A base station resource scheduling method and device, electronic equipment and storage medium

Abstract

The embodiment of the application discloses a base station resource scheduling method and device, electronic equipment and storage medium, belongs to the power grid peak shaving technical field, and can solve the problem of poor actual peak shaving capacity of a virtual power plant. Including: obtaining operation data of each base station included in the virtual power plant, and historical data of the base station; inputting the historical data into a pre-trained load prediction model to output predicted load data corresponding to the base station; determining the constraint condition of the base station and the response information of the base station according to the operation data and the predicted load data; on the basis of meeting the constraint condition, calculating a target scheduling value of the virtual power plant according to the response information of the base station, the target scheduling value including one or more of the following: target base station corresponding income information, target base station peak shaving capacity information and target device priority information; determining resource scheduling information according to the target scheduling value, the resource scheduling information being used for assisting the virtual power plant in scheduling resources of the base stations in the virtual power plant.

Description

Technical Field

[0001] This application relates to the field of power grid peak shaving technology, and in particular to a base station resource scheduling method, apparatus, electronic device and storage medium. Background Technology

[0002] In recent years, virtual power plants, as an emerging product of power market reform and energy internet construction, participate in the peak-shaving ancillary service market in a form similar to traditional power plants by aggregating and dispatching peak-shaving resources on the demand side.

[0003] However, current research on virtual power plants typically relies on long-term statistical data, simply aggregating all available resources without accurately analyzing the differences between distributed resources, leading to poor actual peak-shaving capacity of virtual power plants. Summary of the Invention

[0004] The purpose of this application is to provide a base station resource scheduling method, apparatus, electronic device, and storage medium to solve the problem of poor actual peak-shaving capacity of virtual power plants.

[0005] To solve the above-mentioned technical problems, the embodiments of this application are implemented as follows:

[0006] In a first aspect, embodiments of this application provide a base station resource scheduling method, comprising: acquiring operational data of each base station included in a virtual power plant, and historical data of the base station, wherein the operational data includes data related to the operational status of the base station, and the historical data includes data related to the predicted load data of the base station;

[0007] The historical data is input into a pre-trained load prediction model, which outputs the predicted load data corresponding to the base station.

[0008] Based on the operational data and the predicted load data, the constraints of the base station and the response information of the base station are determined. The response information of the base station includes: the peak shaving capability of the base station and / or the priority of each device corresponding to the base station.

[0009] Based on the constraints, the target scheduling value of the virtual power plant is calculated according to the response information of the base station. The target scheduling value includes one or more of the following: revenue information corresponding to the target base station, peak shaving capacity information of the target base station, and priority information of the target equipment.

[0010] Based on the target scheduling value, resource scheduling information is determined, which is used to assist the virtual power plant in scheduling resources for the base stations within the virtual power plant.

[0011] Secondly, embodiments of this application provide a base station resource scheduling device, including: an acquisition module, used to acquire operating data of each base station included in a virtual power plant, and historical data of the base station, wherein the operating data includes data related to the operating status of the base station, and the historical data includes data related to the predicted load data of the base station;

[0012] The training module is used to input the historical data into a pre-trained load prediction model and output the predicted load data corresponding to the base station.

[0013] The determination module is used to determine the constraints of the base station and the response information of the base station based on the operating data and the predicted load data. The response information of the base station includes: the peak shaving capability of the base station and / or the priority of each device corresponding to the base station.

[0014] The calculation module is used to calculate the target scheduling value of the virtual power plant based on the response information of the base station, on the basis of satisfying the constraints. The target scheduling value includes one or more of the following: revenue information corresponding to the target base station, peak shaving capacity information of the target base station, and priority information of the target equipment.

[0015] The output module is used to determine resource scheduling information based on the target scheduling value. The resource scheduling information is used to assist the virtual power plant in scheduling resources for the base stations in the virtual power plant.

[0016] Thirdly, embodiments of this application provide an electronic device, including a processor and a memory electrically connected to the processor, the memory storing a computer program, and the processor being used to call and execute the computer program from the memory to implement the aforementioned base station resource scheduling method.

[0017] Fourthly, embodiments of this application provide a computer-readable storage medium for storing a computer program that can be executed by a processor to implement the aforementioned base station resource scheduling method.

[0018] Fifthly, embodiments of this application provide a chip, the chip including a processor and a communication interface, the communication interface being coupled to the processor, the processor being used to run programs or instructions to implement the aforementioned base station resource scheduling method.

[0019] The technical solution of this application embodiment acquires the operational data and historical data of each base station included in the virtual power plant. The operational data includes data related to the operating status of the base station, and the historical data includes data related to the predicted load data of the base station. The historical data is input into a pre-trained load prediction model, and the predicted load data corresponding to the base station is output. Based on the operational data and the predicted load data, the constraints and response information of the base station are determined. The response information of the base station includes: the peak-shaving capacity of the base station and / or the priority of each device corresponding to the base station. The acquired operational data and constraints are used to dynamically and quantitatively evaluate the peak-shaving capacity of the base station. Based on satisfying the constraints, the target scheduling value of the virtual power plant is calculated according to the response information of the base station. The target scheduling value includes one or more of the following: the revenue information corresponding to the target base station, the peak-shaving capacity information of the target base station, and the priority information of the target device. Based on the target scheduling value, resource scheduling information is determined. The resource scheduling information is used to assist the virtual power plant in resource scheduling of the base stations in the virtual power plant. It is evident that the calculated target scheduling value not only considers the revenue information and peak-shaving capacity of the target base station, but also the equipment priority of each base station in the virtual power plant. This allows for a refined analysis of the differences among distributed resources, providing auxiliary decision-making for virtual power plants to participate in the peak-shaving market capacity application, and can solve the problem of poor actual peak-shaving capacity of virtual power plants. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in one or more embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in one or more embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a schematic diagram of the structure of a virtual power plant according to an embodiment of this application;

[0022] Figure 2 This is a schematic flowchart of a base station resource scheduling method according to an embodiment of this application;

[0023] Figure 3 This is a schematic flowchart of a base station resource scheduling method according to another embodiment of this application;

[0024] Figure 4 This is a schematic block diagram of a base station resource scheduling device according to an embodiment of this application;

[0025] Figure 5 This is a schematic diagram of the hardware structure of a base station resource scheduling device according to an embodiment of this application. Detailed Implementation

[0026] To enable those skilled in the art to better understand the technical solutions in this application, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this application.

[0027] The base station resource scheduling method provided in this application can be executed by an electronic device or by software installed in an electronic device. Specifically, the electronic device can be a terminal device or a server device. The terminal device can include smartphones, laptops, smart wearable devices, vehicle terminals, etc., and the server device can include an independent physical server, a server cluster composed of multiple servers, or a cloud server capable of cloud computing.

[0028] Figure 1 This is a schematic diagram of a virtual power plant according to an embodiment of this application. Figure 1 As shown, the dispatchable resources in a virtual power plant include multiple base stations and multiple devices in each base station. The dispatchable resources are determined based on the peak-shaving capacity of the base station, the priority of the devices, and the revenue information corresponding to the response capacity of the base station. This fully leverages the peak-shaving potential of the controllable resources within the base station and provides decision support for the virtual power plant to participate in peak-shaving market transactions.

