Energy storage capacity dynamic optimization system for source, grid, load and storage

Abstract

The present invention belongs to the technical field of power grid energy storage, and discloses a source-grid-load-storage energy storage capacity dynamic optimization system; the system comprises: a region labeling module for labeling R power consumption regions and generating corresponding power consumption region values; a first acquisition module for collecting weather, wind speed and temperature data of R power consumption regions in the future; a first processing module for preprocessing future time intervals and extracting date and time period features; a second processing module for extracting weather, wind speed and temperature features and generating environmental feature data; a power generation prediction module for inputting power consumption region values, time features and environmental features into a power consumption prediction model and outputting future power consumption of R regions; a capacity adjustment module for calculating the required capacity of G energy storage regions based on the prediction results and completing charging before the time interval; the present invention can dynamically calculate and adjust the energy storage capacity based on the predicted power consumption, thereby improving the efficiency and stability of power supply.

Description

Technical Field

[0001] The present invention relates to the technical field of power grid energy storage, and more specifically, to a system for dynamically optimizing energy storage capacity of a source-grid-load-storage system. Background Art

[0002] With the continuous growth of electricity demand, the safe operation of transmission lines has become increasingly important. Existing transmission line inspection methods mainly rely on manual inspection and some automated equipment.

[0003] To improve the efficiency and accuracy of transmission line fault detection, a Chinese patent application with publication number CN114693085A proposes a method for the joint optimization scheduling of energy and backup capacity for integrated variable energy sources. The method includes: first, using a convolutional neural network (CNN) to predict load demand using hourly real-time data. Then, by integrating renewable energy (RES) and a battery energy storage system (BSS), energy and spinning reserve capacity are jointly scheduled to meet the predicted load demand. Furthermore, the power generation system is penalized based on the amount of energy demand not met due to generation constraints, using a cost coefficient for unserved load. Furthermore, due to the tilting of thermal units, the available excess power is stored in the backup energy storage system, taking into account the charge state of the energy storage system. A particle swarm optimization algorithm is used to minimize this cost.

[0004] Although the above methods can meet most scenarios, research and practical application of the above methods and existing technologies have revealed that the above methods and existing technologies have at least the following defects:

[0005] Existing technologies use hourly real-time data to predict future electricity consumption, without taking into account the impact of the attributes of the electricity consumption area, the attributes of the future time interval, and the environmental data of the future time on electricity consumption. The attributes of the electricity consumption area include residential areas, industrial areas, office areas, and cultural and tourism areas, the attributes of the future time interval include date attributes (weekdays, weekends, and holidays) and time periods, and environmental data such as weather conditions, temperature, and wind speed. As a result, the accuracy of electricity consumption prediction is not high. When controlling the energy storage capacity based on the predicted electricity consumption, it is not considered whether the electricity demand requires the support of energy storage equipment. For areas that require the support of energy storage equipment, it is impossible to dynamically calculate and adjust the corresponding energy storage capacity based on the predicted electricity consumption, resulting in excessive or insufficient energy storage capacity, thereby wasting or insufficient energy storage resources, and reducing the efficiency and stability of power supply.

[0006] In view of this, the present invention proposes a source-grid-load-storage energy storage capacity dynamic optimization system to solve the above problems. Summary of the Invention

[0007] In order to overcome the above-mentioned defects of the prior art and to achieve the above-mentioned objectives, the present invention provides the following technical solution: a system for dynamically optimizing energy storage capacity of a source-grid-load-storage system, comprising:

[0008] The area labeling module is used to numerically label the R power consumption areas and obtain the power consumption area values ​​corresponding to the R power consumption areas;

[0009] The first collection module is used to collect weather conditions, ambient wind speed and ambient temperature of R power consumption areas in the future time interval;

[0010] The first processing module is used to pre-process the future time interval to obtain the date attribute value and the time period value;

[0011] The second processing module is used to extract features of weather conditions, ambient wind speed and ambient temperature respectively to obtain environmental feature data;

[0012] The power generation prediction module is used to input the R power consumption area values, date attribute values, time period values ​​and environmental characteristic data into the power consumption prediction model corresponding to the power consumption area values, and obtain the predicted power consumption corresponding to the R power consumption areas in the future time interval;

[0013] The capacity adjustment module is used to calculate the energy storage capacity corresponding to the G distributed energy storage areas based on the predicted power consumption of the R power consumption areas, and charge the energy storage devices of the G distributed energy storage areas before the starting point of the future time interval arrives, so that the storage capacity of the G distributed energy storage areas reaches the corresponding energy storage capacity.

