Configuration method considering distributed new energy aggregation

Abstract

The invention provides a distributed new energy aggregation configuration method. The method comprises the following steps: constructing a management and interaction mechanism of a virtual power plant for distributed new energy; constructing a universal flexible resource power generation model according to the power generation characteristics of the distributed new energy; establishing an optimization combination model by taking the maximum expected income as a target; performing aggregation configuration on the distributed new energy according to an output result of the optimization combination model; according to the invention, the multi-type resources of the virtual power plant are deployed in combination with the energy consumption demand of the interactive resources.

Description

Technical Field

[0001] The present invention relates to the technical field of virtual power plants, and in particular to a method for configuring distributed renewable energy aggregation. Background Art

[0002] The rise of distributed new energy, controllable loads, and energy storage on the power distribution and consumption side has gradually transformed the existing large-scale power production and long-distance transmission into distributed new energy utilization. Distributed new energy management has further promoted the optimization of supply and demand coordination, and mobilizing user-side resources has played an increasingly important role in shaving peaks and filling valleys, alleviating the contradiction between power supply and demand, and promoting the consumption of clean energy. System absorption capacity is a necessary condition for the development of new energy. Therefore, it is necessary to strengthen the construction of flexible power supply adjustment in the system and vigorously improve the response capacity of the power demand side. Virtual power plants aggregate user-side resources in the power market environment, optimize power consumption timing and behavior, and provide flexible adjustment capabilities to the power grid, providing a new solution for promoting the consumption of renewable energy. Through advanced information technology and control methods, the virtual power plant can respond to the power grid instructions quickly and accurately, provide auxiliary services such as peak load regulation, frequency regulation, and load adjustment for the power grid, and bring direct and considerable economic benefits. Reasonable planning through the resource combination of virtual power plants can effectively reduce the construction scale of distribution facilities and information systems, thereby reducing the construction cost of distribution network infrastructure. Virtual power plants provide regulation capabilities to the power grid by precisely controlling the electricity consumption side, which helps to improve the power grid's ability to absorb new energy, ensure the safe and stable operation of the power grid, and promote the formation of a new power system with new energy as the main body, which can produce huge environmental benefits.

[0003] (1) The differences in the technical characteristics of interactive resources in virtual power plants are the difficulty in the coordinated operation of virtual power plants. The technical characteristics are divided into two aspects: resource power characteristics and response characteristics. Taking various types of distributed power generation into consideration in the virtual power plant model, a distributed control strategy is proposed to control the output of multiple distributed generators in a coordinated manner, so that these generators are aggregated into a virtual power plant in the distribution network; considering the demand-side resources as interactive resources, a general model of demand-responsive virtual power plants is established, and a model for virtual power plants to participate in day-ahead unit combination is established. However, the model cannot reflect the operating characteristics of different load equipment, and fails to fully consider conventional virtual power plant control resources such as distributed generation and energy storage. A virtual power plant energy management model consisting of dispatchable units, random generators, energy storage equipment and flexible loads is constructed, and a comprehensive resource power characteristic model is established.

[0004] (2) Virtual power plant resource combination belongs to the planning and configuration problem. The combination of resources determines the product characteristics of the virtual power plant. In application, the working process of the virtual power plant can be divided into three stages: planning, operation and settlement. At present, foreign research on virtual power plants is mostly focused on scheduling operation and benefit distribution, and there is less research on planning work. In addition, the electric power and operating costs of distributed renewable energy are changing during operation, which leads to the timeliness of the alliance composed of distributed renewable energy in the virtual power plant. Since the virtual power plant has no mandatory requirements for interactive resources, users can choose whether to join the virtual power plant response. Therefore, when the scale of the virtual power plant is limited, all alliance schemes can still be enumerated, but as the number of alliance members increases, the number of combination schemes explodes exponentially, forming a dimensionality disaster. The enumeration method is inefficient and is not conducive to the efficient operation of the virtual power plant. Therefore, a suitable combination method is a prerequisite for efficient, reliable and economical operation.

