Method for evaluating theoretical line loss of transformer area

Abstract

The invention discloses a transformer area theoretical line loss evaluation method, which comprises the following steps of: 1, acquiring related information of transformer area distribution transformers and users in real time, and calculating through a key factor calculation big data model to obtain each key factor value of a transformer area; and step 2, inputting each key factor value into a theoretical line loss big data model, and calculating the line loss degree of the transformer area through the theoretical line loss big data model. According to the invention, through the key factor calculation big data model and the theoretical line loss big data model, each key factor value of the transformer area and the line loss degree of the transformer area can be rapidly calculated and obtained, and the abnormality reason and suggested measures are given based on the abnormality library, the expert library and the case library. Accurate guidance is provided for power supply facility transformation, reactive compensation, load adjustment, power grid optimization layout and other decisions of the transformer area, and therefore the transformer area line loss is effectively controlled and reduced.

Description

Technical Field

[0001] The present invention relates to the field of line loss detection, and in particular to a method for evaluating theoretical line loss in a transformer area. Background Art

[0002] The causes of abnormal line losses in substations are complex and the influencing coefficients vary. It is necessary to collect, monitor and analyze line loss related data to discover the problems and propose corresponding solutions. This not only places high technical requirements on grassroots personnel, but also requires a lot of time and energy to integrate and analyze the data. Summary of the invention

[0003] In view of the technical defects and technical drawbacks in the prior art, the embodiment of the present invention provides a method for evaluating theoretical line loss in a transformer area to overcome the above problems or at least partially solve the above problems. The specific scheme is as follows:

[0004] A method for evaluating theoretical line loss in a transformer area, characterized in that the method comprises:

[0005] Step 1: Obtain relevant information about the distribution transformer and users in the substation in real time, and calculate the key factor values ​​of the substation through the key factor calculation big data model;

[0006] Step 2: input the values ​​of each key factor into a theoretical line loss big data model, and calculate the line loss degree of the substation through the theoretical line loss big data model.

[0007] Furthermore, in step 1, the relevant information of the distribution transformers and users in each substation includes: daily power consumption data of the terminal, voltage, current, and power collection data of multiple sections of the terminal in one day, statistical line loss data of the substation, GIS data information of the substation, transformer and terminal user table, substation and user file information, and substation line information.

[0008] Furthermore, the key features include: three-phase imbalance, power factor, load rate and load characteristics.

[0009] Furthermore, the theoretical line loss big data model is also used for:

[0010] Obtain the load data of each phase, simulate the current imbalance under different load combinations, and quantify the impact of current imbalance on line loss. For substations where the three-phase imbalance exceeds the preset percentage, predict the additional line loss caused by current imbalance, and combine the corresponding historical data and load forecasts to output recommended measures to improve balance, including optimizing load distribution and adjusting transformer capacity configuration.

[0011] Furthermore, the theoretical line loss big data model is also used for:

[0012] When the power factor is lower than the preset value, the cause of the low power factor is obtained by matching with a large amount of historical data that causes the low power factor value, and recommended measures to improve the power factor are output.

[0013] Furthermore, the theoretical line loss big data model is also used for:

[0014] For the load rate, based on the historical load data of the substation and the current load status, the specific impact of different load rates on line loss is analyzed. When the load rate is higher than the first load rate threshold, an early warning is issued that there is a potential risk of line loss growth, and appropriate capacity expansion strategies are recommended with reference to the corresponding historical data; when the load rate is low, suggestions for the reasonable allocation of power resources are output.

[0015] Furthermore, the theoretical line loss big data model is also used for:

[0016] Based on historical load data, analyze the changing patterns of load characteristics, including the difference between daytime and nighttime loads, seasonal load changes, and load mutation patterns during special holidays, and obtain the relationship between load characteristics and line losses;

[0017] Based on the changing rules of load characteristics and the relationship between load characteristics and line losses, the distribution network management and operation and maintenance strategies are output.

[0018] Furthermore, the theoretical line loss big data model is also used to calculate the daily power loss of a single user on the line through the relevant information of the distribution transformer in the substation area and the user, and calculate the daily loss percentage of all users based on the daily power loss. The formula is as follows:

[0019] Formula 1:

[0020] Formula 2:

[0021] Where ΔA i is the daily power loss of a single user on the line, U is the total meter voltage; Ui is the voltage of the i-th user, M is the total number of users; Ai is the daily power consumption of the i-th user; ΔM i is the daily meter loss of electricity for the ith user; ΔP% is the daily loss percentage for all users.

