Virtual power plant-oriented multi-objective optimization model construction, regulation and control method and system

Abstract

The invention provides a virtual power plant-oriented multi-objective optimization model construction and regulation method and system, and is applied to the technical field of virtual power plant operation optimization. The method comprises the steps of dividing distributed resources in a virtual power plant into continuous adjustment type resources, graded adjustment type resources and switch adjustment type resources based on adjustment characteristics of the distributed resources in the virtual power plant; the aggregation cost and the aggregation income of various adjustment resources aggregated in the process that the virtual power plant aggregator participates in the power market auxiliary service are considered, and a multi-objective optimization objective function is constructed by minimizing the aggregation cost and maximizing the aggregation income; and determining constraint conditions based on the adjustment capability and the adjustment rule constraint of various adjustment resources, the production safety constraint and the power grid operation constraint so as to construct a virtual power plant-oriented multi-target optimization model. The method solves the problems that the participation mode cannot be optimized and the ability of an aggregator to participate in the auxiliary service is relatively low due to the fact that the transaction model of the virtual power plant participating in the auxiliary service market is insufficient in fineness.

Description

Technical Field

[0001] The present invention relates to the technical field of virtual power plant operation optimization, and specifically relates to a method and system for constructing a multi-objective optimization model and regulation for a virtual power plant. Background Art

[0002] In the environment of the new power system, the penetration rate of new energy will gradually increase, and the gap of flexible peak shaving resources will be further expanded. It is necessary to combine the diversified development trend presented by the power sales side market, and comprehensively consider the technical maturity, interest demands and clean contributions of traditional units and emerging market players, so as to design an auxiliary market mechanism and commercial interaction model considering the participation of demand-side resources for constructing an open, shared and competitive auxiliary market ecosystem. A virtual power plant is a system that aggregates scattered demand-side resources to participate in power market transactions and power system operation.

[0003] At present, the construction of the trading model for virtual power plants to participate in the auxiliary service market is not fine enough, resulting in the inability of grid enterprises and cloud-side aggregators to objectively evaluate the value of virtual power plants participating in auxiliary services, thus unable to optimize the participation method, and resulting in a low ability of aggregators to participate in the auxiliary service market. Summary of the Invention

[0004] In order to overcome the problems of inability to optimize the participation method and low ability of aggregators to participate in auxiliary services caused by the insufficient fineness of the trading model for virtual power plants to participate in the auxiliary service market, the present invention provides a method and system for constructing a multi-objective optimization model and regulation for a virtual power plant.

[0005] On the one hand, the present invention provides a method for constructing a multi-objective optimization model for a virtual power plant, including:

[0006] Based on the regulation characteristics of distributed resources in the virtual power plant, the distributed resources in the virtual power plant are divided into continuously adjustable resources, hierarchically adjustable resources and switchable adjustable resources;

[0007] Considering the aggregation cost and aggregation revenue of various types of adjustable resources aggregated during the process of virtual power plant aggregators participating in power market auxiliary services, a multi-objective optimization objective function is constructed to minimize the aggregation cost and maximize the aggregation revenue;

[0008] Based on the regulation capabilities and regulation rule constraints of various types of adjustable resources in the virtual power plant, the production safety constraints of the virtual power plant and the grid operation constraints, the constraint conditions of the multi-objective optimization objective function are determined;

[0009] Based on the multi-objective optimization objective function and the constraint conditions, a multi-objective optimization model for a virtual power plant is constructed.

[0010] Optionally, the continuously adjustable resources include distributed energy and continuously adjustable loads with continuously adjustable power consumption. The distributed energy includes energy storage devices and distributed power sources; the stepwise adjustable resources are stepwise adjustable loads with stepwise and graded adjustable power consumption; the switching adjustable resources are switching adjustable loads with adjustable power consumption of 0 or rated power.

[0011] Optionally, considering the aggregation costs and aggregation revenues of various types of adjustable resources aggregated during the participation of the virtual power plant aggregator in the ancillary services of the power market, a multi-objective optimization objective function is constructed to minimize the aggregation costs and maximize the aggregation revenues, including:

[0012] Based on the regulation operation costs of each distributed energy source, the regulation response costs of various types of adjustable loads, the regulation deviation default costs of the virtual power plant, and the external power purchase costs of the virtual power plant during the participation of the virtual power plant in the ancillary services of the power market, determine the aggregation costs of the virtual power plant aggregator participating in the ancillary services of the power market;

[0013] Based on the revenues from the participation of the virtual power plant in the ancillary services of the power market and the capacity revenues corresponding to the adjustable capacities of various types of adjustable resources aggregated by the virtual power plant aggregator, determine the aggregation revenues of the virtual power plant aggregator participating in the ancillary services of the power market;

[0014] Construct the multi-objective optimization objective function by minimizing the aggregation costs and maximizing the aggregation revenues;

[0015] Among them, the external power purchase costs of the virtual power plant are the costs of purchasing electricity from the power grid when the distributed power sources in the virtual power plant cannot meet the load demand or for economic optimization objectives; various types of adjustable loads include the continuously adjustable loads, the stepwise adjustable loads, and the switching adjustable loads.

[0016] Optionally, the determining the aggregation revenues of the virtual power plant aggregator participating in the ancillary services of the power market based on the revenues from the participation of the virtual power plant in the ancillary services of the power market and the capacity revenues corresponding to the adjustable capacities of various types of adjustable resources aggregated by the virtual power plant aggregator includes:

[0017] Based on the regulated electricity volume and real-time electricity price corresponding to the participation of the virtual power plant aggregator in the ancillary services of the power market, determine the revenues from the participation of the virtual power plant in the ancillary services of the power market;

[0018] Based on the upward real-time capacity price and the upward adjustable power of the virtual power plant, determine the corresponding upward capacity revenue;

[0019] Based on the downward real-time capacity price and the downward adjustable power of the virtual power plant, determine the corresponding downward capacity revenue;

[0020] Based on the upward capacity revenue and the downward capacity revenue, determine the aggregation revenues of the virtual power plant aggregator participating in the ancillary services of the power market;

[0021] Among them, the adjustable-up power and the adjustable-down power are respectively obtained by solving the corresponding sub-optimization problems, and the sub-optimization problems are constructed with the objective of maximizing the difference between the adjustable capacity of the virtual power plant and the corresponding power of the spot market clearing, and with the constraint conditions as the constraints.

[0022] Optionally, the regulation operation costs of the distributed energy sources include the energy storage regulation operation cost and the downward compensation cost of the distributed photovoltaic. The expression of the multi-objective optimization objective function is:

[0023]

[0024] In the formula, C ess is the energy storage device regulation operation cost in the virtual power plant, C pv is the downward compensation cost of the distributed photovoltaic in the virtual power plant, C Load is the regulation response cost of various adjustable loads in the virtual power plant, C Pun is the regulation deviation default cost of the virtual power plant, C buy is the external power purchase cost of the virtual power plant, R aux is the income of the virtual power plant participating in the ancillary service regulation of the power market, R cap is the capacity income of the virtual power plant aggregator participating in the ancillary service of the power market;

[0025]

[0026] In the formula, C loss,i is the regulation operation cost of the i-th energy storage device, N ess is the total number of aggregated energy storage devices in the virtual power plant, μ loss is the battery loss coefficient corresponding to the unit power exchange, P ess,i (t) is the output power of the power converter of the i-th energy storage at time t; Δt is the acquisition time interval, and T is the optimization period;

[0027]

[0028] In the formula, C pv,j is the downward compensation cost of the distributed photovoltaic, N pv is the total number of aggregated distributed photovoltaics in the virtual power plant; P pv,j (t) is the output power of the j-th distributed photovoltaic and its power converter at time t, is the maximum output capacity of the j-th distributed photovoltaic at each moment;

[0029]

[0030] In the formula, C LoadCont,i'is the regulation response cost of the i'-th continuously adjustable load, C LoadLev,i” is the regulation response cost of the i''-th stepwise adjustable load, C LoadSwt,i”' is the regulation response cost of the i'''-th switchable adjustable load, N LoadCont 、N LoadLev 、N LoadSwt are the total numbers corresponding to the continuously adjustable load, stepwise adjustable load, and switchable adjustable load aggregated by the virtual power plant respectively; is the load regulation benchmark for the i'-th continuously adjustable load during production and operation at time t, P LoadCont,i' (t) is the power consumption of the i'-th continuously adjustable load at time t, is the production operation cost loss coefficient of the i'-th continuously adjustable load; is the load regulation benchmark for the i''-th stepwise adjustable load during production and operation at time t, P LoadLev,i” (t) is the power consumption of the i''-th stepwise adjustable load at time t, is the production operation cost loss coefficient of the i''-th stepwise adjustable load; is the load regulation benchmark for the i'''-th switchable adjustable load during production and operation at time t, P LoadSwt,i”' (t) is the power consumption of the i'''-th switchable adjustable load at time t, is the production operation cost loss coefficient of the i'''-th switchable adjustable load;

[0031]

[0032] In the formula, λ is the penalty coefficient; P AGG (t) is the real-time output of the virtual power plant aggregator after aggregating resources at time t; ΔP exc (t) is the regulation command deviation at time t; are the lower limit and upper limit of the allowable deviation of the real-time command respectively; q e (t) is the real-time electricity price at time t; {T aux} is the set of moments of the auxiliary service regulation period;

[0033]

[0034] In the formula, P buy (t) is the real-time power cleared in the day-ahead spot market; ΔP buy (t) is the deviation power during the intraday non-regulation period; ξ is the real-time deviation electricity price coefficient;

[0035]

[0036] In the formula, α is the electricity price revenue multiplier for participating in the auxiliary service regulation; P base(t) is the power generation and consumption baseline of the virtual power plant;

[0037]

[0038] Where: w up (t) is the upward real-time capacity price; w down (t) is the downward real-time capacity price; {Res} is the set of the constraint conditions; is the upward adjustable power at time t, is the downward adjustable power at time t.

