Power grid distribution heavy overload prediction system and method

By obtaining the power consumption difference in the power grid distribution system and using the Prophet-LSTM model, the power consumption trends in the power grid area are analyzed, and the problem of inaccurate prediction of the operating status of the distribution transformer caused by the complex power consumption units in the area is solved. Improved prediction accuracy and reliability.

CN118917455APending Publication Date: 2024-11-08HEFEI SWITCH FACTORY
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410893672.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-04
Publication Date
2024-11-08

AI Technical Summary

Technical Problem

In power distribution work in power grid, due to the complex power consumption units in the region, the problem of predicting the operating status of the distribution transformer based on the community occupancy rate is not accurate enough.

Method used

The power consumption difference is obtained through the target power consumption data module, the associated power consumption data module and the historical reload data module. Combined with the Prophet-LSTM prediction model, the power consumption trends in the same power consumption area are analyzed, interference factors are reduced, and prediction accuracy is improved. .

Benefits of technology

It improves the accuracy of overload prediction of distribution transformers, enhances the reliability of prediction results, reduces prediction interference, and ensures equipment safety and power quality.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118917455A_ABST
    Figure CN118917455A_ABST
Patent Text Reader

Abstract

The invention provides a power grid power distribution heavy overload prediction system and method, and relates to the technical field of power distribution heavy overload prediction, and the system comprises a central processing module, a prediction model, a target power utilization data module, an associated power utilization data module, and a historical heavy overload data module. The target power consumption data module is used for acquiring the type and the power consumption of power consumption equipment of a predicted unit, and the associated power consumption data module is used for acquiring the type and the power consumption of power consumption equipment of units except the predicted unit in the same power consumption area; according to the technical key points, the variance of the difference values of data obtained by a target power utilization data module, data obtained by an associated power utilization data module and data in a historical heavy overload data module is solved, so that the dispersion degree between actual data and historical data is determined, interference factors of a predicted power utilization unit can be eliminated conveniently, and the prediction accuracy is improved. And the trend that the associated power consumption data in the same power consumption area influences the overload of the predicted unit weight is analyzed, so that the accuracy and credibility of the prediction result are improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of power distribution severe overload prediction, and in particular to a power grid power distribution severe overload prediction system and method. Background Art

[0002] The whole of the power system consisting of substations and transmission and distribution lines of various voltages is called the power grid. It includes three units: substation, transmission, and distribution. The task of the power grid is to transmit and distribute electrical energy and change the voltage.

[0003] In the process of power distribution, the power grid receives electric energy from the transmission network or regional power plants, and distributes it locally or step by step according to voltage to various users through distribution facilities. The distribution network is composed of overhead lines, cables, towers, distribution transformers, disconnectors, reactive power compensation capacitors and some ancillary facilities, and plays an important role in distributing electric energy in the power grid.

[0004] The distribution network is simply divided into medium voltage distribution network and low voltage distribution network, with the distribution transformer as the dividing point. The voltage of medium voltage distribution network is commonly 10kV or 20kV, and a few are 35kV; the voltage of low voltage distribution network is 380V (three-phase) and 220V (single-phase).

[0005] The patent document with publication number CN109063922B is a method for predicting heavy overload of distribution transformers based on community occupancy rate. The method obtains detailed data of customer electricity meters and water meters in the community with multiple meters integrated into one by associating them with user numbers from the electricity consumption information collection system of power users; the community occupancy rate is calculated based on the customer electricity and water consumption; finally, the heavy overload judgment rule is used to determine whether there is a risk of overload in the distribution transformer under the saturated occupancy state of the community. This can effectively solve the problem that it is difficult to accurately carry out medium- and long-term heavy overload prediction of distribution transformers due to the sudden change and volatility of the load in the distribution area, and can accurately predict the operating status of distribution transformers, and give early warning of the risk of heavy overload of distribution transformers, which will help to improve the operation and maintenance level of distribution transformers and ensure the safety of equipment and the quality of electricity consumption of community residents.

[0006] However, in the process of implementing the above technical solution, it is found that the above technical solution has the following technical problems:

[0007] The work of power grid distribution is complex, and its objects are spread all over the city. When the heavy overload of transformer is predicted by the occupancy rate of the community, the safety of equipment and the quality of electricity consumption of community residents can be guaranteed. However, in the actual application process, power distribution exists in the form of region, and the electricity users in the region are relatively complex, which can easily affect the prediction of the operating status of distribution transformer by the occupancy rate of the community, resulting in inaccurate prediction. Summary of the invention

[0008] In order to overcome the problem that in the actual application of the existing distribution transformer heavy overload prediction method, since the power distribution exists in the form of regions and the power users in the region are relatively complex, it is easy to affect the prediction of the distribution transformer operating status by the community occupancy rate, resulting in inaccurate prediction, the embodiment of the present application provides a power grid distribution heavy overload prediction system and method, which calculates the variance of the difference between the data obtained from the target power consumption data module, the data obtained from the associated power consumption data module and the data in the historical heavy overload data module to determine the degree of discreteness between the actual data and the historical data, so as to eliminate the interference factors of the predicted power user unit, analyze the trend of the associated power consumption data in the same power consumption area affecting the heavy overload of the predicted unit, and make the prediction result more reliable.

