A Statistical Method for the Mathematical Model of Transformer Heavy Overload

Abstract

The present invention provides a mathematical model statistical method for transformer heavy overload, belonging to the technical field of power grid equipment data calculation, and solves the problems of large workload and low efficiency in the calculation and statistics of heavy overload operation in traditional methods; it includes constructing a mathematical model for the transformer, collecting the load rate values at intervals of one minute and marking them into the model; taking a certain number of consecutive data points to the right of the coordinate origin as the initial sub-model, using the least squares method for curve fitting to obtain the fitted sub-model, and calculating the load average value and the maximum value; decomposing the mathematical model at intervals of a certain number of consecutive data points into multiple calculation sub-models to obtain multiple data groups, each data group includes the load rate average value and the maximum value, and finally obtaining the continuous duration and the highest peak moment of the transformer. After querying using the specified conditions, the statistical results are obtained; based on the statistical principle, the present invention realizes a statistical process of transformer heavy overload with high precision and low computational complexity.

Description

Technical Field

[0001] The present invention belongs to the technical field of power grid equipment data calculation and is applied to the calculation process of transformer heavy overload. Specifically, it is a statistical method for the mathematical model of transformer heavy overload. Background Technique

[0002] With the continuous increase in the social demand for electricity, every time summer comes, accidents of distribution transformers being burned due to heavy overload occur frequently. Therefore, solving the problem of heavy overload operation of distribution transformers, avoiding equipment accidents, and improving the power supply quality and reliability are particularly important in the process of power grid operation.

[0003] During the operation of the power grid, the electricity consumption of users on the terminal side is continuously increasing, resulting in an increasing load on the power grid year by year, which is the main reason for the heavy overload phenomenon of transformers. At present, in the daily operation and maintenance management of substations, the monitoring of the heavy overload of transformer load rates is an important indicator for substation operation and maintenance. However, in the existing calculation methods, the introduction and application of superior calculation methods are lacking, and manual recording and manual calculation are carried out, resulting in a series of problems such as large basic data calculation volume, long calculation time, and deviation of calculation results. How to efficiently and scientifically count the heavy overload data of transformers, apply superior technologies such as statistics, computer assistance, and mathematical model construction to complete the work of transformer heavy overload statistics, and provide valuable data and auxiliary decision-making for management personnel has become an important problem faced in the process of power grid safe operation. Summary of the Invention

[0004] In order to solve the problems mentioned in the background technique, the present invention statistically counts the heavy overload data of transformers in an efficient and scientific way. The basic data of transformer load rates is preprocessed by the least squares method, and a daily load rate table is formed based on the SQL database, and calculations are carried out based on statistical principles, finally realizing a statistical process with high precision and low computation volume.

[0005] The present invention adopts the following technical solutions to achieve the purpose:

[0006] A statistical method for the mathematical model of transformer heavy overload, including the following steps:

[0007] S1. Construct a time-load rate mathematical model for the target transformer, and the time-load rate mathematical model is constructed using a two-dimensional plane rectangular coordinate system;

[0008] S2. At intervals of one minute, collect the load rate values of the target transformer, and mark the load rate values corresponding to the collection time obtained by collection into the time-load rate mathematical model;

[0009] S3. In the two-dimensional plane rectangular coordinate system of the time-load rate mathematical model, starting from the origin of coordinates to the right, a certain number of consecutive time-load rate data points are taken as the initial sub-model;

[0010] S4. Using the least squares method, curve fitting is performed on the consecutive time-load rate data points in the initial sub-model to obtain a fitted sub-model, and the average load rate data and the maximum load rate data in the fitted sub-model are calculated;

[0011] S5. In the two-dimensional plane rectangular coordinate system of the time-load rate mathematical model, starting from the origin of coordinates to the right, with a certain number of consecutive time-load rate data points as the interval, it is decomposed into multiple calculation sub-models;

[0012] S6. The multiple calculation sub-models in step S5 are used as the initial sub-model in step S3, and the content of step S4 is executed to obtain multiple data groups corresponding to the multiple fitted sub-models. Each data group includes the average load rate data and the maximum load rate data;

[0013] S7. Using the statistical grouping method to process the multiple data groups to obtain the duration and the highest peak moment of the target transformer;

[0014] S8. Query the duration and the highest peak moment using specified conditions to obtain the statistical result.

