Power distribution network load prediction and adjustment method

Abstract

The invention discloses a power distribution network load prediction and adjustment method, and relates to the technical field of power distribution network load prediction, and the method comprises the following steps: dividing the use types of power distribution networks based on the management areas of different power distribution networks in a city, including a resident type, a commercial type, an industrial type and a hybrid type, building a city power distribution network database, respectively establishing power distribution network files based on the types; the method comprises the following steps: collecting daily different power distribution network load data in different historical months, calculating to obtain each-time predicted peak load of different regions based on a temperature interval, and obtaining predicted surplus load and predicted deficit load by combining planned loads of different power distribution networks; and dividing surplus load time intervals and deficient load time intervals of different areas in different months. The load of the power distribution network is predicted by combining the characteristics of the management area of the power distribution network, and the profit and loss management pairing is performed on the power distribution network by combining the predicted load, so that the practicability and the functionality are enhanced.

Description

Technical Field

[0001] The present invention relates to the technical field of distribution network load prediction, and in particular to a distribution network load prediction and adjustment method. Background Art

[0002] The power supply load of the distribution network refers to the load that the distribution network bears for external power supply in the region. It is necessary to ensure that the power supply load is greater than the power consumption load at any time to meet the power supply needs of the region. Therefore, it is necessary to predict the power consumption load of the area under the responsibility of the distribution network to ensure the stable operation of the distribution network. A Chinese patent with publication number "CN114819280A" discloses "a distribution network load forecasting method, belonging to the technical field of power load forecasting, the specific steps of the forecasting method are as follows: step one: obtaining historical power data; step two: determining load influencing factors and selecting similar days; step three: determining electric vehicle load data; step four: extracting training sets; step five: building a regional short-term load forecasting model; step six: load forecasting;" Although it can predict the load of the distribution network through the short-term load forecasting model, the regional composition of different distribution networks in actual situations is different. The fluctuation of the distribution network load under different regional compositions is also different under the influence of different temperatures. The existing load forecasting cannot perform detailed analysis and prediction for the areas under the jurisdiction of the distribution network, and cannot perform load transfer and pairing management for the load profit and loss in different regions, and has problems of low practicality and functionality.

[0003] Currently, no effective solution has been proposed for the problems in the related technologies. Summary of the invention

[0004] In view of the problems in the related technology, the present invention proposes a distribution network load forecasting and adjustment method to overcome the above technical problems existing in the existing related technology.

[0005] To this end, the specific technical solution adopted by the present invention is as follows: A method for load forecasting and adjustment of a distribution network, the method comprising the following steps: S1. Based on the management areas of different urban distribution networks, the types of distribution network usage are divided into residential, commercial, industrial and mixed types, and a database of urban distribution networks is established. Distribution network files are established based on the types; S2. Collect the daily load data of different distribution networks in different months in history, calculate the hourly predicted peak loads in different regions based on the temperature range, combine the planned loads of different distribution networks to obtain the predicted surplus load and predicted deficit load, and divide the surplus load time intervals and deficit load time intervals in different regions in different months; S3. Based on the daily surplus load range, predicted surplus load, deficit load range, and predicted deficit load in different regions and different months, daily forecast matching is performed on the distribution network files in different months to make comprehensive adjustments to the distribution network load.

[0006] As a preferred embodiment, the S1 includes the following sub-steps: S11. Based on the management areas of each distribution network under the jurisdiction of the city in the GIS map, obtain the proportion of residential areas in different distribution network management areas , Commercial area ratio , Industrial area ratio , based on the area proportion in different distribution network management areas, the distribution network types are divided, and the steps are as follows: When the proportion of residential area in the distribution network management area ≥70%, it means that the current distribution network type is residential; When the commercial area accounts for the proportion of the distribution network management area ≥60%, it means that the current distribution network type is commercial; When the industrial area accounts for the proportion of the distribution network management area ≥50%, it means that the current distribution network type is industrial; When the proportion of residential area, commercial area and industrial area in the distribution network management area do not meet the above ratios, it means that the current distribution network type is mixed; S12. Establish the urban distribution network database of the current city through MySQL, and establish type groups in the database, including residential, commercial, industrial, and mixed types. Based on the types of each distribution network under the city, establish distribution network files under the corresponding groups, and number the distribution networks in the files in sequence. The distribution network files include the geographical location, planned load, and name of the distribution network.