[0005] In summary, current research on the planning and configuration of virtual power plants is still mainly focused on the static capacity configuration of various power sources. It fails to fully consider the heterogeneity of interactive resources and the timeliness of virtual power plant alliances, and lacks research on the combination method of multi-energy virtual power plant alliances with timeliness. Summary of the invention

[0006] The present invention solves the problem that the existing technology fails to fully consider the heterogeneity of interactive resources and the timeliness of the virtual power plant alliance, and proposes a method for configuring the aggregation of distributed new energy sources, which combines the energy demand of interactive resources and allocates multiple types of resources of the virtual power plant.

[0007] In order to achieve the above objectives, the following technical solutions are proposed: A method for configuring distributed new energy aggregation includes the following steps: S1, building a virtual power plant management and interaction mechanism for distributed renewable energy; S2, building a general flexible resource generation model based on the generation characteristics of distributed renewable energy; S3, establish an optimization portfolio model with the goal of maximizing expected returns; S4, aggregate and configure distributed renewable energy sources according to the output results of the optimization combination model.

[0008] The present invention fully considers the heterogeneity of interactive resources and the timeliness of the virtual power plant alliance, establishes a general flexible resource power generation model, then measures the uncertainty of user response power with combined risk, establishes an optimized combination model with the goal of maximizing expected benefits, and then aggregates and configures distributed new energy. Finally, it verifies the impact of the virtual power plant resource combination method on the virtual power plant's aggregate response capability and economy. The present invention can take into account the wide range of resource types and different technical characteristics, decompose the virtual power plant resource combination problem into the alliance member composition method and member response plan allocation problem, and establish an optimized combination model; establish an optimized combination model with the goal of maximizing expected benefits, and provide a basis for the hybrid virtual power plant resource combination.

[0009] Preferably, the S2 further comprises the following steps: According to the control authority of the virtual power plant over distributed renewable energy, the distributed renewable energy is divided into controllable resources and uncontrollable resources; the risky asset response characteristic model is constructed based on the general flexible resource generation model and the response characteristics of uncontrollable resources; A risk-free asset response characteristic model is constructed based on the general flexible resource generation model and the response characteristics of controllable resources.

[0010] The present invention takes into account that the user's response value is affected by the user's response willingness. Based on the control authority of the virtual power plant over the interactive resources, the interactive resource response electricity is divided into risky and non-risky assets, and the uncertainty of the user's response electricity is measured by the combined risk.

[0011] Preferably, when constructing the risky asset response characteristic model, it is necessary to obtain the electricity purchase demand forecast curve and response range submitted by the user, and the virtual power plant can achieve any response power within the response range by default.

[0012] Preferably, the S1 comprises the following steps: The spatial aggregation characteristics and temporal aggregation characteristics of the virtual power plant are obtained, and distributed new energy sources are managed and interacted based on the spatial aggregation characteristics and temporal aggregation characteristics of the virtual power plant.

[0013] Preferably, S1 includes a constraint condition: the virtual power plant needs to ensure that the unbalanced electricity volume is within the allowable deviation of the electricity market.

[0014] Preferably, the general flexible resource power generation model includes the following formula: Where: r t -i,j,k , They are respectively the upper and lower bounds of the power generation, cumulative power generation and ramp rate of the j-th type and k-numbered flexible resource under the virtual power plant node i in the t-th time period.

[0015] Preferably, the risky asset response characteristic model includes the following formula: Where: C i，DR，k Responding to the cost of risky users, is the electricity purchase demand forecast curve, is the response range.

[0016] As a preference, the risky asset response characteristic model takes uncertainty into account and converts ΔP t i,DR,k Rewritten as: Among them, ΔP t -i,DR,k is the expected value of the response power allocated to the user, The random deviation of the user's response power.

[0017] Preferably, the risk-free asset response characteristic model includes the following formula: The superscript {i,DG,k} indicates that the load participating in the response is the k-th risk-free user at node i, and a 0-1 variable is introduced to indicate whether the risk-free user is cut. Indicates that it is cut in time period t; β i,DG,k Indicates the reduction rate of the load that can be reduced; represents the response costs of risk-free users; It represents the compensation of risk-free user capacity cost; It means that the load electricity price compensation can be reduced; Represents the cost of risk-free user unit response power.