[0009] The technical solution adopted by the embodiment of the present application to solve the technical problem is:

[0010] A power grid distribution heavy overload prediction system, comprising a central processing module, a prediction model, a target power consumption data module, a related power consumption data module and a historical heavy overload data module;

[0011] Wherein, the target power consumption data module is used to obtain the type and power consumption of the predicted unit power consumption equipment;

[0012] The associated power consumption data module is used to obtain the type and power consumption of the unit power consumption equipment other than the predicted unit in the same power consumption area;

[0013] The historical overload data module is used to record the type of electrical equipment and power consumption when the unit is predicted to have circuit overloaded in the past;

[0014] The central processing module obtains the type of power-consuming equipment and power consumption from the target power consumption data module, the associated power consumption data module and the historical heavy overload data module, obtains the power consumption difference, and performs a difference comparison;

[0015] The prediction model predicts the overload condition of the predicted unit according to the power difference value obtained by the central processing module and the difference point obtained by comparison.

[0016] In a possible implementation, an input end of the associated power consumption data module is connected to an associated unit screening module, and the associated unit screening module classifies units other than predicted units in the same power consumption area.

[0017] In a possible implementation, the type of corresponding power-consuming equipment obtained by the target power consumption data module and the associated power consumption data module is also distinguished from the model and power, and the total number is counted; the power consumption of the corresponding power-consuming equipment obtained by the target power consumption data module and the associated power consumption data module includes its power consumption in different seasons of a year and different time periods of a day.

[0018] In a possible implementation, the current power consumption data of the target unit and the associated units, and the difference calculated from the historical heavy overload data are divided into positive and negative types, and are marked separately.

[0019] In a possible implementation, the prediction model is Prophet-LSTM, which makes predictions based on the actual power consumption status of the region and the difference between the current power consumption and the historical heavy overload power consumption.

[0020] In a possible implementation, the prediction items of the prediction model for the predicted unit include the overload period of the target unit, the influencing factors affecting the variation of the overload situation, and the time of severe overload of the distribution power.

[0021] A method for predicting heavy overload of power distribution network, comprising: S1: counting the type, power and quantity of power consumption equipment of a target unit, and obtaining the power consumption of the target unit in different seasons and time periods;

[0022] S2: Count the types of related units in the same power supply area at the current stage, and count the types, power and quantity of electrical equipment of each related unit, and obtain their power consumption in different seasons and time periods;

[0023] S3: extract the current power consumption data of the target unit and the related units, compare them with the historical heavy overload data, and obtain the power consumption difference;

[0024] S4: Screening positive and negative differences, comparing the current power consumption with the historical heavy overload power consumption according to the actual power consumption status of the area;

[0025] S5: Use Prophet-LSTM to predict the overload period of the target unit, the factors affecting the variation of the overload situation, and the time of heavy overload of the distribution within the same power consumption area.

[0026] In one possible implementation, the difference characteristics between the current power consumption in S4 and the historical heavy overload power consumption include the increase or decrease in the number of target unit power equipment, the adjustment of the power consumption time of the target unit power equipment, the change in the number of associated power unit power equipment and the adjustment of the power consumption time of the associated power unit power equipment.

[0027] The beneficial effects of this application are:

[0028] First, in this scheme, by obtaining the power consumption difference between the power consumption data of the predicted unit power consumption equipment obtained by the target power consumption data module and the data in the historical overload data module, and obtaining the difference between the power consumption data of the unit power consumption equipment other than the predicted unit in the same power consumption area obtained by the associated power consumption data module and the data in the historical overload data module, the influence of the power consumption of such equipment on the overall unit power overload can be determined in the subsequent prediction process, which is conducive to reducing prediction interference and improving prediction accuracy;

[0029] Second, in this scheme, the variance of the difference between the data obtained from the target electricity consumption data module, the data obtained from the associated electricity consumption data module and the data in the historical heavy overload data module is calculated to determine the degree of discreteness between the actual data and the historical data, which facilitates the removal of interference factors of the predicted electricity consumption unit and analyzes the trend of the associated electricity consumption data in the same electricity consumption area affecting the heavy overload of the predicted unit, so that the prediction result is more reliable. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 It is a schematic diagram of the overall structure of a power grid distribution heavy overload prediction system and method in use according to the present invention;

