A transaction-driven customized construction method for virtual power plants

Abstract

The present invention discloses a transaction-driven customized construction method for a virtual power plant, (1) DSO performs day-ahead market clearing according to the next-day output interval and quotation reported by the virtual power plant operator, and forms a day-ahead transaction contract with the clearing result and sends it to the virtual power plant operator; (2) The virtual power plant operator publishes the local market initial price to the producer-consumer cluster; (3) The producer-consumer cluster makes the optimal bidding decision based on the local market price and submits the bid to the virtual power plant operator; (4) The virtual power plant operator updates the local market price according to the bidding strategy of the producer-consumer with the goal of maximizing the operating profit and the constraint of meeting the customized output requirements of the virtual power plant, and publishes it to each producer-consumer; (5) If the local market price converges, the producer-consumer signs a day-ahead transaction contract with the virtual power plant operator according to the final decision result of both parties; otherwise, return to (3) cycle. The method provided by the present invention effectively promotes the balance of power supply and demand of the power system and promotes energy consumption, and provides a channel for small-scale distributed resources by organizing local market transactions.

Description

Technical Field

[0001] The present invention relates to a transaction-driven customized construction method of a virtual power plant, and belongs to power systems and automation technologies thereof. Background Art

[0002] With the increasing number of distributed resources such as rooftop photovoltaics, park wind power, energy storage equipment and controllable loads, the role of electricity users has changed from traditional "consumers" to current "producers and consumers". While having their own electricity demand, they also have the ability to generate electricity. According to the relative size of their power generation capacity and electricity demand and after optimal energy management, they can flexibly transform the roles of "producers" and "consumers". They can also use their own power generation capacity and flexible regulation potential to provide services for the power grid, among which distributed power sources can provide electricity services for the power grid, and energy storage equipment and controllable loads can provide auxiliary services for the power grid.

[0003] At present, the surplus electricity of producers and consumers can only be purchased by the power grid at a very low grid-connected electricity price, while the insufficient electricity must be purchased from the electricity seller at the retail electricity price, which has low economic benefits for individuals. With the expansion of the scale of distributed power sources, in the context of deepening the reform of the power system, the National Development and Reform Commission and the National Energy Administration issued the "Notice on the Pilot of Distributed Power Generation Market Trading" in November 2017, proposing three guiding distributed power generation market trading models including direct trading, entrusted sales, and purchase at benchmark electricity prices, aiming to introduce distributed power generation into competitive market transactions through distributed transactions, reduce policy subsidies for distributed power generation projects, and encourage them to improve their competitiveness and achieve sustainable development.

[0004] Virtual power plants use advanced coordination and control technology, intelligent metering technology, and information and communication technology to aggregate multiple distributed resources such as distributed power sources, energy storage systems, and demand response resources to form a unified output to the outside world. As an intermediate unit, they participate in market transactions and grid dispatching, and can achieve interconnection and interaction between many power generation and power consumption equipment. They are a typical example of the ubiquitous power Internet of Things, and provide a market trading platform for many small-scale, dispersed distributed resources.

[0005] Therefore, a transaction-driven customized construction method of virtual power plants is proposed. The clusters of producers and consumers in the same distribution station area are aggregated into a virtual power plant. The virtual power plant participates in day-ahead contract transactions as the application scenario. According to the output requirements stipulated in the contract, the next-day power generation and consumption plan of the producers and consumers within the virtual power plant is determined in a transaction-driven manner to achieve customized construction of virtual power plants. This can not only effectively promote the balance of power supply and demand in the power system and promote the local consumption of new energy, but also provide profit channels for small-scale distributed resources by organizing local market transactions. Summary of the invention

[0006] Purpose of the invention: In order to make full use of the power generation and power regulation capabilities of distributed resources, promote the local consumption of new energy, and ensure the balance of power supply and demand in the power system from the market transaction link, the present invention provides a transaction-driven customized construction method for a virtual power plant, which aggregates clusters of producers and consumers in the same distribution station area into a virtual power plant, and takes the virtual power plant participating in the day-ahead contract transaction as the application scenario. According to the output requirements stipulated in the contract, the next-day power generation and consumption plan of the producers and consumers within the virtual power plant is determined in a transaction-driven manner, thereby realizing the customized construction of the virtual power plant.