[0018] Preferably, the formula of the optimization combination model is: max:E(R p )=C in -E(C cost ); Among them, E(R p ) represents the expected revenue of the virtual power plant after combination; C in represents the expected revenue of the virtual power plant, E(C cost ) is the expected cost, E(C cost ) is the expected cost; h(P t i,j,k )≥0 represents the risk and utility constraints of the optimal combination solution shown in the above formula, u(P t i,j,k)≥0 indicates resource response characteristic constraints, g(P t i,j,k )=0 is the equality constraint representing the aggregated external characteristics of the virtual power plant, s(P t i,j,k )≥0 means application constraints; The expected revenue of the virtual power plant is expressed as: C in =(C r -C w ) aim,t Δt; Among them, C r is the retail electricity price, C w is the wholesale electricity price in the electricity market, P aim,t is the bidding curve; The expected cost is expressed as: Among them: f1 is the incentive cost; f2 is the uncertainty cost, c er is the unit deviation penalty price; f3 is the access cost; C i,j,k is the unit response incentive cost of interactive resources; λ is the access cost required for a single resource to participate in regulation; V i,j,k is a 0 / 1 variable, V i,j,k =1 indicates that the resource participates in the response; the weights of various asset combinations are expressed as

[0019] The beneficial effects of the present invention are as follows: the present invention fully considers the heterogeneity of interactive resources and the timeliness of the virtual power plant alliance, establishes a general flexible resource power generation model, then measures the uncertainty of user response power with combined risk, establishes an optimized combination model with the goal of maximizing expected benefits, and then aggregates and configures distributed new energy, and finally verifies the impact of virtual power plant resource combination methods on the aggregate response capability and economy of virtual power plants. The present invention can take into account the wide range of resource types and different technical characteristics, decompose the virtual power plant resource combination problem into the alliance member composition method and member response plan allocation problem, and establish an optimized combination model; considering that the user's response value is affected by the user's willingness to respond, based on the virtual power plant's control authority over interactive resources, the interactive resource response power is divided into risky and risk-free assets, the user's response power uncertainty is measured with combined risk, and an optimized combination model is established with the goal of maximizing expected benefits, providing a basis for hybrid virtual power plant resource combination. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a flow chart of the method of the present invention.

[0021] Figure 2 It is a schematic diagram of the virtual power plant responding to electricity demand of the present invention.

[0022] Figure 3 It is a schematic diagram of the response range of the virtual power plant of the present invention. DETAILED DESCRIPTION

[0023] Example: This embodiment proposes a method for configuring distributed new energy aggregation. Figure 1 , including the following steps: S1, in order to facilitate the virtual power plant to regulate the distributed renewable energy, a management and interaction mechanism of the virtual power plant for the distributed renewable energy is constructed, so that the virtual power plant can control the resource equipment of the distributed renewable energy and put the distributed renewable energy in a specific working state; S2, in order to overcome the heterogeneity of distributed renewable energy and facilitate unified management, a general flexible resource generation model applicable to each distributed renewable energy is constructed according to the power generation characteristics of distributed renewable energy: The difficulty in unified management of distributed renewable energy lies in its heterogeneity, that is, there are significant differences in the entities, technical characteristics, and business objectives of distributed renewable energy. Distributed renewable energy is located at the end of the measurement system, and the distribution network can only actively manage the power equipment before the meter. For the power grid, the operating status of distributed renewable energy after the meter is uncontrollable and cannot be directly observed.

[0024] In order to overcome the constraints of distributed renewable energy in commercial applications and technology, virtual power plants take advantage of the advantages of multiple types of distributed renewable energy, and the complementary characteristics of the management and interaction mechanisms of distributed renewable energy, and give play to their scale benefits. Based on the flexibility of distributed renewable energy response capabilities, multiple types of distributed renewable energy are aggregated to form a virtual power plant, forming a virtual entity, which presents flexible adjustment characteristics similar to traditional power plants to the outside world, and can participate in the electricity market or grid interaction.

[0025] Virtual power plants aggregate distributed renewable energy through information and communication technology, which can break the limitations of geographical location. Therefore, the participants of virtual power plants are heterogeneous in technical characteristics and dispersed in geographical locations.

