Regional load demand response scheduling method and device and computer equipment

Abstract

The invention relates to a regional load demand response scheduling method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: determining a power operation strategy of a target area in a scheduling time period through carbon emission cost represented by power grid electricity purchasing power of the target area in the scheduling time period; obtaining an actual load demand of the target area in the scheduling time period and a predicted photovoltaic output of the predicted target area in the scheduling time period; and under the condition that the actual photovoltaic output is not matched with the predicted photovoltaic output under the condition that the power operation strategy is executed in the target area, demand response equipment in the target area is determined and scheduled based on the actual photovoltaic output, the actual load demand and the power operation strategy. Therefore, the load demand after the demand response equipment participates in the demand response in the scheduling time period is matched with the actual load demand. By adopting the method, the accuracy of demand response scheduling of the region can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power systems, and particularly relates to a response scheduling method and device for regional load demand, computer equipment, computer readable storage medium and computer program product. BACKGROUND

[0002] Generally, the energy system of a region is the core infrastructure supporting the normal operation of the region, and its core role is to meet the diversified energy demand of enterprises, facilities and users in the region through efficient, stable and sustainable energy production, conversion, distribution and consumption. The carbon management system of the region is the core tool for the region to achieve the goal, and its role is to promote the low-carbon transformation of the region by scientifically monitoring, quantitatively analyzing and dynamically controlling carbon emission activities.

[0003] However, the energy system and the carbon management system of the region are independent of each other, and lack dynamic collaborative optimization capability. When the region responds to the load demand, the accuracy of the demand response scheduling of the region is not high. SUMMARY

[0004] Therefore, it is necessary to provide a response scheduling method and device for regional load demand, computer equipment, computer readable storage medium and computer program product, which can improve the accuracy of the demand response scheduling of the region.

[0005] In a first aspect, the present application provides a response scheduling method for regional load demand, comprising: determining a power operation strategy of a target region in a scheduling period based on a carbon emission cost represented by a power grid purchase power of the target region in the scheduling period; obtaining an actual load demand of the target region in the scheduling period and a predicted photovoltaic output of the target region in the scheduling period; and in a case where an actual photovoltaic output of the target region under the power operation strategy does not match the predicted photovoltaic output, determining and scheduling a demand response device in the target region based on the actual photovoltaic output, the actual load demand and the power operation strategy, so that a load demand after the demand response device participates in the demand response in the scheduling period matches the actual load demand.

[0006] In a second aspect, the application provides a device for responding to regional load demand, the device comprising: a processing module configured to determine a power operation strategy for a target region in a scheduling period based on a carbon emission cost represented by grid electricity purchasing power of the target region in the scheduling period; an acquisition module configured to acquire an actual load demand of the target region in the scheduling period and a predicted photovoltaic output of the target region in the scheduling period; and a scheduling module configured to, in a case where an actual photovoltaic output of the target region under the power operation strategy does not match the predicted photovoltaic output, determine and schedule a demand response device in the target region based on the actual photovoltaic output, the actual load demand and the power operation strategy, so that a load demand after the demand response device participates in demand response in the scheduling period matches the actual load demand.

[0007] In a third aspect, the application provides a computer device comprising a memory and a processor, the memory storing a computer program, and the processor implementing the following steps when executing the computer program: determining a power operation strategy for a target region in a scheduling period based on a carbon emission cost represented by grid electricity purchasing power of the target region in the scheduling period; acquiring an actual load demand of the target region in the scheduling period and a predicted photovoltaic output of the target region in the scheduling period; and in a case where an actual photovoltaic output of the target region under the power operation strategy does not match the predicted photovoltaic output, determining and scheduling a demand response device in the target region based on the actual photovoltaic output, the actual load demand and the power operation strategy, so that a load demand after the demand response device participates in demand response in the scheduling period matches the actual load demand.

[0008] In a fourth aspect, the application provides a computer readable storage medium having a computer program stored thereon, the computer program being executed by a processor to implement the following steps: determining a power operation strategy for a target region in a scheduling period based on a carbon emission cost represented by grid electricity purchasing power of the target region in the scheduling period; acquiring an actual load demand of the target region in the scheduling period and a predicted photovoltaic output of the target region in the scheduling period; and in a case where an actual photovoltaic output of the target region under the power operation strategy does not match the predicted photovoltaic output, determining and scheduling a demand response device in the target region based on the actual photovoltaic output, the actual load demand and the power operation strategy, so that a load demand after the demand response device participates in demand response in the scheduling period matches the actual load demand.

[0009] In a fifth aspect, the present application further provides a computer program product comprising a computer program which, when executed by a processor, implements the following steps: determining a power operation strategy of a target region in a scheduling period based on a carbon emission cost of the target region in the scheduling period represented by a grid power purchase of the target region in the scheduling period; obtaining an actual load demand of the target region in the scheduling period and a predicted photovoltaic output of the target region in the scheduling period; and in a case where an actual photovoltaic output of the target region under the power operation strategy does not match the predicted photovoltaic output, determining and scheduling a demand response device in the target region based on the actual photovoltaic output, the actual load demand and the power operation strategy, so that a load demand after the demand response device participates in demand response in the scheduling period matches the actual load demand.

[0010] The above-mentioned response scheduling method, device, computer equipment, computer readable storage medium and computer program product of the load demand of a region determine a power operation strategy of a target region in a scheduling period based on a carbon emission cost of the target region in the scheduling period represented by a grid power purchase of the target region in the scheduling period, obtain an actual load demand of the target region in the scheduling period and a predicted photovoltaic output of the target region in the scheduling period, and then in a case where an actual photovoltaic output of the target region under the power operation strategy does not match the predicted photovoltaic output, determine and schedule a demand response device in the target region based on the actual photovoltaic output, the actual load demand and the power operation strategy, so that a load demand after the demand response device participates in demand response in the scheduling period matches the actual load demand. Thus, the present application determines the power operation strategy of the target region by considering the carbon emission cost of the target region, combines the energy system and the carbon management system of the target region, and performs load demand scheduling of the target region under the synergy of the energy system and the carbon management system, which can improve the accuracy of demand response scheduling of the region and further improve the accuracy of energy distribution of the region. BRIEF DESCRIPTION OF DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related art, the drawings needed to be used in the description of the embodiments of the present application or the related art will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other related drawings can also be obtained without creative labor.

[0012] Figure 1 A flowchart of a response scheduling method of a load demand of a region in an embodiment;

[0013] Figure 2 A system architecture diagram of a collaborative management system in an embodiment;

[0014] Figure 3A flowchart of a method for responding to regional load demand in an embodiment;

[0015] Figure 4 A block diagram of an apparatus for responding to regional load demand in an embodiment;

[0016] Figure 5 An internal structure diagram of a computer device in an embodiment. DETAILED DESCRIPTION

[0017] In order to make the purposes, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only intended to explain the present application and not to limit the protection scope of the present application.

[0018] The embodiments of the technical solutions of the present application will be described in detail below with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solutions of the present application, and therefore only serve as examples, and cannot limit the protection scope of the present application.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs; the terms used herein are only for the purpose of describing specific embodiments and are not intended to limit the present application; the terms "include" and "have" and any variations thereof in the specification and claims of the present application and the above description of drawings are intended to cover non-exclusive inclusion.

[0020] Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the present application. The occurrence of the phrase "in one embodiment" at various locations in the specification does not necessarily all refer to the same embodiment, nor is it necessary that every embodiment include the particular feature, structure, or characteristic. It will be explicitly understood that the embodiments described herein can be combined with other embodiments.

[0021] The method for responding to regional load demand provided by the embodiments of the present application can be applied to a collaborative management system corresponding to energy management and carbon management. Specifically, by determining the carbon emission cost represented by the power grid purchase power of the target region in the scheduling period, the power operation strategy of the target region in the scheduling period is determined, and the actual load demand of the target region in the scheduling period and the predicted photovoltaic output of the target region in the scheduling period are obtained, and then in the case that the actual photovoltaic output and the predicted photovoltaic output of the target region under the power operation strategy do not match, the demand response device in the target region is determined and dispatched based on the actual photovoltaic output, the actual load demand and the power operation strategy, so that the load demand after the demand response device participates in the demand response in the scheduling period matches the actual load demand.