[0031] Figure 2 A schematic diagram of the system workflow of a power grid distribution heavy overload prediction system and method according to the present invention;

[0032] Figure 3 This is a schematic diagram of a testing method of a power grid distribution heavy overload prediction system and method of the present invention. DETAILED DESCRIPTION

[0033] The technical solution in the embodiment of the present application is to solve the problems of the above-mentioned background technology, and the overall idea is as follows:

[0034] Embodiment 1:

[0035] This embodiment introduces a power grid distribution heavy overload prediction system. Figure 1-Figure 3 As shown, it includes a central processing module, a prediction model (Prophet-LSTM), a target power consumption data module, a related power consumption data module and a historical heavy overload data module;

[0036] Among them, the target power consumption data module mainly obtains the type and power consumption of the predicted unit power consumption equipment, while the associated power consumption data module mainly obtains the type and power consumption of the unit power consumption equipment other than the predicted unit in the same power consumption area. When the target power consumption data module and the associated power consumption data module obtain the type of corresponding power consumption equipment, each type of equipment is also distinguished (the distinction is mainly based on model and power), and different types of power consumption equipment are summarized and summarized to count the number;

[0037] At the same time, when the target power consumption data module and the associated power consumption data module obtain the power consumption of the corresponding power-using equipment, it is necessary to obtain it in different seasons of the year and different time periods of the day. The acquired data covers the whole year and the whole day, which can avoid errors and omissions, and is conducive to screening specific data, providing a certain reference value for subsequent research and prediction;

[0038] Secondly, in order to make similar power users in the same area be studied together when predicting heavy overload of power distribution, an associated unit screening module is connected to the input end of the associated power consumption data module. When the associated unit screening module is used to classify the units other than the predicted units in the same power consumption area, multiple power users in the same power supply area can be classified, and the data can be integrated later;

[0039] Furthermore, the historical overload data module mainly records the type of electrical equipment and power consumption when the predicted unit circuit was overloaded in the past, and provides a reference object for the central processing module to process the data obtained by the target power consumption data module and the associated power consumption data module;

[0040] At the same time, the central processing module obtains the type and power consumption of power consumption equipment from the target power consumption data module, the associated power consumption data module and the historical overload data module. The power consumption data of the predicted unit power consumption equipment obtained by the target power consumption data module is recorded as X1, and the power consumption data of the unit power consumption equipment other than the predicted unit in the same power consumption area obtained by the associated power consumption data module is recorded as y1, while the power consumption data of the predicted unit power consumption equipment in the historical overload data module is X0, and the power consumption data of the unit power consumption equipment other than the predicted unit in the same power consumption area in the historical overload data module is y0. By obtaining the power consumption difference, the evaluation formula is as follows:

[0041] X=X1-X0,y=y1-y0

[0042] The difference obtained by this formula can support the prediction module to compare each data with the standard data, and obtain the difference between the power consumption of a single device and the power consumption under the historical heavy overload state, so as to facilitate the subsequent prediction process to determine the impact of the power consumption of such equipment on the overall unit power overload;

[0043] At the same time, the power consumption of the predicted unit power equipment and the power consumption of the unit power equipment other than the predicted unit in the same power consumption area, compared with the power consumption data of the predicted unit power equipment in the historical heavy overload data module and the power consumption data of the unit power equipment other than the predicted unit in the same power consumption area, are studied by the following formula;

[0044]

[0045] Among them, S 2is the variance, n is the number of electrical equipment providing data, and the data finally calculated by this formula can be used to study the power consumption of the predicted unit power equipment and the power consumption of the unit power equipment other than the predicted unit in the same power consumption area. Compared with the discrete degree of the power consumption data of the predicted unit power equipment in the historical heavy overload data module and the power consumption data of the unit power equipment other than the predicted unit in the same power consumption area, it is convenient to analyze the heavy overload trend of the predicted unit;

[0046] Furthermore, by combining the power consumption difference with the discreteness, the influence of a certain type of equipment on the heavy overload of the power distribution can be judged according to the amplitude change and the differences such as the change in the number and type of each device, so that the prediction model can predict the overload of the predicted unit according to the power consumption difference obtained by the central processing module and the difference point obtained by comparison;

[0047] In some examples, the Prophet-LSTM prediction model makes predictions based on the actual power consumption status of the region, comparing the current power consumption with the historical heavy overload power consumption. The prediction items for the predicted unit include the overload cycle of the target unit, the factors affecting the variation of the overload situation, and the time of heavy overload of the distribution.