[0026] The aggregation characteristics of a virtual power plant are mainly composed of power baseline and regulation boundary profile, among which the analysis of interactive resource power characteristics is the basis for understanding the aggregation characteristics of a virtual power plant.

[0027] For the convenience of expression, the relevant parameters of the virtual power plant are first defined as shown in Table 1.

[0028] Table 1 Parameters of virtual power plant symbol illustrate {t:t=1,...,T} A day is divided into T time periods, where t is the time period number. {i:i∈I} i represents the number of the virtual power plant node <![CDATA[{j:j∈J i }]]> j represents the jth type of flexible power generation resource connected to the virtual power plant node i <![CDATA[{k:k∈K ij }]]> k represents the number of the jth flexible resource connected to the virtual power plant node i W,S,E Respectively represent wind power generation, photovoltaic power generation, energy storage and other power generation resource types It is assumed that the generation power of all flexible resources remains constant over a very short time period of length σ.

[0029] use It represents the power generation of the jth type and numbered k flexible resource connected to the virtual power plant node i in the tth time period. If the power value is positive, it means that the resource sends active power to the grid, and if it is negative, it means that the resource absorbs active power from the grid. It represents the cumulative power generation of the flexible resource of the jth type and numbered k in the virtual power plant node i from the 1st to the tth time period.

[0030] Definition: The generation characteristics of a generic flexible resource in time period t can be expressed as a six-tuple Description. The parameters of this six-tuple represent the active power boundary, cumulative power generation boundary and ramp rate boundary of the flexible resource in period t, namely: In the formula r t -i,j,k , They are respectively the upper and lower bounds of the power generation, cumulative power generation and ramp rate of the j-th type and k-numbered flexible resource under the virtual power plant node i in the t-th time period.

[0031] Since virtual power plants can continuously adjust the output power of distributed power sources, the volatility of the output power of distributed power sources affects virtual power plants mainly due to the difficulty in load forecasting, which leads to an unbalanced power between the actual operating power of the virtual power plant and the predicted power reported to the superior power grid. The flexibility of the virtual power plant's regulation is needed to compensate for this part of the unbalanced power and avoid economic losses, which also leads to a decrease in the flexibility of the virtual power plant.

[0032] The present invention can divide resources into two categories: controllable and uncontrollable according to the control authority of the virtual power plant over resource equipment.

[0033] In the contract signed between the virtual power plant and the user, if the user does not entrust the control authority of the terminal device to the virtual power plant, it can be regarded as a demand response user. The response power provided by such users is uncertain and is regarded as a risky asset. Most users are large-load users, such as campuses, office buildings, etc., which have a variety of equipment with adjustable power consumption characteristics and have their own energy management system. The response action of such users is similar to incentive-based demand response (IBDR). As an aggregator, the virtual power plant does not need to know the specific response method of such users, but only needs to pay attention to the actual response effect of the users. In order to reasonably optimize the user's electricity consumption plan, such users need to submit a power purchase demand forecast curve to the virtual power plant before the scheduling cycle. And response range The virtual power plant assumes that users can achieve any response power within their response capabilities.

[0034] Risky asset response characteristic model: Where ΔP t i,DR,k There is uncertainty, which causes the profit risk of the virtual power plant. The uncertainty of the response of general IBDR users is essentially due to the lack of information, and it is difficult to match their own response capabilities with the response requirements of the system; but under the management mode of the virtual power plant, the virtual power plant comprehensively considers the external demand and the response capabilities of internal resources, and then allocates the response power of each participating user. Therefore, the free response users can achieve global optimization by responding according to the allocation plan of the virtual power plant, and the response deviation will only bring economic losses to the virtual power plant. After considering the uncertainty, ΔP t i,DR,k Rewritten as follows: Where ΔP t -i,DR,k is the expected value of the response power allocated to the user, The random deviation of the user's response power.

[0035] According to the historical response deviation ε of risky asset users participating in the response, the standard deviation of each user's income σ can be obtained i and the correlation between risky assets ρ ij , ρ ij >0 means that the response errors of risky asset user i and risky asset user j are consistent, and they always have over-response or under-response at the same time; otherwise, ρ ij ≤0 means that the response errors between users are complementary.