[0039] Optionally, determining the constraint conditions of the multi-objective optimization objective function based on the regulation capabilities and regulation rule constraints of various regulation resources in the virtual power plant, the production safety constraints of the virtual power plant, and the grid operation constraints includes:

[0040] Determining the regulation capabilities and regulation rule constraints of various regulation resources in the virtual power plant based on the ramping constraints of each distributed energy source and continuously adjustable load in the virtual power plant, the response delay constraints of the hierarchical adjustable load and switchable adjustable load in the virtual power plant, and the regulation duration of various regulation resources in the virtual power plant;

[0041] Determining the production safety constraints of the virtual power plant based on the regulation power constraints of each energy storage in the virtual power plant and the regulation timing constraints of the hierarchical adjustable load;

[0042] Determining the grid operation constraints based on the power balance constraints and power flow constraints between the aggregated resources of the virtual power plant aggregator and the purchased power;

[0043] Taking the regulation capabilities and regulation rule constraints of various regulation resources in the virtual power plant, the production safety constraints of the virtual power plant, and the grid operation constraints as the constraint conditions of the multi-objective optimization objective function.

[0044] Optionally, the expression of the ramping constraints of each distributed energy source and continuously adjustable load is:

[0045]

[0046] Where, P Cont,m (t), P Cont,m (t - 1) are the device powers corresponding to the m-th continuously adjustable resource at the current time and the previous time respectively, and the continuously adjustable resources include energy storage devices, distributed power sources, and continuously adjustable loads; λ UP,m , λ down,m are the upward and downward ramping capabilities corresponding to the m-th continuously adjustable resource respectively;

[0047] The expression of the response delay constraints of the hierarchical adjustable load is:

[0048] t2 -t 1 ≥nΔt;

[0049] Wherein, t 2 is the regulation command response time of the stepwise adjustable load, t 1 is the regulation command release time of the stepwise adjustable load, and n is the number of delay intervals;

[0050] The expression of the response delay constraint of the switchable adjustable load is:

[0051]

[0052] Wherein, t 3 is the adjustment end time of the switchable adjustable load; t' 2 is the regulation command response time of the switchable adjustable load; n 1 is the minimum continuous interval number, n 2 is the maximum continuous interval number;

[0053] The expression of the adjustment timing constraint of the stepwise adjustable load is:

[0054]

[0055] Wherein, P Lev (t Lev,1 ) is the power of the stepwise adjustable load at the first stepwise adjustment timing, P Lev (t Lev,s ) is the power of the stepwise adjustable load at the s-th stepwise adjustment timing, P Lev (t Lev,N ) is the power of the stepwise adjustable load at the N-th stepwise adjustment timing, N is the number of stepwise adjustment timings; Δt Lev,s is the time interval between adjacent steps; n Lev is the minimum time interval number of stepwise adjustment;

[0056] The expression of the power balance constraint between the aggregated resources and the purchased power is:

[0057] P AGG (t) = P buy (t) + ΔP buy (t);

[0058] Wherein, P AGG (t) is the real-time output at time t after the virtual power plant aggregator aggregates resources; P buy (t) is the real-time power cleared by the day-ahead spot market; ΔP buy (t) is the deviation power during the non-regulation period within the day.

[0059] Optionally, after constructing a multi-objective optimization model for a virtual power plant based on the multi-objective optimization objective function and the constraint conditions, the following steps are further included:

[0060] Use the sampling method to discretize the continuous variables corresponding to each continuous regulation type of resource in the multi-objective optimization model.

[0061] On the other hand, the present invention also provides a multi-objective optimization regulation method for a virtual power plant, including:

[0062] Obtain the day-ahead output data of various types of regulation resources in the virtual power plant;

[0063] Solve the multi-objective optimization model constructed above based on the day-ahead output data of various types of regulation resources in the virtual power plant to obtain the intra-day regulation strategies of various types of regulation resources when the virtual power plant participates in ancillary service regulation;

[0064] Conduct operation regulation of the virtual power plant based on the intra-day regulation strategies of various types of regulation resources.

[0065] On the other hand, the present invention also provides a multi-objective optimization model construction system for a virtual power plant, including:

[0066] A resource classification module, configured to divide the distributed resources in the virtual power plant into continuous regulation type resources, hierarchical regulation type resources, and switching regulation type resources based on the regulation characteristics of the distributed resources in the virtual power plant;

[0067] An objective function construction module, configured to consider the aggregation costs and aggregation revenues of various types of regulation resources aggregated during the process of the virtual power plant aggregator participating in the electricity market ancillary services, and construct a multi-objective optimization objective function to minimize the aggregation cost and maximize the aggregation revenue;

[0068] A constraint construction module, configured to determine the constraint conditions of the multi-objective optimization objective function based on the regulation capabilities and regulation rule constraints of various types of regulation resources in the virtual power plant, the production safety constraints of the virtual power plant, and the grid operation constraints;

[0069] An optimization model construction module, configured to construct a multi-objective optimization model for a virtual power plant based on the multi-objective optimization objective function and the constraint conditions.

[0070] On the other hand, the present invention also provides a multi-objective optimization regulation system for a virtual power plant, including:

[0071] A data acquisition module, configured to obtain the day-ahead output data of various types of regulation resources in the virtual power plant;

[0072] A model solving module, configured to solve the multi-objective optimization model constructed above based on the day-ahead output data of various regulation resources in the virtual power plant, and obtain the intra-day regulation strategies of various regulation resources when the virtual power plant participates in ancillary service regulation;

[0073] A regulation module, configured to perform operation regulation of the virtual power plant based on the intra-day regulation strategies of various regulation resources.

[0074] On the other hand, the present invention also provides an electronic device, including: at least one processor and a memory; the memory and the processor are connected through a bus;

[0075] The memory is used to store one or more programs;

[0076] When the one or more programs are executed by the at least one processor, the multi-objective optimization model construction method or regulation method for a virtual power plant described in any one of the above is implemented.

[0077] On the other hand, the present invention also provides a readable storage medium, on which an execution program is stored, and when the execution program is executed, the multi-objective optimization model construction method or regulation method for a virtual power plant described in any one of the above is implemented.

[0078] Compared with the prior art, the beneficial effects of the present invention are:

[0079] A multi-objective optimization model construction method for a virtual power plant provided by the present invention abstracts the regulation capabilities into three categories: continuous regulation, hierarchical regulation, and switching regulation according to the regulation characteristics of different types of resources on the demand side. Based on the regulation capability types, the distributed resources in the virtual power plant are divided into continuous regulation resources, hierarchical regulation resources, and switching regulation resources. Based on this resource classification, all adjustable resources on the demand side can be basically covered, so as to realize the modeling of all adjustable resources in the virtual power plant, improve the refinement degree of the model for the virtual power plant to participate in the ancillary service market, and thus enhance the ability of the virtual power plant aggregator to participate in ancillary service regulation.

[0080] Based on the aggregation costs and aggregation revenues of various regulation resources aggregated in the virtual power plant, the present invention constructs a multi-objective optimization objective function and constraint conditions for various regulation resources with the goal of minimizing the aggregation cost and maximizing the aggregation revenue, realizes the multi-objective optimization modeling of the optimization problem of the virtual power plant aggregator participating in ancillary service regulation, can take into account various regulation resources aggregated in the virtual power plant, accurately evaluate the value of the virtual power plant participating in ancillary services, and improve the revenue and participation enthusiasm of the virtual power plant aggregator participating in ancillary services through the optimization of the participation method, thereby optimizing resource utilization. Description of the Drawings

[0081] Figure 1Flow schematic diagram of a method for constructing a multi-objective optimization model for a virtual power plant according to the present invention;

[0082] Figure 2 Flow schematic diagram of the discretization process of continuous variables according to the present invention;

[0083] Figure 3 Structural schematic diagram of a multi-objective optimization model constructed by an example according to the present invention;

[0084] Figure 4 Structural schematic diagram of the electronic device according to the present invention. Detailed implementation manners

[0085] The following further elaborates on the detailed implementation manners of the present invention with reference to the accompanying drawings.

[0086] Example 1

[0087] A method for constructing a multi-objective optimization model for a virtual power plant provided by the present invention, as shown in the schematic diagram Figure 1 as follows, including:

[0088] Step S110: Based on the regulation characteristics of distributed resources in the virtual power plant, divide the distributed resources in the virtual power plant into continuously adjustable resources, hierarchically adjustable resources, and switchable adjustable resources;

[0089] Step S120: Considering the aggregation costs and aggregation revenues of various types of regulation resources aggregated during the participation of virtual power plant aggregators in electricity market ancillary services, construct a multi-objective optimization objective function to minimize the aggregation cost and maximize the aggregation revenue;

[0090] Step S130: Based on the regulation capabilities and regulation rule constraints of various types of regulation resources in the virtual power plant, the production safety constraints of the virtual power plant, and the grid operation constraints, determine the constraint conditions of the multi-objective optimization objective function;

[0091] Step S140: Based on the multi-objective optimization objective function and the constraint conditions, construct a multi-objective optimization model for the virtual power plant.

[0092] In this exemplary embodiment, the virtual power plant aggregator can be an operator on the cloud side, and the virtual power plant can be edge-side resources. The distributed resources aggregated within the virtual power plant can include distributed power sources (such as solar energy, wind energy), energy storage devices, controllable loads, electric vehicles, and other resources. The regulation characteristics include the regulation method. According to the regulation characteristics of different types of resources on the demand side, the regulation capabilities can be abstracted into three categories: continuous regulation, hierarchical regulation, and switching regulation. Correspondingly, the distributed resources within the virtual power plant are divided into continuous regulation type resources, hierarchical regulation type resources, and switching regulation type resources. The continuous regulation type resources can include distributed energy and continuously adjustable loads with continuously adjustable power consumption. The distributed energy includes energy storage devices and distributed power sources. The continuously adjustable loads can be loads such as electric vehicle charging and swapping stations, variable frequency motors, etc. The hierarchical regulation type resources are loads with power consumption regulated in hierarchical levels and steps, that is, hierarchically adjustable loads, generally electric heating type loads (i.e., regulated by means of series-parallel structures of electric heating wires, changing the number of loop resistors, etc.), such as electric heating for heating, electric heating furnaces, water heater loads, etc. The switching regulation type resources are loads with power consumption adjustable to 0 or the rated power, that is, switchable adjustable loads. Most of the power adjustable loads without refined flexibility transformation are switchable adjustable loads. The power consumption of such loads is only in two states: 0 or the rated power.

[0093] Exemplarily, based on the divided resource types, decision variables of an optimization model are designed for the process of the virtual power plant aggregator aggregating various resources to participate in the ancillary services of the power market. Specifically as follows:

[0094] 1) For the aggregated energy storage devices, the corresponding optimization decision variable is the active power output by each energy storage device. Define the power output by the i-th energy storage device and its power converter at time t as P ess,i (t), where a positive value represents discharging and a negative value represents charging. This decision variable is a continuous regulation variable, and the variable value range is:

[0095]

[0096] where is the maximum output capacity of the i-th energy storage device.