[0029] The base station resource scheduling method provided in this application will be described in detail below with reference to the accompanying drawings, through specific embodiments and application scenarios.

[0030] Figure 2 The diagram illustrates a schematic flowchart of a base station resource scheduling method according to an embodiment of the present invention, the method comprising the following steps:

[0031] S202, Obtain the operational data of each base station included in the virtual power plant, as well as the historical data of the base stations.

[0032] The operational data includes data related to the base station's operational status, while the historical data includes data related to the base station's predicted load data.

[0033] Data related to the base station's operational status includes, but is not limited to, the operating status of the corresponding switching power supply, the current battery charge, and the load status of the communication equipment. Specifically, switching power supply information includes: switching power supply status, battery type, core capacity, float charge voltage, equalization charge voltage, discharge voltage, charging limit voltage, discharge termination voltage, backup power capacity, initial response status, and initial response duration; day-ahead switching power supply data includes: day-ahead switching power supply response status and day-ahead response capacity; switching power supply load forecast data includes: DC load forecast power and baseline load forecast power; historical switching power supply status data includes: historical switching power supply response status and historical DC load; base station user information data includes: number of battery banks, number of switching power supplies, and adjacent response time intervals; base station user electricity price data includes: time-of-use electricity price curves; virtual power plant information includes: invitation type, calculation mode, and number of optimization periods; and peak-shaving auxiliary service clearing information includes: peak-shaving capacity instructions and peak-shaving subsidy prices.

[0034] Data related to the predicted load data of base stations specifically includes imported raw data from the base stations, such as historical electricity consumption, historical weather data, and future forecast weather data. The load of base stations can be predicted based on historical data, including predicted load power and baseline load.

[0035] It should be noted that the virtual power plant comprises multiple base stations in different regions, each corresponding to its own account number. Based on highly intelligent communication technology, operational data is transmitted to the virtual power plant's central management system on a regular basis, ensuring data timeliness and accuracy. By establishing a comprehensive data foundation for periodically acquiring the operational status of base stations, subsequent peak-shaving capacity analysis and decision-making can be conducted, providing a data basis for dynamically and quantitatively assessing the peak-shaving capacity of base stations. This avoids the problem of traditional technologies that typically estimate the response capacity of virtual power plants participating in the peak-shaving ancillary services market based on long-term statistical data and models, neglecting the dynamic demand response of actual base stations.

[0036] S204: Input historical data into the pre-trained load prediction model and output the predicted load data corresponding to the base station.

[0037] By inputting historical data into a pre-trained load prediction model, predicted load data is obtained. Based on the predicted load data, constraints, the peak-shaving capacity of base stations, and the priority of equipment can be calculated. Furthermore, it can more sensitively capture changes and fluctuations in base station load on the invitation day, thereby more accurately reflecting the actual load level and helping to reduce the disturbance of uncertainty to ancillary services.

[0038] S206. Based on operational data and predicted load data, determine the constraints and response information of the base station.

[0039] The response information of the base station includes: the peak shaving capability of the base station and / or the priority of each device corresponding to the base station.

[0040] The constraints include: power consumption constraints of the base station, response capacity constraints of the base station, continuous response constraints of the switching power supply, maximum number of response times of the switching power supply, battery capacity constraints, valid recognition constraints of virtual power plant response, and control constraints of switching power supply commands.

[0041] The peak-shaving capability of a base station includes: the base station's ability to adjust its load fluctuations and changes to ensure that power demand is met and the base station's stability, etc. Specifically, it can include response capacity information, etc. The stronger the peak-shaving capability of a base station, the stronger its ability to resist sudden events, and the better it can meet power demand and improve power quality.

[0042] The priorities of each device corresponding to the base station include: the response priority of each device determined based on the device type, load level, and response time.

[0043] S208, based on the constraints, calculate the target scheduling value of the virtual power plant according to the response information of the base station.

[0044] The target scheduling value includes one or more of the following: revenue information corresponding to the target base station, peak shaving capacity information of the target base station, and priority information of the target equipment.

[0045] Specifically, the target scheduling value is an objective function that comprehensively considers the revenue and priority information of the target base station. It can be combined with ancillary service market rules to maximize peak-shaving subsidies while taking into account the resource characteristics differences among a large number of base stations. Based on the calculation of the target scheduling value, base stations with strong peak-shaving capabilities and high equipment response priority are prioritized for scheduling.

[0046] Based on the constraints, it is shown that the base station has the ability to respond. Furthermore, by combining the priority information of each device corresponding to the base station and balancing the peak-shaving capability of the base station and the priority of the devices, the target scheduling value of the virtual power plant is calculated. When the target scheduling value is the maximum, the target base station corresponding to the target scheduling value is determined.

[0047] S210, determine resource scheduling information based on the target scheduling value.

[0048] Resource scheduling information is used to assist the virtual power plant in scheduling resources for base stations within the virtual power plant.

[0049] Based on the target scheduling value, the target base station information is determined. This information includes, for example, the target base station's capacity response status, corresponding equipment information, and response status. Resource scheduling information is then aggregated and summarized. Specifically, resource scheduling information includes: optimization result data, such as optimization success rate, optimization calculation time, total operating cost, and optimization result messages; a power supply aggregation result table, such as power supply response status and peak-shaving capacity; a base station account aggregation result table, such as account response status, account power consumption plan, and account peak-shaving capacity; and a virtual power plant aggregation result table, such as virtual power plant peak-shaving capacity and virtual power plant power consumption plan.

[0050] Based on the target scheduling value, information from each level of equipment, base station, and virtual power plant can be integrated to ultimately determine resource scheduling information. Based on the resource scheduling information, the virtual power plant performs resource scheduling on the base stations within the virtual power plant.

[0051] The technical solution of this application embodiment acquires the operational data and historical data of each base station included in the virtual power plant. The operational data includes data related to the operating status of the base station, and the historical data includes data related to the predicted load data of the base station. The historical data is input into a pre-trained load prediction model, and the predicted load data corresponding to the base station is output. Based on the operational data and the predicted load data, the constraints and response information of the base station are determined. The response information of the base station includes: the peak-shaving capacity of the base station and / or the priority of each device corresponding to the base station. The acquired operational data and constraints are used to dynamically and quantitatively evaluate the peak-shaving capacity of the base station. Based on satisfying the constraints, the target scheduling value of the virtual power plant is calculated according to the response information of the base station. The target scheduling value includes one or more of the following: the revenue information corresponding to the target base station, the peak-shaving capacity information of the target base station, and the priority information of the target device. Based on the target scheduling value, resource scheduling information is determined. The resource scheduling information is used to assist the virtual power plant in resource scheduling of the base stations in the virtual power plant. It is evident that the calculated target scheduling value not only considers the revenue information and peak-shaving capacity of the target base station, but also the equipment priority of each base station in the virtual power plant. This allows for a refined analysis of the differences among distributed resources, providing auxiliary decision-making for virtual power plants to participate in the peak-shaving market capacity application, and can solve the problem of poor actual peak-shaving capacity of virtual power plants.