[0048] Embodiment 2:

[0049] Based on Example 1, this example introduces a method for predicting heavy overload of power distribution network, including:

[0050] S1: Count the type, power and quantity of the target unit's electrical equipment, and obtain the target unit's electricity consumption in different seasons and time periods;

[0051] S2: Count the types of related units in the same power supply area at the current stage, and count the types, power and quantity of electrical equipment of each related unit, and obtain their power consumption in different seasons and time periods;

[0052] S3: extract the current power consumption data of the target unit and the related units, compare them with the historical heavy overload data, and obtain the power consumption difference;

[0053] S4: Screening positive and negative differences, comparing the current power consumption with the historical heavy overload power consumption according to the actual power consumption status of the area;

[0054] Among them, the difference characteristics between the current power consumption and the historical heavy overload power consumption include the increase or decrease in the number of power consumption equipment of the target unit, the adjustment of the power consumption time of the power consumption equipment of the target unit, the change in the number of power consumption equipment of the associated power consumption unit and the adjustment of the power consumption time of the power consumption equipment of the associated power consumption unit;

[0055] S5: Use Prophet-LSTM to predict the overload period of the target unit, the factors affecting the variation of overload conditions, and the time of heavy overload of distribution within the same power consumption area.

[0056] Finally, it should be noted that: Obviously, the above embodiments are only examples for clearly explaining the present invention, and are not intended to limit the implementation methods. For ordinary technicians in the relevant field, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived from this are still within the scope of protection of the present invention.

Claims

1. A power grid distribution heavy overload prediction system, characterized in that: include: Central processing module, prediction model, target power consumption data module, associated power consumption data module and historical heavy overload data module; Wherein, the target power consumption data module is used to obtain the type and power consumption of the predicted unit power consumption equipment; The associated power consumption data module is used to obtain the type and power consumption of the unit power consumption equipment other than the predicted unit in the same power consumption area; The historical overload data module is used to record the type of electrical equipment and power consumption when the unit is predicted to have circuit overloaded in the past; The central processing module obtains the type of power-consuming equipment and power consumption from the target power consumption data module, the associated power consumption data module and the historical heavy overload data module, obtains the power consumption difference, and performs a difference comparison; The prediction model predicts the overload condition of the predicted unit according to the power difference value obtained by the central processing module and the difference point obtained by comparison.

2. A power grid distribution heavy overload prediction system as claimed in claim 1, characterized in that: The input end of the associated power consumption data module is connected to an associated unit screening module, and the associated unit screening module classifies units other than the predicted units in the same power consumption area.

3. A power grid distribution heavy overload prediction system as claimed in claim 1, characterized in that: The types of corresponding power-consuming devices obtained by the target power consumption data module and the associated power consumption data module are also distinguished from the models and powers, and the total number is counted. The power consumption of the corresponding power-consuming devices obtained by the target power consumption data module and the associated power consumption data module includes the power consumption in different seasons of the year and different time periods of the day.

4. A power grid distribution heavy overload prediction system as claimed in claim 1, characterized in that: The current target unit's electricity consumption data and related unit electricity consumption data are divided into positive and negative differences with the historical heavy overload data and are marked separately.

5. A power grid distribution heavy overload prediction system as claimed in claim 1, characterized in that: The prediction model is Prophet-LSTM, which makes predictions based on the actual power consumption status of the region and the difference between the current power consumption and the historical heavy overload power consumption.

6. A power grid distribution heavy overload prediction system as claimed in claim 1, characterized in that: The prediction items of the prediction model for the predicted unit include the overload period of the target unit, the influencing factors affecting the variation of the overload situation and the time of heavy overload of the distribution power.

7. A method for predicting heavy overload of power distribution network, the method is implemented based on any one of claims 1-6, characterized in that: include: S1: Count the type, power and quantity of the target unit's electrical equipment, and obtain the target unit's electricity consumption in different seasons and time periods; S2: Count the types of related units in the same power supply area at the current stage, and count the types, power and quantity of electrical equipment of each related unit, and obtain their power consumption in different seasons and time periods; S3: extract the current power consumption data of the target unit and the related units, compare them with the historical heavy overload data, and obtain the power consumption difference; S4: Screening positive and negative differences, comparing the current power consumption with the historical heavy overload power consumption according to the actual power consumption status of the area; S5: Use Prophet-LSTM to predict the overload period of the target unit, the factors affecting the variation of overload conditions, and the time of heavy overload of distribution within the same power consumption area.

8. A method for predicting heavy overload of power distribution network according to claim 7, characterized in that: The difference characteristics between the current power consumption in S4 and the historical heavy overload power consumption include the increase or decrease in the number of target unit power consumption equipment, the adjustment of the power consumption time of the target unit power consumption equipment, the change in the number of associated power consumption units and the adjustment of the power consumption time of the associated power consumption units.