[0007] As a preferred embodiment, S2 includes the following sub-steps: S21. Collect the load peak values ​​of different distribution networks every day and every hour in the historical months, and collect the temperature values ​​of different distribution networks every day and every hour in the historical months. Based on different distribution network grouping types, group the working days and weekends in the same month based on the temperature range. S22. Calculate the hourly peak load based on the temperature range grouping, and combine the current meteorological data of the distribution network to obtain the predicted peak load of the distribution network at each moment of the day. Combine the planned loads of different distribution networks to calculate the predicted surplus load and predicted deficit load, and divide the surplus load time interval and the deficit load time interval.

[0008] As a preferred embodiment, the S21 includes the following sub-steps: S211. Collect the temperature data of the same month in the past two years and obtain the temperature value at each hour in the current month , and distinguish between working days and weekends, where n and N represent the month and the day of the month respectively, and P represents the time; S222: Based on different distribution network grouping types, the working days and weekends in the same month are grouped based on the temperature range. The specific steps are as follows: For residential areas, a range of 3°C is used to capture the subtle changes in residential loads under different temperature conditions, including , i represents the highest temperature point, based on the temperature values ​​of the residential distribution network at different times in the current month , summarize the distribution network load peak data at the same time every day in the same month in the same temperature range according to the temperature value, and establish the data file at the current time. The file includes all temperature ranges at this time in the current month and the summarized distribution network load peak data in each temperature range, and separates the working days and weekends into groups; For commercial areas, a 6°C interval is used to capture the load changes in commercial areas under different temperature conditions, including , i represents the highest temperature point, based on the temperature values ​​at different times in the current month of the commercial distribution network , summarize the distribution network load peak data at the same time every day in the same month in the same temperature range according to the temperature value, and establish the data file at the current time. The file includes all temperature ranges at this time in the current month and the summarized distribution network load peak data in each temperature range, and separates the working days and weekends into groups; For industrial type, a range of 10°C is used to capture the overall change trend of industrial area load, including , i represents the highest temperature point, based on the temperature values ​​of the industrial distribution network at different times in the current month , summarize the distribution network load peak data at the same time every day in the same month in the same temperature range according to the temperature value, and establish the data file at the current time. The file includes all temperature ranges at this time in the current month and the summarized distribution network load peak data in each temperature range, and separates the working days and weekends into groups; For the mixed type, a range of 7°C is used to capture the overall change trend of the mixed zone load, including , i represents the highest temperature point, based on the temperature values ​​at different times in the current month of the hybrid distribution network , the distribution network load peak data at the same time every day in the same month are summarized in the same temperature range according to the temperature value, and the data file at the current time is established. The file includes all temperature ranges at this time in the current month and the summarized distribution network load peak data in each temperature range, and the working days and weekends are grouped separately.

[0009] As a preferred embodiment, the S22 includes the following sub-steps: S221. Combined with the data archive, for the working days and weekends in the same month of the same distribution network, the peak load data of the distribution network in all temperature intervals at the same time are taken to obtain the representative loads of different temperature intervals at the same time. The specific algorithm formula is: ; in, represents the peak load data of the nth distribution network in the ith temperature interval at the same time, Represents the representative load of the ith temperature interval at time Q, and calculates the representative load of working days and the representative load of weekends respectively; S222: Based on the meteorological information, the predicted temperature of the current distribution network at each moment of the same day is obtained. For weekdays and weekends, the predicted temperature is matched with the temperature range in the data file at the same moment, and the corresponding representative load is selected as the predicted peak load at the current moment. , Q is the time, and its value range is 0-23; S223, based on the planned load of the current distribution network , calculate the differential load at each moment of the current forecast day , and its algorithm formula is: ; The differential load at each moment Make a judgment to determine the type of the current differential load. The specific steps are as follows: when When ≥0, it means that the difference load at the current moment is the predicted surplus load, which is recorded as ; when When <0, it means that the current differential load is the predicted deficit load, which is recorded as ; Based on the predicted surplus load and predicted deficit load at different times of the day, the surplus load time interval and the deficit load time interval are divided, and the time period corresponding to the predicted surplus load is summarized into the surplus load time interval of the day, and the time period corresponding to the predicted deficit load is summarized into the deficit load time interval of the day.

[0010] As a preferred embodiment, S3 includes the following sub-steps: S31. Based on the meteorological forecast information, the predicted surplus load or predicted deficit load at different times of the day in different months is obtained, and time matching analysis is performed on different regions to obtain a time matching combination; S32. Based on the combined results of the time matching analysis, load matching analysis is performed on different distribution networks, and one-to-one or one-to-many combinations are performed on different distribution network areas to generate a distribution network load pairing strategy.