[0022] In an exemplary embodiment, as shown in Figure 1 A method for responding to regional load demand is provided, which is applied to a collaborative management system as an example, and includes the following steps:

[0023] S102, determining a power operation strategy of the target region in the scheduling period by a carbon emission cost of the target region in the scheduling period represented by a grid purchase power of the target region.

[0024] The target region can refer to a park, and the range of the park is not limited. The target region can also refer to a city, and the form of the target region is not limited.

[0025] The grid purchase power refers to the power purchased by the target region from the grid in the scheduling period, and the unit is kilowatt (kW). The carbon emission cost is used to represent the economic cost of the target region due to the emission of greenhouse gases, and the carbon emission cost can be represented by the grid purchase power of the target region in the scheduling period.

[0026] In an embodiment, the carbon emission cost of the target region in the scheduling period is represented by a carbon tax resource value, a grid carbon emission factor and a grid purchase power of the target region in the scheduling period.

[0027] For example, the carbon tax resource value is used to represent the carbon tax unit price, and the carbon emission cost satisfies:

[0028] P g,t =P grid,t ·τ t ·f grid

[0029] Wherein, P g,t represents the carbon emission cost of the target region in the tth scheduling period, P grid,t represents the grid purchase power of the target region in the tth scheduling period, which is non-negative; f grid represents the grid carbon emission factor, the unit is kg CO2 / kWh (i.e. the carbon dioxide emission per kilowatt-hour of electricity consumed), τ t represents the carbon tax unit price (yuan / kg) of the tth scheduling period.

[0030] The power operation strategy is used to represent the power strategy to be executed by the target region in the scheduling period. For example, the power operation strategy includes a target grid purchase power and a target energy storage power, the target energy storage power can include a target energy storage discharge power and a target energy storage charging power, the target region should purchase the power matched with the target grid purchase power from the grid in the scheduling period, and the target region should discharge the power matched with the target discharge power in the scheduling period, or the target region should charge the power matched with the target charging power in the scheduling period.

[0031] In one embodiment, the preset power operation strategy matching the carbon emission cost of the target region in the scheduling period can be determined as the power operation strategy of the target region in the scheduling period based on a mapping relationship between a preset carbon emission cost range and a preset power operation strategy.

[0032] S104, obtain an actual load demand of the target region in the scheduling period and a predicted photovoltaic output of the target region in the scheduling period.

[0033] The actual load demand is used to represent the total power demand of the target region in the scheduling period. The predicted photovoltaic output refers to the predicted photovoltaic output of the target region in the scheduling period. For example, the predicted photovoltaic output can be predicted based on historical photovoltaic output data of the target region.

[0034] In one embodiment, the historical actual photovoltaic output and the historical predicted photovoltaic output of the target region in the scheduling period are obtained. In the case that the historical predicted photovoltaic output does not match the historical actual photovoltaic output, a first preset adjustment coefficient and a second preset adjustment coefficient matching the target region are obtained. The sum of the first preset adjustment coefficient and the second preset adjustment coefficient is the same as the target coefficient. The historical actual photovoltaic output and the historical predicted photovoltaic output are weighted and summed based on the first preset adjustment coefficient and the second preset adjustment coefficient to obtain the predicted photovoltaic output of the target region in the scheduling period.

[0035] For example, the first preset adjustment coefficient is greater than the second preset adjustment coefficient. The product of the historical predicted photovoltaic output and the first preset adjustment coefficient is determined as the sum of the product of the historical actual photovoltaic output and the second preset adjustment coefficient. For example, the target coefficient is 1, and the second preset adjustment coefficient can be 0.2 or other values.

[0036] S106, in the case that the actual photovoltaic output of the target region under the power operation strategy does not match the predicted photovoltaic output, the demand response device in the target region is determined and dispatched based on the actual photovoltaic output, the actual load demand and the power operation strategy, so that the load demand after the demand response device participates in the demand response in the scheduling period matches the actual load demand.

[0037] The actual photovoltaic output refers to the actual active power output of the target region when the power operation strategy is executed. For example, in the case that the photovoltaic output deviation between the actual photovoltaic output and the predicted photovoltaic output is outside the preset deviation range, it is determined that the actual photovoltaic output does not match the predicted photovoltaic output.

[0038] In an embodiment, the collaborative management system comprises a data collection layer, the data collection layer is deployed with a photovoltaic inverter, and actual photovoltaic output of the target region when performing the power operation strategy in the scheduling period can be collected through the photovoltaic inverter.

[0039] The demand response device refers to an electrical device in the target region. For example, the demand response device includes, but is not limited to, a device for providing landscape lighting, a device for providing indoor lighting, an air conditioner, and the like.

[0040] In an embodiment, based on the actual photovoltaic output, the actual load demand, and the power operation strategy, the load demand deficiency can be determined, and the device in the target garden that matches the load demand deficiency is determined as the demand response device in the target region.

[0041] Based on Figure 1 As shown in the method, the carbon emission cost represented by the grid purchase power of the target region in the scheduling period is used to determine the power operation strategy of the target region in the scheduling period, the actual load demand of the target region in the scheduling period is obtained, and the predicted photovoltaic output of the target region in the scheduling period is predicted. In the case that the actual photovoltaic output of the target region in the scheduling period does not match the predicted photovoltaic output under the power operation strategy, the demand response device in the target region is determined and dispatched based on the actual photovoltaic output, the actual load demand, and the power operation strategy, so that the load demand after the demand response device in the target region participates in the demand response in the scheduling period matches the actual load demand. Therefore, the power operation strategy of the target region is determined by considering the carbon emission cost of the target region, so that the energy system and the carbon management system of the target region are combined, and the load demand of the target region is scheduled under the collaboration of the energy system and the carbon management system. The accuracy of the demand response scheduling of the region can be improved, and the accuracy of the energy distribution of the region can be further improved.

[0042] In an embodiment, before obtaining the carbon emission cost represented by the grid purchase power of the target region in the scheduling period, and determining the power operation strategy of the target region in the scheduling period, the method further comprises:

[0043] S100, based on the energy consumption of each energy type in the target region in the scheduling period and the carbon emission factor of each energy type, the carbon emission amount of the target region in the scheduling period is obtained.

[0044] In one embodiment, the collaborative management system comprises a data collection layer, the data collection layer is deployed with an electric carbon meter, the electric carbon meter can be deployed at each energy consumption node in the target area, such as a factory building, an office building, etc., and the electric carbon meter can collect data such as voltage, power and energy consumption in real time. The electric carbon meter has a built-in carbon emission factor library, and the carbon emission factor library stores carbon emission factors corresponding to each energy type. For example, the carbon emission factor of thermal power is 0.8 kg / kWh (the mass of carbon dioxide emitted per 1 kWh of consumed electric energy), and the carbon emission factor of green electricity is 0.3 kg / kWh.

[0045] In one embodiment, the carbon emission of the target area in the scheduling period satisfies:

[0046]

[0047] wherein E carbon represents the carbon emission, P i represents the energy consumption of the i-th energy type, f i represents the carbon emission factor corresponding to the i-th energy type. The energy types include electric energy (including green electricity, thermal power, etc.), heat, gas, water, etc.

[0048] For example, assuming that the calculation of the carbon emission is represented by electric energy (including green electricity, thermal power, etc.), wherein the thermal power consumption is 100 kWh, the carbon emission factor corresponding to the thermal power is 0.8 kg / kWh, the green electricity consumption is 200 kWh, and the corresponding carbon emission factor is 0.3 kg / kWh, then the carbon emission = 100 kWh x 0.8 kg / kWh + 200 kWh x 0.3 kg / kWh = 140 kg (kilograms).

[0049] In one embodiment, based on a plurality of energy nodes on a preset carbon flow path, a plurality of carbon emissions corresponding to the plurality of energy nodes at a plurality of detection times are obtained, and based on the plurality of carbon emissions, a carbon flow spectrum corresponding to the target area is generated. For example, based on the plurality of carbon emissions, a carbon emission heat map matching the preset carbon flow path is generated, the carbon emission heat map is superimposed on a geographic information system (GIS) map, and a carbon flow spectrum corresponding to the target area is obtained.