[0015] Furthermore, in step S1, the time-load rate mathematical model is constructed using the first quadrant of the two-dimensional plane rectangular coordinate system, where the abscissa of the X-axis represents the time point, and the ordinate of the Y-axis represents the load rate; the unit of the time point is minutes, the load rate is expressed as a percentage, and the point f(x, y) represents the load rate value of the target transformer at the x moment.

[0016] Specifically, in step S2, the load rate value of the target transformer is collected as follows: The load rate value of the target transformer is collected once per minute, continuously for 24 hours, and a total of 1440 load rate values are obtained.

[0017] Specifically, in step S3, in the first quadrant of the two-dimensional plane rectangular coordinate system of the time-load rate mathematical model, starting from the origin of coordinates to the right, 15 consecutive time-load rate data points are taken as the initial sub-model; in step S4, curve fitting is performed on the 15 consecutive time-load rate data points in the initial sub-model.

[0018] Specifically, in step S5, in the first quadrant of the two-dimensional rectangular coordinate system of the time-load rate mathematical model, starting from the origin of coordinates to the right, with 15 consecutive time-load rate data points as an interval, the total 1440 time-load rate data points in the time-load rate mathematical model are divided into 96 parts, forming 96 calculation sub-models, and the 96 calculation sub-models are numbered in sequence.

[0019] Specifically, in step S7, the statistical grouping method is used to process the average load rate data and the maximum load rate data in each of the 96 data groups, and the continuous duration and the highest peak moment of the target transformer are obtained.

[0020] Further, step S8 specifically includes:

[0021] S81. Perform the processing procedures described in steps S1 to S7 on multiple target transformers to obtain the continuous duration and the highest peak moment corresponding to each of the multiple target transformers;

[0022] S82. Use the SQL function row_number() over() to group and sort all the continuous duration and highest peak moment data to obtain a result set, number the result set, and take the maximum value in the result set as the output result for output;

[0023] S83. Use the conditional query method for the output result to form a daily load rate result table, and the daily load rate result table includes the daily maximum load rate and the daily maximum moment.

[0024] Preferably, based on the daily load result table, the monthly load rate result table and the annual load rate result table are obtained through statistical methods.

[0025] In summary, due to the adoption of this technical solution, the beneficial effects of the present invention are as follows:

[0026] In the process of calculating and statistics of the heavy overload operation of transformers in the power grid, the present invention innovatively constructs a reusable mathematical model. Compared with the traditional method of relying on manual comparison and analysis of all collected data, gradually calculating and judging the load rate trend, studying the data law, and searching for data results to obtain the heavy overload conclusion, the present invention is based on statistical principles and database programming, reduces the amount of basic calculation data, and improves the calculation efficiency. Because the mathematical model constructed by the present invention only needs to obtain the collected load rate data, the information such as the daily load rate result table can be obtained intuitively and accurately, greatly simplifying the workload of relevant personnel.

[0027] The present invention introduces the least squares method to preprocess the original data of the load rate, reduces the influence of abnormal data such as short circuits and impacts, and enables the statistical results to more truly and accurately reflect the equipment load situation.

[0028] Through the method of the present invention, the average load rate, the maximum load rate, and the minimum load rate in multiple time dimensions such as daily, monthly, and yearly can be efficiently calculated and statistically obtained, which can guide the reasonable planning and development of the power grid and has practical application value. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 is a schematic flow chart of the method of the present invention;

[0030] Figure 2 is a schematic diagram of the output result using SQL functions;

[0031] Figure 3 is a schematic diagram of the daily load rate result table;

[0032] Figure 4 is a schematic diagram of the yearly load rate result table. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0033] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Usually, the components of the embodiments of the present invention described and illustrated herein can be arranged and designed in various different configurations.

[0034] Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed present invention, but merely represents selected embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts fall within the scope of protection of the present invention.