[0011] As a preferred implementation, the S31 includes the following sub-steps: S311. Obtain a surplus load set based on the daily predicted surplus load or predicted deficit load of different distribution stations or deficit load collection ,in represents the moment of load surplus or load deficit, Represents the surplus load time period of the distribution network numbered n on the day, Represents the time period of the distribution network numbered n with a lack of load on the day; S312, for any two distribution networks i and j, where , when satisfied When , it represents the surplus and deficit time matching of distribution networks i and j. The combinations are recorded separately, and all i that meet the above conditions are excluded from the distribution network number. For the excluded distribution networks 1 to m and j, the combined matching is calculated: When satisfied When , after recording the combination, exclude 1 to m that meet the above from the distribution network number, and repeat the above until there is no combination; Repeat the above until the matching combinations of all distribution network deficit time periods are obtained.

[0012] As a preferred embodiment, the S32 comprises the following steps: S321, calculate the corresponding time for each combination after pairing When all the absolute value differences in the set are greater than or equal to 0, it means that the load pairing result is met. When there are multiple groups of matching results, the absolute value differences in each combination are accumulated and sorted in ascending order. The first combination is selected as the pairing combination of the current distribution network deficit, and all subsequent combination schemes containing the surplus distribution network number in the current pairing combination are deleted; S322, for the remaining distribution networks that meet the time matching but do not meet the load matching, the surplus combinations in the time matching scheme are accumulated and combined, and the corresponding time is calculated after the accumulation. When all the absolute value differences in the set are greater than or equal to 0, it means that the load pairing result is met. When there are multiple sets of matching results, the absolute value differences in each new combination are accumulated and arranged in ascending order. The first new combination is selected as the pairing combination of the current distribution network deficit, and all subsequent combination schemes containing the surplus distribution network number in the current new pairing combination are deleted. When all the absolute value differences in the set are still not greater than or equal to 0 after re-combination, the current distribution network is marked red in the urban distribution network database; S323. After all the distribution networks under the jurisdiction of the urban distribution network database have been paired, the paired distribution networks are associated, and a separate file is created to summarize the load profit and loss complementation, and all the red-marked distribution networks are summarized into the same file; S324, repeat the above operation to perform daily pairing in combination with daily meteorological information.

[0013] The beneficial effects of the present invention are: The present invention sets different temperature interval division strategies for different types of distribution networks under its jurisdiction, combines the load fluctuation characteristics of residential, industrial, commercial and mixed areas, captures the sensitivity of loads in different areas to temperature changes, and makes a precise prediction of the load of the distribution network. The obtained data is processed by the extreme value method, which effectively avoids the deviation of the prediction result caused by short-term load fluctuation, makes the prediction result more stable and reliable, and enhances the practicality of the method. The present invention obtains the temperature value at each moment of the day by combining the meteorological data of the current forecast day, and separately matches each moment in combination with the temperature interval divided by historical data, so as to obtain the predicted peak load at each moment. By calculating the difference between the planned load and the predicted peak load, it can be determined whether the current distribution network needs to adjust the load at the current moment, which is helpful to formulate a more sophisticated and dynamic load dispatching strategy to balance supply and demand and ensure the stable operation of the power grid. The present invention can determine a distribution network with complementary loads by performing time matching and load matching for the surplus time periods and deficit time periods of different distribution networks, and transfer the surplus load to meet the deficit load, thereby effectively balancing the power supply and demand relationship between different regions, avoiding the waste of power resources, improving the utilization efficiency of power resources, and facilitating the management and adjustment of the distribution network under the jurisdiction of the city; The present invention can give priority to the most suitable combination in load matching by accumulating the absolute value differences and arranging them in ascending order when there are multiple groups of matching results, and selecting the first combination, thereby ensuring that the combination with the smallest total difference is selected under the premise of satisfying load matching, thereby achieving efficient allocation of power resources, and summarizing the red-marked distribution networks into the same file, so as to focus on and deal with those distribution networks that cannot achieve effective load matching on the same day, providing data support for subsequent grid transformation, load adjustment and other measures, and enhancing practicality and functionality. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0015] Figure 1 The present invention is a flow chart of a method for load forecasting and adjustment in a distribution network according to an embodiment of the present invention. DETAILED DESCRIPTION

[0016] To further illustrate each embodiment, the present invention provides drawings, which are part of the disclosure of the present invention and are mainly used to illustrate the embodiments and can be used in conjunction with the relevant descriptions in the specification to explain the operating principles of the embodiments. With reference to these contents, ordinary technicians in the field should be able to understand other possible implementations and advantages of the present invention. The components in the figures are not drawn to scale, and similar component symbols are generally used to represent similar components.