[0052] In one embodiment, calculating the target scheduling value of the virtual power plant (i.e., S208) can be performed by the following steps A1-A3:

[0053] Step A1: Within a preset time period, determine the revenue information corresponding to the target base station based on the response capacity in the peak shaving capability of the base station.

[0054] The preset time period represents the time interval for scheduling. In real-time scheduling, it is generally 15 minutes. Within the scheduling time period, the response information of each battery in the base station is obtained based on the response capacity included in the peak-shaving capacity of the base station, thereby determining the revenue information of the target base station. The battery types include lead-acid batteries, nickel-cadmium batteries, nickel-metal hydride batteries, and lithium-ion batteries, etc.

[0055] Step A2: Within a preset time period, determine the priority information corresponding to the target device based on the response status of the devices in the base station.

[0056] During the preset time period in step A1, the device's response status may include: the charging status of the device's power supply during the preset time period: a 0 / 1 variable, where 1 indicates charging and 0 indicates not participating, as well as the device type and the device's load capacity, and determining the base station's priority information.

[0057] Step A3: When the sum of the revenue information and priority information is greater than the preset threshold, the sum of the revenue information and priority information is determined as the target scheduling value.

[0058] The target scheduling value is an objective function, as follows:

[0059]

[0060] Among them, the constants include: T is the time window for scheduling, which is 24 hours in day-ahead scheduling; This indicates the scheduling time interval, which is 15 minutes in real-time scheduling. The demand response subsidy price for time period t is expressed in yuan / kWh. The price for battery charging during time period t is expressed in yuan / kWh. This indicates the device response priority, set based on device type, peak-shaving capacity, and response time. Variables include: This represents the response capacity for user number s during time period t, in kW. Indicates switching power supply During the period Charging status: 0 / 1 variable, where 1 indicates charging and 0 indicates not participating.

[0061] The preset threshold includes a value that is one less than the maximum value, which is the maximum value of the sum of the base station's revenue and priority. This value is used as the target scheduling value. Based on the target scheduling value, the relevant information of the base station is determined. The target scheduling value can correspond to one or more base stations.

[0062] It should be noted that when calculating the target scheduling value, if the total revenue of two base stations is greater than that of a single base station, then two complementary base stations can be automatically selected to ensure maximum total revenue. As an example, when three base stations can be used to calculate the target scheduling value, two of which use lithium-ion batteries and one uses lead-acid batteries, it should be understood that lead-acid batteries have a larger capacity, such as providing a 100kW response capability, while the two lithium-ion batteries can provide a 50kW response capability. In this case, the response capability provided by the two lithium-ion batteries is equivalent to that of the one lead-acid battery. However, the lithium-ion battery type has better performance and a longer response time; therefore, combining two lithium-ion batteries is superior to using a single lead-acid battery.

[0063] In this embodiment, by analyzing the economic aspects and different base station combinations, specifically based on revenue information and base station priority information, the target scheduling value that maximizes the combination of revenue and priority is determined. The resource differences among base stations can be identified through the priority in the objective function, and the total peak-shaving capacity can be calculated from a massive number of base stations. Finally, based on this target scheduling value, the corresponding target base station is determined. This not only identifies the optimal target base station but also achieves the effect of resource complementarity among the base stations in the virtual power plant.

[0064] In one embodiment, the response information of the base station is determined based on the operating data and the predicted load data (i.e., S206), and the following steps B1-B5 can be performed:

[0065] Step B1: Traverse all base stations and the corresponding devices of each base station.

[0066] Obtain relevant information on the operation data and predicted load data of all base stations and corresponding equipment in the virtual power plant.

[0067] Step B2: Based on operational data and predicted load data, obtain the response capacity information corresponding to the base station, and determine the peak shaving capacity of the base station according to the response capacity information of the base station.

[0068] Based on the base station's corresponding operational data and predicted load data, the base station's power consumption, response capacity, and load fluctuation changes are calculated to determine the base station's load level, i.e., response capacity information. Based on the response capacity information, the base station's peak-shaving capability is determined.

[0069] Step B3: Based on the operational data and predicted load data, obtain the equipment type, load level, and response time of each device.

[0070] The equipment types include different types such as lithium-ion batteries and nickel-cadmium batteries; and based on operating data and predicted load data, the load level of the equipment, i.e., the response capacity and response time of each piece of equipment, is calculated, thereby determining the load level and response time of the equipment.

[0071] Step B4: Determine the priority of the equipment based on the equipment type, the equipment load level, and the equipment response time.

[0072] Based on the device type, device load level, and device response time, it is possible to traverse all base stations and devices and determine the response priority of the devices corresponding to the base station. It should be understood that the longer the response time of a device and the stronger its load capacity, the higher the response priority of the corresponding device.

[0073] Step B5: Determine the base station's response information based on the base station's peak shaving capability and / or equipment priority.

[0074] In this embodiment, by traversing the operating data and predicted load data of all base stations and devices, the response information corresponding to the base station is determined. This response information includes the peak-shaving capability of the base station and the priority of the device response determined according to the device type, the device load level and the device response time. This allows for in-depth analysis of the resource characteristics of different base stations and can take into account the actual equipment situation of the base station before making a quantitative evaluation.

[0075] In one embodiment, based on operational data and predicted load data, the constraints of the base station are determined, including response capacity constraints, which can be achieved by performing the following steps C1-C4:

[0076] Step C1: Based on the predicted load data, obtain the predicted power consumption and baseline load power.

[0077] The baseline load in baseline load power refers to a load curve estimated based on the user's historical load data. As an example, when the response date is a weekday, the average load of the five normal working days preceding the invitation date is selected as the peak-shaving baseline; when the response date is a Saturday (Sunday), the average load of the three Saturdays (Sundays) preceding the invitation date is selected as the peak-shaving baseline. The purpose of calculating the baseline is to clarify the electricity consumption when the virtual power plant does not participate in peak shaving, and to accurately assess the peak-shaving capacity that the base station can provide based on this.

[0078] Step C2: Determine the power consumption of the base station based on the predicted power consumption and the power consumption of each battery corresponding to the equipment in the operating data.

[0079] The power consumption of a base station is calculated by taking into account not only its predicted power consumption but also the float charging power provided by the batteries connected to the equipment. The predicted power consumption plus the power consumption of each battery is then used as the base station's power consumption. The specific calculation formula is as follows:

[0080]

[0081] Among them, constants: This represents the predicted power consumption of base station user number s during time period t, in kW; This represents the float charging power of battery i, in kW. Variables: This represents the power consumption of base station user s during time period t, in kW; The charging state of the switching power supply i during time period t is represented by a 0 / 1 variable, where 1 indicates charging and 0 indicates not participating.

[0082] Therefore, the power consumption constraints of a base station include: determining the actual power consumption of each base station based on the sum of the predicted power consumption and the float charging power provided by the battery in the equipment corresponding to the base station.

[0083] Step C3: When the power consumption of the base station is greater than the baseline load power, the response capacity is the difference between the power consumption of the base station and the baseline load power. The difference is used to characterize whether the base station meets the constraint conditions.