[0014] Furthermore, the method of calculating the energy storage capacity corresponding to the G distributed energy storage areas and making the storage capacity of the G distributed energy storage areas reach the corresponding energy storage capacity includes:

[0015] S121: Obtain a power supply topology diagram of K distributed energy storage areas connected to R power consumption areas, and obtain a set of power supply areas respectively supplied by the K distributed energy storage areas based on the power supply topology diagram;

[0016] S122: Let the initial value of i be 1, and the value range of i be ;

[0017] S123: Obtain a power supply area set for the i-th distributed energy storage area, sum the predicted power consumption corresponding to the power consumption areas in the power supply area set, and obtain the predicted total power consumption of the power consumption areas connected to the i-th distributed energy storage area;

[0018] S124: If the predicted total power consumption is greater than or equal to the preset power consumption threshold, the power estimated to be stored in the i-th distributed energy storage area is calculated based on the predicted total power consumption, marked as the energy storage capacity, i is marked as the energy storage area sequence number, i and the energy storage capacity form a storage capacity subset, and the storage capacity subset is added to the storage capacity set. The storage capacity set has a total of G storage capacity subsets, where G is less than or equal to K, and S125 is executed. If the predicted total power consumption is less than the preset power consumption threshold, S125 is directly executed.

[0019] S125: Set i=i+1; if i is less than or equal to K, execute S123 to S124; if i is greater than K, execute S126;

[0020] S126: Before the start point of the future time interval arrives, the energy storage devices in the distributed energy storage area corresponding to the energy storage area number in the storage capacity set are respectively charged so that the energy storage capacity of the energy storage device reaches the energy storage capacity corresponding to the energy storage area number.

[0021] Furthermore, the method for obtaining the set of power supply areas respectively supplied by the K distributed energy storage areas includes:

[0022] S1211: Let the initial value of r be 1, and the value range of r is ;

[0023] S1212: If the rth power consumption area is connected to the ith distributed energy storage area, then add the rth power consumption area to the power supply area set of the ith distributed energy storage area;

[0024] S1213: Let r=r+1. If r is less than or equal to R, execute S1212. If r is greater than R, let i=i+1. If i is less than or equal to K, let r=1 and execute S1212. If i is greater than K, end.

[0025] Furthermore, the method for calculating the energy storage capacity includes:

[0026] S1241: Subtract the predicted total power consumption from the power consumption threshold to obtain the energy storage power supply of the i-th distributed energy storage area;

[0027] S1242: Calculate the energy storage capacity of the i-th distributed energy storage area based on the energy storage power supply, discharge depth, and charge and discharge efficiency;

[0028] The calculation method of the energy storage capacity is:

[0029] ;

[0030] is the energy storage capacity of the i-th distributed energy storage area, is the energy storage power supply of the i-th distributed energy storage area, is the depth of discharge, is the charge and discharge efficiency.

[0031] Furthermore, the method for obtaining the date attribute value includes:

[0032] S101: retain only the year, month, and day in the time interval and mark it as the label date;

[0033] S102: If the label date is a working day, execute S103; if the label date is a non-working day, execute S104;

[0034] S103: Only the hours, minutes and seconds are retained in the time interval, and marked as label time; the working time interval of the preset working day is hour point Seconds to hour point seconds; if the label time is within the working time range, the date attribute value is If the label time is not within the working time range, the date attribute value is ;

[0035] S104: If the tag date is a weekend, mark the date attribute value as If the label date is a holiday, the date attribute value is marked according to the number of days of the holiday. The longer the holiday, the larger the date attribute value.

[0036] Furthermore, the method for obtaining the time period value includes:

[0037] S111: retain the time interval in hours, minutes, and seconds to obtain the time period of the future time interval;

[0038] S112: Pre-constructing a time period value table; the time period value table includes time periods and time period values ​​corresponding to the time periods, and the time periods only include hours, minutes, and seconds; the time period length in the time period value table is greater than or equal to the time period length of the future time interval;

[0039] S113: If the time period of the future time interval is within the time period in the time period value table, the time period value of the future time interval is the corresponding time period value in the time period value table;

[0040] S114: If the time period of the future time interval is not completely included in any time period in the time period value table, that is, the time period of the future time interval does not match any time period in the time period value table, then calculate the proportion of the time period of the future time interval in each time period in the time period value table, and take the time period value corresponding to the time period with the largest proportion as the time period value of the future time interval.

[0041] Furthermore, the training method of the electricity consumption prediction model includes:

[0042] An electricity consumption prediction data set is collected in advance, wherein the electricity consumption prediction data set includes electricity consumption prediction data and regional electricity consumption corresponding to the electricity consumption prediction data; the electricity consumption prediction data includes electricity consumption area values, date attribute values, time period values ​​and environmental characteristic data; the electricity consumption prediction data set is divided into a training set and a test set, the electricity consumption prediction data in the training set is used as the input of the electricity consumption prediction model, and the regional electricity consumption in the training set is used as the output of the electricity consumption prediction model, with minimizing the sum of the prediction accuracies of all predicted regional electricity consumption as the training goal; training is stopped when the sum of the prediction accuracies reaches convergence; the electricity consumption prediction model is a gradient boosting tree model.

[0043] Furthermore, the environmental characteristic data includes weather state mean, weather state standard deviation, wind speed trimmed mean and temperature trimmed mean;

[0044] The method for obtaining the weather state average value and the weather state standard deviation includes:

[0045] Numerical annotation of weather conditions; dividing the future time interval into Q time periods, establishing a weather analysis set based on the weather conditions of the Q time periods, and calculating the average and standard deviation of the weather conditions within the weather analysis set;

[0046] The method for obtaining the trimmed mean wind speed includes:

[0047] A wind speed analysis set is established based on the ambient wind speeds of Q time periods in the future time interval. The maximum and minimum wind speeds are removed from the wind speed analysis set to form a new wind speed analysis set. The average temperature in the new temperature analysis set is calculated to obtain the trimmed temperature average.

