CN118233927A

CN118233927A - Base station resource scheduling method and device, electronic equipment and storage medium

Info

Publication number: CN118233927A
Application number: CN202410288650.4A
Authority: CN
Inventors: 刘国锋; 李建伟; 高健; 杨慧; 林禄辉; 贾军伟; 黄继明; 毛向晖; 王鼎乾; 张学涛
Original assignee: Tower Energy Co ltd; China Tower Co Ltd
Current assignee: Tower Energy Co ltd; China Tower Co Ltd
Filing date: 2024-03-13
Publication date: 2024-06-21

Abstract

The embodiment of the application discloses a base station resource scheduling method, a base station resource scheduling device, electronic equipment and a storage medium, belongs to the technical field of power grid peak shaving, and can solve the problem of poor actual peak shaving capacity of a virtual power plant. Comprising the following steps: acquiring operation data of each base station and historical data of the base station, wherein the operation data are included in the virtual power plant; inputting the historical data into a pre-trained load prediction model, and outputting predicted load data corresponding to the base station; determining constraint conditions of the base station and 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 response information of the base station, wherein the target scheduling value comprises one or more of the following: the method comprises the steps of obtaining income information corresponding to a target base station, peak shaving capacity information of the target base station and priority information of target equipment; and determining resource scheduling information according to the target scheduling value, wherein the resource scheduling information is used for assisting the virtual power plant to schedule resources of the base station in the virtual power plant.

Description

Base station resource scheduling method and device, electronic equipment and storage medium

Technical Field

The present application relates to the field of power grid peak shaving technologies, and in particular, to a base station resource scheduling method, a base station resource scheduling device, an electronic device, and a storage medium.

Background

In recent years, as an emerging product of electric power market reform and energy internet construction, a virtual power plant participates in a peak shaving auxiliary service market in a form similar to a conventional power plant by aggregating and scheduling peak shaving resources on a demand side.

However, the research of the current virtual power plant generally relies on long-term statistical data, and all available resources are simply aggregated, so that the difference of each distributed resource cannot be accurately analyzed, and the problem of poor actual peak shaving capacity of the virtual power plant is caused.

Disclosure of Invention

The embodiment of the application aims to provide a base station resource scheduling method, a base station resource scheduling device, electronic equipment and a storage medium, which are used for solving the problem of poor actual peak regulation capability of a virtual power plant.

In order to solve the technical problems, the embodiment of the application is realized as follows:

In a first aspect, an embodiment of the present application provides a base station resource scheduling method, including: acquiring operation data of each base station included in a virtual power plant, wherein the operation data comprises data related to the operation state of the base station, and historical data of the base station, and the historical data comprises data related to predicted load data of the base station;

Inputting the historical data into a pre-trained load prediction model, and outputting the predicted load data corresponding to the base station;

Determining constraint conditions of the base station and response information of the base station according to the operation data and the predicted load data, wherein the response information of the base station comprises: the peak shaving capacity of the base station and/or the priority of each device corresponding to the base station;

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, wherein the target scheduling value comprises one or more of the following: the method comprises the steps of obtaining income information corresponding to a target base station, peak shaving capacity information of the target base station and priority information of target equipment;

And determining resource scheduling information according to the target scheduling value, wherein the resource scheduling information is used for assisting the virtual power plant to schedule resources of the base station in the virtual power plant.

In a second aspect, an embodiment of the present application provides a base station resource scheduling apparatus, including: an acquisition module, configured to acquire operation data of each base station included in a virtual power plant, and historical data of the base station, where the operation data includes data related to an operation state of the base station, and the historical data includes data related to predicted load data of the base station;

The training module is used for inputting the historical data into a pre-trained load prediction model and outputting the predicted load data corresponding to the base station;

the determining module is configured to determine, according to the operation data and the predicted load data, a constraint condition of the base station and response information of the base station, where 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 calculation module is used for calculating a target scheduling value of the virtual power plant according to the response information of the base station on the basis of meeting the constraint condition, wherein the target scheduling value comprises one or more of the following: the method comprises the steps of obtaining income information corresponding to a target base station, peak shaving capacity information of the target base station and priority information of target equipment;

and the output module is used for determining resource scheduling information according to the target scheduling value, and the resource scheduling information is used for assisting the virtual power plant to schedule the resources of the base stations in the virtual power plant.

In a third aspect, an embodiment of the present application provides an electronic device, including a processor and a memory electrically connected to the processor, where the memory stores a computer program, and the processor is configured to invoke and execute the computer program from the memory to implement a base station resource scheduling method as described above.

In a fourth aspect, an embodiment of the present application provides a computer readable storage medium storing a computer program executable by a processor to implement a base station resource scheduling method as described above.

In a fifth aspect, an embodiment of the present application provides a chip, where the chip includes a processor and a communication interface, where the communication interface is coupled to the processor, and the processor is configured to execute a program or an instruction, to implement the above-mentioned method for scheduling base station resources.

By adopting the technical scheme of the embodiment of the application, the operation data of each base station included in the virtual power plant and the historical data of the base station are obtained, wherein the operation data comprises data related to the operation state of the base station, and the historical data comprises data related to the predicted load data of the base station; inputting the historical data into a pre-trained load prediction model, and outputting predicted load data corresponding to the base station; determining constraint conditions of the base station and response information of the base station according to the operation data and the predicted load data, wherein the response information of the base station comprises: the peak shaving capability of the base station and/or the priority of each device corresponding to the base station, wherein the peak shaving capability of the base station is used for dynamically and quantitatively evaluating the base station according to the acquired operation data and constraint conditions. On the basis of meeting the constraint condition, calculating a target scheduling value of the virtual power plant according to response information of the base station, wherein the target scheduling value comprises one or more of the following: the method comprises the steps of obtaining income information corresponding to a target base station, peak shaving capacity information of the target base station and priority information of target equipment; and determining resource scheduling information according to the target scheduling value, wherein the resource scheduling information is used for assisting the virtual power plant to schedule resources of the base station in the virtual power plant. According to the calculated target scheduling value, not only the income information and the peak shaving capacity of the target base station are focused, but also the equipment priority of each base station in the virtual power plant is focused, the difference of each distributed resource can be finely analyzed, an auxiliary decision is provided for the virtual power plant to participate in peak shaving market capacity reporting, and the problem that the actual peak shaving capacity of the virtual power plant is poor can be solved.