[0007] Technical solution: To achieve the above purpose, the technical solution adopted by the present invention is:

[0008] A transaction-driven virtual power plant customization construction method includes the following steps:

[0009] (1) DSO conducts day-ahead market clearing based on the next-day output range and corresponding quotations reported by the virtual power plant operator, and issues the clearing results as day-ahead transaction contracts to the virtual power plant operator;

[0010] (2) The virtual power plant operator publishes the initial local market price to the prosumer cluster;

[0011] (3) The prosumer cluster makes the best bidding decision based on the local market price and submits the bidding strategy to the virtual power plant operator;

[0012] (4) The virtual power plant operator aims to maximize operating profits and is constrained by meeting the customized output requirements of the virtual power plant. It updates the local market price based on the bidding strategy of the prosumers and publishes it to each prosumer.

[0013] (5) If the local market price converges, the prosumer signs a day-ahead transaction contract with the virtual power plant operator based on the final decision of both parties to complete the customized construction of the virtual power plant; otherwise, the process returns to (3) and repeats the cycle.

[0014] Specifically, in step (1), the DSO performs day-ahead market clearing based on the next-day output range and corresponding quotation reported by the virtual power plant operator, and sends the clearing result to the virtual power plant operator in the form of a day-ahead transaction contract, specifically:

[0015] Based on the next-day output forecast data reported by the prosumer cluster, the virtual power plant operator calculates the output range and corresponding quotation for each period of the next day:

[0016]

[0017]

[0018] Where: P t N is the power of the virtual power plant in the day-ahead contract, Day-ahead contract transaction electricity price; g i,t represents the power generation of prosumer i, d i,t represents the power consumption of prosumer i; the superscripts min and max represent the minimum and maximum values ​​of the corresponding items respectively; and is the quotation coefficient of the virtual power plant operator to the DSO; Ψ represents the set of production-type prosumers, Ω represents the set of consumption-type prosumers, T represents the set of trading periods, d j，t is the internal load power demand of prosumer j in the tth trading period, g j，i is the day-ahead forecast of power generation of prosumer j, and t represents the tth trading period.

[0019] Specifically, in step (2), the virtual power plant operator publishes the local market initial price to the prosumer cluster, specifically:

[0020] (2-1) Initial electricity purchase price in the local market is a random number between the unified power purchase price of the power grid and the contract transaction price, as shown in the following formula

[0021]

[0022] in: Unify the electricity purchase price for the power grid.

[0023] (2-2) Initial electricity sales price in the local market is a random number between the contract transaction price and the retail market price, as shown below

[0024]

[0025] in: The retail market electricity price.

[0026] Specifically, in step (3), the prosumer cluster makes the optimal bidding decision based on the local market price and submits a bidding strategy to the virtual power plant operator, specifically:

[0027] (3-1) Prosumers calculate the power generation-consumption ratio (GDR) according to the following formula: t , and thus determine their own identity - producer or consumer

[0028]

[0029] Where: g t is the electricity generated by the prosumer, d t For electricity consumption by producers and consumers; when GDR t When ≥1, the prosumer is a producer, otherwise it is a consumer.

[0030] (3-2) The producer and consumer uses the following quadratic utility function to calculate their own utility U when consuming different amounts of electricity t

[0031]

[0032]

[0033] Where: μ t is the user preference parameter that characterizes the user's electricity consumption behavior, ν is the parameter that each user has set, and P t con is the electricity consumption of the prosumer, and They are the upper and lower limits of electricity demand of producers and consumers respectively.

[0034] (3-3) Prosumers make optimal bidding decisions based on the local market electricity purchase and sales prices and submit bidding strategies to the virtual power plant operator and

[0035] 1) Prosumers as producers

[0036] Let i be the number of the production-type prosumer. When acting as a producer, the prosumer can sell electricity to the virtual power plant operator on the basis of meeting its own electricity demand. The bidding strategy submitted by prosumer i includes the price sensitivity coefficient a i,t And the power limit of electricity sales Forming a bidding strategy set

[0037] Since the prosumer has flexible loads such as central air conditioning and energy storage equipment with adjustable power, the amount of electricity available for sale is sensitive to price, as shown in the following formula.