[0097] 2) For the aggregated distributed photovoltaics, the corresponding optimization decision variable is the active power output by each distributed photovoltaic. Define the power output by the j-th distributed photovoltaic and its power converter at time t as P pv,j (t), where a positive value represents power generation. This decision variable is a continuous regulation variable, and the variable value range is:

[0098]

[0099] where is the maximum output capacity of the j-th distributed photovoltaic at time t.

[0100] 3) For the aggregated adjustable loads on the user side, they can be discussed in three cases.

[0101] The first type is continuously adjustable loads, and the corresponding decision variable is the power consumption of continuously adjustable loads. Define P LoadCont,i’ (t) as the power consumption of the i'-th continuously adjustable load at time t. This decision variable is a continuous adjustment variable, and the variable value range is:

[0102]

[0103] Among them, is the rated operating power of the i'-th continuously adjustable load.

[0104] The second type is stepwise adjustable loads, and the corresponding decision variable is the power consumption of stepwise adjustable loads. Define P LoadLev,i” (t) as the power consumption of the i''-th stepwise adjustable load at time t. It is a discrete variable, and the variable value range is:

[0105]

[0106] Among them, is the power consumption of the i''-th stepwise adjustable load at the Nm-th adjustment step.

[0107] The third type of load is switchable adjustable loads, and its corresponding decision variable is the power consumption of switchable adjustable loads, which is. Define the power consumption of the i'''-th switchable adjustable load at time t as P LoadSwt,i”' (t). It is a discrete variable, and the variable value range is:

[0108]

[0109] Among them, is the rated power consumption of the i'''-th switchable adjustable load.

[0110] In this example, the distributed resources in the virtual power plant are divided into continuously adjustable resources, stepwise adjustable resources, and switchable adjustable resources based on the type of adjustment ability. Based on this resource classification, all adjustable resources on the demand side can be basically covered, so as to realize the modeling of all adjustable resources in the virtual power plant, improve the refinement degree of the constructed model, and enhance the ability of the virtual power plant aggregator to participate in ancillary service regulation.

[0111] In an exemplary implementation manner, the aggregating cost and aggregating revenue of various types of adjustable resources aggregated during the process of the virtual power plant aggregator participating in the ancillary services of the power market in S120 are considered, and a multi-objective optimization objective function is constructed to minimize the aggregating cost and maximize the aggregating revenue, including:

[0112] Based on the regulation operation costs of various distributed energy resources, the regulation response costs of various adjustable loads, the regulation deviation default costs of the virtual power plant, and the external power purchase costs of the virtual power plant when the virtual power plant participates in the auxiliary service regulation of the power market, determine the aggregation cost of the virtual power plant aggregator participating in the auxiliary service of the power market;

[0113] Based on the revenue of the virtual power plant participating in the auxiliary service regulation of the power market and the capacity revenue corresponding to the adjustable capacity of various regulation resources aggregated by the virtual power plant aggregator, determine the aggregation revenue of the virtual power plant aggregator participating in the auxiliary service of the power market;

[0114] Construct the multi-objective optimization objective function by minimizing the aggregation cost and maximizing the aggregation revenue.

[0115] In the present exemplary embodiment, a model objective function, i.e., a multi-objective optimization objective function, is constructed based on various costs and revenues in the process of the virtual power plant aggregator participating in the auxiliary service of the power market. Various costs (aggregation costs) include the regulation operation costs of various distributed energy resources in the virtual power plant, the regulation response costs of various adjustable loads, the regulation deviation default costs of the virtual power plant, and the external power purchase costs of the virtual power plant. The regulation operation costs of various distributed energy resources include the energy storage regulation operation costs and the downward adjustment compensation costs of distributed photovoltaics. The energy storage regulation operation costs are the battery losses brought by the charge and discharge of the energy storage device; the downward adjustment compensation costs of distributed photovoltaic operation refer to the fees paid to users to compensate for their original rated output capacity after the output of distributed photovoltaics is reduced. The original rated output capacity can be obtained by fitting the simulation data of a certified photovoltaic simulator. The regulation response costs of various adjustable loads refer to the losses incurred when the aggregator aggregates power users to participate in the auxiliary service load regulation and the normal production of products by users is affected due to the regulation of various adjustable loads; various adjustable loads include continuously adjustable loads, stepwise adjustable loads, and switchable adjustable loads. The regulation deviation default costs of the virtual power plant refer to the price penalty paid by the virtual power plant that won the bid in the auxiliary service market the day before for excessive command execution deviation when executing the regulation command during the execution period. The external power purchase costs of the virtual power plant are the costs of purchasing power from the grid when the distributed power sources in the virtual power plant cannot meet the load demand or for the economic optimization goal, that is, when the aggregator / operator is operating, the internal distributed power sources cannot meet the load demand or for the goal of achieving the optimal economy, the cost of purchasing power from the large power grid. The external power purchase costs can include the day-ahead spot power purchase costs and the intra-day real-time spot deviation power purchase costs. The above realizes the fine description of the regulation cost of the virtual power plant participating in the power auxiliary service regulation by considering the aggregation costs of various regulation resources in the virtual power plant from multiple dimensions.

[0116] On this basis, this example also considers various revenues obtained by the virtual power plant participating in power auxiliary service regulation, that is, the aggregated revenue. The aggregated revenue includes the revenue from the virtual power plant participating in the auxiliary service regulation of the power market and the capacity revenue corresponding to the adjustable capacity of the virtual power plant aggregator aggregating various types of regulation resources. The revenue from participating in the auxiliary service regulation of the power market refers to the revenue obtained by the virtual power plant when participating in the regulating auxiliary service market by responding to the regulation instructions of the power grid operation agency. The main calculation principle of this revenue is to pay for the regulating electricity actually generated by the virtual power plant aggregator / operator, and this part of the revenue can be reflected in the form of electricity sales revenue. The capacity revenue refers to the revenue that the adjustable capacity of the user-side energy storage, distributed photovoltaic, and adjustable load aggregated by the virtual power plant aggregator / operator can obtain in the capacity market. By constructing the objective function from the two dimensions of revenue and cost, the value of the virtual power plant participating in the auxiliary service of the power market can be evaluated more accurately.

[0117] Exemplarily, determining the aggregated revenue of the virtual power plant aggregator participating in the auxiliary service of the power market based on the revenue from the virtual power plant participating in the auxiliary service regulation of the power market and the capacity revenue corresponding to the adjustable capacity of the virtual power plant aggregator aggregating various types of regulation resources includes:

[0118] Determining the revenue of the virtual power plant participating in the auxiliary service regulation of the power market based on the regulating electricity volume and real-time electricity price corresponding to the virtual power plant aggregator participating in the auxiliary service regulation of the power market;

[0119] Determining the corresponding upward adjustment capacity revenue based on the upward adjustment real-time capacity price and the upward adjustable power of the virtual power plant;

[0120] Determining the corresponding downward adjustment capacity revenue based on the downward adjustment real-time capacity price and the downward adjustable power of the virtual power plant;

[0121] Determining the aggregated revenue of the virtual power plant aggregator participating in the auxiliary service of the power market based on the upward adjustment capacity revenue and the downward adjustment capacity revenue.

[0122] In this exemplary embodiment, the regulated electricity quantity corresponding to the virtual power plant aggregator's participation in the auxiliary service regulation of the power market can be the difference between the real-time output after participating in the auxiliary service regulation and the power generation and consumption baseline of the virtual power plant. The power generation and consumption baseline can adopt the day-ahead spot clearing curve. The revenue from participating in the auxiliary service regulation is calculated based on the regulated electricity quantity corresponding to the auxiliary service and the real-time electricity price. The capacity revenue can include the upward regulation capacity revenue and the downward regulation capacity revenue. The upward regulation capacity revenue is determined by the upward real-time capacity price and the upward adjustable power of the virtual power plant; the downward regulation capacity revenue is determined by the downward real-time capacity price and the downward adjustable power of the virtual power plant. Among them, the upward adjustable power and the downward adjustable power can be obtained by solving the corresponding sub-optimization problems, and the sub-optimization problem is constructed with the goal of maximizing the difference between the adjustable capacity of the virtual power plant and the corresponding power of the spot clearing, subject to the constraint conditions. That is to say, calculating the capacity revenue requires additionally solving two sub-optimization problems of the upward adjustable power and the downward adjustable power. The solution of this sub-optimization problem can be carried out simultaneously with the multi-objective optimization model, that is, the solution of the liberalization problem is completed during the model solution process.

[0123] Exemplarily, the multi-objective optimization objective function can be expressed by the following formula:

[0124]

[0125] In the formula, C ess is the regulation operation cost of the energy storage device in the virtual power plant, C pv is the downward compensation cost of the distributed photovoltaic in the virtual power plant, C Load is the regulation response cost of various adjustable loads in the virtual power plant, C Pun is the regulation deviation default cost of the virtual power plant, C buy is the external power purchase cost of the virtual power plant, R aux is the revenue of the virtual power plant participating in the auxiliary service regulation of the power market, R cap is the aggregation revenue of the virtual power plant aggregator participating in the auxiliary service of the power market;

[0126]

[0127] In the formula, C loss,i is the regulation operation cost of the i-th energy storage device, N ess is the total number of aggregated energy storage devices in the virtual power plant, μ loss is the battery loss coefficient corresponding to the unit electricity quantity exchange, P ess,i (t) is the output power of the power converter of the i-th energy storage at time t; Δt is the acquisition time interval, and T is the optimization period;

[0128]

[0129] In the formula, Cpv,j To reduce the compensation cost of distributed PV, N pv is the total number of distributed PVs aggregated in the virtual power plant; P pv,j (t) is the output power of the j-th distributed PV and its power converter at time t, and is the maximum output capacity of the j-th distributed PV at each moment.