Drawings

In order to more clearly illustrate one or more embodiments of the present application or the technical solutions in the prior art, the following description will briefly describe the drawings used in the embodiments or the description of the prior art, and it is apparent that the drawings in the following description are only some embodiments described in one or more embodiments of the present application, and other drawings may be obtained according to these drawings without inventive effort to a person skilled in the art.

FIG. 1 is a schematic diagram of a virtual power plant according to an embodiment of the application;

fig. 2 is a schematic flow chart of a base station resource scheduling method according to an embodiment of the present application;

fig. 3 is a schematic flow chart of a base station resource scheduling method according to another embodiment of the present application;

fig. 4 is a schematic block diagram of a base station resource scheduling apparatus according to an embodiment of the present application;

fig. 5 is a schematic hardware structure of a base station resource scheduling apparatus according to an embodiment of the present application.

Detailed Description

The embodiment of the application provides a base station resource scheduling method, a base station resource scheduling device, electronic equipment and a storage medium, which are used for solving the problem of function name leakage in a program file.

In order to make the technical solution of the present application better understood by those skilled in the art, the technical solution of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application, and it is apparent that the described embodiments are only some embodiments of the present application, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, shall fall within the scope of the application.

The base station resource scheduling method provided by the embodiment of the application can be executed by the electronic equipment or software installed in the electronic equipment, and specifically, the electronic equipment can be terminal equipment or service end equipment. The terminal device may include a smart phone, a notebook computer, an intelligent wearable device, an on-board terminal, and the like, and the server device may include an independent physical server, a server cluster composed of a plurality of servers, or a cloud server capable of performing cloud computing.

Fig. 1 is a schematic diagram of a virtual power plant according to an embodiment of the present application. As shown in FIG. 1, the schedulable resources in the virtual power plant comprise a plurality of base stations and a plurality of devices existing in each base station, wherein the schedulable resources are determined according to the peak shaving capacity of the base stations and the priority of the devices and based on the benefit information corresponding to the response capacity of the base stations, the peak shaving potential of the adjustable resources in the base stations is fully exerted, and decision support is provided for the virtual power plant to participate in peak shaving market trade reporting.

The following describes a base station resource scheduling method provided by the embodiment of the application in detail through a specific embodiment and an application scenario thereof with reference to the accompanying drawings.

Fig. 2 is a schematic flow chart of a base station resource scheduling method according to an embodiment of the present invention, the method includes the following steps:

S202, operation data of each base station included in the virtual power plant and historical data of the base station are obtained.

Wherein the operational data comprises data related to an operational state of the base station and the historical data comprises data related to predicted load data of the base station.

The data related to the operation state of the base station includes, but is not limited to, information such as the operation state of a switching power supply corresponding to the equipment, the current electric quantity of a storage battery, the load condition of the communication equipment, and the like, and specifically, the information data of the switching power supply: such as a switching power supply state, a storage battery type, a nuclear capacity, a floating charge voltage, a uniform charge voltage, a discharge voltage, a charge limiting voltage, a discharge termination voltage, a standby capacity, an initial response state, an initial response time length and the like; switching power supply day-ahead data: such as the day-ahead response state, day-ahead response capacity, etc. of the switching power supply; switching power supply load prediction data: such as dc load predicted power, baseline load predicted power, etc.; switching power supply historical state data: such as a historical response state of the switching power supply, a historical direct current load and the like; user number information data of the base station: such as the number of battery packs, the number of switching power supplies, adjacent response time intervals, etc.; user number electricity price data of the base station: such as a time-sharing electricity price curve and the like; information data of the virtual power plant: such as offer type, calculation mode, number of optimization periods, etc.; peak shaving auxiliary service clearing information: peak regulating capacity instruction, peak regulating patch price, etc.

The data related to the predicted load data of the base station specifically comprises the original data imported into the base station, such as the data of historical electricity consumption, historical weather, future predicted weather and the like. The load of the base station may be predicted based on historical data of the base station, wherein the predicted load data is, for example, the predicted load power of the base station, the baseline load of the base station, and the like.

It should be noted that the virtual power plant includes a plurality of base stations in different regions, and each base station corresponds to a respective subscriber number. Based on a highly intelligent communication technology, the operation data is transmitted to a central management system of the virtual power plant at regular time, so that the timeliness and the accuracy of the data can be ensured. By establishing a comprehensive data base, the method is used for periodically acquiring the running state of the base station so as to carry out subsequent peak shaving capacity analysis and decision, and can provide a data base for dynamically and quantitatively evaluating the peak shaving capacity of the base station. The problem that the response capacity of the virtual power plant participating in the peak shaving auxiliary service market is estimated according to long-term statistical data and a model in the traditional technology, and the dynamic demand response condition of an actual base station is ignored is avoided.

S204, inputting the historical data into a pre-trained load prediction model, and outputting predicted load data corresponding to the base station.

The historical data is input into a pre-trained load prediction model to obtain predicted load data, wherein constraint conditions, peak shaving capacity of the base station and equipment priority can be calculated according to the predicted load data, and changes and fluctuation of the base station load on the date of solicitation can be captured more sharply, so that the actual load level is reflected more accurately, and disturbance of uncertainty on auxiliary service is reduced.

S206, determining constraint conditions of the base station and response information of the base station according to the operation data and the predicted load data.

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

Constraint conditions, comprising: the method comprises the steps of restraining power consumption of a base station, restraining response capacity of the base station, restraining continuous response of a switching power supply, restraining maximum response times of the switching power supply, restraining electric quantity of a storage battery, restraining effective identification of response of a virtual power plant and restraining command control of the switching power supply.

The peak shaving capability of the base station comprises the following steps: the base station adjusts the fluctuation and change of the load of the base station, ensures that the power requirement and the stability of the base station are met, and particularly can comprise response capacity information and the like, the stronger the burst resistance of the base station with strong peak regulation capability is, the better the power requirement can be met, and the power quality is improved.

The priority of each device corresponding to the base station comprises: and determining the response priority of each device according to the type, the load level and the response time length of the device.

And S208, calculating a target scheduling value of the virtual power plant according to the response information of the base station on the basis of meeting the constraint condition.

The target schedule value includes one or more of the following: the method comprises the steps of obtaining income information corresponding to a target base station, peak shaving capacity information of the target base station and priority information of target equipment.