[0050] For example, the preset carbon flow path can be constructed based on a life cycle assessment (LCA) model. For example, the preset carbon flow path is photovoltaic, energy storage, A zone factory, etc. in sequence, the carbon emission heat map is superimposed on a GIS map, and the red area can represent high carbon emission (such as a coal-fired boiler room).

[0051] Wherein, the plurality of energy nodes can include source, net and load. For example, the source is used to represent photovoltaic or power grid, the net is used to represent power distribution cabinet, and the load is used to represent generating equipment.

[0052] S101, in the case where the carbon emission is greater than or equal to the carbon emission threshold corresponding to the target region, return to execute the step of determining the power operation strategy of the target region in the scheduling period by the carbon emission cost represented by the grid purchase power of the target region in the scheduling period.

[0053] In one embodiment, in the case where the carbon emission is greater than or equal to the carbon emission threshold corresponding to the target region, the warning information can be pushed, and the traceability report can be generated. Wherein, the carbon emission threshold corresponding to the target region can refer to the upper limit of the single-day carbon emission of the target region, such as the upper limit of 1000kg CO2 (i.e. 1 ton CO2 / day), and other settings can also be provided.

[0054] Exemplarily, the warning information can be pushed to the maintenance terminal in the form of short message, and / or the warning information can be pushed to the application program (APP) corresponding to the target region, so as to maintain in time.

[0055] Exemplarily, the target region can include a plurality of sub-regions, and the traceability report can include the region equipment with large energy consumption and the sub-region where the region equipment is located. For example, based on the traceability report, it can be found that the energy consumption of equipment A in B area is large, which needs to be optimized.

[0056] Based on the contents of S100-S101, the starting condition of regional load demand scheduling is determined from the perspective of carbon emission, so that the timeliness of load demand scheduling can be improved by combining the load scheduling process of the region with the carbon management process.

[0057] In one embodiment, the power operation strategy of the target region in the scheduling period is determined by the carbon emission cost represented by the grid purchase power of the target region in the scheduling period (i.e. S102), which includes the following steps:

[0058] S11, the energy storage loss cost of the target region is represented by the energy storage power of the target region in the scheduling period.

[0059] Wherein, the energy storage power includes energy storage discharge power and energy storage charging power. Exemplarily, the energy storage loss cost satisfies:

[0060]

[0061] Wherein, P s,t represents the energy storage loss cost of the target region in the tth scheduling period, represents the energy storage charging power of the target area in the tth scheduling period. η1 represents the charging loss coefficient, which represents the unit loss coefficient (yuan / kWh, the cost required for each kilowatt-hour of electricity consumed) when charging in the target area. It represents the energy storage discharge power of the target area in the tth scheduling period, η2 represents the discharge loss coefficient, and characterizes the unit loss coefficient (yuan / kWh) when the target area is discharged. The discharge loss coefficient is greater than the charging loss coefficient.

[0062] S12, characterizes the carbon emission cost and energy cost of the target area through the power grid purchase power of the target area during the scheduling period.

[0063] In one embodiment, the energy cost of the target area during the scheduling period is represented based on the power resource value during the scheduling period and the power grid purchased power of the target area during the scheduling period.

[0064] For example, the power resource value represents the electricity price, and the energy cost satisfies:

[0065] P n,t =P grid,t π t

[0066] Among them, P n,t represents the energy cost of the target area in the tth scheduling period, P grid,t represents the power purchased by the target area in the tth dispatch period, π t Represents the electricity price in the tth scheduling period (yuan / kWh).

[0067] The implementation method of the carbon emission cost may refer to the content adaptation description of S102.

[0068] S13, based on the energy storage loss cost, carbon emission cost and energy cost, construct an objective function for characterizing the total regional operation cost of the target area during the scheduling period.

[0069] Among them, the total regional operating cost is the sum of energy storage loss cost, carbon emission cost and energy cost.

[0070] For example, the total regional operating cost P t satisfy:

[0071] P t =P n,t +P g,t +P s,t

[0072] S14, solving the objective function, and obtaining a power operation strategy including a target energy storage power and a target grid purchase power while minimizing the total regional operation cost within the scheduling period.

[0073] The target energy storage power refers to the energy storage power when the total cost of the regional operation is minimized, and the target grid power purchase refers to the grid power purchase when the total cost of the regional operation is minimized.

[0074] Based on the content shown in S11-S14, the power operation strategy is determined from the perspective of the total cost of the regional operation of the target region, so that the accuracy of the load demand scheduling of the target region can be realized based on the power operation strategy while reducing the total cost of the regional operation.

[0075] In one embodiment, the target function can be solved under certain constraints to improve the accuracy of the solution of the target function. For example, the following steps are included:

[0076] S21, based on the relationship between the grid power purchase, the energy storage power, and the predicted photovoltaic output of the target region in the scheduling period, and the predicted load demand of the target region in the scheduling period, the power balance constraint condition is determined.

[0077] In one embodiment, the energy storage power includes energy storage charging power and energy storage discharging power. For example, the power balance constraint condition satisfies:

[0078]

[0079] P grid,t represents the grid power purchase of the target region in the tth scheduling period, represents the predicted photovoltaic output of the target region in the tth scheduling period, represents the energy storage discharging power of the target region in the tth scheduling period, represents the energy storage charging power of the target region in the tth scheduling period, represents the predicted load demand of the target region in the tth scheduling period.

[0080] For example, based on the historical load demand data of the target region, the predicted load demand is obtained.

[0081] S22, the target function is solved under the constraint of the power balance constraint condition, and the power operation strategy including the target energy storage power and the target grid power purchase is obtained when the total cost of the regional operation in the scheduling period is minimized.

[0082] Based on the content shown in S21-S22, by introducing the power balance constraint condition, when the target function is solved under the constraint of the power balance constraint condition, the accuracy of obtaining the target energy storage power can be improved.

[0083] In one embodiment, the energy storage power includes energy storage charging power and energy storage discharging power, and certain constraints can be configured for the energy storage charging power and the energy storage discharging power to improve the accuracy of solving the objective function. For example, the following steps can be included:

[0084] S31, determining a target state of charge of the target area in the scheduling period based on a historical state of charge of the target area in a previous scheduling period of the scheduling period, the energy storage discharging power and the energy storage charging power.

[0085] For example, the target state of charge satisfies:

[0086]

[0087] wherein E bat,t represents the target state of charge of the target area in the tth scheduling period, E bat,t-1 represents the historical state of charge of the target area in the previous scheduling period (t-1) of the scheduling period, η ch represents the energy storage charging efficiency, η dis represents the energy storage discharging efficiency, represents the energy storage charging power of the target area in the tth scheduling period, represents the energy storage discharging power of the target area in the tth scheduling period, and Δt represents the length of the scheduling period, in hours.

[0088] In some embodiments, when t is 1, i.e., the target area performs load demand scheduling for the first time, is a preset initial state of charge.

[0089] In some embodiments, the number of scheduling periods can be T, i.e., t = 1, 2, 3,..., T, and then E bat,T is the target state of charge of the Tth scheduling period, is a preset final state of charge.

[0090] S32, determining a state of charge constraint condition based on the target state of charge and a preset state of charge range.

[0091] For example, the state of charge constraint condition satisfies:

[0092]

[0093] wherein, represents a lower limit value of the preset state of charge range, represents an upper limit value of the preset state of charge range, and the upper limit value and the lower limit value can be in kWh, which is usually related to the capacity of the energy storage battery.

[0094] S33, determine a discharging power constraint condition based on the energy storage discharging power and the preset discharging power range, and determine a charging power constraint condition based on the energy storage charging power and the preset charging power range.

[0095] For example, the discharging power constraint condition satisfies:

[0096]

[0097] For example, the charging power constraint condition satisfies:

[0098]

[0099] wherein, represents an upper limit value of the preset charging power range, represents an upper limit value of the preset discharging power range.

[0100] S34, based on the energy storage discharging power and the upper limit value of the preset discharging power range, the energy storage charging power and the upper limit value of the preset charging power range, construct a limit charging and discharging constraint condition.