[0130] To ensure accurate external response to the auxiliary service regulation command, it is necessary to utilize the adjustable load regulation ability to respond to the demand for increasing or decreasing the load. The adjustable load response cost is the loss incurred when the aggregator aggregates power users to participate in the auxiliary service load regulation, which affects the normal production of users due to the load regulation. Specifically as follows:

[0131]

[0132] In the formula, C LoadCont,i' is the regulation response cost of the i'-th continuously adjustable load, C LoadLev,i” is the regulation response cost of the i''-th stepwise adjustable load, C LoadSwt,i”' is the regulation response cost of the i'''-th switchable adjustable load, N LoadCont 、N LoadLev 、N LoadSwt are the corresponding total numbers of continuously adjustable loads, stepwise adjustable loads, and switchable adjustable loads aggregated in the virtual power plant respectively; is the load regulation reference for the production operation of the i'-th continuously adjustable load at time t, P LoadCont,i' (t) is the power consumption of the i'-th continuously adjustable load at time t, is the production operation cost loss coefficient of the i'-th continuously adjustable load; is the load regulation reference for the production operation of the i''-th stepwise adjustable load at time t, P LoadLev,i” (t) is the power consumption of the i''-th stepwise adjustable load at time t, is the production operation cost loss coefficient of the i''-th stepwise adjustable load; is the load regulation reference for the production operation of the i'''-th switchable adjustable load at time t, P LoadSwt,i”' (t) is the power consumption of the i'''-th switchable adjustable load at time t, is the production operation cost loss coefficient of the i'''-th switchable adjustable load. The load regulation reference of each type of adjustable load can be obtained by decomposing the user's day-ahead spot market reporting volume curve.

[0133] The regulation deviation default cost of the virtual power plant is specifically as follows:

[0134]

[0135] where λ is the penalty coefficient; P AGG (t) is the real-time output at time t after the virtual power plant aggregator aggregates resources; ΔP exc (t) is the regulation command deviation at time t; are respectively the lower limit and upper limit of the allowable deviation of the real-time command; q e (t) is the real-time electricity price at time t; {T aux} is the set of times in the auxiliary service regulation period;

[0136]

[0137] where P buy (t) is the real-time power cleared by the day-ahead spot market; ΔP buy (t) is the deviation power during the intraday non-regulation period; ξ is the real-time deviation electricity price coefficient; the day-ahead spot regulation deviation is not evaluated during the auxiliary service regulation period.

[0138]

[0139] where α is the electricity price revenue multiple participating in the auxiliary service regulation; P base (t) is the power generation and consumption baseline of the virtual power plant; in practice, the day-ahead spot clearing curve can be adopted.

[0140]

[0141] where: w up (t) is the real-time capacity price for upward adjustment; w down (t) is the real-time capacity price for downward adjustment; {Res} is the set of the above-mentioned constraint conditions, that is, the constraint conditions of the multi-objective optimization objective function; is the power that can be adjusted upward at time t, is the power that can be adjusted downward at time t. According to Equation (13), it can be seen that calculating the operator's capacity revenue requires additionally solving two sub-optimization problems of the power that can be adjusted upward and the power that can be adjusted downward.

[0142] In some exemplary embodiments, the determining of the constraint conditions of the multi-objective optimization objective function based on the regulation capabilities and regulation rule constraints of various regulation resources in the virtual power plant, the production safety constraints of the virtual power plant, and the grid operation constraints in S130 includes:

[0143] Based on the ramping constraints of each distributed energy source and continuously adjustable load in the virtual power plant, the response delay constraints of the hierarchical adjustable load and switchable adjustable load in the virtual power plant, and the regulation duration of various regulation resources in the virtual power plant, determine the regulation capabilities and regulation rule constraints of various regulation resources in the virtual power plant;

[0144] Based on the regulation power constraints of each energy storage in the virtual power plant and the regulation time sequence constraints of the hierarchical adjustable load, determine the production safety constraints of the virtual power plant;

[0145] Based on the power balance constraints and power flow constraints of the aggregated resources and purchased power of the virtual power plant aggregator, determine the grid operation constraints;

[0146] Take the regulation capabilities and regulation rule constraints of various regulation resources in the virtual power plant, the production safety constraints of the virtual power plant, and the grid operation constraints as the constraint conditions of the multi-objective optimization objective function.

[0147] In the present exemplary embodiment, the constraint conditions of the multi-objective optimization objective function include the regulation capabilities and regulation rule constraints of various regulation resources in the virtual power plant, the production safety constraints of the virtual power plant, and the grid operation constraints. Specifically as follows:

[0148] 1) Construction of the regulation capabilities and regulation rule constraints of various regulation resources in the virtual power plant: Since there are various types of demand-side resources, their regulation characteristics are extremely complex. The types of their regulation capabilities and regulation rule constraints can be mainly divided into ramp constraints, delay constraints, and continuous duration constraints.

[0149] The ramp constraints mainly target continuously adjustable variables, such as user-side energy storage devices, distributed power sources (such as distributed photovoltaics), continuously adjustable loads, etc. There is a certain ramp rate in their regulation process. Therefore, the ramp constraints are modeled as follows:

[0150]

[0151] In the formula, P Cont,m (t), P Cont,m (t - 1) are the device powers corresponding to the mth continuously adjustable resource at the current time and the previous time respectively. The continuously adjustable resources include energy storage devices, distributed power sources, and continuously adjustable loads; λ UP,m , λ down,m are the up-ramp and down-ramp capabilities corresponding to the mth continuously adjustable resource respectively; in this example, the corresponding ramp constraints can be constructed for energy storage devices, distributed photovoltaics, and continuously adjustable loads respectively. P Cont,m (t) can specifically be one of P ess,i (t), P pv,j (t), P LoadCont,i' (t).

[0152] The delay constraints mainly target switch-type and hierarchical adjustment-type variables. The delay constraint refers to the time interval from when the operator issues a regulation instruction to when the device responds to the regulation instruction. The delay constraints are modeled as follows:

[0153] t 2 -t 1 ≥nΔt (15)

[0154] Wherein, t 2 is the regulation command response time of the step - adjustable load, t 1 is the regulation command issuance time of the step - adjustable load, and n is the number of delay intervals.

[0155] The duration constraint is applicable to all adjustable resources on the demand side and includes two types: the minimum duration and the maximum duration. The duration constraint is modeled as follows:

[0156]

[0157] Wherein, t 3 is the regulation end time of the switch - type adjustable load; t' 2 is the regulation command response time of the switch - type adjustable load; n 1 is the minimum number of continuous intervals, and n 2 is the maximum number of continuous intervals.

[0158] 2) Construction of the production safety constraints of the virtual power plant:

[0159] For the adjustable resources of the user - side energy storage type, that is, energy storage devices, to ensure safety, their regulation processes need to meet the constraints of the energy storage power. Therefore, the production safety constraints of each energy storage device in the virtual power plant are modeled as:

[0160]

[0161] Wherein: is the discharge power of the i - th energy storage device at time t; is the charging power of the i - th energy storage device at time t; E ess,i (t) is the energy storage power of the i - th energy storage device at time t; η ch is the charging efficiency; η dis is the discharge efficiency; are respectively the minimum and maximum energy storage powers of the i - th energy storage device.

[0162] Furthermore, the production safety constraints also include regulation timing constraints. The regulation timing constraints mainly target some step - regulation - type variables, which must be regulated in a specific order (such as the sorting order) during the step - regulation process, and the adjacent step - regulations must meet the minimum continuous interval. The regulation timing constraints are modeled as follows:

[0163]

[0164] Wherein, P Lev (t Lev,s ) is the power of the step - adjustable load under the s - th step - regulation timing, P Lev (t Lev,N) is the power of the hierarchically adjustable load under the Nth hierarchical adjustment time sequence, where N is the number of hierarchical adjustment time sequences; Δt Lev,s is the time interval between adjacent hierarchies; n Lev is the minimum number of time intervals for hierarchical adjustment.

[0165] 3) Grid operation constraint construction:

[0166] Grid operation constraints include system power balance constraints and system power flow constraints, etc.

[0167] The power balance constraint means that the virtual power plant operator / aggregator on the cloud side needs to ensure the balance of the active power of its aggregated resources and the power purchased from the grid at any time. The specific constraint is:

[0168] P AGG (t) = P buy (t) + ΔP buy (t); (19)

[0169] In the formula, P AGG (t) is the real-time output of the virtual power plant aggregator after aggregating resources at time t; P buy (t) is the real-time power cleared by the day-ahead spot market; ΔP buy (t) is the deviation power during the non-regulation period within the day.

[0170] The system power flow constraint means that during the actual operation of the distributed power sources and adjustable loads aggregated by the operator, the transmission power of each power transmission and distribution branch needs to be ensured within a certain range. To simplify the model construction, the power flow constraint is approximated by a linearized power flow constraint, and its approximation error is small within the given operation range. The specific constraint form is:

[0171]

[0172] In the formula: P trans,uv (t) is the active power transmitted from node u to node v in the power grid, g uv , b uv are the conductance and susceptance between nodes u and v respectively, U u (t), U v (t) are the voltages at node u and node v at time t respectively, δ u (t), δ v (t) are the phase angles at node u and node v at time t respectively; Q trans,uv (t) is the reactive power transmitted from node u to node v in the power grid; are the upper and lower limits of the power of the line transmitted from node u to node v respectively; are the upper and lower limits of the voltage of node u respectively, N net is the set of nodes in the power grid.

[0173] The above equations (1)-(20) can be used as a complete multi-objective optimization model for virtual power plant operators / aggregators to participate in the ancillary services of the power market. This model can be regarded as a two-layer optimization model, that is, it includes the inner-layer optimization of the two sub-optimization problems in equation (13) and the outer-layer optimization of the multi-objective optimization problem.

[0174] In some embodiments, if the real-time capacity price mechanism is not considered, the multi-objective optimization model can degenerate from a two-layer optimization model to a single-layer optimization model, that is, it does not include the sub-optimization problem corresponding to equation (13).

[0175] In some embodiments, since the optimization objective variables include three types: continuous optimization variables, hierarchical discrete optimization variables, and switching optimization variables, and there are also discrete constraints in the constraint conditions, such as the three discrete constraints in equations (15), (16), and (18), which result in a large number of mixed integer, continuous, discrete constraints, and variable couplings in this optimization model, making it difficult to handle and solve. Therefore, after constructing a multi-objective optimization model for a virtual power plant based on the multi-objective optimization objective function and the constraint conditions, it further includes:

[0176] Using the sampling method to discretize the continuous variables corresponding to each continuous regulation type of resource in the multi-objective optimization model.

[0177] In the exemplary embodiment of the present invention, the continuous variables in the optimization model are transformed into discrete variables by the sampling method. The specific flowchart is as Figure 2 shown, that is, first determine the number of discrete sampling points N for the continuous regulation interval corresponding to the continuous variable d , determine the maximum value P max and the minimum value P min of the original value range of the continuous variable to be discretized, and normalize the value range of the continuous variable to the interval [0,1]; discretize the interval [0,1] to form a set P′∈{0, 1 / N d , 2 / N d , …, 1}; for each sampling point value p′ in the discretized set P′, perform discretization recovery. The discretization recovery formula is: p = P min +(P max -P min )*p′, where p is the discrete variable obtained after the discretization recovery of p′, that is, the recovery result. After the discretization process, the optimization model is transformed into a completely discrete combinatorial optimization model, and the heuristic algorithm can be directly called to solve the model.