Specifically, the target scheduling value is an objective function, the benefit information and the priority information of the target base station are comprehensively considered, and the resource characteristic difference of a large number of base stations can be considered while peak shaving and patch are maximized in combination with the auxiliary service market rule. And according to the calculation of the target scheduling value, the peak shaving capacity of the priority call is strong, and meanwhile, the equipment responds to the base station with high priority.

On the basis of meeting constraint conditions, the base station is provided with response capability, and the target scheduling value of the virtual power plant is calculated by balancing the peak regulation capability of the base station and the priority of the equipment by combining the priority information of each equipment corresponding to the base station, and when the target scheduling value is maximum, the target base station corresponding to the target scheduling value is determined.

S210, determining resource scheduling information according to the target scheduling value.

The resource scheduling information is used for assisting the virtual power plant in scheduling resources for the base stations in the virtual power plant.

And determining target base station information of scheduling according to the target scheduling value, wherein the target base station information comprises capacity response conditions of the target base station, equipment information corresponding to the target base station, response states of the target base station and the like, and the resource scheduling information gathers the capacity response conditions. Specifically, the resource scheduling information includes: optimizing result information data such as whether optimization is successful, optimizing calculation time, total running cost, optimizing result information and the like; a switching power supply aggregation result table, such as a switching power supply response state, a switching power supply peak regulation capacity and the like; a family number aggregation result table of the base station, such as family number response state, family number power consumption plan, family number peak regulation capacity and the like; and virtual power plant aggregation result tables such as virtual power plant peak shaving capacity, virtual power plant power consumption plans and the like.

According to the target scheduling value, the information of each level of the equipment, the base station and the virtual power plant can be integrated respectively, the resource scheduling information is finally determined, and the virtual power plant performs resource scheduling on the base station in the virtual power plant based on the resource scheduling information.

In one embodiment, the target schedule value for the virtual power plant is calculated (i.e., S208), steps A1-A3 may be performed as follows:

Step A1: and in a preset time period, determining the corresponding benefit information of the target base station based on the response capacity in the peak shaving capacity of the base station.

The method comprises the steps of presetting a time period, representing a scheduling time interval, wherein the time interval is generally 15 minutes in real-time scheduling, acquiring response information of each storage battery in a base station according to response capacity included in peak shaving capacity of the base station in the scheduling time period, and further determining income information of a target base station, wherein the storage battery comprises the following types: lead acid batteries, nickel cadmium batteries, nickel hydrogen batteries, lithium ion batteries, and the like.

Step A2: and in a preset time period, determining priority information corresponding to the target equipment based on the response condition of the equipment in the base station.

Within the preset period of time in step A1, the response conditions of the device may include: the charging state of a switching power supply of the device in a preset time period: a 0/1 variable, where 1 indicates charging, 0 indicates no participation, and the type of device and the load capacity of the device, determines priority information of the base station.

Step A3: and when the sum of the benefit information and the priority information is larger than a preset threshold value, determining the sum of the benefit information and the priority information as a target scheduling value.

The target scheduling value is an objective function as follows:

Wherein the constant comprises: t is a time window of scheduling, and is 24 hours in the day-ahead scheduling; Δt represents a scheduled time interval, which is 15min in real-time scheduling; The demand response subsidy price for the time period t is expressed in yuan/kWh; lambda _t represents the battery charge price in yuan/kWh for period t; omega _i represents the device response priority, and is set according to the device type, peak shaving capability and response time length. Variables include: /(I) The response capacity of the user number s in the period t is expressed in kW; beta _i,t denotes the state of charge of the switching power supply i during the period t: a 0/1 variable, wherein 1 represents charge and 0 represents no participation.

The preset threshold value comprises a value which is one less than the maximum value, namely, the maximum value of the sum of the benefits and the priorities of the base stations is determined, the maximum value is used as a target scheduling value, and relevant information of the base stations is determined according to the target scheduling value, wherein the target scheduling value can correspond to one or more base stations.

It should be noted that, when the target scheduling value is specifically calculated, if the total gain of two base stations is greater than that of one base station, then two complementary base stations can be automatically optimized, so as to ensure that the total gain is maximized. As an example, there are three base stations capable of calculating the target scheduling value, where two base stations correspond to a device type that is a lithium ion battery and one base station corresponds to a lead acid battery, and it should be understood that the capacity of the lead acid battery is relatively large, for example, 100kw of response capability may be provided, and two other lithium ion batteries may provide 50kw of response capability, where the response capability provided by two lithium ion batteries corresponds to that of one lead acid battery, but the device type performance of the lithium ion battery is better and the corresponding response time is longer, and thus, the two lithium ion lead acid batteries are combined more than a single lead acid battery.

In this embodiment, the maximum target scheduling value after the combination of the benefit and the priority is determined by analyzing the combination of the economy level and different base stations, that is, according to the benefit information and the priority information of the base stations. The resource difference of each base station can be divided through the priority in the objective function, the total peak shaving capacity is calculated from a large number of base stations, and finally, the corresponding objective base station is determined according to the objective scheduling value. The optimal target base station can be determined, and the effect that corresponding resources among all base stations in the virtual power plant can be complemented can be achieved.

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

Step B1: and traversing all the base stations and all the devices corresponding to the base stations.

And acquiring the operation data and the related information of the predicted load data of all the base stations and corresponding equipment in the virtual power plant.

Step B2: based on the operation data and the predicted load data, response capacity information corresponding to the base station is obtained, and the peak shaving capacity of the base station is determined according to the response capacity information of the base station.

And calculating the power consumption, the response capacity and the load fluctuation change of the base station according to the operation data and the predicted load data corresponding to the base station, determining the load level of the base station, namely the response capacity information, and determining the peak shaving capacity of the base station based on the response capacity information.

Step B3: and acquiring the equipment type, the load level and the response time of the equipment corresponding to each equipment based on the operation data and the predicted load data.

The device types comprise lithium ion batteries, nickel cadmium batteries and other different types; and calculating the load level of the equipment, namely the response capacity, the response time length and other information of each equipment according to the operation data and the predicted load data, and further determining the load level of the equipment and the response time length of the equipment.

Step B4: and determining the priority of the equipment according to the equipment type, the load level of the equipment and the response time length of the equipment.

Based on the type of the device, the load level of the device and the response time length of the device, all base stations and devices can be traversed, and the response priority of the device corresponding to the base station is judged.

Step B5: and determining response information of the base station based on the peak shaving capability of the base station and/or the priority of the equipment.