[0036] Risk-free asset response characteristic model: Risk-free asset users, hereinafter referred to as DG users. DG users include distributed photovoltaic, distributed wind power generation, small hydropower, biomass power generation, etc. Among them, distributed photovoltaic and wind power generation are restricted by meteorological conditions. For example, the DC side of the photovoltaic grid-connected inverter usually adopts the MPPT strategy to control the active power output to the grid connection point. The MPPT power represents the maximum output of DG.

[0037] Under the maximum power limit, DG has a similar response characteristic to the load that can be reduced, and can reduce the active output below the MPPT power. However, from the perspective of the power grid, the reduction in DG output is equivalent to the increase in the amount of electricity purchased by the virtual power plant, which is regarded as providing positive response electricity. The response characteristics of DG are: The superscript {i, DG, k} indicates that the load participating in the response is the kth DG at node i. A 0-1 variable is introduced to indicate whether the DG is cut. Indicates that it is cut in time period t; β i,DG,k Indicates the reduction rate of the load that can be reduced; represents the response cost of DG; represents DG capacity fee compensation; It means that the load electricity price compensation can be reduced; Represents the cost of DG unit response power.

[0038] S3, establish an optimization portfolio model with the goal of maximizing expected returns: The resource optimization combination problem of virtual power plants is similar to the unit combination problem, but the unit combination problem focuses on the power generation capacity of the unit, while the virtual power plant focuses on the response capacity of the resources. An inappropriate virtual power plant resource combination method cannot meet the application requirements for its product characteristics, and it is difficult to obtain good economic efficiency. In this invention, a virtual power plant participating in the day-ahead energy market is taken as an example to establish an optimization combination model: max:E(R p )=C in -E(C cost ); In the model, the goal is to maximize the expected return. The objective function E(R p ) represents the expected revenue of the virtual power plant after combination; C in represents the expected revenue of the virtual power plant, and determines the bidding curve P when the market clears aim,t After that, the expected revenue from the energy market is C in =(C r -C w ) aim,t Δt is a fixed constant value, and the profit margin comes from the retail electricity price C r and wholesale electricity price C w The difference between E(C cost ) is the expected cost, which consists of three parts and is expressed as follows: Where f1 is the incentive cost required to encourage users to change their output plans and provide responsive power; f2 is the uncertainty cost caused by the deviation between the actual power of the virtual power plant and the target power, which corresponds to the penalty cost from the power market during the settlement stage; c er is the unit deviation penalty price; f3 is the access cost of communication and computing resources required to maintain users’ participation in virtual power plant interaction. i,j,kis the unit response incentive cost of interactive resources, i.e., the internal transaction price; λ is the access cost required for a single resource to participate in regulation; V i,j,k is a 0 / 1 variable, V i,j,k =1 indicates that the resource participates in the response.

[0039] Among the constraints, h(P t i,j,k )≥0 represents the risk and utility constraints of the optimal combination solution shown in the above formula, u(P t i,j,k )≥0 represents the resource response characteristic constraint shown in equations (4-5) to (4-34), g(P t i,j,k )=0 is the equality constraint representing the aggregated external characteristics of the virtual power plant, s(P t i,j,k )≥0 indicates application constraints. The weights of various asset portfolios can be expressed as Virtual power plant aggregation characteristics: Virtual power plants are software-defined resources, and their aggregation characteristics depend on the characteristics of the various distributed resources involved in the combination, as well as the resource control strategy within the virtual power plant. Virtual power plants need to adjust the output or working status of interactive resources within the allowable range to make the overall power of the system close to the target curve. Since interactive resources respond to virtual power plant control instructions and have certain power regulation characteristics, after aggregation, the power regulation capability of virtual power plants will produce a scale effect, and the power curve will be smoother.

[0040] The spatial area to which the user belongs is divided based on the grid connection point, and the spatial aggregation characteristics of the virtual power plant are obtained: After aggregating the response characteristics distributed in different geographical locations, the time aggregation characteristics of the virtual power plant are obtained: Where ΔP VPP,t represents the response value provided by the virtual power plant in period t, P VPP,t Represents the actual aggregate load value of the virtual power plant in period t after the response.