[0048] Furthermore, the method for obtaining the temperature trimmed mean value includes:

[0049] A temperature analysis set is established based on the ambient temperature of Q time periods in the future time interval. The maximum temperature and the minimum temperature are removed from the temperature analysis set to form a new temperature analysis set. The average temperature in the new temperature analysis set is calculated to obtain the trimmed temperature average.

[0050] The technical effects and advantages of the source-grid-load-storage energy storage capacity dynamic optimization system of the present invention are as follows: different electricity consumption areas have different electricity consumption characteristics, so the present invention numerically labels the electricity consumption areas to distinguish different electricity consumption characteristics; the present invention also takes into account multiple variables such as weather conditions, ambient wind speed, ambient temperature, date attributes and time periods; the electricity consumption area values ​​and multiple variables are combined with a machine learning model to more accurately predict electricity consumption; based on accurate electricity consumption predictions, the system dynamically calculates the required energy storage capacity of each distributed energy storage area, and charges the energy storage equipment in advance, thereby improving the dynamic adjustment effect of the energy storage capacity; this method avoids the situation of excessive or insufficient energy storage capacity, improves resource utilization and the reliability of power supply; dynamically calculates and adjusts the corresponding energy storage capacity according to the predicted electricity consumption, solves the problem of waste or insufficient energy storage resources, and improves the efficiency and stability of power supply. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 Schematic diagram of a system for dynamically optimizing energy storage capacity of a source-grid-load-storage system according to Example 1 of the present invention;

[0052] Figure 2 Schematic diagram of a system for dynamic optimization of energy storage capacity of source, grid, load and storage according to embodiment 2 of the present invention;

[0053] Figure 3 This is a flow chart of a method for dynamically optimizing energy storage capacity of source, grid, load and storage according to embodiment 3 of the present invention;

[0054] Figure 4 Flowchart of a method for dynamically achieving a corresponding energy storage capacity for storage capacity. DETAILED DESCRIPTION

[0055] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0056] Example 1

[0057] See also Figure 1 As shown, this embodiment provides a dynamic optimization system for energy storage capacity of source, grid, load and storage, including a regional labeling module, a first acquisition module, a first processing module, a second processing module, a power generation prediction module and a capacity adjustment module; each module realizes data transmission through wired and / or wireless connections.

[0058] The area labeling module is used to numerically label the R power consumption areas and obtain the power consumption area values ​​corresponding to the R power consumption areas.

[0059] The method for obtaining the power consumption area value includes:

[0060] The electricity consumption areas are divided into residential areas, commercial areas, office areas and cultural and tourism areas, and the residential areas, commercial areas, office areas and cultural and tourism areas are numerically labeled to obtain the electricity consumption area values.

[0061] For example, the electricity consumption area value of the residential area is marked as 1, the electricity consumption area value of the commercial area is marked as 2, the electricity consumption area value of the office area is marked as 3, and the electricity consumption area value of the cultural and tourism area is marked as 4; it should be noted that different electricity consumption areas show different electricity consumption characteristics at different times. For example, during working hours, the electricity consumption in the office area is higher, and the electricity consumption in the commercial area, residential area and cultural and tourism area is lower; for example, during non-working hours, the electricity consumption in the commercial area, residential area and cultural and tourism area is higher than that in the office area.

[0062] The first collection module is used to collect weather conditions, ambient wind speed and ambient temperature of R power consumption areas in a future time interval.

[0063] The temperature and weather conditions in the future time interval are obtained through a weather station; the weather conditions include sunny, sunny to cloudy, cloudy, overcast, light rain, light snow, fog, haze, moderate snow, heavy snow, thunderstorm and rainstorm.

[0064] The time interval is a continuous time period from the start point of the future time interval to the end point of the future time interval, and the time interval includes year, month, day, hour, minute, and second. For example, if the future time interval starts at 12:01:10 on May 20, 2023, and ends at 13:00:30 on May 20, 2023, the time interval is from 12:01:10 on May 20, 2023 to 13:00:30 on May 20, 2023.

[0065] The first processing module is used to pre-process the future time interval to obtain a date attribute value and a time period value.

[0066] The method for obtaining the date attribute value includes:

[0067] S101: retain only the year, month, and day in the time interval and mark it as the label date;

[0068] S102: If the label date is a working day, execute S103; if the label date is a non-working day, execute S104;

[0069] S103: Only the hours, minutes and seconds are retained in the time interval, and marked as label time; the working time interval of the preset working day is hour point Seconds to hour point seconds; if the label time is within the working time range, the date attribute value is If the label time is not within the working time range, the date attribute value is For example, the working time range is from 8:30:00 to 18:00:00.

[0070] S104: If the tag date is a weekend, mark the date attribute value as If the label date is a holiday, the date attribute value is marked according to the number of days of the holiday. The longer the holiday, the larger the date attribute value. For example, if it is New Year's Day, Qingming Festival, Dragon Boat Festival or Mid-Autumn Festival, the date attribute value is marked as ; If the label date is Labor Day, then the date attribute value is marked as ; If the label date is National Day, the date attribute value is marked as ; If the label date is Spring Festival, the date attribute value is marked as ; .