In this embodiment, by traversing the operation data and the predicted load data of all the base stations and the devices, response information corresponding to the base stations is determined, where the response information includes: the peak shaving capability of the base station, and the priority of the equipment response determined according to the equipment type, the load level of the equipment and the response time of the equipment can deeply analyze the resource characteristics of different base stations, and can take the actual equipment condition of the base station into consideration for quantitative evaluation.

In one embodiment, determining constraints of the base station based on the operational data and the predicted load data comprises: in response to the capacity constraint, the following steps C1-C4 may be performed:

Step C1: and obtaining the predicted electric power and the baseline load power according to the predicted load data.

The baseline load in the baseline load power refers to 1 load curve estimated from the historical load data of the user. As an example, when the response day is the working day, the load average value of the first 5 normal working days of the solicited day is selected as the peak regulation baseline, when the response day is the Saturday (day), the load average value of the first 3 Saturday (day) of the solicited day is selected as the peak regulation baseline, wherein the role of calculating the baseline is to determine the electricity consumption condition when the virtual power plant does not participate in peak regulation, and the peak regulation capacity which can be provided by the base station is accurately estimated based on the electricity consumption condition.

Step C2: and determining the power consumption of the base station according to the predicted power consumption and the power consumption of each storage battery corresponding to the equipment in the operation data.

The power consumption of the base station is based on the predicted power consumption, and the power consumption of the base station is obtained by taking the predicted power consumption and the power consumption of each storage battery into consideration the floating charge power provided by the storage battery to which the equipment belongs, wherein the specific calculation formula is as follows:

Wherein the constant: The predicted power consumption of the base station with the user number s in the period t is represented by kW; /(I) The float charge power of the battery i is shown in kW. The variables: p _s,t represents the power consumption of the base station for the user number s in the period t, and the unit is kW; beta _i,t denotes the state of charge of the switching power supply i in period t: a 0/1 variable, wherein 1 represents charge and 0 represents no participation.

It follows that the power constraints of the base station include: and determining the actual electric power of each base station according to the sum of the predicted electric power and the floating charge power provided by the storage battery in the equipment corresponding to the base station.

Step C3: when the power consumption of the base station is larger than the baseline load power, the response capacity is the difference value between the power consumption of the base station and the baseline load power, and the difference value is used for representing that the base station meets constraint conditions.

When the power consumption of the base station is increased above the baseline load, that is, the power consumption of the base station calculated in the step C2 is subtracted by the baseline load power in the step C1, if the calculated difference is greater than 0, the response of the corresponding base station is effective, the response capacity is the difference between the power consumption and the baseline load power, the base station is determined to meet the constraint condition, and the base station meeting the constraint condition can be used as a screening base station for calculating the target scheduling value.

The specific calculation formula is as follows:

Wherein the constant comprises: The base station with the user number S is represented by a base line load with the time period t, and the unit is kW; m represents a sufficiently large positive number, usually 10 ⁶. Variables include: beta _s,t denotes a device with a switching power supply i, a charging state at a period t: a 0/1 variable, wherein 1 represents charging and 0 represents no participation; gamma _s,t denotes the response valid state of the base station S in period t: a 0/1 variable, wherein 1 represents valid and 0 represents invalid; /(I) Expressed as the response capacity of the base station S in kW during the period t; p _s,t denotes the power used by the base station S in kW during the period t.

Step C4: when the power consumption of the base station is smaller than the baseline load power, the response capacity is invalid, and the base station does not meet the constraint condition.

When the power consumption of the base station is smaller than the baseline load, the response of the base station is invalid, the response capacity of the base station is 0, and the base station does not meet the constraint condition, so that the base station can not be used as a screening base station for calculating the target scheduling value.

Besides the constraint conditions, the continuous response constraint of the switching power supply, the maximum response frequency constraint of the switching power supply, the storage battery electric quantity constraint, the virtual power plant response effective identification constraint and the switching power supply instruction control constraint are also included, and the specific calculation method is as follows:

(1) The switching power supply continues to respond to constraints: once the switching power supply participates in the response, the response can be stopped after the response-capable duration of continuous response is necessary, wherein the response condition of the switching power supply corresponds to the response condition of the equipment. The calculation formula is as follows:

Constant: t _i ^dr represents the responseable duration of the switching power supply i, and the unit is a period (15 min). The variables: beta _i,t denotes the state of charge of the switching power supply i in period t: a 0/1 variable, wherein 1 represents charge and 0 represents no participation. That is, in the case where each device is in a charged state for the period of time, the corresponding base station satisfies the constraint condition.

(2) Maximum response times constraint of switching power supply: the maximum response times of the switching power supply in the demand response of the single offer are 1 time. The calculation formula is as follows:

μ_i,t-υ_i,t＝β_i,t-β_i,t-1

μ_i,t+υ_i,t≤1

The variables: beta _i,t denotes the state of charge of the switching power supply i in period t: a 0/1 variable, wherein 1 represents charging and 0 represents no participation; mu _i,t represents a start response state of the switching power supply i in a period t, 0/1 variable, where 1 represents a start response and 0 represents no start response; v _i,t represents the stop response state of the switching power supply i in the period t, 0/1 variable, wherein 1 represents the stop response and 0 represents the non-stop response.

(3) And (3) constraint of the electric quantity of the storage battery: the battery's charge at that time is equal to its charge at that time plus the charge supplemented by the participation demand response. The calculation formula is as follows:

V_i ^rsv≤ESS_i,t≤V_i

Wherein the constant comprises: The floating charge power of the storage battery i is represented by kW; v _i ^rsv denotes the minimum capacity allowed for battery i in kWh; v _i denotes the maximum capacity allowed for battery i in kWh. Variables include: ESS _i,t represents the charge of battery i during period t in kWh; beta _i,t denotes the state of charge of the switching power supply i in period t: a 0/1 variable, wherein 1 represents charge and 0 represents no participation.

(4) The virtual power plant responds to the active identification constraint: the virtual power plant response is effectively identified as: the minimum load is greater than the baseline minimum load; and the cumulative load is greater than the baseline cumulative load. The calculation formula is as follows:

Wherein the constant comprises: a baseline load representing the base station s over a period t, in kW; /(I) Representing a minimum baseline load of a virtual power plant offer window in kW; m represents a sufficiently large positive number, usually 10 ⁶. Variables include: /(I)Representing the virtual power plant response valid state: a 0/1 variable, wherein 1 represents valid and 0 represents invalid; /(I)Representing the minimum power consumption of a virtual power plant offer window, wherein the unit is kW; p _s,t denotes the power consumption of the base station S in kW during the period t.