[0017] According to an embodiment of the present invention, a method for load forecasting and adjustment in a distribution network is provided.

[0018] The present invention is further described with reference to the accompanying drawings and specific embodiments: Example 1: Figure 1 As shown, according to a distribution network load forecasting and adjustment method according to an embodiment of the present invention, the method comprises the following steps: S1. Based on the management areas of different urban distribution networks, the types of distribution network usage are divided into residential, commercial, industrial and mixed types, and a database of urban distribution networks is established. Distribution network files are established based on the types; S11. Based on the management areas of each distribution network under the jurisdiction of the city in the GIS map, obtain the proportion of residential areas in different distribution network management areas , Commercial area ratio , Industrial area ratio , based on the area proportion in different distribution network management areas, the distribution network types are divided, and the steps are as follows: When the proportion of residential area in the distribution network management area ≥70%, it means that the current distribution network type is residential; When the commercial area accounts for the proportion of the distribution network management area ≥60%, it means that the current distribution network type is commercial; When the industrial area accounts for the proportion of the distribution network management area ≥50%, it means that the current distribution network type is industrial; When the proportion of residential area, commercial area and industrial area in the distribution network management area do not meet the above ratios, it means that the current distribution network type is mixed; S12. Establish the urban distribution network database of the current city through MySQL, and establish type groups in the database, including residential, commercial, industrial, and mixed types. Based on the types of each distribution network under the city, establish distribution network files under the corresponding groups, and number the distribution networks in the files in sequence. The distribution network files include the geographical location, planned load, and name of the distribution network.

[0019] It should be noted that the numbering format of the distribution network takes the residential type as an example, which is JM-001, JM-002, ..., JM-00N, where N is the total number of distribution networks under the current residential type grouping. By dividing the types of distribution networks, data support can be provided for temperature range division in subsequent load forecasts, and the planned load is the maximum load set when the current distribution network is designed.

[0020] S2. Collect the daily load data of different distribution networks in different months in history, calculate the hourly predicted peak loads in different regions based on the temperature range, combine the planned loads of different distribution networks to obtain the predicted surplus load and predicted deficit load, and divide the surplus load time intervals and deficit load time intervals in different regions in different months; S21. Collect the load peak values ​​of different distribution networks every day and every hour in the historical months, and collect the temperature values ​​of different distribution networks every day and every hour in the historical months. Based on different distribution network grouping types, group the working days and weekends in the same month based on the temperature range. S211. Collect the temperature data of the same month in the past two years and obtain the temperature value at each hour in the current month , and distinguish between working days and weekends, where n and N represent the month and the day of the month respectively, and P represents the time; S222: Based on different distribution network grouping types, the working days and weekends in the same month are grouped based on the temperature range. The specific steps are as follows: For residential areas, a range of 3°C is used to capture the subtle changes in residential loads under different temperature conditions, including , i represents the highest temperature point, based on the temperature values ​​of the residential distribution network at different times in the current month , summarize the distribution network load peak data at the same time every day in the same month in the same temperature range according to the temperature value, and establish the data file at the current time. The file includes all temperature ranges at this time in the current month and the summarized distribution network load peak data in each temperature range, and separates the working days and weekends into groups; For commercial areas, a 6°C interval is used to capture the load changes in commercial areas under different temperature conditions, including , i represents the highest temperature point, based on the temperature values ​​at different times in the current month of the commercial distribution network , summarize the distribution network load peak data at the same time every day in the same month in the same temperature range according to the temperature value, and establish the data file at the current time. The file includes all temperature ranges at this time in the current month and the summarized distribution network load peak data in each temperature range, and separates the working days and weekends into groups; For industrial type, a range of 10°C is used to capture the overall change trend of industrial area load, including , i represents the highest temperature point, based on the temperature values ​​of the industrial distribution network at different times in the current month , summarize the distribution network load peak data at the same time every day in the same month in the same temperature range according to the temperature value, and establish the data file at the current time. The file includes all temperature ranges at this time in the current month and the summarized distribution network load peak data in each temperature range, and separates the working days and weekends into groups; For the mixed type, a range of 7°C is used to capture the overall change trend of the mixed zone load, including , i represents the highest temperature point, based on the temperature values ​​at different times in the current month of the hybrid distribution network , the distribution network load peak data at the same time every day in the same month are summarized in the same temperature range according to the temperature value, and the data file at the current time is established. The file includes all temperature ranges at this time in the current month and the summarized distribution network load peak data in each temperature range, and the working days and weekends are grouped separately.