[0022] Furthermore, the method further comprises:

[0023] Step 101, obtaining relevant information of historical distribution transformers and users in the substation area, and calculating the key factor values ​​of the substation area through the key factor calculation big data model as the sample set data of the theoretical line loss big data model;

[0024] Step 102, dividing the sample set data into sample data for training and sample data for testing;

[0025] Step 103: Using the substation sample data used for training as a benchmark, a random forest algorithm is used to perform machine learning training to obtain a theoretical line loss big data model to be verified;

[0026] Step 104: The theoretical line loss value of the test sample data is calculated by the theoretical line loss big data model to be verified, and whether the current theoretical line loss big data model meets the preset requirements is judged based on the correlation between the theoretical line loss big data model and the statistical line loss. If so, the verified theoretical line loss big data model is output; otherwise, the relevant parameters of the random forest algorithm are adjusted to retrain the theoretical line loss big data model until the theoretical line loss big data model meets the preset requirements.

[0027] The present invention has the following beneficial effects:

[0028] The present invention can quickly calculate the key factor values ​​of the substation and the degree of line loss in the substation through the key factor calculation big data model and the theoretical line loss big data model, and give the abnormal causes and suggested measures based on the abnormal library, expert library and case library, so as to provide precise guidance for the decisions on power supply facility transformation, reactive power compensation, load adjustment, and optimal grid layout in the substation, thereby achieving effective control and reduction of line loss in the substation. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 A schematic diagram of a flow chart of a method for evaluating theoretical line loss in a transformer area provided by an embodiment of the present invention;

[0030] Figure 2 A method for training a theoretical line loss big data model is provided for an embodiment of the invention. DETAILED DESCRIPTION

[0031] 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 present invention, not all 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.

[0032] Figure 1 The following is a flow chart of a method for evaluating theoretical line loss in a transformer area provided by an embodiment of the present invention. Figure 1 As shown, the method includes:

[0033] Step 1: Obtain relevant information about the distribution transformer and users in the substation in real time, and calculate the key factor values ​​of the substation through the key factor calculation big data model;

[0034] Step 2: input the values ​​of each key factor into a theoretical line loss big data model, and calculate the line loss degree of the substation through the theoretical line loss big data model.

[0035] In the above embodiment, the key factor calculation big data model is a model pre-trained with historical substation distribution and user-related data, and the theoretical line loss big data model is a model pre-trained with historical key factor values, wherein the theoretical line loss big data model is also connected to an abnormality library, an expert library and a case library, wherein the abnormality library, the expert library and the case library record various historical data, the causes corresponding to the abnormal data and suggested measures, and can be continuously improved based on new cases.

[0036] In some embodiments, the relevant information of the substation distribution transformer and users includes: daily terminal electricity consumption data, voltage, current, and power collection data of multiple sections of the terminal in one day, substation statistical line loss data, GIS data information of substations, transformers and terminal user tables, substation and user profile information, substation line information, and the key features include: three-phase imbalance, power factor, load rate and load characteristics; the key factor calculation big data model is a model that has been pre-trained with relevant data of historical substation distribution transformers and users.

[0037] In some embodiments, the theoretical line loss big data model is also used to perform three-phase unbalance analysis, specifically including:

[0038] Obtain the load data of each phase, simulate the current imbalance under different load combinations, and quantify the impact of current imbalance on line loss. For the substation where the three-phase imbalance exceeds the preset percentage, predict the additional line loss caused by current imbalance, and perform similarity analysis with the historical data of current imbalance recorded in the anomaly library, expert library and case library to obtain the causes of the corresponding line loss and recommended measures, such as optimizing load distribution, adjusting transformer capacity configuration, etc., to effectively reduce line loss and improve power supply reliability.

[0039] In some embodiments, the theoretical line loss big data model is also used to perform power factor analysis, specifically including:

[0040] When the power factor is lower than the preset value, the model performs similarity analysis with a large amount of historical data that causes low power factor values ​​recorded in the abnormal database, expert database, and case database to obtain the cause of the low power factor and output recommended measures to improve the power factor, such as the presence of a large amount of reactive loads. Based on massive historical data and real-time monitoring data, the model analyzes the corresponding causes and recommended measures to improve the power factor, reduce ineffective power consumption, thereby reducing line losses and improving power transmission efficiency.