[0041] In my country's day-ahead energy market, virtual power plants, as independent third-party entities, compete for quantity but not price, and submit bid curves as price receivers in the market. Generally speaking, after weighing factors such as the power generation forecast error and load forecast error of internal resources, virtual power plant operators use the predicted virtual power plant aggregate power baseline as the bidding curve to participate in the bidding in the energy market. Through this strategy, ideally, the power baseline forecast value, the clearing power curve, and the actual power baseline value are completely consistent, and there is no need to adjust the user's electricity consumption plan. In this ideal case, the virtual power plant operator does not need to spend the incentive cost to compensate users for adjusting their electricity consumption plans, and the revenue can be increased; for users, the electricity consumption plan does not need to be adjusted, and the electricity consumption satisfaction is the highest. However, in reality, the market compares the bids of multiple participating entities and then clears, and it cannot be guaranteed that the clearing curve is consistent with the bidding curve. If the user's electricity consumption plan is not adjusted, there will be an unbalanced electricity between the actual power baseline and the clearing curve, resulting in the virtual power plant having to pay the market a corresponding amount according to the amount of unbalanced electricity, resulting in penalty costs.

[0042] According to the current implementation rules of the power spot market transactions in Jiangsu, Shandong, Guangdong and other places in my country, there is a clear market deviation processing mechanism in the settlement stage. For example, the processing mechanism for deviation electricity in the power spot market transaction rules of Shandong Province is as follows: In the energy market, the deviation between the electricity demand curve reported by the power sales company and wholesale users in the day-ahead electricity energy market and their actual electricity demand curve shall not exceed the allowable deviation range. When the actual deviation rate is higher than the maximum allowable reported deviation, the corresponding spot electricity energy market settlement income shall be recovered.

[0043] Based on this, the present invention sets the following background: on the premise that the power baseline forecast value and the clearing power have been determined, a combination of users participating in the response is selected and a user response plan is formulated, and combined with the above-mentioned power market transaction implementation rules, external characteristic requirements are used as application constraints for virtual power plants to participate in the day-ahead electricity energy market. In addition, referring to the transaction implementation rules, in the internal settlement principle of the virtual power plant, the present invention uses the penalty cost f2 in the objective function to characterize the size of the revenue recovery of the virtual power plant in the market settlement. According to market transaction rules, the virtual power plant needs to ensure that the unbalanced electricity volume is within the allowable deviation ε of the power market. VPP Inside: -ε VPP ≤P VPP,t Δt-P aim,t Δt≤ε VPP ; S4, aggregate and configure distributed renewable energy sources according to the output results of the optimization combination model.

[0044] The present invention also analyzes the impact of virtual power plant resource integration methods on the aggregate response capability and economy of virtual power plants. The target response power that a virtual power plant needs to provide during the dispatch cycle is as follows: Figure 2As shown, 1408kWh of response power and -808.4kWh of negative response power are required during the dispatch period, totaling 2216.41kWh of response power. Among them, in time period T1 = {2,3,4,5,6,8,10,11,12,13,14,15,16}, the response power is positive, and the demand is relatively large in time period 10-15; in time period T2 = {7,9,17,18,19,20,21,22,23,24}, the response power is negative, and the demand is relatively large in time period 18, 20-24. According to the contract signed with the user, the basic information of risk-free assets is obtained: (1) 30 single EV users with transferable load, charging power of 7kW; (2) There are 3 transferable load ESS users, with a maximum allowed charging / discharging power of 400 kW and a total installed capacity of 1 MW; (3) 14 load users can be reduced, a 10% reduction ratio is allowed, and a response power of -400 to 0 kW can be provided; (4) There are 5 distributed PV users, and a 5% reduction ratio is allowed, which can provide a response value of 0 to 285 kW. The response power of the free response users is settled in units of grid connection points. In the simulation example, it is assumed that there are 14 types of risky assets. The incentive unit prices of various users of the virtual power plant are shown in Table 2.