[0021] It should be noted that i represents the highest temperature point. The specific settings can be flexibly set according to the monthly climate in different regions. Residential loads are mainly composed of household electricity consumption, with obvious peaks in the morning and evening, and large seasonal load fluctuations. A finer temperature range is used to capture subtle changes in residential loads under different temperature conditions. Commercial loads are mainly composed of office and commercial activities, with peaks during the day and troughs at night, and small seasonal load fluctuations. A medium-width temperature range is used to capture changes in commercial loads under different temperature conditions. Industrial loads are mainly composed of production equipment, which are usually relatively stable and have no obvious peak hours. Seasonal load fluctuations are small. A wider temperature range is used to capture the overall change trend of industrial loads. Mixed loads are composed of various types, including residential buildings, commercial facilities, and industrial production equipment. The load curve is more complex. A temperature range between commercial and industrial types is selected, and a distinction is made between weekdays and weekends to improve the representativeness and accuracy of subsequent predicted loads.

[0022] S22, based on the temperature interval grouping calculation to obtain the hourly peak load, combined with the current meteorological data of the distribution network, to obtain the predicted peak load of the distribution network at each moment of the day, combined with the planned loads of different distribution networks, to calculate the predicted surplus load and the predicted deficit load, and divide the surplus load time interval and the deficit load time interval; S221. Combined with the data archive, for the working days and weekends in the same month of the same distribution network, the peak load data of the distribution network in all temperature intervals at the same time are taken to obtain the representative loads of different temperature intervals at the same time. The specific algorithm formula is: ; in, represents the peak load data of the nth distribution network in the ith temperature interval at the same time, Represents the representative load of the ith temperature interval at time Q, and calculates the representative load of working days and the representative load of weekends respectively; It should be noted that the sources of the distribution network load peak data on weekdays and weekends are respectively weekdays and weekends. The representative load at each moment on weekdays and weekends is calculated separately to improve the accuracy of subsequent predicted loads. By taking the largest value, load fluctuations can be prevented from affecting the subsequent distribution network pairing.

[0023] S222: Based on the meteorological information, the predicted temperature of the current distribution network at each moment of the same day is obtained. For weekdays and weekends, the predicted temperature is matched with the temperature range in the data file at the same moment, and the corresponding representative load is selected as the predicted peak load at the current moment. , Q is the time, and its value range is 0-23; S223, based on the planned load of the current distribution network , calculate the differential load at each moment of the current forecast day , and its algorithm formula is: ; The differential load at each moment Make a judgment to determine the type of the current differential load. The specific steps are as follows: when When ≥0, it means that the difference load at the current moment is the predicted surplus load, which is recorded as ; when When <0, it means that the current differential load is the predicted deficit load, which is recorded as ; Based on the predicted surplus load and predicted deficit load at different times of the day, the surplus load time interval and the deficit load time interval are divided, and the time period corresponding to the predicted surplus load is summarized into the surplus load time interval of the day, and the time period corresponding to the predicted deficit load is summarized into the deficit load time interval of the day.

[0024] It should be noted that by combining the meteorological data of the current forecast day, the temperature value at each moment of the day is obtained, and each moment is matched separately in combination with the divided temperature interval, so as to obtain the predicted peak load at each moment. By calculating the difference between the planned load and the predicted peak load, it can be determined whether the current distribution network needs to adjust the load at the current moment.

[0025] Embodiment 2: S3, based on the daily surplus load interval, predicted surplus load, deficit load interval, and predicted deficit load in different months in different regions, daily forecast matching is performed on the distribution network files in different months, and the distribution network load is comprehensively adjusted; S31. Based on the meteorological forecast information, the predicted surplus load or predicted deficit load at different times of the day in different months is obtained, and time matching analysis is performed on different regions to obtain a time matching combination; S311. Obtain a surplus load set based on the daily predicted surplus load or predicted deficit load of different distribution stations or deficit load collection ,in represents the moment of load surplus or load deficit, Represents the surplus load time period of the distribution network numbered n on the day, Represents the time period of the distribution network numbered n with a lack of load on the day; S312, for any two distribution networks i and j, where , when satisfied When , it represents the surplus and deficit time matching of distribution networks i and j. The combinations are recorded separately, and all i that meet the above conditions are excluded from the distribution network number. For the excluded distribution networks 1 to m and j, the combined matching is calculated: When satisfied When , after recording the combination, exclude 1 to m that meet the above from the distribution network number, and repeat the above until there is no combination; Repeat the above until the matching combinations of all distribution network deficit time periods are obtained.