[0101] The limit charging and discharging constraint condition is used to avoid the target area from simultaneously charging and discharging.

[0102] For example, the limit charging and discharging constraint condition satisfies:

[0103]

[0104] S35, solve the target function under the constraints of the power balance constraint condition, the discharging power constraint condition, the charging power constraint condition and the limit charging and discharging constraint condition, and obtain a power operation strategy containing the target energy storage power and the target grid power purchase power when the total cost of the region operation in the dispatching period is minimized.

[0105] Based on the contents shown in S31-S35, in the case of energy storage power including energy storage charging power and energy storage discharging power, by introducing the discharging power constraint condition, the charging power constraint condition and the limit charging and discharging constraint condition, when solving the target function, the accuracy of obtaining the target energy storage power can be improved.

[0106] From the above, the above contents describe how to determine the power operation strategy of the target area in the dispatching period. The following contents will describe the process of realizing the load demand response scheduling of the target area based on the power operation strategy.

[0107] In one embodiment, based on the actual photovoltaic output, the actual load demand and the power operation strategy, the determination and scheduling of the demand response equipment in the target area (i.e. S106) is implemented, including the following steps:

[0108] S41, determine a power adjustment strategy matched with the target region based on the actual photovoltaic output and the actual load demand, to adjust the power operation strategy.

[0109] In one embodiment, the preset power adjustment strategy matched with the actual photovoltaic output and the actual load demand can be determined as the power adjustment strategy matched with the target region based on a mapping relationship among the preset actual photovoltaic output, the preset actual load demand, and the preset power adjustment strategy.

[0110] S42, determine and dispatch the demand response device in the target region based on the adjusted power operation strategy, the actual photovoltaic output, and the actual load demand.

[0111] In one embodiment, the power operation strategy includes a target energy storage power and a target grid power purchase.

[0112] For example, the current load of the target region in the dispatch period is obtained based on the actual photovoltaic output, the adjusted target energy storage power, and the adjusted target grid power purchase; the dispatch reward of the target region in the dispatch period is obtained based on the carbon emission weight coefficient, the carbon emission factor, and the adjusted target grid power purchase; and the demand response device matched with the load shortage between the actual load demand and the current load is determined and dispatched in the case that the dispatch reward is non-negative.

[0113] Specifically, the current load of the target region in the dispatch period is used to represent the power supply that the target region can currently provide. The current load is the sum of the actual photovoltaic output, the adjusted target energy storage power, and the adjusted target grid power purchase.

[0114] Specifically, the load shortage is the difference between the actual load demand and the current load. For example, if the power consumed by the device for landscape lighting matches the load shortage, the device for landscape lighting is turned off, so that the adjusted actual photovoltaic output of the target region is less than the unadjusted actual photovoltaic output, and the adjusted current load of the target region matches the actual load demand, so as to meet the supply-demand balance of the target region.

[0115] For example, the dispatch reward satisfies:

[0116] R t =-(ΔC energy +λ·ΔC carbon )-μ·V t

[0117] ΔC energy =(P′ grid,t -P grid,t )·π t

[0118] ΔCcarbon = f grid · τ t · (P' grid,t - P grid,t )

[0119] wherein, R t represents the scheduling reward of the target area in the tth scheduling period, P' grid,t represents the adjusted grid power purchase of the target area in the tth scheduling period, λ represents the carbon emission weight coefficient. V t represents the constraint violation penalty value, if the target state of charge of the target area in the tth scheduling period is outside the preset state of charge range, the constraint violation penalty value is non-0, otherwise 0; μ is the penalty weight coefficient.

[0120] Based on the content shown in S41-S42, by determining the load shortage, the demand response equipment in the target area can be accurately determined, and the demand response equipment is scheduled in the case of non-negative scheduling reward. It can be explained that the demand response process after scheduling the demand response equipment is a positive feedback process, so that the timeliness and accuracy of the load demand response of the target area can be improved.

[0121] In one embodiment, based on the actual photovoltaic output and the actual load demand, a power adjustment strategy matched with the target area is determined to adjust the power operation strategy (i.e., S41), including the following steps:

[0122] S411, in the case of taking the actual photovoltaic output, the actual load demand, the target state of charge of the target area in the scheduling period and the carbon tax resource value, and the prediction deviation absolute value between the actual photovoltaic output and the predicted photovoltaic output as the model input, updating the preset reinforcement learning model to obtain a target reinforcement learning model matched with the target area.

[0123] Illustratively, based on the actual photovoltaic output, the actual load demand, the target state of charge of the target area in the scheduling period and the carbon tax resource value, and the prediction deviation absolute value between the actual photovoltaic output and the predicted photovoltaic output, a target state space vector matched with the target area is constructed; the state space of the preset reinforcement learning model is updated based on the target state space vector, and the updated reinforcement learning model is determined as a target reinforcement learning model matched with the target area.

[0124] S412, at least two groups of preset action space vectors matched with the target area are obtained; each group of preset action space vectors includes an energy storage power adjustment amount and a grid power purchase adjustment amount.

[0125] Illustratively, the preset action space vector satisfies:

[0126] a t = (ΔP, ΔPgrid )

[0127] wherein, a t represents a preset action space vector, ΔP represents an energy storage power adjustment amount, and ΔP grid represents a grid power purchase power adjustment amount.

[0128] It should be noted that the target energy storage power includes a target energy storage discharge power and a target energy storage charging power, and the equipment in the target area will not discharge and charge at the same time, so when the target energy storage discharge power is 0, the target energy storage charging power is non-0, and further, the preset action space vector includes an energy storage power adjustment amount and a grid power purchase power adjustment amount, specifically, the preset action space vector includes an energy storage charging power adjustment amount and a grid power purchase power adjustment amount. Similarly, when the target energy storage discharge power is non-0, the target energy storage charging power is 0, and further, the preset action space vector includes an energy storage power adjustment amount and a grid power purchase power adjustment amount, specifically, the preset action space vector includes an energy storage discharge power adjustment amount and a grid power purchase power adjustment amount.

[0129] S413, input each preset action space vector into the target reinforcement learning model to obtain an expected value corresponding to each preset action space vector.

[0130] Illustratively, the target reinforcement learning model includes a policy network (Deep Q-Network, DQN), and based on the policy network, each preset action space vector is decided to obtain an expected value corresponding to each preset action space vector.

[0131] S414, obtaining a power adjustment strategy including an energy storage power adjustment amount and a grid power purchase power adjustment amount in the target vector; the target vector is a preset action space vector corresponding to the maximum expected value in the expected values.

[0132] In one embodiment, the power operation strategy is adjusted based on the power adjustment strategy. Illustratively, the power adjustment strategy includes a target energy storage power and a target grid power purchase power, the power adjustment strategy includes an energy storage power adjustment amount and a grid power purchase power adjustment amount, and the sum of the target energy storage power and the energy storage power adjustment amount is taken as an adjusted target energy storage power, and the sum of the target grid power purchase power and the grid power purchase power adjustment amount is taken as an adjusted target grid power purchase power.

[0133] Based on the content shown in S411-S414, by obtaining the power adjustment strategy including the energy storage power adjustment amount and the grid power purchase power adjustment amount in the target vector from the reinforcement learning model, the accuracy of the obtained power adjustment strategy can be improved.

[0134] In one embodiment, the following steps can also be included:

[0135] S51, in the case of scheduling a demand response device, based on the post-scheduling state of charge, determine the constraint penalty parameter matched with the next scheduling period of the scheduling period; the post-scheduling state of charge is determined based on the target state of charge of the target region in the scheduling period and the adjusted target energy storage power.

[0136] Exemplarily, the constraint penalty parameter satisfies:

[0137]

[0138] Wherein, P l represents the constraint penalty parameter matched with the next scheduling period of the scheduling period, E represents the post-scheduling state of charge, represents the lower limit value of the preset state of charge range, represents the upper limit value of the preset state of charge range.

[0139] Wherein, the post-scheduling state of charge can be described based on the content of determining the target state of charge in S31.