[0045] Table 2 Virtual power plant resource incentive costs Comparative analysis of combination methods: According to the combination method of alliance members, the following two virtual power plants are used for comparison: Virtual Power Plant 1 selects alliance members according to the optimized combination model proposed in the invention scheme; Virtual Power Plant 2 is a large alliance and does not select preferred combination members.

[0046] According to the target power regulation of the virtual power plant, two types of virtual power plants that meet the regulation requirements are obtained. The number of each type of user in the alliance members is shown in Table 3.

[0047] Table 3 Number of alliance members Distributed Power Demand Response Transferable load Translable load Load reduction possible total Virtual Power Plant 1 0 14 3 18 11 46 Virtual Power Plant 2 5 14 3 30 14 66 The regulation boundaries of the two virtual power plants are as follows: Figure 3As shown in Figure 1, ignoring the uncertainty of the response of demand response users, the regulation capabilities of both meet the regulation requirements. In periods 1 to 8, the target regulation power is positive, and the upward adjustment boundaries of the two virtual power plants are approximately the same, mainly provided by the demand response load; in periods 9 to 18, due to the photovoltaic output, the upward adjustment boundary of virtual power plant 2 is significantly higher than that of virtual power plant 1; in periods 19 to 24, the upward adjustment boundaries of the two virtual power plants are approximately the same. The downward adjustment capacity of the virtual power plant is provided by demand response users, transferable load users, shiftable load users and load reduction users. The shiftable load and load reduction users in virtual power plant 2 are less than those in virtual power plant 1. In periods 20 to 24, the downward adjustment capacity of virtual power plant 1 is slightly lower than that of virtual power plant 2. Another meaning of the combination is the asset allocation weight. The product costs provided by different types of users are regarded as assets. The asset allocation weights of the two virtual power plants are shown in Table 4.

[0048] Table 4 Virtual power plant asset allocation weights According to the previous formula, the expected cost of the product consists of three parts: incentive cost f1, penalty cost f2 and access cost f3. Among them, the amount of response power determines the size of f1; the size of the response deviation determines the value of f2. The adjustment range of the two virtual power plants completely covers the target response power. Therefore, there is no penalty cost caused by insufficient response capacity. It may only be the uncertainty of the response power of the demand response user that causes the response deviation. f2 is also the only variable with uncertainty in the objective function; the number of alliance members directly determines the value of f3. The more alliance members there are, the greater the access cost f3.

[0049] In summary, although the response method of virtual power plant 2 is more flexible and the response power is less, its cost f3 is higher than that of virtual power plant 1 due to the large number of members. In addition, the cost f1 reduced by virtual power plant 2 due to the reduction in response power fails to cover the increase in access cost f3 caused by the increase in the number of alliance members, resulting in the maximum expected cost of virtual power plant 2.

[0050] In addition, the alliance members of virtual power plant 1 are selected based on the response demand and resource response capability. DG is not included, and the response capability meets the requirements, which means that the access of DG in virtual power plant 2 only causes the access cost to increase, and does not participate in the response, resulting in low utilization of interactive resources. This also further illustrates that the appropriate combination is a necessary condition for achieving good economic efficiency.

[0051] Return-risk comparison analysis: Under the alliance member combination mode of virtual power plant 1, this scheme further compares and analyzes the impact of risk and utility constraints on asset allocation weights.

[0052] The risky allocation weight is reduced to 42.3%. The weights of the third, seventh and tenth asset combinations are relatively high, among which the risky asset allocation weight of the tenth category is the highest, only 6.37%. The asset allocation is relatively dispersed, and the correlation coefficient between the three is ρ 3，7 =-0.68,ρ 3，10 =-0.14, the strong negative correlation between Category 3 and Category 7 assets can effectively diversify the risk value of the portfolio. The risk-free allocation weight is 57.6%, of which the EV response plan for transferable loads reduces the portfolio weight to 13.24%, the ESS portfolio weight for transferable loads reduces to 15.72%, and the weight for curtailable loads is 21.34%. The expected return of the portfolio is reduced to 13,523 yuan, and the Sharpe ratio is 191.06. After taking into account the risk-utility constraint, the expected return increases by about 236 yuan for each additional unit of risk in the portfolio.