[0026] It should be noted that by giving priority to any two distribution networks i and j, To perform pairing, first obtain and exclude the one-to-one pairing situation, and then obtain the one-to-many pairing situation by combining and excluding multiple distribution network surpluses and a single deficit, which is convenient for subsequent load matching analysis; S32. Based on the combined results of the time matching analysis, load matching analysis is performed on different distribution networks, and one-to-one or one-to-many combinations are performed on different distribution network areas to generate a distribution network load pairing strategy; S321, calculate the corresponding time for each combination after pairing When all the absolute value differences in the set are greater than or equal to 0, it means that the load pairing result is met. When there are multiple groups of matching results, the absolute value differences in each combination are accumulated and sorted in ascending order. The first combination is selected as the pairing combination of the current distribution network deficit, and all subsequent combination schemes containing the surplus distribution network number in the current pairing combination are deleted; S322, for the remaining distribution networks that meet the time matching but do not meet the load matching, the surplus combinations in the time matching scheme are accumulated and combined, and the corresponding time is calculated after the accumulation. When all the absolute value differences in the set are greater than or equal to 0, it means that the load pairing result is met. When there are multiple sets of matching results, the absolute value differences in each new combination are accumulated and arranged in ascending order. The first new combination is selected as the pairing combination of the current distribution network deficit, and all subsequent combination schemes containing the surplus distribution network number in the current new pairing combination are deleted. When all the absolute value differences in the set are still not greater than or equal to 0 after re-combination, the current distribution network is marked red in the urban distribution network database; It should be noted that when there are multiple groups of matching results, the absolute value differences are accumulated and arranged in ascending order, and the first combination is selected. This can give priority to the most suitable combination in load matching, ensuring that while satisfying the load matching, the combination with the smallest total difference is selected, thereby achieving efficient allocation of power resources.

[0027] S323. After all the distribution networks under the jurisdiction of the urban distribution network database have been paired, the paired distribution networks are associated, and a separate file is created to summarize the load profit and loss complementation, and all the red-marked distribution networks are summarized into the same file; S324, repeat the above operation to perform daily pairing in combination with daily meteorological information.

[0028] It should be noted that all red-marked distribution networks are grouped into the same file to facilitate focused attention and processing of those distribution networks that cannot achieve effective load matching on the day, providing data support for subsequent grid transformation, load adjustment and other measures. Repeating the above operations for daily pairing in combination with daily meteorological information can dynamically adjust the pairing combination of distribution networks according to changes in meteorological conditions.

[0029] In summary, the present invention sets different temperature interval division strategies for the types of distribution networks under its jurisdiction, combines the load fluctuation characteristics of residential, industrial, commercial and mixed areas, captures the sensitivity of loads in different areas to temperature changes, and makes a fine prediction of the load of the distribution network. The obtained data is processed by the method of taking extreme values, effectively avoiding the deviation of the prediction results caused by short-term load fluctuations, making the prediction results more stable and reliable. By combining the meteorological data of the current prediction day, the temperature value at each moment of the day is obtained, and the temperature intervals divided by historical data are matched separately at each moment to obtain the predicted peak load at each moment. By calculating the difference between the planned load and the predicted peak load, it can be determined whether the current distribution network needs to adjust the load at the current moment, which is helpful to formulate a more refined and dynamic load scheduling strategy to balance supply and demand and ensure the stable operation of the power grid. By performing time matching and load matching for the surplus time periods and deficit time periods of different distribution networks, we can determine a distribution network with complementary loads and transfer the surplus load to meet the deficit load. This can effectively balance the power supply and demand relationship between different regions, avoid the waste of power resources, improve the utilization efficiency of power resources, and facilitate the management and adjustment of the distribution network under the jurisdiction of the city.