[0140] S52, based on the constraint penalty parameter and the scheduling reward of the target region in the scheduling period, update the target reinforcement learning model to obtain a new target reinforcement learning model, so as to determine the power adjustment strategy of the target region in the next scheduling period based on the new target reinforcement learning model.

[0141] Wherein, the content of obtaining the scheduling reward of the target region in the scheduling period can be described with reference to the content in S42.

[0142] Based on the content shown in S51-S52, the target reinforcement learning model is updated by combining the constraint penalty parameter and the scheduling reward, so that the scheduling accuracy of the load demand of the target region can be further improved on the basis of the optimization model.

[0143] In one embodiment, the predicted photovoltaic output of the target region in the next scheduling period of the scheduling period can also be obtained based on the actual photovoltaic output and the predicted photovoltaic output of the target region in the scheduling period. Specifically, the first preset adjustment coefficient and the second preset adjustment coefficient matched with the target region are obtained; the sum of the first preset adjustment coefficient and the second preset adjustment coefficient is the same as the target coefficient; the actual photovoltaic output and the predicted photovoltaic output are weighted and summed based on the first preset adjustment coefficient and the second preset adjustment coefficient, to obtain the predicted photovoltaic output of the target region in the next scheduling period of the scheduling period.

[0144] Exemplarily, the predicted photovoltaic output of the next scheduling period satisfies:

[0145]

[0146] Wherein, a predicted photovoltaic output of the target region in a next scheduling period (t+1) of the tth scheduling period, a predicted photovoltaic output of the target region in a next scheduling period (t+1) of the tth scheduling period, an actual photovoltaic output of the target region in the tth scheduling period, and a is a second preset adjustment coefficient, a is 0.2 or other values.

[0147] In combination with the above, as shown in Figure 2 a schematic diagram of a system architecture for cooperatively managing synergy is provided, wherein the system architecture for cooperatively managing synergy includes a data acquisition layer, an algorithm optimization layer, and an application layer.

[0148] The data acquisition layer is deployed with an electric carbon meter, and devices such as a photovoltaic inverter, an energy storage controller, and a heat sensor. The electric carbon meter (i.e., an electric power carbon emission meter) is an intelligent device integrating electric power metering and carbon emission accounting functions. The electric carbon meter can be deployed at each energy consumption node of the target region, such as a factory building, an office building, etc., to collect electric energy data such as voltage, power, current, and energy consumption in real time. The electric carbon meter has a built-in carbon emission factor library that stores carbon emission factors corresponding to each energy type. The photovoltaic inverter is used to collect photovoltaic power generation data and power generation data of the target region. The energy storage controller is used to collect state of charge data of the energy storage battery of the target region, receive charge and discharge instructions, and control the battery action. The heat sensor is used to collect data such as steam consumption and hot water consumption of the target region. Then, each data is uploaded to the algorithm optimization layer based on a communication mode of a Long Range Wide Area Network (LoRaWAN). The data includes data for calculating carbon emissions.

[0149] The algorithm optimization layer includes a data preprocessing module, an instruction issuing module, and a multi-energy flow dynamic optimization core algorithm. The data preprocessing module is deployed with a data cleaning unit and a protocol conversion gateway. The protocol conversion gateway is used to convert each data in the form of Modbus, RS485, etc. uploaded by the data acquisition layer into JSON format, such as {"device":"PV1","power":150}). Then, the data cleaning unit is used to clean each data in JSON format, such as eliminating abnormal values (e.g., negative power or out-of-range data), to input the cleaned data into a carbon emission calculation module, calculate carbon emission results, and write them into a database. The multi-energy flow dynamic optimization core algorithm can be configured with a Mixed-Integer Linear Programming (MILP) solver and a reinforcement learning model. The MILP solver is used to solve the objective function, and the reinforcement learning model is used to determine the power adjustment strategy.

[0150] The application layer includes a carbon image generation module, an environment (Environment), a social (Social Responsibility) and a governance (Governance) corresponding report (i.e. ESG report) generation module, and a user terminal. The carbon image generation module is deployed with a carbon flow atlas visualization interface for visualizing and displaying the carbon flow atlas corresponding to the target region based on the carbon flow atlas visualization interface after generating the carbon flow atlas corresponding to the target region. The carbon image generation module is also deployed with an LCA model engine for calculating a plurality of full-chain carbon emissions of the target region, combining a plurality of energy nodes corresponding to the carbon emission amount at a plurality of detection times to generate a carbon flow atlas, and marking a high-carbon emission area (such as red for high emission and green for low emission) in the form of a heat map on a GIS map. The instruction issuing module is used to send a control instruction to a control device corresponding to a demand response device in the target region, so that the control device controls the demand response device to perform an action matched with the control instruction. For example, the control instruction is used to instruct to cut off the power supply of the demand response device.

[0151] The ESG report generation module includes a rule engine and a blockchain storage interface, the rule engine is used to encode the battery regulations and other standards into SQL query logic, automatically extract data and generate PDF reports, and attach the blockchain hash value (using SHA-256 algorithm) obtained based on the blockchain storage interface on the PDF report to ensure tamper-proofing. Wherein, the type of rule engine is not limited, for example, the rule engine can be Drools engine or other types of engine. For example, by checking whether the generated ESG report contains the field required by the EU battery regulations (such as "supply chain carbon footprint traceability"), and checking whether the report is attached with a blockchain hash value (such as an Ethereum transaction ID), it can be determined whether the produced ESG report is accurate.

[0152] The user terminal is used to receive the early warning information pushed when the carbon emission amount of the target region in the scheduling period is greater than or equal to the carbon emission amount threshold corresponding to the target region, and is also used to display a management interface to manage the carbon emission and energy of the target region.

[0153] Referring to the foregoing content, in one embodiment, as shown in Figure 3 A flowchart of a regional load demand response scheduling method is provided, which is taken as an example of the method applied to a collaborative management system corresponding to energy management and carbon management, including the following steps:

[0154] S302, the energy storage loss cost of the target region is represented by the energy storage power of the target region in the scheduling period.

[0155] S304, the carbon emission cost and energy cost of the target region are represented by the grid power purchase power of the target region in the scheduling period.

[0156] S306, based on the energy storage loss cost, the carbon emission cost and the energy cost, a target function for representing a total cost of regional operation of the target region in the scheduling period is constructed.

[0157] S308, a target constraint condition matched with the target function is constructed.

[0158] The target constraint condition includes a power balance constraint condition, a discharging power constraint condition, a charging power constraint condition and a limit charging and discharging constraint condition. The content of the target constraint condition can refer to the foregoing content.

[0159] S310, the target function is solved under the constraint of the target constraint condition, and when the total cost of regional operation in the scheduling period is minimized, a power operation strategy containing a target energy storage power and a target grid power purchase power is obtained.

[0160] S312, actual load demand of the target region in the scheduling period and predicted photovoltaic output of the target region in the scheduling period are obtained.

[0161] S314, in a case where actual photovoltaic output under the power operation strategy executed in the target region does not match the predicted photovoltaic output, the actual photovoltaic output, the actual load demand, a target state of charge of the target region in the scheduling period and a carbon tax resource value, and a predicted deviation absolute value between the actual photovoltaic output and the predicted photovoltaic output are taken as model inputs, a preset reinforcement learning model is updated to obtain a target reinforcement learning model matched with the target region.

[0162] S316, at least two groups of preset action space vectors matched with the target region are obtained; each group of preset action space vectors includes an energy storage power adjustment amount and a grid power purchase power adjustment amount.

[0163] S318, each preset action space vector is input into the target reinforcement learning model to obtain an expected value corresponding to each preset action space vector.

[0164] S320, a power adjustment strategy containing the energy storage power adjustment amount and the grid power purchase power adjustment amount in a target vector is obtained to adjust the power operation strategy; the target vector is a preset action space vector corresponding to a maximum expected value among the expected values.

[0165] S322, based on the actual photovoltaic output, the adjusted target energy storage power and the adjusted target grid power purchase power, a current load of the target region in the scheduling period is obtained.

[0166] S324, based on the load shortage between the actual load demand and the current load, determining and scheduling a demand response device in the target region matched with the load shortage, so that the load demand after the demand response device participates in the demand response in the scheduling period matches the actual load demand.