[0178] At present, the construction of the trading model for virtual power plants to participate in the ancillary service market lacks refinement, and the optimization objectives cannot take into account many adjustment objectives during the adjustment process of virtual power plants, resulting in the inability of grid enterprises and cloud-side operator enterprises to objectively evaluate the participation value of virtual power plants in ancillary services and optimize the participation methods. Chinese Patent with the publication number CN 118917491A proposes a combined optimization method and device for the joint clearing problem of virtual power plants in the electricity energy market and the ancillary service market. The combined optimization method is adopted to achieve the joint trading clearing of the electricity energy market and the ancillary service market. However, its research focus is on solving the dynamic combined optimization problem, and the decision-making objective is only to maximize the overall market participation revenue, without giving a detailed method for constructing the model of the operator aggregating the demand-side resources to participate in the ancillary service market. Chinese Patent with the publication number CN118569430A proposes a distributed operation optimization method and related device for virtual power plants, establishes a user electricity consumption satisfaction model, and considers the user satisfaction as part of the adjustment cost during the interaction process. However, its method does not further consider the characteristics of different adjustment variables (such as continuous adjustment variables, discrete adjustment variables, switch-type adjustment variables) and their processing methods in the optimization model. Therefore, it is necessary to construct a multi-objective optimization model construction method for the cloud-side operator to aggregate the edge-side to participate in the ancillary service market. As Figure 3 shown, the multi-objective optimization model construction method for virtual power plants of the present invention mainly includes three parts: the construction of decision variables, the construction of objective functions, and the construction of constraint conditions. The decision variables include continuous adjustment variables, hierarchical adjustment variables, and switch adjustment variables. The objective functions include resource operation cost, default cost, purchased electricity cost, adjustment revenue, and capacity revenue. The constraint conditions include adjustment capacity constraint, production safety constraint, and power grid operation constraint. In view of the adjustment characteristics of different types of resources on the demand side, the present invention abstracts the adjustment capacity into three categories: continuous adjustment, step adjustment, and switch adjustment, which can effectively cover the adjustment characteristics of most adjustable resources on the demand side. A calculation method considering various costs and revenues existing in practice and a multi-objective optimization model construction method are designed for the problem of the cloud-side aggregator of virtual power plants aggregating edge-side resources to participate in ancillary service optimization, ensuring the optimization effect of the model. During the construction process of the optimization model for the cloud-side aggregator of virtual power plants aggregating edge-side resources to participate in ancillary services, multi-variable constraints such as adjustment capacity and adjustment rule constraints, user safety constraints, and power grid operation constraints are modeled, showing strong adaptability.

[0179] In the present invention, a multi-objective optimization model for the cloud-side virtual power plant operator containing heterogeneous demand-side resources to participate in the power ancillary service market is finally formed, which can improve the ability and revenue level of the operator to participate in the ancillary service market and enhance the economy of the virtual power plant operator to participate in the ancillary service market.

[0180] The model constructed by the present invention can be applied to the decision-making process of aggregating various types of resources on the demand side, such as virtual power plants, power load aggregators, and microgrid system operators, to participate in the operation regulation of the power auxiliary service market.

[0181] Embodiment 2

[0182] Based on the same inventive concept, the present invention also provides a multi-objective optimization regulation method for a virtual power plant, including:

[0183] Obtain the day-ahead output data of various types of regulation resources in the virtual power plant;

[0184] Solve the multi-objective optimization model constructed in any one of Embodiment 1 based on the day-ahead output data of various types of regulation resources in the virtual power plant, and obtain the intra-day regulation strategies of various types of regulation resources when the virtual power plant participates in auxiliary service regulation;

[0185] Conduct the operation regulation of the virtual power plant based on the intra-day regulation strategies of various types of regulation resources.

[0186] In the present exemplary embodiment, when the virtual power plant participates in auxiliary service regulation, it is necessary to obtain the day-ahead output data of various types of regulation resources. The day-ahead output data may include the day-ahead power generation and consumption data of distributed photovoltaics, energy storage devices, and various adjustable loads. The data period may not be longer than 15 minutes. As the regulation end, parameter data and instruction data of the virtual power plant are pre-existing. The parameter data includes the capacity, power, and efficiency of the energy storage, the network parameters of the power grid, the rated output power of distributed photovoltaics, the ramp rate and regulation delay of various adjustable loads, etc. The instruction data includes the intra-day regulation instruction curve, spot market electricity price curve, or time-of-use electricity price curve issued by the power grid. A heuristic algorithm can be called to solve the multi-objective optimization model. The heuristic algorithm can be one of a genetic algorithm (or multi-objective genetic algorithm NSGA2), artificial immune algorithm, simulated annealing algorithm, etc. The solution result is the intra-day regulation strategies of various types of regulation resources.

[0187] Exemplarily, based on the day-ahead output data of various types of regulation resources in the virtual power plant, the genetic algorithm is used to solve the multi-objective optimization model. Among them, the discrete variables corresponding to the output data of various types of regulation resources in the multi-objective optimization model are compiled into individual DNA sequences, and the constraint conditions are added as penalty functions to the objective function of the multi-objective optimization model to form a fitness function, and the discrete variables corresponding to the output data of various types of regulation resources are iteratively optimized through the operations of individual selection, crossover, and mutation.

[0188] In this exemplary embodiment, when all the optimization variables are transformed into discrete variable constraints, the discrete variables can be directly binary-coded and compiled into individual DNA sequences, while the constraint conditions can be added to the objective function in the form of penalty functions as the fitness function of the individual, or the constraint conditions can be used to design the elimination rules for the individuals, so as to ensure that the individual DNA sequences selected by the genetic algorithm meet the constraint conditions. After the final optimization is completed, the individual DNA sequence with the optimal fitness is used as the optimization solution result, and the final optimization solution result is obtained through the decompilation method.

[0189] The optimization solution result of the present invention can be used as the regulation strategy for demand-side resource aggregation entities such as virtual power plant operators and load aggregators to participate in the ancillary service market, and form a regulation decision-making scheme for various aggregated adjustable resources from day-ahead to intra-day. The specific processes of the design of various decision variables, the design of the objective function, and the design of the constraint conditions mentioned in the present invention are given, and the processing method of non-continuous variables is provided after forming a cloud-side operator aggregation edge participation ancillary service market model with mixed parameter multi-objective optimization.

[0190] Embodiment 3

[0191] Based on the same inventive concept, the present invention also provides a multi-objective optimization model construction system for a virtual power plant, including:

[0192] A resource classification module, configured to divide the distributed resources in the virtual power plant into continuously adjustable resources, hierarchically adjustable resources, and switchable adjustable resources based on the regulation characteristics of the distributed resources in the virtual power plant;

[0193] An objective function construction module, configured to consider the aggregation costs and aggregation revenues of various types of adjustable resources aggregated during the participation of the virtual power plant aggregator in the ancillary services of the power market, and construct a multi-objective optimization objective function to minimize the aggregation cost and maximize the aggregation revenue;

[0194] A constraint construction module, configured to determine the constraint conditions of the multi-objective optimization objective function based on the regulation capabilities and regulation rule constraints of various types of adjustable resources in the virtual power plant, the production safety constraints of the virtual power plant, and the grid operation constraints;

[0195] An optimization model construction module, configured to construct a multi-objective optimization model for a virtual power plant based on the multi-objective optimization objective function and the constraint conditions.

[0196] In a possible implementation manner, the continuously adjustable resources include distributed energy and continuously adjustable loads with continuously adjustable power consumption, and the distributed energy includes energy storage devices and distributed power sources; the hierarchically adjustable resources are hierarchically adjustable loads with hierarchically adjustable power consumption; the switchable adjustable resources are switchable adjustable loads with adjustable power consumption to 0 or rated power.

[0197] In a possible implementation manner, the objective function construction module includes:

[0198] A cost determination sub-module, configured to determine the aggregation cost of the virtual power plant aggregator participating in the auxiliary service regulation of the power market based on the regulation operation costs of various distributed energy resources, the regulation response costs of various adjustable loads, the regulation deviation default cost of the virtual power plant, and the external power purchase cost of the virtual power plant;

[0199] A revenue determination sub-module, configured to determine the aggregation revenue of the virtual power plant aggregator participating in the auxiliary service regulation of the power market based on the revenue from the virtual power plant participating in the auxiliary service regulation of the power market and the capacity revenue corresponding to the adjustable capacity of the virtual power plant aggregator aggregating various regulation resources;

[0200] An optimization objective construction sub-module, configured to construct the multi-objective optimization objective function by minimizing the aggregation cost and maximizing the aggregation revenue;

[0201] Wherein, the external power purchase cost of the virtual power plant is the cost of purchasing power from the power grid when the distributed power sources in the virtual power plant cannot meet the load demand or for the economic optimization objective; various adjustable loads include the continuously adjustable load, the hierarchical adjustable load, and the switchable adjustable load.

[0202] In a possible implementation manner, the revenue determination sub-module is specifically configured to:

[0203] Determine the revenue from the virtual power plant participating in the auxiliary service regulation of the power market based on the regulation power and the real-time electricity price corresponding to the virtual power plant aggregator participating in the auxiliary service regulation of the power market;

[0204] Determine the corresponding upward regulation capacity revenue based on the upward real-time capacity price and the upward regulation power of the virtual power plant;

[0205] Determine the corresponding downward regulation capacity revenue based on the downward real-time capacity price and the downward regulation power of the virtual power plant;

[0206] Determine the aggregation revenue of the virtual power plant aggregator participating in the auxiliary service regulation of the power market based on the upward regulation capacity revenue and the downward regulation capacity revenue;

[0207] Wherein, the upward regulation power and the downward regulation power are obtained by solving the corresponding sub-optimization problem, and the sub-optimization problem is constructed with the objective of maximizing the difference between the adjustable capacity of the virtual power plant and the power corresponding to the spot clearing and with the constraint conditions as the constraints.