[0041] In some embodiments, the theoretical line loss big data model is also used to perform load factor analysis, specifically including:

[0042] For the load rate, according to the historical load data of the substation and the current load status, the specific impact of different load rates on line loss is analyzed. When the load rate is higher than the first load rate threshold, an early warning is issued that there is a potential risk of line loss growth, and a similarity analysis is performed with the historical data of the load rate recorded in the anomaly library, expert library and case library, and an appropriate capacity expansion strategy is output; when the load rate is low, suggestions for the reasonable allocation of power resources are output to avoid energy waste, while ensuring that the utilization rate of power supply facilities is maintained in the high-efficiency range.

[0043] In some embodiments, the theoretical line loss big data model is also used to perform load characteristic analysis, specifically including:

[0044] Based on historical load data, analyze the changing patterns of load characteristics, including the difference between daytime and nighttime loads, seasonal load changes, and load mutation patterns during special holidays, and obtain the relationship between load characteristics and line losses;

[0045] Based on the changing rules of load characteristics and the relationship between load characteristics and line loss, the corresponding line loss and expected line loss change of the current load characteristic value are output, and a similarity analysis is performed with the historical data of line loss and expected line loss change recorded in the anomaly library, expert library and case library, and the distribution network management and operation and maintenance strategy is output, providing a scientific basis for formulating flexible and efficient distribution network management and operation and maintenance strategies, thereby effectively controlling line loss and improving the operating efficiency of the power grid.

[0046] In some embodiments, the theoretical line loss big data model is also used to analyze the structure characteristics of the power grid.

[0047] In the analysis of grid structure characteristics, the model comprehensively analyzes the impact of grid structure on line losses by integrating information such as the substation topology, line parameters, and equipment attributes. Through detailed modeling and simulation of factors such as grid nodes, line lengths, and conductor materials, it reveals potential structural bottlenecks and optimization directions, such as line optimization and equipment replacement, to improve the overall performance of the grid, reduce line losses, and increase power supply reliability and economy.

[0048] In some embodiments, the theoretical line loss big data model is also used for:

[0049] The daily power loss of a single user on the line is calculated through the relevant information of the distribution transformer in the substation area and the user, and the daily loss percentage of all users is calculated based on the daily power loss. The formula is as follows:

[0050] Formula 1:

[0051] Formula 2:

[0052] Where ΔA iis the daily power loss of a single user on the line, U is the total meter voltage; Ui is the voltage of the i-th user, M is the total number of users; Ai is the daily power consumption of the i-th user; ΔM i is the daily meter loss of electricity for the ith user; ΔP% is the daily loss percentage for all users.

[0053] The data processed by the theoretical line loss big data model is used to generate a line loss analysis report, which provides strong data support for decisions such as power supply facility transformation, reactive power compensation, load adjustment, and grid optimization layout in the substation, thereby achieving effective control and reduction of line losses in the substation.

[0054] like Figure 2 In some embodiments, the method further comprises:

[0055] Step 101, obtaining relevant information of historical distribution transformers and users in the substation area, and calculating the key factor values ​​of the substation area through the key factor calculation big data model as the sample set data of the theoretical line loss big data model;

[0056] Step 102, dividing the sample set data into sample data for training and sample data for testing;

[0057] Step 103: Using the substation sample data used for training as a benchmark, a random forest algorithm is used to perform machine learning training to obtain a theoretical line loss big data model to be verified;

[0058] Step 104: The theoretical line loss value of the test sample data is calculated by the theoretical line loss big data model to be verified, and whether the current theoretical line loss big data model meets the preset requirements is judged based on the correlation between the theoretical line loss big data model and the statistical line loss. If so, the verified theoretical line loss big data model is output; otherwise, the relevant parameters of the random forest algorithm are adjusted to retrain the theoretical line loss big data model until the theoretical line loss big data model meets the preset requirements.

[0059] Among them, the residual is calculated based on the theoretical line loss and statistical line loss of the test sample. If the distribution ratio of the residual in the preset interval reaches the expected value, it is judged that the correlation between the two meets the preset requirements.

[0060] 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 principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for evaluating theoretical line loss in a transformer area, characterized in that: The method comprises: Step 1: Obtain relevant information about the distribution transformer and users in the substation in real time, and calculate the key factor values ​​of the substation through the key factor calculation big data model; Step 2: input the values ​​of each key factor into a theoretical line loss big data model, and calculate the line loss degree of the substation through the theoretical line loss big data model.