[0071] It should be noted that the date attribute value has an important impact on the predicted power consumption. When the date attribute value changes, the power consumption behavior also changes. For example, the date attribute value is , which is the working time on weekdays. People go from residential areas to office areas to work. The electricity consumption in office areas is higher than that in residential areas, commercial areas, and cultural and tourism areas. The date attribute value is When is the non-working time on weekdays, people finish work and go from office areas to residential areas and commercial areas. The electricity consumption in residential areas and commercial areas is higher than that in office areas and cultural and tourism areas. The date attribute value is When it is the weekend, people are resting, the flow of people in the commercial and cultural tourism areas increases, and the electricity consumption increases; the date attribute value is 、 or During holidays, the flow of people in commercial and cultural tourism areas is higher than on weekends, and the electricity consumption is higher than on weekends. The number of people traveling increases, and the electricity consumption in residential areas decreases accordingly. The larger the date attribute value, the longer the corresponding holiday time, and people's activity choices and activity time will increase. At the same time, the electricity consumption also increases.

[0072] The method for obtaining the time period value includes:

[0073] S111: retain the time interval in hours, minutes, and seconds to obtain the time period of the future time interval;

[0074] S112: Pre-constructing a time period value table; the time period value table includes time periods and time period values ​​corresponding to the time periods, and the time periods only include hours, minutes, and seconds; the time period length in the time period value table is greater than or equal to the time period length of the future time interval;

[0075] S113: If the time period of the future time interval is within the time period in the time period value table, the time period value of the future time interval is the corresponding time period value in the time period value table;

[0076] S114: If the time period of the future time interval is not completely included in any time period in the time period value table, that is, the time period of the future time interval does not match any time period in the time period value table, then calculate the proportion of the time period of the future time interval in each time period in the time period value table, and take the time period value corresponding to the time period with the largest proportion as the time period value of the future time interval.

[0077] For example, if the time period of the future time interval is from 18:41:00 to 20:40:00, and the time period value table contains the time period from 18:00:01 to 20:00:00 with a time period value of 10, and the time period from 20:00:01 to 22:00:00 with a time period value of 11, then 40 minutes is in the time period from 20:00:01 to 22:00:00, and the proportion is , 80 minutes in the time period 18:00:01 to 20:00:00, accounting for , that is, the time period value of the future time interval is 10.

[0078] An example of a time period value table is shown in Table 1:

[0079] Table 1 Time period value table

[0080] Time period Time period value 00：00：01-02：00：00 1 02：00：01-04：00：00 2 04：00：01-06：00：00 3 06：00：01-08：00：00 4 08：00：01-10：00：00 5 10：00：01-12：00：00 6 12：00：01-14：00：00 7 14：00：01-16：00：00 8 16：00：01-18：00：00 9 18：00：01-20：00：00 10 20：00：01-22：00：00 11 22：00：01-24：00：00 12

[0081] It's important to note that electricity demand differs between specific weekday activities (such as office work and industrial production) and non-weekday activities (such as family entertainment, shopping, and travel), exhibiting distinct cyclical patterns. Using date attribute values ​​and time period values ​​can effectively identify these cyclical patterns, thereby improving the accuracy of electricity consumption forecasts for specific areas. For example, during the day on weekends, electricity consumption in commercial and cultural tourism areas increases. However, on weekend evenings, as people return to residential areas from these areas, electricity consumption in these areas decreases, while electricity consumption in residential areas increases.

[0082] The second processing module is used to extract features of weather conditions, ambient wind speed and ambient temperature respectively to obtain environmental feature data; the environmental feature data includes weather condition average value, weather condition standard deviation, wind speed trimmed average value and temperature trimmed average value.

[0083] The method for obtaining the weather state average value and the weather state standard deviation includes:

[0084] Numerical annotation of weather conditions;

[0085] For example, sunny is marked as 12, sunny to cloudy is marked as 11, cloudy is marked as 10, overcast is marked as 9, light rain is marked as 8, light snow is marked as 7, fog is marked as 6, haze is marked as 5, moderate snow is marked as 4, heavy snow is marked as 3, thunderstorm is marked as 2, and heavy rain is marked as 1.

[0086] The future time interval is divided into Q time periods, and the weather conditions of the Q time periods are used to establish a weather analysis set. The average value and standard deviation of the weather conditions in the weather analysis set are calculated. The specific method is as follows:

[0087] The calculation method of the weather state average value includes:

[0088] ;

[0089] The calculation method of the weather state standard deviation includes:

[0090] ;

[0091] in, is the average weather condition; is the weather status of the qth time period in the weather analysis set; The standard deviation of weather conditions reflects the concentration of changes in the average weather conditions. The larger the average weather conditions and the smaller the standard deviation, the better the weather conditions. During non-working days, it is more conducive to people's travel, and vice versa, it is not conducive to people's travel, thus affecting the power consumption in the power consumption area.