[0084] When the power consumption of a base station increases above the baseline load, that is, the difference between the power consumption of the base station calculated in step C2 and the baseline load power in step C1 is greater than 0, then the response of the corresponding base station is valid. Its response capacity is the difference between the power consumption and the baseline load power. Then it is determined that the base station meets the constraint conditions. The base station that meets the constraint conditions can be used as the screening base station for calculating the target scheduling value.

[0085] The specific calculation formula is as follows:

[0086]

[0087] The constants include: This represents the baseline load of base station with user number S during time period t, in kW; M represents a sufficiently large positive number, usually taken as... The variables include: The charging state of device i with switching power supply during time period t is represented by a 0 / 1 variable, where 1 represents charging and 0 represents not participating. This represents the valid response status of base station S during time period t: a 0 / 1 variable, where 1 indicates valid and 0 indicates invalid; This represents the response capacity of base station S during time period t, expressed in kW. This represents the power consumption of base station S during time period t, expressed in kW.

[0088] Step C4: When the power consumption of the base station is less than the baseline load power, the response capacity is invalid, and the base station does not meet the constraint conditions.

[0089] When the power consumption of a base station is less than the baseline load, the base station's response is invalid, its response capacity is 0, and the base station does not meet the constraints, so it cannot be used as a screening base station for calculating the target scheduling value.

[0090] It should be noted that, in addition to the constraints mentioned above, there are also constraints on the continuous response of the switching power supply, the maximum number of responses of the switching power supply, the battery capacity, the valid recognition of the virtual power plant response, and the switching power supply command control. The specific calculation methods are shown below:

[0091] (1) Continuous Response Constraint of Switching Power Supply: Once the switching power supply participates in the response, it must continuously respond for the specified response time before stopping. The response status of the switching power supply corresponds to the response status of the device. The calculation formula is shown below:

[0092]

[0093] constant: This represents the response time of switching power supply i, in time intervals (15 minutes). Variables: The charging state of switching power supply i during time period t is represented by a 0 / 1 variable, where 1 indicates charging and 0 indicates not participating. In other words, under this time duration, if every device is in a charging state, the corresponding base station satisfies the constraint condition.

[0094] (2) Maximum number of responses from the switching power supply: The maximum number of responses from the switching power supply in a single demand request is 1. The calculation formula is shown below:

[0095]

[0096] variable: The charging state of the switching power supply i during time period t is represented by a 0 / 1 variable, where 1 indicates charging and 0 indicates not participating. The variable represents the initial response state of the switching power supply i during time period t: a 0 / 1 variable, where 1 indicates the start of response and 0 indicates no start of response; The variable represents the stop response state of the switching power supply i during time period t: a 0 / 1 variable, where 1 indicates a stop response and 0 indicates no stop response.

[0097] (3) Battery capacity constraint: The battery capacity at any given moment equals its current capacity plus the capacity replenished by demand response charging. The calculation formula is shown below:

[0098]

[0099] The constants include: This represents the float charging power of battery i, in kW; This indicates the minimum allowable capacity of battery i, in kWh; This represents the maximum allowable capacity of battery i, in kWh. Variables include: This represents the charge of battery i during time period t, in kWh. The charging state of the switching power supply i during time period t is represented by a 0 / 1 variable, where 1 indicates charging and 0 indicates not participating.

[0100] (4) Constraints for Valid Virtual Power Plant Response: The criteria for valid virtual power plant response are: minimum load is greater than baseline minimum load; and cumulative load is greater than baseline cumulative load. The calculation formula is shown below:

[0101]

[0102] The constants include: This represents the baseline load of base station s during time period t, in kW; This represents the minimum baseline load of the virtual power plant invitation window, in kW; M represents a sufficiently large positive number, usually taken as... The variables include: The virtual power plant response status is represented by a 0 / 1 variable, where 1 indicates valid and 0 indicates invalid. This indicates the minimum power consumption of the virtual power plant invitation window, in kW. This represents the power consumption of base station S during time period t, expressed in kW.

[0103] (5) Switching power supply command control constraints:

[0104] ① Before the invitation period, the switching power supply control command is set to no charging and no discharging voltage.

[0105]

[0106] The constants include: This represents the voltage value of the switching power supply i when it is neither charging nor discharging, in volts (V). Variables include: This indicates the control power command for switching power supply i during time period t, in volts (V).

[0107] ② During the invitation period, the switching power supply sets control commands based on its response status:

[0108] i. The switching power supply maintains a non-charging / non-discharging voltage before participating in the response.

[0109]

[0110] The constants include: This represents the voltage value of the switching power supply i when it is neither charging nor discharging, in volts (V). Variables include: This indicates the control power command for switching power supply i during time period t, in volts (V).

[0111] ii. During the response period, and while the device is in a charging state, i.e. The switching power supply control command is set to the float charge voltage:

[0112]

[0113] The constants include: This represents the float charge voltage value of the switching power supply i, in volts (V); variables include: This indicates the control power command for switching power supply i during time period t, in volts (V).

[0114] iii. Once the battery participates in the response, set the float charge voltage:

[0115]

[0116] The constants include: This represents the float charge voltage value of the switching power supply i, in volts (V); variables include: This indicates the control power command for switching power supply i during time period t, in volts (V).

[0117] ③ After the invitation period, the switching power supply issues a recovery command to set the float charge voltage:

[0118]

[0119] The constants include: This represents the float charge voltage value of the switching power supply i, in volts (V); variables include: This indicates the control power command for switching power supply i during time period t, in volts (V).

[0120] In this embodiment, various constraints are used to determine whether the base station's response capability meets the constraints. The response capability of the base station is dynamically and quantitatively evaluated. This quantitative evaluation considers the actual equipment conditions of the base station, such as the status of the power supply, the charging power of the battery, and its response capacity. Only when the constraints are met can the base station be used to calculate the objective function. This is equivalent to screening the base stations in the virtual power plant. The screened base stations can be used to calculate the target scheduling value, improving the calculation efficiency of the target scheduling value and more quickly determining the base stations to participate in the virtual power plant's peak shaving.

[0121] In one embodiment, the training of the load forecasting model can be performed using the following steps D1-D3:

[0122] Step D1: Obtain historical sample data and sample load data for each base station.

[0123] Historical data of the sample includes factors such as the historical load of the same base station at various times the day before the prediction period, such as the load at different times on different dates, and the impact of meteorological factors such as temperature, precipitation, and wind speed on the electricity load.

[0124] Sample load data, including load data of the corresponding base stations determined based on historical sample data.

[0125] Step D2: Perform data preprocessing on the historical sample data to obtain the historical sample data feature vector, and input the historical sample data feature vector into the load prediction model.

[0126] Data preprocessing includes cleaning the imported base station raw dataset, selecting prediction features strongly correlated with the load based on correlation analysis, and processing the data for duplicates, outliers, and missing values ​​in sequence.

[0127] Based on the preprocessing of the data, the preprocessed historical data feature vectors are obtained and used as the input feature vectors for the load forecasting model, thus forming the overall training dataset. Similarly, the sample load data feature vectors of the forecast points at different dates and times are constructed to form the overall test dataset.