[0208] In a possible implementation manner, the regulation operation costs of various distributed energy resources include the energy storage regulation operation cost and the downward compensation cost of distributed photovoltaics, and the expression of the objective function is:

[0209]

[0210] Wherein, C ess is the regulation operation cost of the energy storage device in the virtual power plant, C pv is the downward compensation cost of the distributed photovoltaic in the virtual power plant, C Load is the regulation response cost of various adjustable loads in the virtual power plant, C Pun is the regulation deviation default cost of the virtual power plant, C buy is the external power purchase cost of the virtual power plant, R aux is the income of the virtual power plant participating in the auxiliary service regulation of the power market, R cap is the aggregation income of the virtual power plant aggregator participating in the auxiliary service of the power market;

[0211]

[0212] Wherein, C loss,i is the regulation operation cost of the i-th energy storage device, N ess is the total number of aggregated energy storage devices in the virtual power plant, μ loss is the battery loss coefficient corresponding to the unit power exchange, P ess,i (t) is the output power of the power converter of the i-th energy storage at time t; Δt is the acquisition time interval, and T is the optimization period;

[0213]

[0214] Wherein, C pv,j is the downward compensation cost of the distributed photovoltaic, N pv is the total number of aggregated distributed photovoltaics in the virtual power plant; P pv,j (t) is the output power of the j-th distributed photovoltaic and its power converter at time t, is the maximum output capacity of the j-th distributed photovoltaic at each moment;

[0215]

[0216] Wherein, C LoadCont,i' is the regulation response cost of the i'-th continuously adjustable load, C LoadLev,i” is the regulation response cost of the i''-th stepwise adjustable load, C LoadSwt,i”' is the regulation response cost of the i'''-th switchable adjustable load, N LoadCont 、N LoadLev 、N LoadSwt are the total numbers corresponding to the continuously adjustable load, stepwise adjustable load, and switchable adjustable load aggregated by the virtual power plant respectively; is the load regulation reference for the production operation of the i'-th continuously adjustable load at time t, P LoadCont,i'$(t)$ is the power consumption of the $i^{th}$ continuously adjustable load at time $t$. is the production operation cost loss coefficient of the $i^{th}$ continuously adjustable load; is the load regulation reference for the production operation of the $i^{th}$ step - adjustable load at time $t$, $P$ LoadLev,i” $(t)$ is the power consumption of the $i^{th}$ step - adjustable load at time $t$. is the production operation cost loss coefficient of the $i^{th}$ step - adjustable load; is the load regulation reference for the production operation of the $i^{th}$ switch - type adjustable load at time $t$, $P$ LoadSwt,i”' $(t)$ is the power consumption of the $i^{th}$ switch - type adjustable load at time $t$. is the production operation cost loss coefficient of the $i^{th}$ switch - type adjustable load;

[0217]

[0218] In the formula, $\lambda$ is the penalty coefficient; $P$ AGG $(t)$ is the real - time output of the virtual power plant aggregator after aggregating resources at time $t$; $\Delta P$ exc $(t)$ is the regulation command deviation at time $t$. are the lower limit and upper limit of the real - time command allowable deviation respectively; $q$ e $(t)$ is the real - time electricity price at time $t$; $\{T$ aux} is the set of moments in the auxiliary service regulation period.

[0219]

[0220] In the formula, $P$ buy $(t)$ is the real - time power cleared by the day - ahead spot market; $\Delta P$ buy $(t)$ is the deviation power during the non - regulation period within the day; $\xi$ is the real - time deviation electricity price coefficient.

[0221]

[0222] In the formula, $\alpha$ is the electricity price revenue multiple for participating in the auxiliary service regulation; $P$ base $(t)$ is the power generation and consumption baseline of the virtual power plant.

[0223]

[0224] In the formula: $w$ up $(t)$ is the real - time capacity price for upward adjustment; $w$ down $(t)$ is the real - time capacity price for downward adjustment; $\{Res\}$ is the set of the said constraint conditions; is the power that can be adjusted upward at time $t$, is the power that can be adjusted downward at time $t$.

[0225] In a possible implementation manner, the constraint construction module is specifically configured to:

[0226] Based on the ramping constraints of each distributed energy source and continuously adjustable load in the virtual power plant, the response delay constraints of the hierarchical adjustable load and the switching-type adjustable load in the virtual power plant, and the adjustment duration of various types of adjustment resources in the virtual power plant, determine the adjustment capabilities and adjustment rule constraints of various types of adjustment resources in the virtual power plant;

[0227] Based on the adjustment power constraints of each energy storage in the virtual power plant and the adjustment timing constraints of the hierarchical adjustable load, determine the production safety constraints of the virtual power plant;

[0228] Based on the power balance constraint and the power flow constraint between the aggregated resources of the virtual power plant aggregator and the purchased power, determine the grid operation constraints;

[0229] Take the adjustment capabilities and adjustment rule constraints of various types of adjustment resources in the virtual power plant, the production safety constraints of the virtual power plant, and the grid operation constraints as the constraint conditions of the multi-objective optimization objective function.

[0230] In a possible implementation manner, the expression of the ramping constraint of each distributed energy source and continuously adjustable load is:

[0231]

[0232] In the formula, P Cont,m (t), P Cont,m (t - 1) are the device powers corresponding to the m-th continuously adjustable resource at the current moment and the previous moment respectively. The continuously adjustable resources include energy storage devices, distributed power sources, and continuously adjustable loads; λ UP,m , λ down,m are the up-ramping and down-ramping capabilities corresponding to the m-th continuously adjustable resource respectively;

[0233] The expression of the response delay constraint of the hierarchical adjustable load is:

[0234] t 2 -t 1 ≥nΔt;

[0235] In the formula, t 2 is the adjustment command response time of the hierarchical adjustable load, t 1 is the adjustment command release time of the hierarchical adjustable load, and n is the number of delay intervals;

[0236] The expression of the response delay constraint of the switching-type adjustable load is:

[0237]

[0238] In the formula, t 3is the adjustment end time of the switch-type adjustable load; t' 2 is the adjustment command response time of the switch-type adjustable load; n 1 is the minimum number of continuous intervals, n 2 is the maximum number of continuous intervals;

[0239] The expression of the adjustment timing constraint of the hierarchical adjustable load is:

[0240]

[0241] In the formula, P Lev (t Lev,s ) is the power of the hierarchical adjustable load at the s-th hierarchical adjustment timing, P Lev (t Lev,N ) is the power of the hierarchical adjustable load at the N-th hierarchical adjustment timing, N is the number of hierarchical adjustment timings; Δt Lev,s is the time interval between adjacent hierarchies; n Lev is the minimum number of time intervals;

[0242] The expression of the power balance constraint between the aggregated resources and the purchased power is:

[0243] P AGG (t) = P buy (t) + ΔP buy (t);

[0244] In the formula, P AGG (t) is the real-time output of the virtual power plant aggregator after aggregating resources at time t; P buy (t) is the real-time power cleared by the day-ahead spot market; ΔP buy (t) is the deviation power during the intraday non-adjustment period.

[0245] In a possible implementation manner, the system further includes:

[0246] A discretization module, configured to discretize the continuous variables corresponding to the continuous adjustment type resources in the multi-objective optimization model by using the sampling method.

[0247] Embodiment 4

[0248] Based on the same inventive concept, the present invention further provides a multi-objective optimization control system for a virtual power plant, including:

[0249] A data acquisition module, configured to acquire the day-ahead output data of various adjustment resources in the virtual power plant;

[0250] A model solution module for solving the multi-objective optimization model constructed in Embodiment 1 based on the day-ahead output data of various regulation resources in the virtual power plant, so as to obtain the intra-day regulation strategies of various regulation resources when the virtual power plant participates in ancillary service regulation;

[0251] A regulation module for performing operation regulation of the virtual power plant based on the intra-day regulation strategies of various regulation resources.

[0252] Embodiment 5

[0253] As Figure 4 shown, the present invention also provides an electronic device, which may be a computer device, a single-chip microcomputer device, an intelligent mobile device, etc. The electronic device in this embodiment may include a processor, a memory, a transceiver component, etc. The memory, the processor, and the transceiver component are connected through a bus; the memory can be used to store an execution program, and an exemplary execution program may include instructions; the processor is used to execute the instructions stored in the memory. The memory can also be used to store data, and the data can be called and / or modified when the instructions are executed.

[0254] The processor may be a Central Processing Unit (CPU), or may also be other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, and is suitable for implementing one or more instructions. Specifically, it is suitable for loading and executing one or more instructions in the storage medium to implement the corresponding method flow or corresponding function, so as to implement the steps of a multi-objective optimization model construction method and a multi-objective optimization regulation method for a virtual power plant in the above embodiments.

[0255] Embodiment 6

[0256] Based on the same inventive concept, the present invention also provides a readable storage medium, specifically an electronic device-readable storage medium (Memory). The electronic device-readable storage medium is a memory device in an electronic device, used to store programs and data. It can be understood that the storage medium here can include both the built-in storage medium in the electronic device and, of course, the extended storage medium supported by the electronic device. The storage medium provides a storage space, and this storage space stores the operating system of the terminal. Moreover, in this storage space, there is also stored one or more instructions suitable for being loaded and executed by the processor, and these instructions can be one or more execution programs (including program codes). It should be noted that the storage medium here can be a high-speed RAM memory or a non-volatile memory, such as at least one disk memory. By the processor loading and executing one or more instructions stored in the storage medium, the steps of a multi-objective optimization model construction method and a multi-objective optimization regulation method for a virtual power plant in the above embodiments can be realized.

[0257] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memories, CD-ROMs, optical memories, etc.) containing computer-usable program codes.

[0258] The present invention is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram, as well as the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for realizing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0259] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device, and the instruction device realizes the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1The functions specified in one or more boxes.

[0260] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process. Thus, the instructions executed on the computer or other programmable device provide for implementing the steps of the functions specified in Figure 1 one process or more processes and / or boxes Figure 1 the functions specified in one box or more boxes.

[0261] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit the scope of its protection. Although the present invention has been described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that after reading the present invention, various changes, modifications or equivalent replacements can still be made to the specific implementation manners of the application. However, these changes, modifications or equivalent replacements are all within the scope of the protection of the claims pending for the application.

Claims

1. A method for constructing a multi-objective optimization model for a virtual power plant, characterized in that: include: Based on the regulation characteristics of distributed resources in the virtual power plant, the distributed resources in the virtual power plant are divided into continuous regulation resources, hierarchical regulation resources and switch regulation resources; Considering the aggregated costs and benefits of various types of regulation resources aggregated by virtual power plant aggregators in the process of participating in ancillary services in the power market, a multi-objective optimization objective function is constructed to minimize the aggregated cost and maximize the aggregated benefit. Determining the constraint conditions of the multi-objective optimization objective function based on the regulation capabilities and regulation rule constraints of various regulation resources in the virtual power plant, the production safety constraints of the virtual power plant and the grid operation constraints; Based on the multi-objective optimization objective function and the constraints, a multi-objective optimization model for virtual power plants is constructed.