[0092] The method for obtaining the trimmed mean wind speed includes:

[0093] The ambient wind speeds in Q time periods in the future time interval are used to establish a wind speed analysis set. The maximum and minimum wind speeds are removed from the wind speed analysis set to form a new wind speed analysis set. The average wind speed in the new wind speed analysis set is calculated to obtain the trimmed wind speed average. The specific method is as follows:

[0094] ;

[0095] in, is the trimmed mean wind speed, is the mth wind speed in the wind speed analysis set; it should be noted that the trimmed mean wind speed calculated after removing the maximum and minimum wind speeds can more accurately reflect the impact of wind speed on people's travel; the larger the trimmed mean wind speed, the greater the wind speed and the greater the impact on people's travel.

[0096] The method for obtaining the temperature trimmed average value includes:

[0097] A temperature analysis set is established based on the ambient temperature of Q time periods in the future time interval. The maximum and minimum temperature values ​​are removed from the temperature analysis set to form a new temperature analysis set. The average temperature in the new temperature analysis set is calculated to obtain the trimmed temperature average. The specific method is as follows:

[0098] ;

[0099] in, is the trimmed mean temperature, is the mth temperature in the temperature analysis set; it should be noted that the trimmed mean temperature calculated after removing the maximum and minimum temperature values ​​can more accurately reflect the impact of temperature on people's travel; the larger the trimmed mean temperature, the higher the temperature and the greater the impact on people's travel.

[0100] The power generation prediction module is used to input the numerical values ​​of R power consumption areas, date attribute values, time period values ​​and environmental characteristic data into the power consumption prediction model corresponding to the power consumption area values, and obtain the predicted power consumption corresponding to the R power consumption areas in the future time interval.

[0101] The training method of the electricity consumption prediction model includes:

[0102] An electricity consumption prediction data set is collected in advance, wherein the electricity consumption prediction data set includes electricity consumption prediction data and regional electricity consumption corresponding to the electricity consumption prediction data; the electricity consumption prediction data includes electricity consumption area values, date attribute values, time period values ​​and environmental characteristic data; the electricity consumption prediction data set is divided into a training set and a test set, the electricity consumption prediction data in the training set is used as the input of the electricity consumption prediction model, and the regional electricity consumption in the training set is used as the output of the electricity consumption prediction model, with minimizing the sum of the prediction accuracies of all predicted regional electricity consumption as the training goal; training is stopped when the sum of the prediction accuracies reaches convergence; the electricity consumption prediction model is a gradient boosting tree model.

[0103] The calculation formula for prediction accuracy is: ,in, For prediction accuracy, For the The predicted regional electricity consumption corresponding to the group electricity consumption forecast data, For the The actual regional electricity consumption corresponding to the group electricity consumption forecast data.

[0104] It should be noted that the power consumption area values, date attribute values, time period values ​​and environmental characteristic data cover the main factors affecting power consumption forecasts. The rationality of these data is reflected in the fact that they jointly describe different aspects that affect power consumption. The lack of any factor will lead to information loss in the power consumption forecast model, affecting the prediction accuracy and reliability of the power consumption forecast model.

[0105] For example, the power consumption region value reflects the power demand characteristics of different power consumption regions. If this feature is missing, the power consumption forecast model will not be able to distinguish the differences between different power consumption regions, resulting in prediction bias. The date attribute value can reflect the impact of weekdays and holidays on power consumption in different regions. If this feature is missing, the power consumption forecast model will not be able to accurately reflect the power consumption changes caused by date changes. The time period value reflects the power consumption at different time periods of the day. For example, power consumption during the morning and evening peak periods is higher than that during the night or off-peak periods. If this feature is missing, the power consumption forecast model will not be able to accurately reflect the power consumption changes caused by time period changes. The average weather state value has a significant impact on regional power consumption. For example, in rainy and snowy weather conditions, people will go out less, which will cause power consumption changes in different regions. If this feature is missing, the power consumption forecast model will not be able to accurately predict power consumption changes caused by weather changes. The weather state standard deviation can reflect the stability or volatility of the weather, which is crucial for predicting power consumption during extreme weather or drastic weather changes. If this feature is missing, the power consumption forecast model will not be able to cope with abnormal power consumption caused by weather fluctuations. The trimmed mean wind speed reflects the overall wind speed of the environment and also affects regional electricity consumption forecasts. For example, when wind speeds are too high, people tend to go out less, causing electricity consumption to vary across different regions. Removing this feature would prevent the electricity consumption forecast model from accurately predicting changes in electricity consumption caused by wind speed fluctuations. The trimmed mean temperature is a key factor affecting electricity consumption, especially when high or low temperatures cause a significant increase in electricity consumption (e.g., due to air conditioning or heating use). Removing this feature would prevent the electricity consumption forecast model from accurately predicting changes in electricity consumption caused by temperature fluctuations.

[0106] The capacity adjustment module is used to calculate the energy storage capacity corresponding to the G distributed energy storage areas based on the predicted power consumption of the R power consumption areas, and charge the energy storage devices of the G distributed energy storage areas before the starting point of the future time interval arrives, so that the storage capacity of the G distributed energy storage areas reaches the corresponding energy storage capacity.