[0035] A mathematical model statistical method for transformer heavy overload can be referred to Figure 1 , and includes the following steps:

[0036] Step 1: Construct a time-load rate mathematical model for the target transformer. The time-load rate mathematical model is constructed using a two-dimensional plane rectangular coordinate system.

[0037] In this embodiment, the time-load rate mathematical model is constructed using the first quadrant of the two-dimensional plane rectangular coordinate system, where the abscissa of the X-axis represents the time point, and the ordinate of the Y-axis represents the load rate; the unit of the time point is minutes, and the load rate is expressed as a percentage. The point f(x, y) represents the load rate value of the target transformer at the x moment; the completed time-load rate mathematical model can be denoted as model O.

[0038] Step 2: Collect the load rate values of the target transformer at intervals of one minute, and mark the load rate values corresponding to the collection time into the time-load rate mathematical model.

[0039] In this embodiment, collecting the load rate values of the target transformer specifically means: collecting the load rate values of the target transformer once per minute, continuously collecting for 24 hours, and obtaining a total of 1440 load rate values.

[0040] Step 3: In the two-dimensional plane rectangular coordinate system of model O, starting from the coordinate origin to the right, take a certain number of consecutive time-load rate data points as the initial sub-model.

[0041] In this embodiment, in the first quadrant of the two-dimensional plane rectangular coordinate system of model O, starting from the coordinate origin to the right, take 15 consecutive time-load rate data points as the initial sub-model, and denote the initial sub-model as model Oα.

[0042] Step 4: Use the least squares method to perform curve fitting on the 15 consecutive time-load rate data points in model Oα to obtain the fitted sub-model, denote the fitted sub-model as model Oβ, calculate the average load rate favg data and the maximum load rate fmax data in model Oβ, and denote them as Oβ-AVG and Oβ-MAX respectively, thus forming a data group.

[0043] Step 5: In the first quadrant of the two-dimensional plane rectangular coordinate system of model O, starting from the coordinate origin to the right, with 15 consecutive time-load rate data points as the interval, divide the total 1440 time-load rate data points in model O into 96 parts, forming 96 calculation sub-models, and number the 96 calculation sub-models in sequence, denoted as model Oα1 to model Oα96 respectively.

[0044] Step 6: Execute the content of Step 4 on the models Oα1 to Oα96 in Step 5 to obtain the data groups corresponding to each model Oβ, denoted as: {Oβ1-AVG, Oβ1-MAX}, {Oβ2-AVG, Oβ2-MAX}, ……, {Oβ96-AVG, Oβ96-MAX}; each data group includes the average load rate favg data and the maximum load rate fmax data.

[0045] Step 7: Use the statistical grouping method to process the 96 data groups, process the average load rate favg data and the maximum load rate fmax data in each group, and obtain the continuous duration Tα and the highest peak moment Tβ of the target transformer.

[0046] Step 8: Perform the processing procedures as in Steps 1 to 7 on multiple target transformers to obtain the duration Tα and the highest peak moment Tβ corresponding to each of the multiple target transformers; group and sort all the duration and highest peak moment data using the SQL function row_number() over() to obtain a result set, number the result set, and take the maximum value in the result set as the output result for output. Refer to Figure 2 for illustration.

[0047] Finally, use the conditional query method on the output result to form a daily load rate result table, where the daily load rate result table includes the daily maximum load rate and the daily maximum moment. Refer to Figure 3 for illustration.

[0048] Based on the daily load result table, the monthly load rate result table and the annual load rate result table can be obtained through statistical methods. Refer to Figure 4 for illustration; the load rate result tables at each time scale can include information such as the maximum load rate, average load rate information, maximum moment, and number of days of heavy overload according to actual application requirements, enabling relevant personnel to have an accurate basis for the heavy overload operation situation and subsequent management of transformers; at the same time, since the method in this embodiment is based on statistical principles and database programming, the amount of basic calculation data is reduced and the calculation efficiency is improved; for the same calculation and statistical content, compared with the traditional method, the data for calculating the daily load rate is reduced from about 210,000 to 14,000, and the calculation efficiency is improved by about 93%. The same applies to the monthly load rate and the annual load rate.