[0167] The contents of S302-S324 can be adapted to the foregoing description.

[0168] Referring to the foregoing description, the following will describe an example of scheduling devices in a target region. Taking the starting time of the scheduling period as 8:00 and the length of the scheduling period as 15 minutes as an example, it is assumed that the predicted photovoltaic output is 150kW, the predicted load demand is 200kW, the real-time electricity price is 1.2 yuan / kWh, the carbon tax unit price is 0.6 yuan / kg, the carbon emission factor of the power grid is 0.8 kg / kWh, the energy storage charging efficiency and the energy storage discharging efficiency are both 95%, the lower limit value of the preset state of charge range is 100kWh (i.e., 20%), the upper limit value of the preset state of charge range is 450kWh (i.e., 90%), and the upper limit value of the preset discharging power range and the preset charging power range is 50kW. In the case of ignoring the battery loss cost, the total cost of the target region is the sum of the energy cost and the carbon tax cost, i.e., the target function is:

[0169] P t =1.2×P grid,1 +0.6×P grid,1 ×0.8

[0170] The power balance constraint condition is:

[0171]

[0172] The state of charge constraint condition is:

[0173]

[0174] Wherein, 0.25h is the hour number converted from 15 minutes (i.e., 15 / 60=0.25h).

[0175] The state of charge constraint condition is:

[0176] 100≤E bat,1 ≤450

[0177] The discharging power constraint condition is:

[0178]

[0179] The charging power constraint condition is:

[0180]

[0181] The limit charging and discharging constraint condition is:

[0182]

[0183] Solving the objective function under the above constraint conditions, the solution result is:

[0184]

[0185] Based on the above solution result, the target power grid power purchase power is 30 kW, the target energy storage discharging power is 20 kW, and the target energy storage charging power is 0 kW, so that the power operation strategy of the target area can be obtained.

[0186] Further, in the case of E bat,0 = 150, the target state of charge of the target area is:

[0187]

[0188] The total area operation cost of the target area is:

[0189] P t = 1.2 * 30 + 0.6 * 30 * 0.8 = 50.4 yuan

[0190] In the case of executing the power operation strategy of the target area, it is assumed that the actual photovoltaic output is detected to be 120 kW at 8:02, which is 30 kW lower than the predicted photovoltaic output of 150 kW; and the actual load demand is detected to be 210 kW, which is 10 kW higher than the predicted load demand of 200 kW. If the target area continues to work according to the previous power operation strategy, the power shortage is 40 kW, i.e. power shortage = (210 kW - 200 kW) + (150 kW - 120 kW).

[0191] Based on this, at this time, the power grid power purchase power needs to be increased, and the increased power grid power purchase power is 70 kW, i.e. increased power grid power purchase power = 30 kW + 40 kW = 70 kW, then the total area operation cost P t is 117.6 yuan, i.e. P t = 1.2 * 70 + 0.6 * 70 * 0.8 = 117.6 yuan, the cost increases by 67.2 yuan (i.e. 117.6 yuan - 50.4 yuan = 67.2 yuan).

[0192] In view of this, the actual photovoltaic output, the actual load demand, the target state of charge, the carbon tax unit price, and the prediction deviation absolute value between the actual photovoltaic output and the predicted photovoltaic output are taken as model inputs, the preset reinforcement learning model is updated, and a target reinforcement learning model matched with the target region is obtained. At least two groups of preset action space vectors matched with the target region are obtained, and the preset action space vector includes the energy storage discharge power adjustment amount and the grid power purchase power adjustment amount. For example, it is assumed that the preset action space vector matched with the target region is shown in Table 1, wherein:

[0193] Table 1

[0194] a t ]]> Expected value (-10,10) 82.1 (+5,-5) 85.1 √ (0,0) 67.2 (-5,+5) 45.6 (-10,+10) 32.1

[0195] It can be known from Table 1 that the energy storage discharge power adjustment amount and the grid power purchase power adjustment amount in the target vector are “+5” and “-5” respectively, so that the adjusted target energy storage discharge power is 25 kW, that is, 20 kW+5 kW=25 kW, and the adjusted target grid power purchase power is 30 kW-5 kW=25 kW.

[0196] Further, the current load of the target region is 170 kW, that is, 25 kW (that is, the adjusted grid power purchase power)+120 kW (that is, the actual photovoltaic output)+25 kW (that is, the adjusted energy storage discharge power)=170 kW. Compared with the actual load demand 210 kW, the load shortage is 40 kW, that is, 210 kW-170 kW=40 kW. The actual load demand can be met by switching the non-critical load, for example, cutting off the power consumed by the equipment for landscape lighting. In the case of cutting off the non-critical load, the total cost P t of the target region is 42 yuan, that is, P t =1.2×25+0.6×25×0.8=42 yuan.

[0197] In the case of no constraint violation and the carbon emission weight coefficient being 1, the scheduling reward R t is:

[0198] R t =-(42-50.4)-0=8.4

[0199] Therefore, although 40 kW of load is cut off, this part of load is not punished (that is, it can be defined as interruptible load), and the total cost of the region operation is reduced by 75.6 yuan compared with 117.6 yuan when there is no adjustment.

[0200] Wherein, the constraint penalty parameter of the next scheduling period is:

[0201]

[0202] The target state of charge of the next scheduling period is:

[0203]

[0204] Therefore, the target state of charge of the next scheduling period is also within the preset state of charge range, and thus the service constraint violation penalty value.

[0205] wherein the predicted photovoltaic output of the next scheduling period is:

[0206]

[0207] In combination with the above, the method provided by the application can realize collaborative scheduling of energy and carbon emissions, solve the data integration problem through the electricity-carbon meter and carbon flow atlas, adapt international standards through the rule engine, and form a "collection-optimization-output" closed-loop management to solve the problems of disconnection between energy management and carbon management, low efficiency of international standard adaptation, and difficulty in cross-department data integration. Among them, through the "software and hardware collaboration + rule engine" architecture, the application can deeply integrate energy scheduling, carbon emission tracking and international standard adaptation to form a closed-loop management. Specifically, in a first aspect, by optimizing energy and carbon emissions collaboratively, the overall efficiency can be improved, and energy data such as photovoltaic and energy storage can be integrated. The MILP solver and reinforcement learning algorithm are combined to realize dynamic scheduling, and the carbon emission cost is included in the objective function. By real-time coordination of the interaction relationship of the energy supply side (source), the transmission and distribution network (network), the energy load (load) and the energy storage device (storage), the minute-level scheduling of multiple energy forms is realized, which significantly improves the regional comprehensive energy utilization efficiency, reduces the carbon emission intensity, improves the accuracy of demand response scheduling in the target area, and realizes the balance of "source-network-load-storage" supply and demand. In a second aspect, an integrated metering device (electricity-carbon meter) is used. The electricity-carbon meter and the standard adaptation engine are integrated with current sensors and a multi-energy carbon emission factor library to track the correlation between energy consumption and carbon emissions in real time. Through the collaborative design of hardware and software, the energy consumption data is mapped to carbon emissions in real time. Combined with the life cycle model, a visual carbon flow atlas is generated. Therefore, based on the real-time data and the dynamic distribution of carbon emissions generated by the life cycle model, the high-carbon emission link is located and the optimization path is recommended. Further, high-precision carbon emission monitoring and tracing are realized to support second-level response to abnormal carbon emission warning and improve management timeliness. In a third aspect, the international standards (such as the European Union battery regulations) are coded into executable logic based on the rule engine to realize automatic generation of reports and compliance verification, and reduce the barriers to cross-border trade of enterprises. That is, the system architecture provided by the application has an international standard adaptation module built-in, which converts regulatory requirements into automated rules to directly generate compliance reports and carbon labels, thereby significantly shortening the compliance review period of enterprise export products and reducing the cost of manual intervention.

[0208] It should be understood that, although the various steps in the flowcharts involved in the various embodiments described above are displayed in sequence according to the instructions of the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the flowcharts involved in the various embodiments described above can include multiple steps or multiple stages, and these steps or stages are not necessarily executed and completed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of steps or stages in other steps.