2. The method according to claim 1, characterized in that The continuously adjustable resources include distributed energy and continuously adjustable loads whose electric power can be continuously adjusted. The distributed energy includes energy storage devices and distributed power sources. The graded adjustment resources are graded adjustable loads whose electric power can be adjusted in grades and levels. The switch adjustment resources are switch adjustable loads whose electric power can be adjusted to 0 or rated power.

3. The method according to claim 2, characterized in that The above-mentioned multi-objective optimization objective function is constructed by considering the aggregated costs and aggregated benefits of various types of regulating resources aggregated by virtual power plant aggregators in the process of participating in auxiliary services in the power market in order to minimize the aggregated costs and maximize the aggregated benefits, including: Based on the regulation and operation costs of each distributed energy source, the regulation and response costs of various adjustable loads, the regulation deviation default costs of the virtual power plant and the purchased electricity costs of the virtual power plant when the virtual power plant participates in the regulation of ancillary services in the power market, the aggregation costs of the virtual power plant aggregator participating in the ancillary services in the power market are determined; Based on the revenue of virtual power plants participating in the regulation of ancillary services in the power market and the capacity revenue corresponding to the adjustable capacity of various regulation resources aggregated by virtual power plant aggregators, determine the aggregated revenue of virtual power plant aggregators participating in ancillary services in the power market; The multi-objective optimization objective function is constructed by minimizing the aggregation cost and maximizing the aggregation benefit; Among them, the cost of purchasing electricity from outside the virtual power plant is the cost of purchasing electricity from the power grid when the distributed power sources within the virtual power plant cannot meet the load demand or for the economic optimization goal; various types of adjustable loads include the continuously adjustable load, the stepped adjustable load and the switch-type adjustable load.

4. The method according to claim 3, characterized in that The method of determining the aggregated revenue of the virtual power plant aggregator participating in the power market ancillary service regulation based on the revenue of the virtual power plant participating in the power market ancillary service regulation and the capacity revenue corresponding to the adjustable capacity of various regulation resources aggregated by the virtual power plant aggregator includes: Based on the regulated electricity volume and real-time electricity price corresponding to the virtual power plant aggregator's participation in the regulation of ancillary services in the power market, determine the benefits of the virtual power plant's participation in the regulation of ancillary services in the power market; Determine the corresponding increase in capacity revenue based on the increase in real-time capacity price and the increase in power of the virtual power plant; Based on the reduction of the real-time capacity price and the adjustable power of the virtual power plant, the corresponding reduction capacity benefits are determined; Based on the upward capacity revenue and the downward capacity revenue, determine the aggregated revenue of the virtual power plant aggregator participating in the auxiliary services in the electricity market; Among them, the adjustable power and the adjustable power are obtained by solving the corresponding sub-optimization problems respectively, and the sub-optimization problems are constructed with the constraint conditions as constraints, with the goal of maximizing the difference between the adjustable capacity of the virtual power plant and the corresponding power of spot clearing.

5. The method according to claim 4, characterized in that The regulation and operation costs of each distributed energy source include the energy storage regulation and operation costs and the downward compensation costs of distributed photovoltaics. The expression of the multi-objective optimization objective function is: In the formula, C ess Adjust the operating cost of the energy storage device in the virtual power plant, C pv is the downward compensation cost of distributed photovoltaic in the virtual power plant, C Load is the control response cost of various adjustable loads in the virtual power plant, C Pun is the default cost of the virtual power plant’s regulation deviation, C buy is the cost of purchased electricity for the virtual power plant, R aux is the revenue of virtual power plants participating in the regulation of ancillary services in the electricity market, R cap Capacity revenue for virtual power plant aggregators participating in ancillary services in the electricity market; In the formula, C loss,i is the regulation and operation cost of the i-th energy storage device, N ess is the total number of energy storage devices aggregated in the virtual power plant, μ loss is the battery loss coefficient corresponding to the unit power exchange, P ess,i (t) is the output power of the i-th energy storage power converter at time t; Δt is the acquisition time interval, and T is the optimization period; In the formula, C pv,j is the downward adjustment compensation cost of distributed photovoltaics, N pv is the total number of distributed photovoltaics aggregated in the virtual power plant; P pv,j (t) is the output power of the jth distributed photovoltaic and its power converter at time t, is the maximum output capacity of the jth distributed photovoltaic at each moment; In the formula, C LoadCont,i' is the control response cost of the i'th continuously adjustable load, C LoadLev,i” is the control response cost of the i-th grade adjustable load, C LoadSwt,i”' is the control response cost of the ith switch-type adjustable load, N LoadCont 、N LoadLev 、N LoadSwt They are the total number of continuously adjustable loads, graded adjustable loads, and switch-type adjustable loads aggregated by the virtual power plant; is the load regulation benchmark of the i'th continuously adjustable load at time t, P LoadCont,i' (t) is the power consumption of the i'th continuously adjustable load at time t, is the production and operation cost loss coefficient of the i'th continuously adjustable load; is the load regulation benchmark for the i-th graded adjustable load at time t, P LoadLev,i” (t) is the power consumption of the i-th graded adjustable load at time t, is the production and operation cost loss coefficient of the i-th grade adjustable load; is the load regulation benchmark of the ith switch type adjustable load at time t, P LoadSwt,i”' (t) is the power consumption of the ith switch type adjustable load at time t, is the production and operation cost loss coefficient of the ith switch type adjustable load; Where λ is the penalty coefficient; P AGG (t) is the real-time output at time t after the virtual power plant aggregator aggregates resources; ΔP exc (t) is the adjustment instruction deviation at time t; They are the lower limit and upper limit of the allowable deviation of the real-time instruction respectively; q e (t) is the real-time electricity price at time t; {T aux } is a time set for auxiliary service adjustment period; Where P buy (t) is the real-time power cleared in the day-ahead spot market; ΔP buy (t) is the deviation power during the non-adjustment period of the day; ξ is the real-time deviation electricity price coefficient; Where α is the electricity price benefit ratio for participating in ancillary service regulation; P base (t) is the power generation and consumption baseline of the virtual power plant; Where: w up (t) is to increase the real-time capacity price; w down (t) is to lower the real-time capacity price; {Res} is the set of constraints; is the power that can be increased at time t, The power can be reduced at time t.

6. The method according to claim 1, characterized in that The constraint conditions of the multi-objective optimization objective function are determined based on the regulation capabilities and regulation rule constraints of various regulation resources in the virtual power plant, the production safety constraints of the virtual power plant, and the grid operation constraints, including: Based on the ramp constraints of distributed energy sources and continuously adjustable loads in the virtual power plant, the response delay constraints of graded adjustable loads and switch-type adjustable loads in the virtual power plant, and the duration of adjustment of various adjustment resources in the virtual power plant, the adjustment capacity and adjustment rule constraints of various adjustment resources in the virtual power plant are determined; Based on the regulation power constraints of each energy storage in the virtual power plant and the regulation timing constraints of the graded adjustable loads, the production safety constraints of the virtual power plant are determined; Determine grid operation constraints based on power balance constraints and flow constraints of the aggregated resources and purchased power of the virtual power plant aggregator; The regulation capabilities and regulation rule constraints of various regulation resources in the virtual power plant, the production safety constraints of the virtual power plant and the power grid operation constraints are used as constraints of the multi-objective optimization objective function.

7. The method according to claim 6, characterized in that The expressions of the ramp constraints of the distributed energy sources and the continuously adjustable loads are as follows: Where P Cont,m (t), P Cont,m (t-1) is the device power corresponding to the mth continuous regulation resource at the current moment and the previous moment respectively. The continuous regulation resources include energy storage devices, distributed power sources and continuously adjustable loads; λ UP,m , down,m are the up-climbing and down-climbing capabilities corresponding to the mth continuous regulation resource respectively; The expression of the response delay constraint of the hierarchical adjustable load is: t2-t1≥nΔt; Where, t2 is the response time of the adjustment command of the graded adjustable load, t1 is the release time of the adjustment command of the graded adjustable load, and n is the number of delay intervals; The expression of the response delay constraint of the switch type adjustable load is: Wherein, t3 is the adjustment end time of the switch type adjustable load; t'2 is the adjustment instruction response time of the switch type adjustable load; n1 is the minimum continuous interval number, n2 is the maximum continuous interval number; The expression of the adjustment timing constraint of the hierarchical adjustable load is: Where P Lev (t Lev,1 ) is the power of the graded adjustable load under the first graded adjustment sequence, P Lev (t Lev,s ) is the power of the graded adjustable load under the sth graded adjustment sequence, P Lev (t Lev,N ) is the power of the graded adjustable load under the Nth graded adjustment sequence, N is the number of graded adjustment sequences; Δt Lev,s is the time interval between adjacent classifications; n Lev The minimum number of time intervals for graded adjustment; The expression of the power balance constraint of the aggregated resources and purchased power is: P AGG (t)=P buy (t)+ΔP buy (t); Where P AGG (t) is the real-time output at time t after the virtual power plant aggregator aggregates resources; P buy (t) is the real-time power cleared in the day-ahead spot market; ΔP buy (t) is the deviation power during the non-adjustment period of the day.

8. The method according to any one of claims 1 to 7, characterized in that: After constructing a multi-objective optimization model for a virtual power plant based on the multi-objective optimization objective function and the constraint conditions, the method further includes: The continuous variables corresponding to each continuous adjustment resource in the multi-objective optimization model are discretized using a sampling method.

9. A multi-objective optimization control method for virtual power plants, characterized in that: include: Obtain the day-ahead output data of various regulating resources in the virtual power plant; Solve the multi-objective optimization model constructed in any one of claims 1 to 8 based on the day-ahead output data of various regulating resources in the virtual power plant to obtain the intraday regulation strategy of various regulating resources when the virtual power plant participates in the regulation of auxiliary services; The operation of the virtual power plant is regulated based on the intraday regulation strategy of various regulation resources.