2. The method for evaluating the theoretical line loss of a transformer area according to claim 1, characterized in that: In step 1, the relevant information of the distribution transformers and users in each substation includes: daily power consumption data of the terminal, voltage, current, and power collection data of multiple sections of the terminal in one day, statistical line loss data of the substation, GIS data information of the substation, transformer and terminal user table, substation and user file information, and substation line information.

3. The method for evaluating the theoretical line loss of a transformer area according to claim 1, characterized in that: The key features include: three-phase imbalance, power factor, load rate and load characteristics. The theoretical line loss big data model is connected to an abnormality library, an expert library and a case library. The abnormality library, the expert library and the case library record various historical data, the causes corresponding to the abnormal historical data and suggested measures.

4. The method for evaluating the theoretical line loss of a transformer area according to claim 3, characterized in that: The theoretical line loss big data model is also used for: Obtain the load data of each phase, simulate the current imbalance under different load combinations, and quantify and analyze the impact of current imbalance on line loss. For substations where the three-phase imbalance exceeds the preset percentage, predict the additional line loss caused by current imbalance, and perform similarity analysis with the historical data of current imbalance recorded in the anomaly library, expert library, and case library to obtain the causes of corresponding line losses and recommended measures, including optimizing load distribution and adjusting transformer capacity configuration.

5. The method for evaluating the theoretical line loss of a transformer area according to claim 1, characterized in that: The theoretical line loss big data model is also used for: When the power factor is lower than the preset value, the cause of the low power factor is obtained by performing similarity analysis with a large amount of historical data on low power factor values ​​recorded in the abnormality library, expert library, and case library, and recommended measures to improve the power factor are output.

6. The method for evaluating the theoretical line loss of a transformer area according to claim 1, characterized in that: The theoretical line loss big data model is also used for: For the load rate, according to the historical load data of the substation and the current load status, the specific impact of different load rates on line loss is analyzed. When the load rate is higher than the first load rate threshold, an early warning is issued that there is a potential risk of line loss growth. A similarity analysis is performed with the historical data of the load rate recorded in the anomaly library, expert library and case library, and an appropriate capacity expansion strategy is output; when the load rate is low, recommendations for the rational allocation of power resources are output.

7. The method for evaluating the theoretical line loss of a transformer area according to claim 1, characterized in that: The theoretical line loss big data model is also used for: Based on historical load data, analyze the changing patterns of load characteristics, including the difference between daytime and nighttime loads, seasonal load changes, and load mutation patterns during special holidays, and obtain the relationship between load characteristics and line losses; Based on the changing rules of load characteristics and the relationship between load characteristics and line loss, the corresponding line loss and expected line loss change of the current load characteristic value are output, and a similarity analysis is performed with the historical data of line loss and expected line loss change recorded in the anomaly library, expert library and case library, and the distribution network management and operation and maintenance strategy is output.

8. The method for evaluating the theoretical line loss of a transformer area according to claim 1, characterized in that: The theoretical line loss big data model is also used to calculate the daily power loss of a single user on the line through the relevant information of the distribution transformer in the substation area and the user, and calculate the daily loss percentage of all users based on the daily power loss. The formula is as follows: Formula 1: Formula 2: Where ΔA i is the daily power loss of a single user on the line, U is the total meter voltage; Ui is the voltage of the i-th user, M is the total number of users; Ai is the daily power consumption of the i-th user; ΔM i is the daily meter loss of electricity for the ith user; ΔP% is the daily loss percentage for all users.

9. The method for evaluating the theoretical line loss of a transformer area according to claim 1, characterized in that: The method further comprises: Step 101, obtaining relevant information of historical distribution transformers and users in the substation area, and calculating the key factor values ​​of the substation area through the key factor calculation big data model as the sample set data of the theoretical line loss big data model; Step 102, dividing the sample set data into sample data for training and sample data for testing; Step 103: Using the substation sample data used for training as a benchmark, a random forest algorithm is used to perform machine learning training to obtain a theoretical line loss big data model to be verified; Step 104: The theoretical line loss value of the test sample data is calculated by the theoretical line loss big data model to be verified, and whether the current theoretical line loss big data model meets the preset requirements is judged based on the correlation between the theoretical line loss big data model and the statistical line loss. If so, the verified theoretical line loss big data model is output; otherwise, the relevant parameters of the random forest algorithm are adjusted to retrain the theoretical line loss big data model until the theoretical line loss big data model meets the preset requirements.

Images