(5) Switching power supply command control constraints:

① Before the offer period, the regulating instruction of the switching power supply is set to be not charged and discharged

Wherein the constant comprises: The voltage value indicating that the switching power supply i is not charged and discharged is expressed as V. Variables include: mu _i,t denotes a control power supply instruction of the switching power supply i in the period t, in V.

② During the offer period, the switching power supply sets a control instruction according to the response state:

i. maintaining no charge and no discharge voltage of switching power supply before participation in response

Response period, and is state of charge, i.e. β _i,t =1, switching power supply regulation command is set to float voltage:

Wherein the constant comprises: The floating charge voltage value of the switching power supply i is represented by V; variables include: mu _i,t denotes a control power supply instruction of the switching power supply i in the period t, in V.

After the battery participates in the response, set to float voltage:

Wherein the constant comprises: : The floating charge voltage value of the switching power supply i is represented by V; variables include: mu _i,t denotes a control power supply instruction of the switching power supply i in the period t, in V.

③ After the offer period, the switching power supply issues a recovery instruction, and the recovery instruction is set as floating charge voltage:

In this embodiment, whether the response capability of the base station meets the constraint condition is determined by various constraint conditions, and the response capability is dynamically evaluated for the base station in a quantitative manner, wherein the quantitative evaluation considers the actual equipment conditions of the base station, such as the state of a switching power supply, the charging power of a storage battery, the response capacity thereof, and the like. And when the constraint condition is met, the method can be used for calculating an objective function, which is equivalent to screening the base stations in the virtual power plant once, the screened base stations can be used for calculating the target scheduling value, the calculation efficiency of the target scheduling value can be improved, and the base stations participating in the peak shaving of the virtual power plant can be determined more quickly.

In one embodiment, training of the load prediction model may be performed specifically as follows steps D1-D3:

Step D1: sample history data and sample load data of each base station are acquired.

The sample historical data comprises factors such as historical loads at various times of the day before the same base station in a prediction period, such as load amounts at different times of different dates, and influences of meteorological factors such as temperature, precipitation, wind speed and the like on the power utilization load.

Sample load data including load data of a corresponding base station determined from the sample history data.

Step D2: and carrying out data preprocessing on the sample historical data to obtain sample historical data feature vectors, and inputting the sample historical data feature vectors into a load prediction model.

The data preprocessing comprises the steps of cleaning data of an original data set imported into a base station, selecting prediction features strongly related to load according to correlation analysis, and processing repeated values, abnormal values and missing values of the data in sequence.

Based on preprocessing the data, obtaining a history data feature vector after preprocessing, and constructing an input feature vector of a load prediction model so as to form an overall training data set; likewise, sample load data feature vectors of the predicted points at different dates and moments are constructed, so that an overall test data set is formed.

In addition, since the data of different base stations need to be subjected to a mutual separation test process, the overall data set is divided into a plurality of small data sets according to the serial numbers (IDs) of the base stations, and the number of IDs of all the base stations is uniformly divided into a plurality of data sets by using a multithreading technology.

Step D3: the load prediction model is trained based on the predicted load data and the sample load data.

The structure of the load prediction model is XGBoost, parameters of XGBoost regression model are set, including learning rate, number of trees, depth of trees and the like, and calculation learning is performed on the input sample historical data feature vector based on XGBoost. Wherein the input sample history data feature vector of the load prediction model comprises: the time characteristic of the predicted point, the historical value characteristic of the same time of the D-1 day, is output as a predicted value of the load, namely predicted load data, wherein the load prediction model is trained, and parameters in XGBoost are adjusted to enable the predicted load data to be consistent with sample load data, so that model training is successful.

In addition, the prediction results obtained by different threads are summarized, the evaluation results are evaluated, the average error of the single target variable is evaluated, and finally a short-term prediction algorithm is formed and used for predicting the load of the base station.

In this embodiment, sample historical data and sample load data of each base station are obtained, based on a load prediction model of XGBoost structures, the sample historical data is trained, and finally predicted load data is output, and a multi-thread training method can be constructed for base stations with different IDs, so that training speed is improved.

In one embodiment, the sample history data is subjected to data preprocessing to obtain a sample history data feature vector (i.e. step D2), and the following steps E1-E3 may be performed:

Step E1: and identifying abnormal values of the sample historical data according to the sample historical data of each base station.

The abnormal values of the sample history data specifically include the following cases: the load data at a point is much greater or less than the previous and next data points; the load data of a certain day of the same site is far greater than or less than the load data of other dates; the problem of load data collection by the base station may result in the load data of a certain day being either invalid NULL or having only a few non-NULL values.

The sample history data set is filtered for the above cases and an index of the locked outliers is output.

Step E2: an outlier of the sample history data is obtained.

Step E3: and correcting the abnormal value by a calculation method of the quartile range to obtain a sample historical data feature vector.

After abnormal value identification is carried out by adopting a calculation method of four bit distances (interquartile range, IQR), abnormal values of sample historical data are obtained, sample load data are divided according to base station IDs, missing values are removed from a sample historical data set corresponding to each ID, and the processed abnormal values are calculated by the IQR.

Specifically, the basis for defining outliers is to define normal ranges above and below the dataset, i.e., upper and lower bounds (considering 1.5×iqr values):

upper＝Q3+1.5*IQR

lower＝Q1–1.5*IQR

Among them, IQR, also called quartile, is a method in descriptive statistics to determine the difference between the third quartile and the first quartile. As with variance and standard deviation, the variance and standard deviation represent the variance of each variable in the statistics, but the fourth difference is more a robust statistic, and index index_list of outliers is recorded.

The correction method of the index index_list of the abnormal value obtained in the above case is as follows:

(1) Linear interpolation: if the continuous abnormality/deletion does not exceed 1h of data, linear interpolation is adopted for supplementation.

(2) Interpolation of similar days: if the continuous abnormality/deletion does not exceed 2 hours of data, interpolation is adopted for interpolation by adopting similar days.

(3) Deletion: if the continuous abnormality/deletion exceeds 2h of data, no interpolation significance exists; the day's load data is deleted from the sample training set.