[0030] The above are only preferred embodiments of the present invention and are 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 load forecasting and adjustment of a distribution network, characterized in that: The method comprises the following steps: S1. Based on the management areas of different urban distribution networks, the types of distribution network usage are divided into residential, commercial, industrial and mixed types, and a database of urban distribution networks is established. Distribution network files are established based on the types; S2. Collect the daily load data of different distribution networks in different months in history, calculate the hourly predicted peak loads in different regions based on the temperature range, combine the planned loads of different distribution networks to obtain the predicted surplus load and predicted deficit load, and divide the surplus load time intervals and deficit load time intervals in different regions in different months; S3. Based on the daily surplus load range, predicted surplus load, deficit load range, and predicted deficit load in different regions and different months, daily forecast matching is performed on the distribution network files in different months to make comprehensive adjustments to the distribution network load.

2. A method for load forecasting and adjustment of distribution network according to claim 1, characterized in that: The S1 comprises the following sub-steps: S11. Based on the management areas of each distribution network under the jurisdiction of the city in the GIS map, obtain the proportion of residential areas in different distribution network management areas , Commercial area ratio , Industrial area ratio , based on the area proportion in different distribution network management areas, the distribution network types are divided, and the steps are as follows: When the proportion of residential area in the distribution network management area ≥70%, it means that the current distribution network type is residential; When the commercial area accounts for the proportion of the distribution network management area ≥60%, it means that the current distribution network type is commercial; When the industrial area accounts for the proportion of the distribution network management area ≥50%, it means that the current distribution network type is industrial; When the proportion of residential area, commercial area and industrial area in the distribution network management area do not meet the above ratios, it means that the current distribution network type is mixed; S12. Establish the urban distribution network database of the current city through MySQL, and establish type groups in the database, including residential, commercial, industrial, and mixed types. Based on the types of each distribution network under the city, establish distribution network files under the corresponding groups, and number the distribution networks in the files in sequence. The distribution network files include the geographical location, planned load, and name of the distribution network.

3. A method for load forecasting and adjustment of distribution network according to claim 1, characterized in that: The S2 comprises the following sub-steps: S21. Collect the load peak values ​​of different distribution networks every day and every hour in the historical months, and collect the temperature values ​​of different distribution networks every day and every hour in the historical months. Based on different distribution network grouping types, group the working days and weekends in the same month based on the temperature range. S22. Calculate the hourly peak load based on the temperature range grouping, and combine the current meteorological data of the distribution network to obtain the predicted peak load of the distribution network at each moment of the day. Combine the planned loads of different distribution networks to calculate the predicted surplus load and predicted deficit load, and divide the surplus load time interval and the deficit load time interval.

4. A method for load forecasting and adjustment of distribution network according to claim 3, characterized in that: The S21 comprises the following sub-steps: S211. Collect the temperature data of the same month in the past two years and obtain the temperature value at each hour in the current month , and distinguish between working days and weekends, where n and N represent the month and the day of the month respectively, and P represents the time; S222: Based on different distribution network grouping types, the working days and weekends in the same month are grouped based on the temperature range. The specific steps are as follows: For residential areas, a range of 3°C is used to capture the subtle changes in residential loads under different temperature conditions, including , i represents the highest temperature point, based on the temperature values ​​of the residential distribution network at different times in the current month , summarize the distribution network load peak data at the same time every day in the same month in the same temperature range according to the temperature value, and establish the data file at the current time. The file includes all temperature ranges at this time in the current month and the summarized distribution network load peak data in each temperature range, and separates the working days and weekends into groups; For commercial areas, a 6°C interval is used to capture the load changes in commercial areas under different temperature conditions, including , i represents the highest temperature point, based on the temperature values ​​at different times in the current month of the commercial distribution network , summarize the distribution network load peak data at the same time every day in the same month in the same temperature range according to the temperature value, and establish the data file at the current time. The file includes all temperature ranges at this time in the current month and the summarized distribution network load peak data in each temperature range, and separates the working days and weekends into groups; For industrial type, a range of 10°C is used to capture the overall change trend of industrial area load, including , i represents the highest temperature point, based on the temperature values ​​of the industrial distribution network at different times in the current month , summarize the distribution network load peak data at the same time every day in the same month in the same temperature range according to the temperature value, and establish the data file at the current time. The file includes all temperature ranges at this time in the current month and the summarized distribution network load peak data in each temperature range, and separates the working days and weekends into groups; For the mixed type, a range of 7°C is used to capture the overall change trend of the mixed zone load, including , i represents the highest temperature point, based on the temperature values ​​at different times in the current month of the hybrid distribution network , the distribution network load peak data at the same time every day in the same month are summarized in the same temperature range according to the temperature value, and the data file at the current time is established. The file includes all temperature ranges at this time in the current month and the summarized distribution network load peak data in each temperature range, and the working days and weekends are grouped separately.

