared with the second standard threshold value respectively.

5. The method for preprocessing missing data of a substation according to claim 4, characterized in that: The substation switch cabinet data with a missing degree higher than the second standard threshold is filled by the spline interpolation method, and the processing method is as follows: Connecting the unit time intervals using a polynomial; The terminal time series of each unit time interval is used as the spline node of the spline equation; Construct a cubic spline function, solve the spline nodes based on the cubic spline function, and obtain the value to be inserted.

6. The method for preprocessing missing data of a substation according to claim 5, characterized in that: The substation switch cabinet data with missing degree lower than the second standard threshold is filled by Newton interpolation method, and the processing method is as follows: The substation switch cabinet data with a missing degree lower than the second standard threshold is placed in the second sub-switch cabinet data set; Based on the second sub-switch cabinet data set, calculate the quotient difference of the missing data; Construct a Newton interpolation polynomial, and substitute the calculated difference quotient into the Newton interpolation polynomial to obtain the value to be filled.

7. The method for preprocessing missing data of a substation according to claim 6, characterized in that: The total variation is calculated as follows: Where: TV is the total variation, m is the total number of data in the one-dimensional switchgear data set, is the total variation, is the total variation of the segment.

8. The method for preprocessing missing data of a substation according to claim 7, characterized in that: The linear interpolation formula is specifically calculated as follows: Where: are the missing values ​​to be interpolated, is the posterior data set corresponding to the i-th missing data, is the previous data group corresponding to the i-th missing data.

9. The method for preprocessing missing data of a substation according to claim 8, characterized in that: The missing degree is calculated as follows: Where: Q is the missing degree calculation formula, is the number of missing items in a unit time interval, It is the total number of substation switch cabinet data whose total variation within a unit time interval is lower than the first standard threshold.

10. The method for preprocessing missing data of a substation according to claim 9, characterized in that: The Newton interpolation polynomial is constructed as follows: Where: is the Newton interpolation polynomial, is the first-order quotient difference, is the second-order quotient difference, is the value corresponding to the position number of the second sub-switch cabinet data set, is the missing data in the second sub-switch cabinet dataset.

Images