[0107] like Figure 4As shown, the method for calculating the energy storage capacity corresponding to the G distributed energy storage areas and making the storage capacity of the G distributed energy storage areas reach the corresponding energy storage capacity includes:

[0108] S121: Obtain a power supply topology diagram of K distributed energy storage areas connected to R power consumption areas, and obtain a set of power supply areas respectively supplied by the K distributed energy storage areas based on the power supply topology diagram;

[0109] S122: Let the initial value of i be 1, and the value range of i be ;

[0110] S123: Obtain a power supply area set for the i-th distributed energy storage area, sum the predicted power consumption corresponding to the power consumption areas in the power supply area set, and obtain the predicted total power consumption of the power consumption areas connected to the i-th distributed energy storage area;

[0111] S124: If the predicted total power consumption is greater than or equal to the preset power consumption threshold, the power estimated to be stored in the i-th distributed energy storage area is calculated based on the predicted total power consumption, marked as the energy storage capacity, i is marked as the energy storage area sequence number, i and the energy storage capacity form a storage capacity subset, and the storage capacity subset is added to the storage capacity set. The storage capacity set has a total of G storage capacity subsets, where G is less than or equal to K, and S125 is executed. If the predicted total power consumption is less than the preset power consumption threshold, S125 is directly executed.

[0112] It's important to note that if the predicted total power consumption is greater than or equal to the preset power consumption threshold, it's considered a peak, requiring the distributed energy storage equipment to supply power. If the predicted total power consumption is less than the preset power consumption threshold, it's considered a peak, requiring no distributed energy storage equipment to supply power. This dual loop ensures a complete match between the R power consumption areas and the K distributed energy storage areas, avoiding mismatches between the power consumption areas and the distributed energy storage areas.

[0113] The storage capacity set examples are as follows:

[0114] ;

[0115] in, is the sequence number of the distributed energy storage area in the Gth storage capacity subset, is the energy storage capacity of the distributed energy storage area in the Gth storage capacity subset.

[0116] S125: Set i=i+1; if i is less than or equal to K, execute S123 to S124; if i is greater than K, execute S126;

[0117] S126: Before the start point of the future time interval arrives, the energy storage devices in the distributed energy storage area corresponding to the energy storage area number in the storage capacity set are respectively charged so that the energy storage capacity of the energy storage device reaches the energy storage capacity corresponding to the energy storage area number.

[0118] The method for obtaining the set of power supply areas respectively supplied by the K distributed energy storage areas includes:

[0119] S1211: Let the initial value of r be 1, and the value range of r is ;

[0120] S1212: If the rth power consumption area is connected to the ith distributed energy storage area, then add the rth power consumption area to the power supply area set of the ith distributed energy storage area;

[0121] S1213: Let r=r+1. If r is less than or equal to R, execute S1212. If r is greater than R, let i=i+1. If i is less than or equal to K, let r=1 and execute S1212. If i is greater than K, end.

[0122] The calculation method of the energy storage capacity includes:

[0123] S1241: Subtract the predicted total power consumption from the power consumption threshold to obtain the energy storage power supply of the i-th distributed energy storage area;

[0124] S1242: Calculate the energy storage capacity of the i-th distributed energy storage area based on the energy storage power supply, discharge depth, and charge and discharge efficiency;

[0125] The calculation method of the energy storage capacity is:

[0126] ;

[0127] in, is the energy storage capacity of the i-th distributed energy storage area, is the energy storage power supply of the i-th distributed energy storage area, is the depth of discharge, =Charge / discharge efficiency. It should be noted that to protect energy storage devices (such as batteries) and extend their lifespan, not all of the energy is released; instead, some energy is retained. The depth of discharge refers to the percentage of the energy storage device's total capacity released during a single discharge. Charge / discharge efficiency refers to the efficiency of energy conversion during the energy storage system's charge and discharge processes. Energy storage devices experience energy losses during the charging and discharging process, so the actual power supply will be less than the storage capacity. For example, if the energy storage capacity is 100 MWh, the depth of discharge is 0.95, and the charge / discharge efficiency is 0.95, the maximum available power is: Maximum available power = 100 MWh × 0.95 × 0.95 = 90.25 MWh.

[0128] Example 2

[0129] See also Figure 2 As shown, this embodiment provides a system for dynamically optimizing energy storage capacity of a source, grid, load, and storage, further comprising:

[0130] The threshold prediction module is used to input the predicted power consumption, power consumption area value, date attribute value, time period value and environmental characteristic data into the threshold prediction model to obtain the predicted power consumption threshold;

[0131] The training method of the threshold prediction model includes:

[0132] A threshold prediction data set is collected in advance, wherein the threshold prediction data set includes threshold prediction data and a power consumption threshold corresponding to the threshold prediction data; the threshold prediction data includes predicted power consumption, power consumption area value, date attribute value, time period value and environmental characteristic data; the threshold prediction data set is divided into a training set and a test set, a classifier is constructed, the threshold prediction data in the training set is used as the input of the threshold prediction model, and the power consumption threshold in the training set is used as the output of the threshold prediction model, the classifier is trained to obtain an initial classifier, the initial classifier is tested using the test set, and a classifier that meets the preset accuracy is output as the threshold prediction model, and the threshold prediction model is a naive Bayes model or a support vector machine model.