5. A method for load forecasting and adjustment of distribution network according to claim 4, characterized in that: The S22 comprises the following sub-steps: S221. Combined with the data archive, for the working days and weekends in the same month of the same distribution network, the peak load data of the distribution network in all temperature intervals at the same time are taken to obtain the representative loads of different temperature intervals at the same time. The specific algorithm formula is: ； in, represents the peak load data of the nth distribution network in the ith temperature interval at the same time, Represents the representative load of the ith temperature interval at time Q, and calculates the representative load of working days and the representative load of weekends respectively; S222: Based on the meteorological information, the predicted temperature of the current distribution network at each moment of the same day is obtained. For weekdays and weekends, the predicted temperature is matched with the temperature range in the data file at the same moment, and the corresponding representative load is selected as the predicted peak load at the current moment. , Q is the time, and its value range is 0-23; S223, based on the planned load of the current distribution network , calculate the differential load at each moment of the current forecast day , and its algorithm formula is: ； The differential load at each moment Make a judgment to determine the type of the current differential load. The specific steps are as follows: when When ≥0, it means that the difference load at the current moment is the predicted surplus load, which is recorded as ; when When <0, it means that the current differential load is the predicted deficit load, which is recorded as ; Based on the predicted surplus load and predicted deficit load at different times of the day, the surplus load time interval and the deficit load time interval are divided, and the time period corresponding to the predicted surplus load is summarized into the surplus load time interval of the day, and the time period corresponding to the predicted deficit load is summarized into the deficit load time interval of the day.

6. A method for load forecasting and adjustment of distribution network according to claim 1, characterized in that: The S3 comprises the following sub-steps: S31. Based on the meteorological forecast information, the predicted surplus load or predicted deficit load at different times of the day in different months is obtained, and time matching analysis is performed on different regions to obtain a time matching combination; S32. Based on the combined results of the time matching analysis, load matching analysis is performed on different distribution networks, and one-to-one or one-to-many combinations are performed on different distribution network areas to generate a distribution network load pairing strategy.

7. A method for load forecasting and adjustment of distribution network according to claim 6, characterized in that: The S31 includes the following sub-steps: S311. Obtain a surplus load set based on the daily predicted surplus load or predicted deficit load of different distribution stations or deficit load collection ,in represents the moment of load surplus or load deficit, Represents the surplus load time period of the distribution network numbered n on the day, Represents the time period of the distribution network numbered n with a lack of load on the day; S312, for any two distribution networks i and j, where , when satisfied When , it represents the surplus and deficit time matching of distribution networks i and j. The combinations are recorded separately, and all i that meet the above conditions are excluded from the distribution network number. For the excluded distribution networks 1 to m and j, the combined matching is calculated: When satisfied When , after recording the combination, exclude 1 to m that meet the above from the distribution network number, and repeat the above until there is no combination; Repeat the above until the matching combinations of all distribution network deficit time periods are obtained.

8. A method for load forecasting and adjustment of distribution network according to claim 7, characterized in that: The S32 comprises the following steps: S321, calculate the corresponding time for each combination after pairing When all the absolute value differences in the set are greater than or equal to 0, it means that the load pairing result is met. When there are multiple groups of matching results, the absolute value differences in each combination are accumulated and sorted in ascending order. The first combination is selected as the pairing combination of the current distribution network deficit, and all subsequent combination schemes containing the surplus distribution network number in the current pairing combination are deleted; S322, for the remaining distribution networks that meet the time matching but do not meet the load matching, the surplus combinations in the time matching scheme are accumulated and combined, and the corresponding time is calculated after the accumulation. When all the absolute value differences in the set are greater than or equal to 0, it means that the load pairing result is met. When there are multiple sets of matching results, the absolute value differences in each new combination are accumulated and arranged in ascending order. The first new combination is selected as the pairing combination of the current distribution network deficit, and all subsequent combination schemes containing the surplus distribution network number in the current new pairing combination are deleted. When all the absolute value differences in the set are still not greater than or equal to 0 after re-combination, the current distribution network is marked red in the urban distribution network database; S323. After all the distribution networks under the jurisdiction of the urban distribution network database have been paired, the paired distribution networks are associated, and a separate file is created to summarize the load profit and loss complementation, and all the red-marked distribution networks are summarized into the same file; S324, repeat the above operation to perform daily pairing in combination with daily meteorological information.

Images