[0128] In addition, since the data from different base stations need to be separated for testing, the overall dataset is divided into multiple smaller datasets according to the base station's serial number (Identification, ID). Multi-threading technology is used to evenly divide the number of all base station IDs into multiple datasets.

[0129] Step D3: Train the load prediction model based on the predicted load data and sample load data.

[0130] The load forecasting model is specifically structured as XGBoost. Parameters for the XGBoost regression model are set, including the learning rate, the number of trees, and the tree depth. Based on XGBoost, the model learns by calculating the feature vector of the input historical sample data. The input historical sample data feature vector of the load forecasting model includes: the time features of the forecast point and the historical value features of the same time on day D-1. The output is the predicted load value, i.e., the predicted load data. Training the load forecasting model involves adjusting the parameters in XGBoost to ensure that the predicted load data matches the sample load data; once trained, the model is successfully trained.

[0131] In addition, the prediction results obtained from different threads are summarized and evaluated. The average error of a single objective variable is evaluated, and finally a short-term prediction algorithm is formed to predict the load of the base station.

[0132] In this embodiment, sample historical data and sample load data of each base station are obtained. Based on the load prediction model of the XGBoost structure, the sample historical data is trained, and the predicted load data is finally output. Furthermore, a multi-threaded training method can be constructed for base stations with different IDs to improve the training speed.

[0133] In one embodiment, preprocessing the historical sample data to obtain the historical sample data feature vector (i.e., step D2) can be performed by executing the following steps E1-E3:

[0134] Step E1: Identify outliers in the historical sample data based on the historical sample data of each base station.

[0135] The outliers in the historical sample data include the following situations: the load data at a certain point is much greater or less than the previous and next data points; the load data of the same site on a certain day is much greater or less than the load data on other days; the load data collection problem of the base station may cause the load data of a certain day to be all invalid NULL, or only a few non-NULL values.

[0136] The above-mentioned cases are filtered in the historical sample dataset, and the index of the locked outliers is output.

[0137] Step E2: Obtain outliers from the historical data of the sample.

[0138] Step E3: Correct outliers using the interquartile range calculation method to obtain the feature vector of the sample's historical data.

[0139] After outlier identification using the interquartile range (IQR) calculation method, outliers in the sample historical data are obtained. The sample load data is divided according to the base station ID, and missing values ​​are removed from the sample historical dataset corresponding to each ID. The processed outliers are then used to calculate the IQR.

[0140] Specifically, defining outliers is based on defining the range above and below the normal range of the dataset, i.e., upper and lower bounds. (Consider an upper bound of 1.5 * IQR value).

[0141] upper = Q3 + 1.5 * IQR

[0142] lower = Q1 – 1.5 * IQR

[0143] IQR, also known as interquartile range, is a method in descriptive statistics used to determine the difference between the third quartile and the first quartile. Like variance and standard deviation, it indicates the dispersion of variables in statistical data. However, IQR is more of a robust statistic, serving as an index (index_list) to record outliers.

[0144] The method for correcting the index_list of outliers obtained in the above situations is as follows:

[0145] (1) Linear interpolation: If the continuous abnormal / missing data does not exceed 1 hour, linear interpolation is used to supplement it.

[0146] (2) Similar day interpolation: If the consecutive abnormal / missing data does not exceed 2 hours, similar day interpolation is used to supplement it.

[0147] (3) Deletion: If there are more than 2 hours of continuous abnormal / missing data, there is no interpolation significance; delete the load data of that day from the sample training set.

[0148] (4) Based on the daily load values ​​of the base stations, if only a small number of them are NULL, they may be erroneous values. Fill the erroneous values ​​with 0 values. If all the load values ​​of the base stations on a certain day are NULL, then the data for that day has no interpolation meaning. Delete the load data for that day from the sample set.

[0149] In addition, due to measurement or transmission issues, some historical data may show a situation where most of the record points are 0 throughout the day. Such samples will negatively affect the prediction performance when trained together with normal samples. Therefore, before training the model, the total load for each day is summed, and a threshold filtering method is used. The threshold is set to 96 * 0.05 to filter out those samples where most of the record points are 0, thus filtering the historical data of the samples.

[0150] In this embodiment, by acquiring outliers in the historical sample data and performing filtering, removal, interpolation, and other processing on the outliers, the historical sample data is made relatively complete. This makes the feature vector of the historical sample data input into the predictive load model more comprehensive, and the output test results more accurate.

[0151] In one embodiment, after preprocessing the historical sample data to obtain the historical sample data feature vector (i.e., step D2), the following steps F1-F3 can also be performed:

[0152] Step F1: Calculate the quotient between the historical data and sample loading data of the sample using the Pearson correlation coefficient.

[0153] Specifically, the Pearson correlation coefficient measures the strength and direction of the linear relationship between two continuous variables. It provides an indicator of the strength of this linear relationship, ranging from -1 to 1, where -1 represents a perfect negative correlation, 0 represents no linear correlation, and 1 represents a perfect positive correlation (an increase in one variable also increases the other). Based on this coefficient, one can determine the strength or weakness of the correlation between variables, offering interpretability and intuitiveness. For large datasets, calculating the correlation coefficient is relatively computationally intensive, allowing for quick results.

[0154] The Pearson correlation coefficient method is used to identify the dominant factors in load forecasting. The Pearson correlation coefficient between two variables X and Y is defined as the quotient of the covariance and standard deviation between the two variables.

[0155]

[0156] In the above formula: X represents the feature vector of the historical sample data corresponding to the historical sample data; Y represents the feature vector of the load sample data corresponding to the load sample data. Covariance of two data columns; These are the standard deviations of the feature vectors of the historical sample data and the feature vectors of the sample loading data, respectively. , For the sample The Pearson formula, which calculates the mean and sample standard deviation, involves summing the squares of the deviations of the variables and summing the related products.

[0157] Step F2: Determine the correlation value between the historical data and the sample loading data based on the quotient.

[0158] The degree of linear correlation between the historical data and sample loading data of the two variables is determined by the statistic of F1 in the above step.

[0159] Step F3: Determine the historical data of samples with correlation values ​​greater than a preset threshold as the historical data feature vector of the samples.

[0160] Based on the degree of correlation, historical sample data that are highly correlated with the sample load data are identified as the input sample historical data feature vectors for the input predictive load data model.

[0161] It should be noted that the input data for the predictive load data model, in addition to considering correlation calculations, also needs to consider other features included in the historical data of the sample, such as the following features:

[0162] (1) Date characteristics

[0163] The time of the data point: This feature is mainly used to distinguish different times of the same day. The load at different times shows either continuity or distinction. That is, at some consecutive times of the same day, the load is the same or similar, but at other times, the load shows obvious switching, or suddenly increases or suddenly decreases. Therefore, the time feature is an important factor for load prediction.

[0164] As an example, consider the day of the week for the current date, and the categorical features of different days in the week; whether the current day is a workday or not, discretize the data during processing, and set the feature value to 1 for workdays and holidays, and set the feature value to 0 for non-workdays (weekends) and non-holidays.