10. A multi-objective optimization model construction system for virtual power plants, characterized in that: include: A resource classification module is used to classify the distributed resources in the virtual power plant into continuous regulation resources, hierarchical regulation resources and switch regulation resources based on the regulation characteristics of the distributed resources in the virtual power plant; The objective function construction module is used to consider the aggregated costs and aggregated benefits of various types of regulation resources aggregated by virtual power plant aggregators in the process of participating in power market ancillary services, and to construct a multi-objective optimization objective function by minimizing the aggregated costs and maximizing the aggregated benefits; A constraint construction module, used to determine the constraint conditions of the multi-objective optimization objective function based on the regulation capabilities and regulation rule constraints of various regulation resources in the virtual power plant, the production safety constraints of the virtual power plant and the grid operation constraints; The optimization model construction module is used to construct a multi-objective optimization model for a virtual power plant based on the multi-objective optimization objective function and the constraint conditions.

11. The system according to claim 10, characterized in that The continuously adjustable resources include distributed energy and continuously adjustable loads whose electric power can be continuously adjusted. The distributed energy includes energy storage devices and distributed power sources. The graded adjustment resources are graded adjustable loads whose electric power can be adjusted in grades and levels. The switch adjustment resources are switch adjustable loads whose electric power can be adjusted to 0 or rated power.

12. The system according to claim 11, characterized in that The objective function building module includes: The cost determination submodule is used to determine the aggregation cost of the virtual power plant aggregator participating in the power market ancillary service based on the regulation and operation cost of each distributed energy source, the regulation and response cost of various adjustable loads, the regulation deviation default cost of the virtual power plant and the purchased electricity cost of the virtual power plant when the virtual power plant participates in the power market ancillary service regulation; A revenue determination submodule, used to determine the aggregated revenue of the virtual power plant aggregator participating in the power market ancillary service regulation based on the revenue of the virtual power plant participating in the power market ancillary service regulation and the capacity revenue corresponding to the adjustable capacity of various regulation resources aggregated by the virtual power plant aggregator; An optimization target construction submodule, used to construct the multi-objective optimization objective function by minimizing the aggregation cost and maximizing the aggregation benefit; Among them, the cost of purchasing electricity from outside the virtual power plant is the cost of purchasing electricity from the power grid when the distributed power sources within the virtual power plant cannot meet the load demand or for the economic optimization goal; various types of adjustable loads include the continuously adjustable load, the stepped adjustable load and the switch-type adjustable load.

13. The system according to claim 12, characterized in that The revenue determination submodule is specifically used for: Based on the regulated electricity volume and real-time electricity price corresponding to the virtual power plant aggregator's participation in the regulation of ancillary services in the power market, determine the benefits of the virtual power plant's participation in the regulation of ancillary services in the power market; Determine the corresponding increase in capacity revenue based on the increase in real-time capacity price and the increase in power of the virtual power plant; Based on the reduction of the real-time capacity price and the adjustable power of the virtual power plant, the corresponding reduction capacity benefits are determined; Based on the upward capacity revenue and the downward capacity revenue, determine the aggregated revenue of the virtual power plant aggregator participating in the auxiliary services in the electricity market; Among them, the adjustable power and the adjustable power are obtained by solving the corresponding sub-optimization problems respectively, and the sub-optimization problems are constructed with the constraint conditions as constraints, with the goal of maximizing the difference between the adjustable capacity of the virtual power plant and the corresponding power of spot clearing.

14. The system according to claim 13, characterized in that The regulation and operation costs of each distributed energy source include the energy storage regulation and operation costs and the downward compensation costs of distributed photovoltaics. The expression of the objective function is: In the formula, C ess Adjust the operating cost of the energy storage device in the virtual power plant, C pv is the downward compensation cost of distributed photovoltaic in the virtual power plant, C Load is the control response cost of various adjustable loads in the virtual power plant, C Pun is the default cost of the virtual power plant’s regulation deviation, C buy is the cost of purchased electricity for the virtual power plant, R aux is the revenue of virtual power plants participating in the regulation of ancillary services in the electricity market, R cap Aggregate revenues for virtual power plant aggregators participating in ancillary services in the electricity market; In the formula, C loss,i is the regulation and operation cost of the i-th energy storage device, N ess is the total number of energy storage devices aggregated in the virtual power plant, μ loss is the battery loss coefficient corresponding to the unit power exchange, P ess,i (t) is the output power of the i-th energy storage power converter at time t; Δt is the acquisition time interval, and T is the optimization period; In the formula, C pv,j is the downward adjustment compensation cost of distributed photovoltaics, N pv is the total number of distributed photovoltaics aggregated in the virtual power plant; P pv,j (t) is the output power of the jth distributed photovoltaic and its power converter at time t, is the maximum output capacity of the jth distributed photovoltaic at each moment; In the formula, C LoadCont,i' is the control response cost of the i'th continuously adjustable load, C LoadLev,i” is the control response cost of the i-th grade adjustable load, C LoadSwt,i”' is the control response cost of the ith switch-type adjustable load, N LoadCont 、N LoadLev 、N LoadSwt They are the total number of continuously adjustable loads, graded adjustable loads, and switch-type adjustable loads aggregated by the virtual power plant; is the load regulation benchmark of the i'th continuously adjustable load at time t, P LoadCont,i' (t) is the power consumption of the i'th continuously adjustable load at time t, is the production and operation cost loss coefficient of the i'th continuously adjustable load; is the load regulation benchmark for the i-th graded adjustable load at time t, P LoadLev,i” (t) is the power consumption of the i-th graded adjustable load at time t, is the production and operation cost loss coefficient of the i-th grade adjustable load; is the load regulation benchmark of the ith switch type adjustable load at time t, P LoadSwt,i”' (t) is the power consumption of the ith switch type adjustable load at time t, is the production and operation cost loss coefficient of the ith switch type adjustable load; Where λ is the penalty coefficient; P AGG (t) is the real-time output at time t after the virtual power plant aggregator aggregates resources; ΔP exc (t) is the adjustment instruction deviation at time t; They are the lower limit and upper limit of the allowable deviation of the real-time instruction respectively; q e (t) is the real-time electricity price at time t; {T aux } is a time set for auxiliary service adjustment period; Where P buy (t) is the real-time power cleared in the day-ahead spot market; ΔP buy (t) is the deviation power during the non-adjustment period of the day; ξ is the real-time deviation electricity price coefficient; Where α is the electricity price benefit ratio for participating in ancillary service regulation; P base (t) is the power generation and consumption baseline of the virtual power plant; Where: w up (t) is to increase the real-time capacity price; w down (t) is to lower the real-time capacity price; {Res} is the set of constraints; is the power that can be increased at time t, The power can be reduced at time t.

15. The system according to claim 10, characterized in that The constraint building module is specifically used for: Based on the ramp constraints of distributed energy sources and continuously adjustable loads in the virtual power plant, the response delay constraints of graded adjustable loads and switch-type adjustable loads in the virtual power plant, and the duration of adjustment of various adjustment resources in the virtual power plant, the adjustment capacity and adjustment rule constraints of various adjustment resources in the virtual power plant are determined; Based on the regulation power constraints of each energy storage in the virtual power plant and the regulation timing constraints of the graded adjustable loads, the production safety constraints of the virtual power plant are determined; Determine grid operation constraints based on power balance constraints and flow constraints of the aggregated resources and purchased power of the virtual power plant aggregator; The regulation capabilities and regulation rule constraints of various regulation resources in the virtual power plant, the production safety constraints of the virtual power plant and the power grid operation constraints are used as constraints of the multi-objective optimization objective function.

16. The system according to claim 15, characterized in that The expressions of the ramp constraints of the distributed energy sources and the continuously adjustable loads are as follows: Where P Cont,m (t), P Cont,m (t-1) is the device power corresponding to the mth continuous regulation resource at the current moment and the previous moment respectively. The continuous regulation resources include energy storage devices, distributed power sources and continuously adjustable loads; λ UP,m , down,m are the up-climbing and down-climbing capabilities corresponding to the mth continuous regulation resource respectively; The expression of the response delay constraint of the hierarchical adjustable load is: t2-t1≥nΔt; Where, t2 is the response time of the adjustment command of the graded adjustable load, t1 is the release time of the adjustment command of the graded adjustable load, and n is the number of delay intervals; The expression of the response delay constraint of the switch type adjustable load is: Wherein, t3 is the adjustment end time of the switch type adjustable load; t'2 is the adjustment instruction response time of the switch type adjustable load; n1 is the minimum continuous interval number, n2 is the maximum continuous interval number; The expression of the adjustment timing constraint of the hierarchical adjustable load is: P Lev (t Lev,1 )→...→P Lev (t Lev,s )→...→P Lev (t Lev,N ); Δt Lev,s ≥n Lev Δt Where P Lev (t Lev,1 ) is the power of the graded adjustable load under the first graded adjustment sequence, P Lev (t Lev,s ) is the power of the graded adjustable load under the sth graded adjustment sequence, P Lev (t Lev,N ) is the power of the graded adjustable load under the Nth graded adjustment sequence, N is the number of graded adjustment sequences; Δt Lev,s is the time interval between adjacent classifications; n Lev is the minimum number of time intervals; The expression of the power balance constraint of the aggregated resources and purchased power is: P AGG (t)=P buy (t)+ΔP buy (t); Where P AGG (t) is the real-time output at time t after the virtual power plant aggregator aggregates resources; P buy (t) is the real-time power cleared in the day-ahead spot market; ΔP buy (t) is the deviation power during the non-adjustment period of the day.

17. The system according to any one of claims 10 to 16, characterized in that: The system further comprises: The discretization module is used to discretize the continuous variables corresponding to each continuous adjustment resource in the multi-objective optimization model by using a sampling method.

18. A multi-objective optimization control system for virtual power plants, characterized in that: include: The data acquisition module is used to obtain the day-ahead output data of various regulating resources in the virtual power plant; A model solving module, used to solve the multi-objective optimization model constructed according to any one of claims 1 to 8 based on the day-ahead output data of various regulating resources in the virtual power plant, and obtain the intraday regulation strategy of various regulating resources when the virtual power plant participates in the regulation of auxiliary services; The control module is used to control the operation of the virtual power plant based on the intraday control strategy of various regulation resources.

19. An electronic device, characterized in that: include: at least one processor and memory; The memory and the processor are connected via a bus; The memory is used to store one or more programs; When the one or more programs are executed by the at least one processor, the method according to any one of claims 1 to 9 is implemented.

20. A readable storage medium, characterized in that: An execution program is stored thereon, and when the execution program is executed, the method according to any one of claims 1 to 9 is implemented.

Images