(4) According to the daily load value condition of the inquiring base station, if only a small part of the conditions are NULL values, the conditions can be error values, and a 0-value filling method is used for filling the error values; if the load value of the base station on a certain day is NULL, the day data has no interpolation meaning, and the load data on the day is deleted from the sample set.

In addition, since the problem in measurement or transmission may cause that most of the record points in one day are 0 in some of the sample history data, such samples can have negative influence on the prediction performance when training together with normal samples, so that the total load of each day is summed up before model training, 96×0.05 is set as a threshold value by using a threshold value filtering method, samples with most of the record points being 0 are filtered, and the sample history data is subjected to a screening operation.

In this embodiment, by acquiring an abnormal value in the sample history data and performing processing such as screening, removing and interpolating on the abnormal value, the sample history data is relatively complete, the feature vector of the sample history data in the prediction load model can be more comprehensive, and the output test result is more accurate.

In one embodiment, the sample history data is subjected to data preprocessing to obtain a sample history data feature vector (i.e. step D2), and the following steps F1-F3 may be further executed:

Step F1: and calculating the quotient between the two variables through the pearson correlation coefficient by using the sample history data and the sample load data.

In particular, the pearson correlation coefficient is used to measure the strength and direction of the linear relationship between two successive variables, and provides an indicator of the strength of the linear relationship between two variables, which can range from-1 to 1, where-1 indicates a complete negative correlation, 0 indicates no linear correlation, and 1 indicates a complete positive correlation, i.e., one variable increases while the other increases. According to the coefficient, the correlation between the variables is judged to be strong or weak, the interpretation and the intuitiveness are realized, the calculated amount of calculating the correlation coefficient is relatively small for a large-scale data set, and the result can be obtained quickly.

The dominant factor of load prediction is identified by adopting a pearson correlation coefficient method, and the pearson correlation coefficient between two variables X, Y is defined as the quotient of covariance and standard deviation between the two variables:

In the above formula: x represents a sample history data feature vector corresponding to the sample history data; y represents a sample load data characteristic vector corresponding to the sample load data; cov (X, Y) covariance of two columns of data; σxσy is the standard deviation of the sample history data feature vector and the sample load data feature vector used, respectively; sigma X is the mean value of sample X _i and the standard deviation of the sample, and the calculation of the pearson equation involves summing the squares of the deviations of the variables, and summing the products of the correlations.

Step F2: and determining a correlation value of the sample history data and the sample load data according to the quotient value.

And determining the linear correlation degree between the two variable sample historical data and the sample load data through the statistic of the step F1.

Step F3: and determining the sample historical data with the correlation value larger than a preset threshold value as a sample historical data feature vector.

And according to the correlation degree, determining the sample historical data which has strong correlation with the sample load data as an input sample historical data characteristic vector of an input prediction load data model.

In addition to the calculation of correlation, the input data of the predictive load data model needs to consider other features included in the sample history data, such as the following features:

(1) Date feature

The time at which the data point is located: this feature is mainly used to distinguish between different moments of the same day, where the load is represented as a continuation or distinction, i.e. in some consecutive moments of the same day, where the load is the same or similar, but also in some moments, where the load is represented as a distinct switch, or suddenly increases or suddenly decreases, so that the moment feature is an important factor for the prediction of the load.

As one example, consider category characteristics for different days of the week, such as the day of the week being the day of the week; and if the day is a working day, discretizing operation is carried out during processing, and the characteristic values corresponding to the working day and the holiday are 1, and the characteristic values corresponding to the non-working day (weekend) and the non-holiday are 0.

(2) Historical value characteristics

As can be seen from an examination of the data, the load fluctuation of each base station between different days is small, and even the load amounts for consecutive days are all the same, the load amounts for the next day, the next three days, and the next five days can be taken into consideration as one feature information because the load amounts change only slightly even if the load amounts change. Such as a historical value of one day before the same time, a load average value of nearly three days at the same time, a load average value of nearly five days at the same time, and the like.

(3) Meteorological characteristics (optional)

The research on the influence of meteorological factors on the load characteristics of the power grid shows that the meteorological factors such as temperature, precipitation, wind speed and the like can also influence the power consumption load. Specifically, the temperature is a meteorological factor with the greatest influence on the load, and as people use air conditioners more, the influence of the air temperature on the load is larger and larger, and meanwhile, the precipitation and the wind speed also have a certain influence on the electric load. Also, the meteorological data may also have an influence on the electricity load prediction of the base station, so that the meteorological features can also be used as a part of characteristic factors, such as temperature, precipitation, wind speed and the like.

In this embodiment, by screening the sample history data and calculating the correlation between the sample history data and the sample load data, sample history feature data with strong correlation influencing factors can be selected from a large number of sample history data, and the sample history feature data is input into the predicted load model, so that the efficiency and accuracy of load prediction can be improved, and the load level can be reflected more accurately.

Fig. 3 is a schematic flow chart of a base station resource scheduling method according to another embodiment of the present application, as shown in fig. 3, the method includes the steps of:

S301, operation data and historical data of all base stations in the virtual power plant are obtained.

The historical data includes data corresponding to the base station prior to the date of the offer, which is used to predict load data.

S302, the historical data of all the base stations are input into a load prediction model.

The load prediction model performs data preprocessing on the historical data, and selects a historical data feature vector with strong correlation through pearson correlation analysis and inputs the historical data feature vector into the load prediction model.

S303, outputting predicted load data of the corresponding base station by the load prediction model.

S304, determining peak shaving capacity of each base station in the virtual power plant according to the operation data and the predicted load data.

S305, based on the operation data and the predicted load data, determining the type of the equipment, the load level of the equipment and the response time length of the equipment of each equipment in the base station, and further determining the priority of the equipment.

S306, determining constraint conditions of the base station according to the operation data and the predicted load data.

Specifically, the base station is dynamically and quantitatively evaluated according to the operation data and the predicted load data.

S307, determining the corresponding benefit information of the base station through the peak shaving capacity of the base station and the priority information of the equipment on the basis of meeting the constraint condition.

And S308, calculating the maximum value of the target scheduling value based on the benefit information, the peak shaving capacity of the base station and the priority of the equipment.

S309, according to the maximum value of the target scheduling value, corresponding base stations and resource scheduling information are determined.

S310, acquiring related content of schedulable resources of the virtual power plant by utilizing the resource scheduling information.

The specific processes of S301 to S310 are described in detail in the above embodiments, and are not described here again.