[0165] (2) Historical value characteristics

[0166] Observation of the data reveals that the load fluctuations of each base station are relatively small across different days, with some even exhibiting the same load for several consecutive days. Even when variations occur, they are minor. Therefore, the load over the past day, three days, and five days can also be considered as a characteristic feature. For example, the historical value of the previous day at the same time, the average load over the past three days at the same time, and the average load over the past five days at the same time.

[0167] (3) Meteorological characteristics (optional)

[0168] Studies on the impact of meteorological factors on power grid load characteristics show that meteorological factors such as temperature, precipitation, and wind speed also affect electricity load. Specifically, temperature is the meteorological factor with the greatest impact on load; as people use air conditioning more, the influence of temperature on load is increasing. Meanwhile, precipitation and wind speed also have some influence on electricity load. Similarly, these meteorological data may also affect the prediction of base station electricity load; therefore, these meteorological characteristics, such as temperature, precipitation, and wind speed, can also be considered as part of the characteristic factors.

[0169] In this embodiment, by screening historical sample data and calculating the correlation between historical sample data and sample load data, it is possible to select historical sample characteristic data with strong correlation and large influencing factors from a large amount of historical sample data and input them into the load prediction model, which can improve the efficiency and accuracy of load prediction and thus more accurately reflect the load level.

[0170] Figure 3 This is a schematic flowchart of a base station resource scheduling method according to another embodiment of this application, such as... Figure 3 As shown, the method includes the following steps:

[0171] S301, obtain the operational and historical data of all base stations in the virtual power plant.

[0172] Historical data includes data for base stations prior to the invitation date, which is used to predict load data.

[0173] S302, input all historical data of base stations into the load prediction model.

[0174] The load forecasting model preprocesses historical data and selects highly correlated historical data feature vectors through Pearson correlation analysis, which are then input into the load forecasting model.

[0175] S303, the load forecasting model outputs the predicted load data for the corresponding base station.

[0176] S304, based on operational data and predicted load data, determines the peak-shaving capacity of each base station in the virtual power plant.

[0177] S305, based on operational data and predicted load data, determines the type, load level, and response time of each device in the base station, and then determines the priority of the devices.

[0178] S306. Based on operational data and predicted load data, determine the constraints for the base station.

[0179] Specifically, base stations are dynamically and quantitatively evaluated based on operational data and predicted load data.

[0180] S307, based on the constraints, determines the revenue information corresponding to the base station by using the base station's peak-shaving capability and the equipment's priority information.

[0181] S308 calculates the maximum value of the target scheduling value based on revenue information, the peak-shaving capacity of the base station, and the priority of the equipment.

[0182] S309: Based on the maximum value of the target scheduling value, determine the corresponding base station and resource scheduling information.

[0183] S310 uses resource scheduling information to obtain relevant information about the schedulable resources of the virtual power plant.

[0184] The specific processes from S301 to S310 have been described in detail in the above embodiments and will not be repeated here.

[0185] The technical solution of this application embodiment acquires the operational data and historical data of each base station included in the virtual power plant. The operational data includes data related to the operating status of the base station, and the historical data includes data related to the predicted load data of the base station. The historical data is input into a pre-trained load prediction model, and the predicted load data corresponding to the base station is output. Based on the operational data and the predicted load data, the constraints and response information of the base station are determined. The response information of the base station includes: the peak-shaving capacity of the base station and / or the priority of each device corresponding to the base station. The acquired operational data and constraints are used to dynamically and quantitatively evaluate the peak-shaving capacity of the base station. Based on satisfying the constraints, the target scheduling value of the virtual power plant is calculated according to the response information of the base station. The target scheduling value includes one or more of the following: the revenue information corresponding to the target base station, the peak-shaving capacity information of the target base station, and the priority information of the target device. Based on the target scheduling value, resource scheduling information is determined. The resource scheduling information is used to assist the virtual power plant in resource scheduling of the base stations in the virtual power plant. It is evident that the calculated target scheduling value not only considers the revenue information and peak-shaving capacity of the target base station, but also the equipment priority of each base station in the virtual power plant. This allows for a refined analysis of the differences among distributed resources, providing auxiliary decision-making for virtual power plants to participate in the peak-shaving market capacity application, and can solve the problem of poor actual peak-shaving capacity of virtual power plants.

[0186] In summary, specific embodiments of this subject matter have been described. Other embodiments are within the scope of the appended claims. In some cases, the actions recited in the claims can be performed in a different order and still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing can be advantageous.

[0187] The above is a base station resource scheduling method provided by the embodiments of this application. Based on the same idea, the embodiments of this application also provide a base station resource scheduling device.

[0188] Figure 4 This is a schematic diagram of a base station resource scheduling device according to an embodiment of the present invention. Figure 4 As shown, the base station resource scheduling device includes: an acquisition module 41, a training module 42, a determination module 43, a calculation module 44, and an output module 45.

[0189] The acquisition module 41 is used to acquire the operating data of each base station included in the virtual power plant, as well as the historical data of the base station. The operating data includes data related to the operating status of the base station, and the historical data includes data related to the predicted load data of the base station.

[0190] Training module 42 is used to input historical data into the pre-trained load prediction model and output the predicted load data corresponding to the base station.

[0191] The determination module 43 is used to determine the constraints and response information of the base station based on the operating data and predicted load data. The response information of the base station includes: the peak shaving capability of the base station and / or the priority of each device corresponding to the base station.

[0192] The calculation module 44 is used to calculate the target scheduling value of the virtual power plant based on the response information of the base station, on the basis of satisfying the constraints. The target scheduling value includes one or more of the following: revenue information corresponding to the target base station, peak shaving capacity information of the target base station, and priority information of the target equipment.

[0193] The output module 45 is used to determine resource scheduling information based on the target scheduling value. The resource scheduling information is used to assist the virtual power plant in scheduling resources for the base stations in the virtual power plant.

[0194] In one embodiment, the computing module 44 includes:

[0195] The first determining unit is used to determine the revenue information corresponding to the target base station based on the response capacity in the peak shaving capability of the base station within a preset time period.

[0196] The second determining unit is used to determine the priority information corresponding to the target device based on the response status of the devices in the base station within a preset time period.

[0197] The third determining unit is used to determine the sum of the revenue information and priority information as the target scheduling value when the sum of the revenue information and priority information is greater than a preset threshold.

[0198] In one embodiment, determining module 43 includes:

[0199] The traversal unit is used to traverse all base stations and the corresponding devices of each base station;

[0200] The acquisition unit is used to acquire the response capacity information corresponding to the base station based on the operating data and predicted load data, determine the peak shaving capacity of the base station based on the response capacity information of the base station, and acquire the equipment type, load level and response time of each device based on the operating data and predicted load data.

[0201] Determine equipment priority based on equipment type, equipment load level, and equipment response time;

[0202] The response information of the base station is determined based on the base station's peak shaving capability and / or the priority of the equipment.

[0203] In one embodiment, the determining module 43 further includes: a response capacity constraint module.

[0204] Based on the predicted load data, obtain the predicted power consumption and baseline load power;

[0205] The power consumption of the base station is determined based on the predicted power consumption and the power consumption of each battery corresponding to the equipment in the operating data.