[0038]

[0039]

[0040] in: is the electricity sales power of prosumer i in the local market; is the price elasticity coefficient of electricity sold by producer / consumer i in the local market, and the purchase price in the local market can be known. The higher it is, the more electricity prosumer i is willing to sell in the local market; represents the willingness of prosumer i to sell electricity in the local market. The premise for prosumer i to choose to sell electricity in the local market is that the purchase price in the local market is Not less than the unified purchase price of the power grid From this we can get is the maximum power sold by producer / consumer i, which should meet the minimum power demand of internal loads Consider selling electricity to the outside world on the basis of i,t is the day-ahead forecast value of power generation of prosumer i.

[0041] The benefit Π of producer-consumer i i,t It includes the following items: the utility of its own electricity consumption, the income from selling electricity in the local market, the income from connecting the surplus electricity to the grid, and the cost of its own power generation. Prosumer i performs day-ahead optimization with the goal of maximizing utility. The objective function is shown in the following formula.

[0042]

[0043] in: is the power of the central air conditioning system of prosumer i; U i,t is the electricity consumption of prosumer i; i,t is the unit electricity generation cost of prosumer i; g i,t is the electricity generation of prosumer i; It is the power value uniformly purchased by the power grid for prosumer i.

[0044] st

[0045] a) Power balance constraint: The sum of the power consumed by producer i and the power sold in the local market should be equal to the total power generated.

[0046]

[0047]

[0048] b) Flexible load constraint. The power consumed by the prosumer i is equal to the power of the air conditioning system And uncontrollable load power Taking into account the adjustable potential of flexible load resources, the central air-conditioning power consumption of prosumer i should be between the upper and lower limits of its own power demand.

[0049]

[0050]

[0051]

[0052]

[0053] in: is the total power consumption of prosumer i; and are the minimum and maximum power of the central air-conditioning system of prosumer i respectively.

[0054] c) Price elasticity constraint. The price elasticity of producer-consumer i is constrained by the user's willingness to use and sell electricity, and should not exceed the maximum price elasticity coefficient of each user.

[0055]

[0056] 2) Prosumers as consumers

[0057] Let j be the consumer prosumer ID. When the prosumer acts as a consumer, the bidding strategy submitted includes the price sensitivity coefficient and power purchase power limit Forming a bidding strategy set

[0058] In this paper, the virtual power plant operator sells electricity to prosumers at a local market price between the retail market price and the contract price. By adjusting the local market price, the power consumption of the prosumers is changed, so that the virtual power plant can meet customized demand for external power generation. The relationship between the power consumption of the prosumers and the local power price is as follows.

[0059]

[0060]

[0061] in: is the power purchased by prosumer j in the local market; is the price elasticity coefficient of electricity purchased by producer / consumer j in the local market, and the electricity selling price in the local market can be known. The higher it is, the more electricity prosumer j is willing to sell in the local market; represents the willingness of prosumer j to purchase electricity in the local market. The premise for prosumer j to choose to purchase electricity in the local market is that the electricity price in the local market is Not higher than the retail market electricity price From this we can get is the maximum power purchase power of prosumer j, i.e., the upper limit of its power demand

[0062] The benefit of producer / consumer j j,t It includes the following items: its own electricity utility, the cost of purchasing electricity in the local market, and its own power generation cost. Prosumer j performs day-ahead optimization with the goal of maximizing utility. The objective function is shown in the following formula.

[0063]

[0064] in: is the power of the central air conditioning system of producer / consumer j; U j,t is the electricity consumption of producer / consumer j; j,tis the unit electricity generation cost of prosumer j; g j,t is the electricity generation of prosumer j.

[0065] st

[0066] a) Power balance constraint: The sum of the power purchased by producer / consumer j in the local market and its own power generation should be equal to the total power consumption.

[0067]

[0068] in: is the total power consumed by prosumer j.

[0069] b) Flexible load constraints: Constraints when the prosumer acts as a producer.

[0070] c) Price elasticity constraint. The price elasticity of producer-consumer j is constrained by the user's own willingness to use electricity and willingness to purchase electricity, and should not be lower than the minimum price elasticity coefficient of each user.