Claims

1. A mathematical model statistical method for transformer heavy overload, characterized in that, It includes the following steps: S1. Construct a time-load rate mathematical model for the target transformer, and the time-load rate mathematical model is constructed using a two-dimensional rectangular coordinate system; S2. At intervals of one minute, collect the load rate values of the target transformer, and mark the load rate values corresponding to the collection time obtained by collection into the time-load rate mathematical model; S3. In the two-dimensional rectangular coordinate system of the time-load rate mathematical model, starting from the coordinate origin to the right, take a certain number of consecutive time-load rate data points as the initial sub-model; S4. Use the least squares method to perform curve fitting on the consecutive time-load rate data points in the initial sub-model to obtain a fitted sub-model, and calculate the average load rate data and the maximum load rate data in the fitted sub-model; S5. In the two-dimensional rectangular coordinate system of the time-load rate mathematical model, starting from the coordinate origin to the right, divide it into multiple calculation sub-models at intervals of a certain number of consecutive time-load rate data points; S6. Take the multiple calculation sub-models in step S5 as the initial sub-models in step S3, and execute the content of step S4 to obtain multiple data groups corresponding to multiple fitted sub-models, and each data group includes average load rate data and maximum load rate data; S7. Use the statistical grouping method to process the multiple data groups to obtain the duration and the highest peak moment of the target transformer; S8. Query the duration and the highest peak moment using specified conditions to obtain the statistical result.

2. The mathematical model statistical method for transformer heavy overload according to claim 1, characterized in that: In step S1, the time-load rate mathematical model is constructed using the first quadrant of the two-dimensional rectangular coordinate system, where the abscissa of the X-axis represents the time point, and the ordinate of the Y-axis represents the load rate; the unit of the time point is minute, the load rate is expressed as a percentage, and the point f(x,y) represents the load rate value of the target transformer at the x moment.

3. A statistical method for the mathematical model of transformer heavy overload according to claim 2, characterized in that: In step S2, collecting the load rate value of the target transformer specifically means: collecting the load rate value of the target transformer once per minute, continuously collecting for 24 hours, and a total of 1440 load rate values are obtained.

4. A statistical method for the mathematical model of transformer heavy overload according to claim 3, characterized in that: In step S3, in the first quadrant of the two-dimensional rectangular coordinate system of the time-load rate mathematical model, starting from the coordinate origin to the right, take 15 consecutive time-load rate data points as the initial sub-model; in step S4, perform curve fitting on the 15 consecutive time-load rate data points in the initial sub-model.

5. A statistical method for the mathematical model of transformer heavy overload according to claim 4, characterized in that: In step S5, in the first quadrant of the two-dimensional rectangular coordinate system of the time-load rate mathematical model, starting from the coordinate origin to the right, divide the total 1440 time-load rate data points in the time-load rate mathematical model into 96 parts at intervals of 15 consecutive time-load rate data points to form 96 calculation sub-models, and number the 96 calculation sub-models in sequence.

6. A statistical method for the mathematical model of transformer heavy overload according to claim 5, characterized in that: In step S7, use the statistical grouping method to process the average load rate data and the maximum load rate data in each of the 96 data groups to obtain the duration and the highest peak moment of the target transformer.

7. A statistical method for the mathematical model of transformer heavy overload according to claim 6, characterized in that, Step S8 specifically includes: S81. Perform the processing procedures described in steps S1 to S7 on multiple target transformers to obtain the duration and the moment of the highest peak corresponding to each of the multiple target transformers; S82. Use the SQL function row_number() over() to group and sort all the duration and the moment of the highest peak data, obtain a result set, number the result set, and take the maximum value in the result set as the output result for output; S83. Use the method of conditional query for the output result to form a daily load factor result table, where the daily load factor result table includes the daily maximum load factor and the daily maximum moment.

8. A statistical method for the mathematical model of transformer heavy overload according to claim 7, characterized in that: According to the daily load factor result table, obtain a monthly load factor result table and an annual load factor result table through statistical methods.

Images