[0206] When the power consumption of the base station is greater than the baseline load power, the response capacity is the difference between the power consumption of the base station and the baseline load power. The difference is used to characterize whether the base station meets the constraint conditions.

[0207] When the power consumption of the base station is less than the baseline load power, the response capacity is invalid, and the base station does not meet the constraint conditions.

[0208] In one embodiment, training module 42 includes:

[0209] Obtain historical sample data and sample load data for each base station;

[0210] The historical data of the sample is preprocessed to obtain the feature vector of the historical data of the sample, and the feature vector of the historical data of the sample is input into the load prediction model;

[0211] The load prediction model is trained based on predicted load data and sample load data.

[0212] In one embodiment, data preprocessing is performed on the historical sample data to obtain a feature vector of the historical sample data, including:

[0213] Based on the historical sample data of each base station, outliers in the historical sample data are identified.

[0214] Identify outliers in historical sample data;

[0215] Outliers are corrected by calculating the interquartile range, resulting in a feature vector of the sample's historical data.

[0216] In one embodiment, preprocessing historical sample data to obtain a feature vector of historical sample data further includes:

[0217] The historical data and sample loading data of the sample are used to calculate the quotient between the two variables using the Pearson correlation coefficient;

[0218] Based on the quotient, determine the correlation value between the historical data of the sample and the sample loading data;

[0219] Historical data of samples with correlation values ​​greater than a preset threshold are identified as feature vectors of historical data.

[0220] The technical solution of this application embodiment acquires the operational data and historical data of each base station included in the virtual power plant. The operational data includes data related to the operating status of the base station, and the historical data includes data related to the predicted load data of the base station. The historical data is input into a pre-trained load prediction model, and the predicted load data corresponding to the base station is output. Based on the operational data and the predicted load data, the constraints and response information of the base station are determined. The response information of the base station includes: the peak-shaving capacity of the base station and / or the priority of each device corresponding to the base station. The acquired operational data and constraints are used to dynamically and quantitatively evaluate the peak-shaving capacity of the base station. Based on satisfying the constraints, the target scheduling value of the virtual power plant is calculated according to the response information of the base station. The target scheduling value includes one or more of the following: the revenue information corresponding to the target base station, the peak-shaving capacity information of the target base station, and the priority information of the target device. Based on the target scheduling value, resource scheduling information is determined. The resource scheduling information is used to assist the virtual power plant in resource scheduling of the base stations in the virtual power plant. It is evident that the calculated target scheduling value not only considers the revenue information and peak-shaving capacity of the target base station, but also the equipment priority of each base station in the virtual power plant. This allows for a refined analysis of the differences among distributed resources, providing auxiliary decision-making for virtual power plants to participate in the peak-shaving market capacity application, and can solve the problem of poor actual peak-shaving capacity of virtual power plants.

[0221] Those skilled in the art will understand that Figure 4 The base station resource scheduling device in the document can be used to implement the base station resource scheduling method described above. The detailed description should be similar to that in the method section above. To avoid being too complicated, it will not be repeated here.

[0222] Based on the same technical concept, embodiments of this application also provide an electronic device for executing the aforementioned base station resource scheduling method. Figure 5 This is a schematic diagram of the structure of an electronic device to implement various embodiments of this application. The electronic device can vary significantly due to differences in configuration or performance, and may include a processor 510, a communications interface 520, a memory 530, and a communication bus 540. The processor 510, communications interface 520, and memory 530 communicate with each other via the communication bus 540. The processor 510 can call a computer program stored in the memory 530 and executable on the processor 510 to perform the following steps:

[0223] Obtain operational data for each base station included in the virtual power plant, as well as historical data for the base stations. Operational data includes data related to the operational status of the base stations, and historical data includes data related to the predicted load data of the base stations.

[0224] Historical data is input into a pre-trained load prediction model, which outputs the predicted load data for the base station.

[0225] Based on operational data and predicted load data, determine the constraints and response information of the base station. The response information of the base station includes: the peak shaving capacity of the base station and / or the priority of each device corresponding to the base station.

[0226] Based on the constraints, the target scheduling value of the virtual power plant is calculated according to the response information of the base station. The target scheduling value includes one or more of the following: revenue information corresponding to the target base station, peak shaving capacity information of the target base station, and priority information of the target equipment.

[0227] Based on the target scheduling value, resource scheduling information is determined. This resource scheduling information is used to assist the virtual power plant in scheduling resources for base stations within the virtual power plant.

[0228] The technical solution of this application embodiment acquires the operational data and historical data of each base station included in the virtual power plant. The operational data includes data related to the operating status of the base station, and the historical data includes data related to the predicted load data of the base station. The historical data is input into a pre-trained load prediction model, and the predicted load data corresponding to the base station is output. Based on the operational data and the predicted load data, the constraints and response information of the base station are determined. The response information of the base station includes: the peak-shaving capacity of the base station and / or the priority of each device corresponding to the base station. The acquired operational data and constraints are used to dynamically and quantitatively evaluate the peak-shaving capacity of the base station. Based on satisfying the constraints, the target scheduling value of the virtual power plant is calculated according to the response information of the base station. The target scheduling value includes one or more of the following: the revenue information corresponding to the target base station, the peak-shaving capacity information of the target base station, and the priority information of the target device. Based on the target scheduling value, resource scheduling information is determined. The resource scheduling information is used to assist the virtual power plant in resource scheduling of the base stations in the virtual power plant. It is evident that the calculated target scheduling value not only considers the revenue information and peak-shaving capacity of the target base station, but also the equipment priority of each base station in the virtual power plant. This allows for a refined analysis of the differences among distributed resources, providing auxiliary decision-making for virtual power plants to participate in the peak-shaving market capacity application, and can solve the problem of poor actual peak-shaving capacity of virtual power plants.

[0229] The specific execution steps can be found in the various steps of the above-described base station resource scheduling method embodiment, and can achieve the same technical effect. To avoid repetition, they will not be repeated here.

[0230] It should be noted that the electronic devices in the embodiments of this application include: servers, terminals, or other devices besides terminals.

[0231] The above electronic device structure does not constitute a limitation on the electronic device. An electronic device may include more or fewer components than illustrated, or combine certain components, or arrange them differently. For example, an input unit may include a Graphics Processing Unit (GPU) and a microphone, and a display unit may use a liquid crystal display (LCD), organic light-emitting diode (OLED), or other similar display panels. User input units include at least one of a touch panel and other input devices. A touch panel is also called a touchscreen. Other input devices may include, but are not limited to, physical keyboards, function keys (such as volume control buttons, power buttons, etc.), trackballs, mice, and joysticks, which will not be elaborated further here.

[0232] Memory can be used to store software programs and various data. Memory can primarily include a first storage area for storing programs or instructions and a second storage area for storing data. The first storage area can store the operating system, application programs or instructions required for at least one function (such as sound playback, image playback, etc.). Furthermore, memory can include volatile memory or non-volatile memory, or both. Non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. Volatile memory can be random access memory (RAM), static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous linked dynamic random access memory (Synchlink DRAM, SLDRAM), and direct memory bus RAM (DRRAM).