Claims

1. A method for configuring distributed new energy aggregation, characterized in that: The following steps are involved: S1, building a virtual power plant management and interaction mechanism for distributed renewable energy; S2, building a general flexible resource generation model based on the generation characteristics of distributed renewable energy; S3, establish an optimization portfolio model with the goal of maximizing expected returns; S4, aggregate and configure distributed renewable energy sources according to the output results of the optimization combination model.

2. According to claim 1, a method for configuring distributed new energy aggregation is characterized in that: The S2 further comprises the following steps: According to the control authority of the virtual power plant over distributed renewable energy, distributed renewable energy is divided into controllable resources and uncontrollable resources; Construct a risky asset response characteristic model based on the general flexible resource generation model and the response characteristics of uncontrollable resources; A risk-free asset response characteristic model is constructed based on the general flexible resource generation model and the response characteristics of controllable resources.

3. According to claim 2, a method for configuring distributed new energy aggregation is characterized in that: When constructing the risky asset response characteristic model, it is necessary to obtain the electricity purchase demand forecast curve and response range submitted by the user. The virtual power plant can achieve any response power within the response range by default.

4. According to claim 1, a method for configuring distributed new energy aggregation is characterized in that: The S1 comprises the following steps: The spatial aggregation characteristics and temporal aggregation characteristics of the virtual power plant are obtained, and distributed new energy sources are managed and interacted based on the spatial aggregation characteristics and temporal aggregation characteristics of the virtual power plant.

5. The method for configuring distributed new energy aggregation according to claim 1 is characterized in that: S1 includes constraints: the virtual power plant must ensure that the unbalanced electricity volume is within the allowable deviation of the electricity market.

6. The method for configuring distributed new energy aggregation according to claim 1 is characterized in that: The general flexible resource generation model includes the following formula: Where: They are respectively the upper and lower bounds of the power generation, cumulative power generation and ramp rate of the j-th type and k-numbered flexible resource under the virtual power plant node i in the t-th time period.

7. The method for configuring distributed new energy aggregation according to claim 2 is characterized in that: The risky asset response characteristic model includes the following formula: Where: C i，DR，k Responding to the cost of risky users, is the electricity purchase demand forecast curve, is the response range.

8. The method for configuring distributed new energy aggregation according to claim 7 is characterized in that: The risky asset response characteristic model takes uncertainty into account. Rewritten as: in, is the expected value of the response power allocated to the user, The random deviation of the user's response power.

9. The method for configuring distributed new energy aggregation according to claim 2 is characterized in that: The risk-free asset response characteristic model includes the following formula: The superscript {i,DG,k} indicates that the load participating in the response is the k-th risk-free user at node i, and a 0-1 variable is introduced to indicate whether the risk-free user is cut. Indicates that it is cut in time period t; β i,DG,k Indicates the reduction rate of the load that can be reduced; represents the response costs of risk-free users; It represents the compensation of risk-free user capacity cost; It means that the load electricity price compensation can be reduced; Represents the cost of risk-free user unit response power.

10. A method for configuring distributed new energy aggregation according to any one of claims 1 to 9, characterized in that: The formula of the optimization combination model is: max:E(R p )=C in -E(C cost ); Among them, E(R p ) represents the expected revenue of the virtual power plant after combination; C in represents the expected revenue of the virtual power plant, E(C cost ) is the expected cost, E(C cost ) is the expected cost; represents the risk and utility constraints of the optimal combination solution shown in the above formula, represents resource response characteristic constraints, To express the equality constraints of the aggregated external characteristics of the virtual power plant, Indicates application constraints; The expected revenue of the virtual power plant is expressed as: C in =(C r -C w )P aim,t Δt; Among them, C r is the retail electricity price, C w is the wholesale electricity price in the electricity market, P aim,t is the bidding curve; The expected cost is expressed as: Among them: f1 is the incentive cost; f2 is the uncertainty cost, c er is the unit deviation penalty price; f3 is the access cost; C i,j,k is the unit response incentive cost of interactive resources; λ is the access cost required for a single resource to participate in regulation; V i,j,k is a 0 / 1 variable, V i,j,k =1 indicates that the resource participates in the response; the weights of various asset combinations are expressed as

Images