In summary, particular embodiments of the present subject matter have been described. Other embodiments are within the scope of the following claims. In some cases, the actions recited in the claims can be performed in a different order and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may be advantageous.

The above method for scheduling base station resources provided by the embodiment of the present application is based on the same thought, and the embodiment of the present application further provides a base station resource scheduling device.

Fig. 4 is a schematic structural diagram of a base station resource scheduling apparatus according to an embodiment of the present invention. As shown in fig. 4, the base station resource scheduling apparatus includes: acquisition module 41, training module 42, determination module 43, calculation module 44, output module 45:

An acquisition module 41, configured to acquire operation data of each base station included in the virtual power plant, and historical data of the base station, where the operation data includes data related to an operation state of the base station, and the historical data includes data related to predicted load data of the base station;

The training module 42 is configured to input the historical data into a pre-trained load prediction model, and output predicted load data corresponding to the base station;

a determining module 43, configured to determine, according to the operation data and the predicted load data, constraint conditions of the base station and response information of the base station, where the response information of the base station includes: peak shaving capability of the base station and/or priority of each device corresponding to the base station;

The calculating module 44 is configured to calculate, based on the response information of the base station, a target scheduling value of the virtual power plant, where the target scheduling value includes one or more of the following: the method comprises the steps of obtaining income information corresponding to a target base station, peak shaving capacity information of the target base station and priority information of target equipment;

And the output module 45 is configured to determine resource scheduling information according to the target scheduling value, where the resource scheduling information is used to assist the virtual power plant in scheduling resources for the base station in the virtual power plant.

In one embodiment, the computing module 44 includes:

the first determining unit is used for determining the benefit information corresponding to the target base station based on the response capacity in the peak shaving capacity of the base station in a preset time period;

the second determining unit is used for determining priority information corresponding to the target equipment based on the response condition of the equipment in the base station in a preset time period;

and a third determining unit configured to determine the sum of the benefit information and the priority information as the target scheduling value when the sum of the benefit information and the priority information is greater than a preset threshold.

In one embodiment, the determining module 43 includes:

The traversing unit is used for traversing all the base stations and all the devices corresponding to the base stations;

The acquisition unit is used for acquiring response capacity information corresponding to the base station based on the operation data and the predicted load data, and determining the peak shaving capacity of the base station according to the response capacity information of the base station; based on the operation data and the predicted load data, acquiring the equipment type, the load level and the response time of the equipment corresponding to each equipment;

determining the priority of the equipment according to the equipment type, the load level of the equipment and the response time length of the equipment;

and determining response information of the base station based on the peak shaving capability of the base station and/or the priority of the equipment.

In one embodiment, the determining module 43 further includes: and a response capacity constraint module:

acquiring predicted electric power and baseline load power according to the predicted load data;

determining the power consumption of the base station according to the predicted power consumption and the power consumption of each storage battery corresponding to the equipment in the operation data;

When the power consumption of the base station is larger than the baseline load power, the response capacity is the difference value between the power consumption of the base station and the baseline load power, and the difference value is used for representing that the base station meets constraint conditions;

When the power consumption of the base station is smaller than the baseline load power, the response capacity is invalid, and the base station does not meet the constraint condition.

In one embodiment, training module 42 includes:

acquiring sample historical data and sample load data of each base station;

Carrying out data preprocessing on the sample historical data to obtain sample historical data feature vectors, and inputting the sample historical data feature vectors into a load prediction model;

The load prediction model is trained based on the predicted load data and the sample load data.

In one embodiment, the data preprocessing is performed on the sample history data to obtain a sample history data feature vector, which includes:

Identifying abnormal values of the sample historical data according to the sample historical data of each base station;

Acquiring an abnormal value of sample historical data;

and correcting the abnormal value by a calculation method of the quartile range to obtain a sample historical data feature vector.

In one embodiment, the data preprocessing is performed on the sample history data to obtain a sample history data feature vector, and the method further includes:

calculating the quotient between the two variables through the pearson correlation coefficient by using the sample history data and the sample load data;

determining a correlation value of the sample history data and the sample load data according to the quotient value;

And determining the sample historical data with the correlation value larger than a preset threshold value as a sample historical data feature vector.

It should be understood by those skilled in the art that the base station resource scheduling apparatus in fig. 4 can be used to implement the base station resource scheduling method described above, and the detailed description thereof should be similar to that of the method section above, so as to avoid complexity and avoid redundancy.

Based on the same technical concept, the embodiment of the application also provides an electronic device, which is used for executing the base station resource scheduling method, and fig. 5 is a schematic structural diagram of an electronic device for implementing the embodiments of the application. The electronic devices may be configured or configured differently, and may include a processor (processor) 510, a communication interface (Communications Interface) 520, a memory (memory) 530, and a communication bus 540, where the processor 510, the communication interface 520, and the memory 530 communicate with each other via the communication bus 540. Processor 510 may invoke a computer program stored in memory 530 and executable on processor 510 to perform the steps of:

Acquiring operation data of each base station included in the virtual power plant, wherein the operation data comprises data related to the operation state of the base station, and historical data of the base station, and the historical data comprises data related to predicted load data of the base station;

inputting the historical data into a pre-trained load prediction model, and outputting predicted load data corresponding to the base station;

Determining constraint conditions of the base station and response information of the base station according to the operation data and the predicted load data, wherein the response information of the base station comprises: peak shaving capability of the base station and/or priority of each device corresponding to the base station;

On the basis of meeting the constraint condition, calculating a target scheduling value of the virtual power plant according to response information of the base station, wherein the target scheduling value comprises one or more of the following: the method comprises the steps of obtaining income information corresponding to a target base station, peak shaving capacity information of the target base station and priority information of target equipment;

The specific execution steps can refer to the steps of the base station resource scheduling method embodiment, and the same technical effects can be achieved, so that repetition is avoided, and the description is omitted here.

It should be noted that, the electronic device in the embodiment of the present application includes: a server, a terminal, or other devices besides a terminal.

The above electronic device structure does not constitute a limitation of the electronic device, and the electronic device may include more or less components than illustrated, or may combine some components, or may be different in arrangement of components, for example, an input unit, may include a graphics processor (Graphics Processing Unit, GPU) and a microphone, and a display unit may configure a display panel in the form of a liquid crystal display, an organic light emitting diode, or the like. The user input unit includes at least one of a touch panel and other input devices. Touch panels are also known as touch screens. Other input devices may include, but are not limited to, a physical keyboard, function keys (e.g., volume control keys, switch keys, etc.), a trackball, a mouse, a joystick, and so forth, which are not described in detail herein.