[0233] The processor may include one or more processing units; optionally, the processor integrates an application processor and a modem processor, wherein the application processor mainly handles operations related to the operating system, user interface, and applications, while the modem processor mainly handles wireless communication signals, such as a baseband processor. It is understood that the aforementioned modem processor may also not be integrated into the processor.

[0234] This application also provides a readable storage medium storing a program or instructions. When the program or instructions are executed by a processor, they implement the various processes of the above-described base station resource scheduling method embodiments and achieve the same technical effects. To avoid repetition, they will not be described again here.

[0235] The processor is the processor in the electronic device described in the above embodiments. The readable storage medium includes computer-readable storage media, such as computer read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk.

[0236] This application embodiment also provides a chip, which includes a processor and a communication interface. The communication interface and the processor are coupled. The processor is used to run programs or instructions to implement the various processes of the above-described base station resource scheduling method embodiments and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0237] It should be understood that the chip mentioned in the embodiments of this application may also be referred to as a system-on-a-chip, system chip, chip system, or system-on-a-chip, etc.

[0238] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. Furthermore, it should be noted that the scope of the methods and apparatuses in the embodiments of this application is not limited to performing functions in the order shown or discussed, but may also include performing functions substantially simultaneously or in the reverse order, depending on the functions involved. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.

[0239] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0240] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of this application.

Claims

1. A base station resource scheduling method, characterized in that, The method includes: The system acquires operational data for each base station included in the virtual power plant, as well as historical data for the base stations. The operational data includes data related to the operational status of the base stations, and the historical data includes data related to the predicted load data of the base stations. The historical data is input into a pre-trained load prediction model, which outputs the predicted load data corresponding to the base station. Based on the operational data and the predicted load data, the constraints and response information of the base station are determined. The response information of the base station includes: the peak shaving capacity of the base station and the priority of each device corresponding to the base station. The constraints include response capacity constraints. Based on the constraints, the target scheduling value of the virtual power plant is calculated according to the response information of the base station. The target scheduling value includes: revenue information corresponding to the target base station, peak shaving capacity information of the target base station, and priority information of the target equipment. Based on the target scheduling value, resource scheduling information is determined, and the resource scheduling information is used to assist the virtual power plant in scheduling resources for the base stations in the virtual power plant; The calculation of the target scheduling value of the virtual power plant includes: Within a preset time period, based on the response capacity in the peak-shaving capability of the base station, the revenue information corresponding to the target base station is determined; Within the preset time period, the priority information corresponding to the target device is determined based on the response status of the device in the base station; The maximum sum of the base station's revenue and priority is determined as the target scheduling value. ; Among them, the constants include: T is the scheduling time window, Indicates the time interval for scheduling. This represents the demand response subsidy price for time period t. This represents the battery charging price during time period t. This indicates the device response priority, set based on device type, peak-shaving capacity, and response time; variables include: This indicates the response capacity for user number s during time period t. Indicates switching power supply During the period Charging status.

2. The method according to claim 1, characterized in that, Determining the base station's response information based on the operational data and the predicted load data includes: Iterate through all the base stations and each device corresponding to the base station; Based on the operational data and the predicted load data, obtain the response capacity information corresponding to the base station, and determine the peak shaving capacity of the base station according to the response capacity information of the base station. Based on the operational data and the predicted load data, the equipment type, load level, and response time of each device are obtained. The priority of the device is determined based on the device type, the device load level, and the device response time; The response information of the base station is determined based on the peak shaving capability of the base station and the priority of the device.

3. The method according to claim 1, characterized in that, The step of determining the constraints of the base station based on the operational data and the predicted load data includes: response capacity constraints. Based on the predicted load data, the predicted power consumption and baseline load power are obtained; The power consumption of the base station is determined based on the predicted power consumption and the power consumption of each battery corresponding to the device in the operating data. When the power consumption of the base station is greater than the baseline load power, the response capacity is the difference between the power consumption of the base station and the baseline load power, and the difference is used to characterize that the base station meets the constraint condition. When the power consumption of the base station is less than the baseline load power, the response capacity is invalid, and the base station does not meet the constraint condition.

4. The method according to claim 1, characterized in that, The training of the load forecasting model includes: Obtain historical sample data and sample load data for each of the base stations; The historical data of the sample is preprocessed to obtain the historical data feature vector of the sample, and the historical data feature vector of the sample is input into the load prediction model; The load prediction model is trained based on the predicted load data and the sample load data.

5. The method according to claim 4, characterized in that, The step of preprocessing the historical sample data to obtain a feature vector of the historical sample data includes: Based on the historical sample data of each of the base stations, outliers in the historical sample data are identified. Obtain outliers from the historical data of the sample; The outliers are corrected by calculating the interquartile range to obtain the feature vector of the sample's historical data.

6. The method according to claim 4, characterized in that, The step of preprocessing the historical sample data to obtain the historical sample data feature vector further includes: The quotient between the historical data and the sample loading data is calculated using the Pearson correlation coefficient. Based on the quotient, determine the correlation value between the historical data of the sample and the sample load data; The historical data of the samples whose correlation values ​​are greater than a preset threshold are determined as the feature vectors of the historical data of the samples.

7. A base station resource scheduling device, characterized in that, The device includes: The acquisition module is used to acquire the operating data of each base station included in the virtual power plant, as well as the historical data of the base station. The operating data includes data related to the operating status of the base station, and the historical data includes data related to the predicted load data of the base station. The training module is used to input the historical data into a pre-trained load prediction model and output the predicted load data corresponding to the base station. The determination module is used to determine the constraints and response information of the base station based on the operating data and the predicted load data. The response information of the base station includes the peak shaving capacity of the base station and the priority of each device corresponding to the base station. The constraints include response capacity constraints. The calculation module is used to calculate the target scheduling value of the virtual power plant based on the response information of the base station, on the basis of satisfying the constraints. The target scheduling value includes: revenue information corresponding to the target base station, peak shaving capacity information of the target base station, and priority information of the target equipment. The output module is used to determine resource scheduling information based on the target scheduling value. The resource scheduling information is used to assist the virtual power plant in scheduling resources for the base stations in the virtual power plant. The computing module includes: The first determining unit is used to determine the revenue information corresponding to the target base station based on the response capacity in the peak-shaving capability of the base station within a preset time period. The second determining unit is used to determine the priority information corresponding to the target device based on the response status of the device in the base station within the preset time period. The third determining unit is used to determine the maximum value of the sum of the base station's revenue and priority, as the target. The target scheduling value is the target scheduling value. ; Among them, the constants include: T is the scheduling time window, Indicates the time interval for scheduling. This represents the demand response subsidy price for time period t. This represents the battery charging price during time period t. This indicates the device response priority, set based on device type, peak-shaving capacity, and response time; variables include: This indicates the response capacity for user number s during time period t. Indicates switching power supply During the period Charging status.

8. An electronic device, characterized in that, The system includes a processor and a memory electrically connected to the processor, the memory storing a computer program, and the processor being configured to call and execute the computer program from the memory to implement a base station resource scheduling method as described in any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The storage medium is used to store a computer program that can be executed by a processor to implement a base station resource scheduling method as described in any one of claims 1-6.

Images