[0209] Based on the same inventive concept, embodiments of the present application also provide a regional load demand response scheduling device for implementing the aforementioned regional load demand response scheduling method. The solution provided by this device is similar to the solution described in the aforementioned method. Therefore, the specific limitations of one or more regional load demand response scheduling device embodiments provided below can be found in the aforementioned limitations of the regional load demand response scheduling method and will not be further elaborated here.

[0210] In an exemplary embodiment, Figure 4 As shown, a regional load demand response scheduling device is provided, including: a processing module 402, an acquisition module 404 and a scheduling module 406, wherein: the processing module 402 is used to determine the power operation strategy of the target area during the scheduling period through the carbon emission cost represented by the power purchased from the power grid of the target area during the scheduling period; the acquisition module 404 is used to obtain the actual load demand of the target area during the scheduling period, and the predicted photovoltaic output of the target area during the scheduling period; the scheduling module 406 is used to determine and schedule the demand response equipment in the target area based on the actual photovoltaic output, actual load demand and power operation strategy when the actual photovoltaic output under the power operation strategy executed in the target area does not match the predicted photovoltaic output, so that the load demand after the demand response equipment participates in the demand response during the scheduling period matches the actual load demand.

[0211] In one embodiment, the processing module 402 is further used to: characterize the energy storage loss cost of the target area through the energy storage power of the target area during the scheduling period; characterize the carbon emission cost and energy cost of the target area through the power purchased from the power grid of the target area during the scheduling period; construct an objective function for characterizing the total regional operation cost of the target area during the scheduling period based on the energy storage loss cost, carbon emission cost and energy cost; solve the objective function to obtain a power operation strategy including the target energy storage power and the target power purchased from the power grid while minimizing the total regional operation cost during the scheduling period.

[0212] In one of the embodiments, the processing module 402 is further configured to determine a power balance constraint condition based on the grid purchase power, the energy storage power, and a relationship between the predicted photovoltaic output of the target region in the dispatch period and the predicted load demand of the target region in the dispatch period; and solve the objective function under the constraint of the power balance constraint condition to obtain the power operation strategy including the target energy storage power and the target grid purchase power when the total cost of the region operation in the dispatch period is minimized.

[0213] In one of the embodiments, the energy storage power includes an energy storage discharge power and an energy storage charge power; the processing module 402 is further configured to determine a target state of charge of the target region in the dispatch period based on a historical state of charge of the target region in a previous dispatch period of the dispatch period, the energy storage discharge power, and the energy storage charge power; determine a state of charge constraint condition based on the target state of charge and a preset state of charge range; determine a discharge power constraint condition based on the energy storage discharge power and a preset discharge power range, and determine a charge power constraint condition based on the energy storage charge power and a preset charge power range; construct a charge and discharge limiting constraint condition based on an upper limit value of the preset discharge power range, an upper limit value of the preset charge power range, the energy storage discharge power, and the energy storage charge power; and solve the objective function under the constraint of the power balance constraint condition, the discharge power constraint condition, the charge power constraint condition, and the charge and discharge limiting constraint condition to obtain the power operation strategy including the target energy storage power and the target grid purchase power when the total cost of the region operation in the dispatch period is minimized.

[0214] In one of the embodiments, the dispatch module 406 is further configured to determine a power adjustment strategy matched with the target region based on the actual photovoltaic output and the actual load demand to adjust the power operation strategy; and determine and dispatch the demand response equipment in the target region based on the adjusted power operation strategy, the actual photovoltaic output, and the actual load demand.

[0215] In one of the embodiments, the dispatch module 406 is further configured to update a preset reinforcement learning model to obtain a target reinforcement learning model matched with the target region in a case where the actual photovoltaic output, the actual load demand, the target state of charge of the target region in the dispatch period, and a predicted deviation absolute value between the actual photovoltaic output and the predicted photovoltaic output are taken as model inputs; obtain at least two groups of preset action space vectors matched with the target region; each preset action space vector includes an energy storage power adjustment amount and a grid purchase power adjustment amount; input each preset action space vector into the target reinforcement learning model to obtain an expected value corresponding to each preset action space vector respectively; and obtain the power adjustment strategy including the energy storage power adjustment amount and the grid purchase power adjustment amount in the target vector; the target vector is a preset action space vector corresponding to a maximum expected value among the expected values.

[0216] In one of the embodiments, the power operation strategy includes the target energy storage power and the target grid power purchase power; the scheduling module 406 is further configured to: in the case of scheduling the demand response device, determine a constraint penalty parameter matched with a next scheduling period of the scheduling period based on the post-scheduling state of charge; the post-scheduling state of charge is determined based on the target state of charge of the target region in the scheduling period and the adjusted target energy storage power; and update the target reinforcement learning model based on the constraint penalty parameter and the scheduling reward of the target region in the scheduling period to obtain a new target reinforcement learning model, so as to determine the power adjustment strategy of the target region in the next scheduling period based on the new target reinforcement learning model.

[0217] In one of the embodiments, the power operation strategy includes the target energy storage power and the target grid power purchase power; the scheduling module 406 is further configured to: obtain the current load of the target region in the scheduling period based on the actual photovoltaic output, the adjusted target energy storage power and the adjusted target grid power purchase power; obtain the scheduling reward of the target region in the scheduling period based on the carbon emission weight coefficient, the carbon emission factor and the adjusted target grid power purchase power; and in the case that the scheduling reward is non-negative, determine and schedule the demand response device matched with the load shortage between the actual load demand and the current load in the target region.

[0218] In one of the embodiments, the processing module 402 is further configured to: obtain a first preset adjustment coefficient and a second preset adjustment coefficient matched with the target region; the sum of the first preset adjustment coefficient and the second preset adjustment coefficient is the same as the target coefficient; and perform weighted summation on the actual photovoltaic output and the predicted photovoltaic output based on the first preset adjustment coefficient and the second preset adjustment coefficient to obtain the predicted photovoltaic output of the target region in the next scheduling period of the scheduling period.

[0219] In one of the embodiments, the processing module 402 is further configured to: obtain the carbon emission amount of the target region in the scheduling period based on the energy consumption amount of each energy type corresponding to the target region in the scheduling period and the carbon emission factor corresponding to each energy type; and in the case that the carbon emission amount is greater than or equal to the carbon emission threshold value corresponding to the target region, return to execute the step of determining the power operation strategy of the target region in the scheduling period by the carbon emission cost represented by the grid power purchase power of the target region in the scheduling period.

[0220] The above-mentioned modules in the response scheduling device for the regional load demand can be all or partially realized by software, hardware and combinations thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in hardware form, or can be stored in the memory in the computer device in software form, so as to be called and executed by the processor to perform the operations corresponding to the above-mentioned modules.

[0221] In an example embodiment, a computer device, which can be a server, is provided, and an internal structure diagram of the computer device can be as shown in FIG. 1. Figure 5 The computer device includes a processor, a memory, an input / output interface (I / O), and a communication interface. The processor, the memory, and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for running the operating system and the computer program in the non-volatile storage medium. The database of the computer device is configured to store time photovoltaic output and other data. The input / output interface of the computer device is configured to exchange information between the processor and external devices. The communication interface of the computer device is configured to communicate with terminals outside through a network connection. The computer program is executed by the processor to implement a method for responding to regional load demand.

[0222] Those skilled in the art can understand that Figure 5 The structure shown in the above embodiment is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device can include more or fewer components than those shown in the diagram, or combine certain components, or have a different arrangement of components.

[0223] In an example embodiment, a computer device is provided, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps in the above method embodiments.

[0224] In an example embodiment, a computer readable storage medium is provided, which stores a computer program. The computer program is executed by a processor to implement the steps in the above method embodiments.

[0225] In an example embodiment, a computer program product is provided, which includes a computer program. The computer program is executed by a processor to implement the steps in the above method embodiments.

[0226] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the user or authorized by all parties, and the collection, use, and processing of the related data need to comply with relevant regulations.

[0227] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when executed, can include the processes of the above-mentioned embodiment methods. Any reference to memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. The non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical storage, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. The volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration but not limitation, the RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The database involved in the embodiments provided in the present application can include at least one of a relational database and a non-relational database. The non-relational database can include a distributed database based on a block chain, etc., without being limited thereto. The processor involved in the embodiments provided in the present application can be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., without being limited thereto.