The memory may be used to store software programs as well as various data. The memory may mainly include a first memory area storing programs or instructions and a second memory area storing data, wherein the first memory area may store an operating system, application programs or instructions (such as a sound playing function, an image playing function, etc.) required for at least one function, and the like. Further, the memory may include volatile memory or nonvolatile memory, or the memory may include both volatile and nonvolatile memory. The nonvolatile Memory may be a Read-Only Memory (ROM), a Programmable ROM (PROM), an Erasable PROM (EPROM), an Electrically Erasable EPROM (EEPROM), or a flash Memory. The volatile memory may be random access memory (Random Access Memory, RAM), static random access memory (STATIC RAM, SRAM), dynamic random access memory (DYNAMIC RAM, DRAM), synchronous Dynamic Random Access Memory (SDRAM), double data rate Synchronous dynamic random access memory (Double DATA RATE SDRAM, DDRSDRAM), enhanced Synchronous dynamic random access memory (ENHANCED SDRAM, ESDRAM), synchronous link dynamic random access memory (SYNCH LINK DRAM, SLDRAM), and Direct random access memory (DRRAM).

The processor may include one or more processing units; optionally, the processor integrates an application processor that primarily processes operations involving an operating system, user interface, application programs, and the like, and a modem processor that primarily processes wireless communication signals, such as a baseband processor. It will be appreciated that the modem processor described above may not be integrated into the processor.

The embodiment of the application also provides a readable storage medium, on which a program or an instruction is stored, which when executed by a processor, implements each process of the above-mentioned base station resource scheduling method embodiment, and can achieve the same technical effect, so that repetition is avoided, and no further description is given here.

Wherein the processor is a processor in the electronic device described in the above embodiment. The readable storage medium includes a computer readable storage medium such as a Read-Only Memory (ROM), a random access Memory (Random Access Memory, RAM), a magnetic disk or an optical disk, and the like.

The embodiment of the application further provides a chip, the chip comprises a processor and a communication interface, the communication interface is coupled with the processor, the processor is used for running programs or instructions, the processes of the base station resource scheduling method embodiment can be realized, the same technical effect can be achieved, and the repetition is avoided, and the description is omitted here.

It should be understood that the chips referred to in the embodiments of the present application may also be referred to as system-on-chip chips, chip systems, or system-on-chip chips, etc.

It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element. Furthermore, it should be noted that the scope of the methods and apparatus in the embodiments of the present application is not limited to performing the functions in the order shown or discussed, but may also include performing the functions in a substantially simultaneous manner or in an opposite order depending on the functions involved, e.g., the described methods may be performed in an order different from that described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.

From the above description of the embodiments, it will be clear to those skilled in the art that the above-described embodiment method may be implemented by means of software plus a necessary general hardware platform, but of course may also be implemented by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium (e.g. ROM/RAM, magnetic disk, optical disk) comprising instructions for causing a terminal (which may be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to perform the method according to the embodiments of the present application.

The embodiments of the present application have been described above with reference to the accompanying drawings, but the present application is not limited to the above-described embodiments, which are merely illustrative and not restrictive, and many forms may be made by those having ordinary skill in the art without departing from the spirit of the present application and the scope of the claims, which are to be protected by the present application.

Claims

1. A method for scheduling base station resources, the method comprising:

Acquiring operation data of each base station included in a virtual power plant, wherein the operation data comprises data related to the operation state of the base station, and historical data of the base station, and the historical data comprises data related to predicted load data of the base station;

2. The method of claim 1, wherein the calculating the target schedule value for the virtual power plant comprises:

Determining the corresponding benefit information of the target base station based on the response capacity in the peak shaving capacity of the base station in a preset time period;

Determining the priority information corresponding to the target equipment based on the response condition of the equipment in the base station in the preset time period;

and when the sum of the benefit information and the priority information is larger than a preset threshold value, determining the sum of the benefit information and the priority information as the target scheduling value.

3. The method of claim 1, wherein said determining response information for said base station based on said operational data and said predicted load data comprises:

traversing all the base stations and all the devices corresponding to the base stations;

Based on the operation data and the predicted load data, response capacity information corresponding to the base station is obtained, and peak shaving capacity of the base station is determined according to the response capacity information of the base station;

Acquiring equipment types corresponding to the equipment, the load level of the equipment and the response time of the equipment based on the operation data and the predicted load data;

4. The method of claim 1, wherein said determining constraints of said base station based on said operational data and said predicted load data comprises: responding to capacity constraints:

when the power consumption of the base station is larger than the baseline load power, the response capacity is the difference value between the power consumption of the base station and the baseline load power, and the difference value is used for representing that the base station meets the constraint condition;

and when the power consumption of the base station is smaller than the baseline load power, the response capacity is invalid, and the base station does not meet the constraint condition.

5. The method of claim 1, wherein the training of the load prediction model comprises:

Acquiring sample history data and sample load data of each base station;

Performing data preprocessing on the sample historical data to obtain sample historical data feature vectors, and inputting the sample historical data feature vectors into the load prediction model;

training the load prediction model based on the predicted load data and the sample load data.

6. The method of claim 5, wherein the performing data preprocessing on the sample history data to obtain a sample history data feature vector comprises:

acquiring an abnormal value of the sample history data;

And correcting the abnormal value by a calculation method of the quartile range to obtain the characteristic vector of the sample historical data.

7. The method of claim 5, wherein the performing data preprocessing on the sample history data to obtain a sample history data feature vector, further comprises:

Calculating the quotient between the two variables through the pearson correlation coefficient by using the sample historical data and the sample load data;

and determining the sample historical data with the correlation value larger than a preset threshold value as the sample historical data feature vector.

8. A base station resource scheduling apparatus, the apparatus comprising:

An acquisition module, configured to acquire operation data of each base station included in a virtual power plant, and historical data of the base station, where the operation data includes data related to an operation state of the base station, and the historical data includes data related to predicted load data of the base station;

9. An electronic device comprising a processor and a memory electrically connected to the processor, the memory storing a computer program, the processor being configured to invoke and execute the computer program from the memory to implement a base station resource scheduling method according to any of claims 1-7.

10. A computer readable storage medium storing a computer program executable by a processor to implement a base station resource scheduling method according to any one of claims 1-7.