[0133] Existing technologies fail to consider the varying power supply capabilities of the main grid to each power consumption area during different time periods (i.e., varying power consumption thresholds). Energy storage devices are used when they are not required, resulting in unnecessary energy conversion losses. Therefore, it is necessary to dynamically set power consumption thresholds for each power consumption area for different time periods. Power consumption thresholds are related to multiple factors, including predicted power consumption, power consumption area values, date attribute values, time period values, and environmental characteristic data. This embodiment proposes a method for dynamically predicting power consumption thresholds. By integrating multiple data sources, including power consumption area values, date attribute values, time period values, predicted power consumption, and environmental characteristic data, a power consumption threshold prediction model is established. This model can dynamically predict corresponding power consumption thresholds based on the actual conditions of different time periods and power consumption areas.

[0134] When the predicted power consumption is higher than the set power consumption threshold, the system will dispatch the energy storage device to discharge power to meet the power demand and relieve the pressure on the grid. When the predicted power consumption is lower than the power consumption threshold, the system will use the main grid power to charge the energy storage device, preparing for the peak period in advance.

[0135] This method not only maximizes the utilization of energy storage devices and enables refined scheduling, but also dynamically adjusts power consumption thresholds based on real-time conditions, improving the economic benefits and energy efficiency of the power system. For example, during peak hours or inclement weather, the power consumption threshold can be appropriately lowered to ensure timely intervention of energy storage devices. Meanwhile, during low-consumption periods, such as at night, the power consumption threshold can be appropriately raised to reduce unnecessary energy conversion losses. By dynamically predicting appropriate power consumption thresholds, it helps improve the flexibility and stability of smart grids, optimize energy storage resource allocation, reduce operating costs, promote efficient energy utilization, and achieve sustainable development.

[0136] Example 3

[0137] See also Figure 3 As shown, this embodiment provides a method for dynamically optimizing energy storage capacity of a source-grid-load-storage system, including:

[0138] Numerical marking is performed on the R power consumption areas to obtain the power consumption area values ​​corresponding to the R power consumption areas;

[0139] Collect weather conditions, ambient wind speed, and ambient temperature of R power consumption areas in the future time interval;

[0140] Preprocess the future time interval to obtain the date attribute value and time period value;

[0141] Extract features of weather conditions, ambient wind speed, and ambient temperature respectively to obtain environmental feature data;

[0142] Input the R power consumption area values, date attribute values, time period values ​​and environmental characteristic data into the power consumption prediction model corresponding to the power consumption area values, and obtain the predicted power consumption corresponding to the R power consumption areas in the future time interval;

[0143] Based on the predicted power consumption of R power consumption areas, the energy storage capacity corresponding to the G distributed energy storage areas is calculated, and before the starting point of the future time interval arrives, the energy storage devices of the G distributed energy storage areas are charged so that the storage capacity of the G distributed energy storage areas reaches the corresponding energy storage capacity.

[0144] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention should be included in the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

[0145] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A dynamic optimization system for energy storage capacity of source, grid, load and storage, characterized by: include: The area labeling module is used to numerically label the R power consumption areas and obtain the power consumption area values ​​corresponding to the R power consumption areas; The first collection module is used to collect weather conditions, ambient wind speed and ambient temperature of R power consumption areas in the future time interval; The first processing module is used to pre-process the future time interval to obtain the date attribute value and the time period value; The second processing module is used to extract features of weather conditions, ambient wind speed and ambient temperature respectively to obtain environmental feature data; The power generation prediction module is used to input the R power consumption area values, date attribute values, time period values ​​and environmental characteristic data into the power consumption prediction model corresponding to the power consumption area values, and obtain the predicted power consumption corresponding to the R power consumption areas in the future time interval; The capacity adjustment module is used to calculate the energy storage capacity corresponding to the G distributed energy storage areas based on the predicted power consumption of the R power consumption areas, and charge the energy storage devices of the G distributed energy storage areas before the start point of the future time interval arrives, so that the storage capacity of the G distributed energy storage areas reaches the corresponding energy storage capacity; The method of calculating the energy storage capacity corresponding to the G distributed energy storage areas and making the storage capacity of the G distributed energy storage areas reach the corresponding energy storage capacity includes: S121: Obtain a power supply topology diagram of K distributed energy storage areas connected to R power consumption areas, and obtain a set of power supply areas respectively supplied by the K distributed energy storage areas based on the power supply topology diagram; S122: Let the initial value of i be 1, and the value range of i be ; S123: Obtain a power supply area set for the i-th distributed energy storage area, sum the predicted power consumption corresponding to the power consumption areas in the power supply area set, and obtain the predicted total power consumption of the power consumption areas connected to the i-th distributed energy storage area; S124: If the predicted total power consumption is greater than or equal to the preset power consumption threshold, the power estimated to be stored in the i-th distributed energy storage area is calculated based on the predicted total power consumption, marked as the energy storage capacity, i is marked as the energy storage area sequence number, i and the energy storage capacity form a storage capacity subset, and the storage capacity subset is added to the storage capacity set. The storage capacity set has a total of G storage capacity subsets, where G is less than or equal to K, and S125 is executed. If the predicted total power consumption is less than the preset power consumption threshold, S125 is directly executed. S125: Set i=i+1; if i is less than or equal to K, execute S123 to S124; if i is greater than K, execute S126; S126: Before the start point of the future time interval arrives, the energy storage devices in the distributed energy storage area corresponding to the energy storage area number in the storage capacity set are respectively charged so that the energy storage capacity of the energy storage device reaches the energy storage capacity corresponding to the energy storage area number.