[0228] The technical features of the above embodiments can be combined arbitrarily. In order to make the description simple, all possible combinations of the technical features in the above embodiments are not described, however, as long as the combinations of the technical features do not exist contradictory, they should be considered as the scope of the present application.

[0229] The above-described embodiments are merely illustrative of several embodiments of the present application, and the description is relatively specific and detailed, but should not be understood as a limitation on the scope of the patent. It should be noted that for those skilled in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, which are all within the scope of the present application. Therefore, the scope of protection of the present application should be subject to the appended claims.

Claims

1. A method for responding to regional load demand, characterized in that: The method comprises: Determining a power operation strategy for the target area during the scheduling period based on the carbon emission cost represented by the power purchased from the power grid in the target area during the scheduling period; Obtaining an actual load demand of the target area during the scheduling period and a predicted photovoltaic output of the target area during the scheduling period; In a case where the actual photovoltaic output under the power operation strategy executed in the target area does not match the predicted photovoltaic output, the demand response equipment in the target area is determined and scheduled based on the actual photovoltaic output, the actual load demand and the power operation strategy, so that the load demand of the demand response equipment after participating in the demand response during the scheduling period matches the actual load demand.

2. The method according to claim 1, characterized in that The determining of the power operation strategy of the target area during the scheduling period based on the carbon emission cost represented by the power purchased from the power grid of the target area during the scheduling period includes: Characterizing the energy storage loss cost of the target area by the energy storage power of the target area during the scheduling period; The carbon emission cost and energy cost of the target area are represented by the power purchased from the power grid of the target area during the scheduling period; constructing an objective function for characterizing the total regional operation cost of the target area during the scheduling period based on the energy storage loss cost, the carbon emission cost, and the energy cost; The objective function is solved to obtain a power operation strategy including a target energy storage power and a target power purchased from the power grid while minimizing the total regional operation cost within the scheduling period.

3. The method according to claim 2, characterized in that The method further comprises: Determining a power balance constraint condition based on a relationship between the power purchased from the power grid, the energy storage power, the predicted photovoltaic output of the target area during the scheduling period, and the predicted load demand of the target area during the scheduling period; The solving of the objective function to obtain a power operation strategy including the energy storage power and the power purchased from the power grid while minimizing the total regional operation cost within the scheduling period includes: The objective function is solved under the constraints of the power balance constraint condition, and when the total regional operation cost within the scheduling period is minimized, a power operation strategy including the target energy storage power and the target grid purchase power is obtained.

4. The method according to claim 3, characterized in that The energy storage power includes energy storage discharge power and energy storage charging power; the method further includes: determining a target state of charge of the target area during the scheduling period based on a historical state of charge of the target area in a previous scheduling period, the energy storage discharge power, and the energy storage charging power; Determining a state of charge constraint condition based on the target state of charge and a preset state of charge range; Determining a discharge power constraint based on the energy storage discharge power and a preset discharge power range, and determining a charging power constraint based on the energy storage charging power and a preset charging power range; Constructing a charge and discharge restriction constraint based on the energy storage discharge power and the upper limit of the preset discharge power range, and the energy storage charging power and the upper limit of the preset charging power range; Solving the objective function under the constraints of the power balance constraint condition, and obtaining a power operation strategy including the target energy storage power and the target grid purchased power when the total regional operation cost within the scheduling period is minimized, includes: The objective function is solved under the constraints of the power balance constraint, the discharge power constraint, the charging power constraint, and the limited charge and discharge constraint, and when the total regional operation cost of the scheduling period is minimized, a power operation strategy including the target energy storage power and the target grid purchase power is obtained.

5. The method according to claim 1, wherein The determining and scheduling of the demand response device in the target area based on the actual photovoltaic output, the actual load demand and the power operation strategy includes: Determining a power adjustment strategy that matches the target area based on the actual photovoltaic output and the actual load demand, so as to adjust the power operation strategy; Demand response equipment in the target area is determined and dispatched based on the adjusted power operation strategy, the actual photovoltaic output, and the actual load demand.

6. The method according to claim 5, characterized in that The determining of a power adjustment strategy based on the actual photovoltaic output and the actual load demand includes: updating a preset reinforcement learning model to obtain a target reinforcement learning model that matches the target area, taking the actual photovoltaic output, the actual load demand, the target state of charge and carbon tax resource value of the target area during the scheduling period, and the absolute value of the predicted deviation between the actual photovoltaic output and the predicted photovoltaic output as model inputs; Obtain at least two sets of preset action space vectors matching the target area; each set of preset action space vectors includes an energy storage power adjustment amount and a grid-purchased power adjustment amount; Inputting each of the preset action space vectors into a target reinforcement learning model to obtain an expected value corresponding to each of the preset action space vectors; A power adjustment strategy including the energy storage power adjustment amount and the grid purchase power adjustment amount in a target vector is obtained; the target vector is a preset action space vector corresponding to the maximum expected value among the expected values.

7. The method according to claim 6, characterized in that The power operation strategy includes a target energy storage power and a target power purchased from the power grid; the method further includes: In the case of scheduling the demand response device, determining a constraint penalty parameter that matches a next scheduling period of the scheduling period based on a post-scheduling state of charge; the post-scheduling state of charge is determined based on a target state of charge of the target area in the scheduling period and the adjusted target energy storage power; Based on the constraint penalty parameter and the scheduling reward of the target area in the scheduling period, the target reinforcement learning model is updated to obtain a new target reinforcement learning model, so as to determine the power adjustment strategy of the target area in the next scheduling period based on the new target reinforcement learning model.

8. The method according to claim 5, characterized in that The power operation strategy includes target energy storage power and target grid purchased power; The determining and scheduling of the demand response device in the target area based on the adjusted power operation strategy, the actual photovoltaic output, and the actual load demand includes: Obtaining a current load of the target area during the scheduling period based on the actual photovoltaic output, the adjusted target energy storage power, and the adjusted target grid-purchased power; Obtaining a dispatch reward for the target area during the dispatch period based on the carbon emission weight coefficient, the carbon emission factor, and the adjusted target grid purchased power; When the dispatch reward is non-negative, based on the load shortfall between the actual load demand and the current load, a demand response device in the target area that matches the load shortfall is determined and dispatched.

9. The method according to claim 1, characterized in that The method further comprises: Obtaining a first preset adjustment coefficient and a second preset adjustment coefficient that match the target area; the sum of the first preset adjustment coefficient and the second preset adjustment coefficient is the same as the target coefficient; Based on the first preset adjustment coefficient and the second preset adjustment coefficient, a weighted sum is performed on the actual photovoltaic output and the predicted photovoltaic output to obtain the predicted photovoltaic output of the target area in the next scheduling period of the scheduling period.

10. The method according to any one of claims 1 to 9, characterized in that The method further comprises: Obtaining the carbon emissions of the target area during the scheduling period based on the energy consumption corresponding to each energy type in the target area during the scheduling period and the carbon emission factors corresponding to each energy type; When the carbon emissions are greater than or equal to the carbon emissions threshold corresponding to the target area, return to the step of executing the carbon emission cost represented by the power purchased from the power grid of the target area during the scheduling period to determine the power operation strategy of the target area during the scheduling period.

11. A regional load demand response dispatching device, characterized in that: The device comprises: a processing module, configured to determine a power operation strategy for the target area during the scheduling period based on the carbon emission cost represented by the power purchased from the power grid of the target area during the scheduling period; an acquisition module, configured to acquire an actual load demand of the target area during the scheduling period and a predicted photovoltaic output of the target area during the scheduling period; A scheduling module is used to determine and schedule the demand response equipment in the target area based on the actual photovoltaic output, the actual load demand and the power operation strategy when the actual photovoltaic output under the power operation strategy executed in the target area does not match the predicted photovoltaic output, so that the load demand after the demand response equipment participates in the demand response during the scheduling period matches the actual load demand.

12. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 10 are implemented.

13. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 10 are implemented.

14. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 10 are implemented.

Images