2. The source-grid-load-storage energy storage capacity dynamic optimization system according to claim 1 is characterized in that: The method for obtaining the set of power supply areas respectively supplied by the K distributed energy storage areas includes: S1211: Let the initial value of r be 1, and the value range of r is ; S1212: If the rth power consumption area is connected to the ith distributed energy storage area, then add the rth power consumption area to the power supply area set of the ith distributed energy storage area; S1213: Let r=r+1. If r is less than or equal to R, execute S1212. If r is greater than R, let i=i+1. If i is less than or equal to K, let r=1 and execute S1212. If i is greater than K, end.

3. The source-grid-load-storage energy storage capacity dynamic optimization system according to claim 2 is characterized in that: The calculation method of the energy storage capacity includes: S1241: Subtract the predicted total power consumption from the power consumption threshold to obtain the energy storage power supply of the i-th distributed energy storage area; S1242: Calculate the energy storage capacity of the i-th distributed energy storage area based on the energy storage power supply, discharge depth, and charge and discharge efficiency.

4. The source-grid-load-storage energy storage capacity dynamic optimization system according to claim 3 is characterized in that: The method for obtaining the date attribute value includes: S101: retain only the year, month, and day in the time interval and mark it as the label date; S102: If the label date is a working day, execute S103; if the label date is a non-working day, execute S104; S103: Only the hours, minutes and seconds are retained in the time interval, and marked as label time; the working time interval of the preset working day is hour point Seconds to hour point seconds; if the label time is within the working time range, the date attribute value is If the label time is not within the working time range, the date attribute value is ; S104: If the tag date is a weekend, mark the date attribute value as If the label date is a holiday, the date attribute value is marked according to the number of days of the holiday. The longer the holiday, the larger the date attribute value.

5. The source-grid-load-storage energy storage capacity dynamic optimization system according to claim 4 is characterized in that: The method for obtaining the time period value includes: S111: retain the time interval in hours, minutes, and seconds to obtain the time period of the future time interval; S112: Pre-constructing a time period value table; the time period value table includes time periods and time period values ​​corresponding to the time periods, and the time periods only include hours, minutes, and seconds; the time period length in the time period value table is greater than or equal to the time period length of the future time interval; S113: If the time period of the future time interval is within the time period in the time period value table, the time period value of the future time interval is the corresponding time period value in the time period value table; S114: If the time period of the future time interval is not completely included in any time period in the time period value table, that is, the time period of the future time interval does not match any time period in the time period value table, then calculate the proportion of the time period of the future time interval in each time period in the time period value table, and take the time period value corresponding to the time period with the largest proportion as the time period value of the future time interval.

6. The source-grid-load-storage energy storage capacity dynamic optimization system according to claim 5, characterized in that: The training method of the electricity consumption prediction model includes: An electricity consumption prediction data set is collected in advance, wherein the electricity consumption prediction data set includes electricity consumption prediction data and regional electricity consumption corresponding to the electricity consumption prediction data; the electricity consumption prediction data includes electricity consumption area values, date attribute values, time period values ​​and environmental characteristic data; the electricity consumption prediction data set is divided into a training set and a test set, the electricity consumption prediction data in the training set is used as the input of the electricity consumption prediction model, and the regional electricity consumption in the training set is used as the output of the electricity consumption prediction model, with minimizing the sum of the prediction accuracies of all predicted regional electricity consumption as the training goal; training is stopped when the sum of the prediction accuracies reaches convergence; the electricity consumption prediction model is a gradient boosting tree model.

7. The source-grid-load-storage energy storage capacity dynamic optimization system according to claim 6, characterized in that: The environmental characteristic data include weather state mean, weather state standard deviation, wind speed trimmed mean and temperature trimmed mean; The method for obtaining the weather state average value and the weather state standard deviation includes: Numerical annotation of weather conditions; dividing the future time interval into Q time periods, establishing a weather analysis set based on the weather conditions of the Q time periods, and calculating the average and standard deviation of the weather conditions within the weather analysis set; The method for obtaining the wind speed trimmed mean value includes: A wind speed analysis set is established based on the ambient wind speeds of Q time periods in the future time interval. The maximum wind speed and the minimum wind speed are removed from the wind speed analysis set to form a new wind speed analysis set. The average wind speed in the new wind speed analysis set is calculated to obtain the wind speed trimmed average.

8. The source-grid-load-storage energy storage capacity dynamic optimization system according to claim 7, characterized in that: The method for obtaining the temperature trimmed average value includes: A temperature analysis set is established based on the ambient temperature of Q time periods in the future time interval. The maximum temperature and the minimum temperature are removed from the temperature analysis set to form a new temperature analysis set. The average temperature in the new temperature analysis set is calculated to obtain the trimmed temperature average.

Images