[0071]

[0072] (3-4) The step-size control method is used to further constrain the decision-making results of the bidding coefficient to prevent it from changing significantly. The specific formula is as follows:

[0073]

[0074] Δ=|RMP t (r+1)·a i,t (r)| (24)

[0075] Where: r is the number of iterations; Δ is the step size during the iteration, a i,t (r) is the decision value of the bidding coefficient obtained in the rth round of iteration, is the optimal value of the bidding coefficient obtained in the r+1th round of iteration, RMP t ∈[0,1] is the climbing coefficient. This step is equivalent to correcting the bidding coefficient obtained by local optimization according to the step size control requirements, so as to obtain the decision value of the bidding coefficient of this round.

[0076] Specifically, in step (4), the virtual power plant operator takes maximizing operating revenue as the goal and satisfying the customized output requirements of the virtual power plant as the constraint, updates the local market price according to the bidding strategy of the prosumer, and publishes it to each prosumer, specifically:

[0077] (4-1) Multi-scenario technology is used to simulate the uncertainty of renewable energy output, user load demand and penalty for breach of trading contracts during actual operation, and the stochastic optimization problem is transformed into a deterministic optimization problem.

[0078] (4-2) Conditional value at risk is used to measure the risk of the default part of the virtual power plant transaction contract, and to reasonably weigh the operating benefits and potential risks of the virtual power plant.

[0079] (4-3) The virtual power plant operator makes a day-ahead optimization decision based on the bidding strategy submitted by the prosumer with the goal of minimizing the customization cost of the virtual power plant. The optimization model is constructed as follows:

[0080]

[0081]

[0082] Where: VPP Customizing costs for virtual power plants; is the electricity purchase cost of the virtual power plant operator in the xth prosumer electricity sales scenario; is the electricity sales revenue of the virtual power plant operator in the yth prosumer electricity purchasing scenario; R con Revenue from signing power contracts for virtual power plant operators; x and ω y are the scenario probabilities corresponding to the xth prosumer electricity sales scenario and the yth prosumer electricity purchase scenario, respectively. There are X prosumer electricity sales scenarios and Y electricity purchase scenarios; γ is the risk aversion coefficient of the virtual power plant operator; V risk is the conditional risk value of the virtual power plant customization problem; Ξ is the decision variable set of the optimization model; ξ is the risk value of the virtual power plant customization problem; δ xyh For linearization V risk The auxiliary variables introduced.

[0083] The calculation of each part in the objective function is as follows:

[0084]

[0085] in: is the amount of electricity purchased by the virtual power plant operator from prosumer i in the xth prosumer electricity sales scenario; is the amount of electricity sold by the virtual power plant operator to prosumer j in the yth prosumer electricity purchasing scenario; Δt is the duration of the unit transaction period; α is the confidence level; ω h is the scenario probability corresponding to the h-th contract breach penalty price, and there are H contract breach penalty price scenarios in total.

[0086] st

[0087] a) Power balance constraint: The net external output of the virtual power plant should not be lower than the customized output value specified in the contract.

[0088]

[0089] b) Constraints on electricity purchase and sales prices. The electricity purchase price of the virtual power plant operator for production-type prosumers should be between the unified purchase price of the power grid and the contract transaction price, and the electricity sales price of the virtual power plant operator for consumption-type prosumers should be between the wholesale market price and the retail market price.

[0090]

[0091] c) Risk constraints. This paper believes that if the actual power delivered by the virtual power plant operator is lower than the value specified in the contract or exceeds the value specified in the contract and the deviation is greater than ΔP max When the deviation occurs, the contract breach penalty needs to be paid for the deviation. risk The auxiliary variable δ in xyh The following constraints should be met.

[0092]

[0093]

[0094]

[0095]

[0096]

[0097] in: The contract breach penalty for the virtual power plant operator under the xth prosumer electricity sales scenario, the yth prosumer electricity purchase scenario, and the hth contract breach penalty price scenario; and They are the actual power purchase deviation ratio and actual power sales deviation ratio of virtual power plant operators respectively; and are the h-th contract breach penalty prices in the case of insufficient and excessive delivery, respectively; ΔP max Indicates the maximum deviation allowed for overdelivery; [v] + means taking the larger value between v and 0, where v refers to and

[0098] (4-4) The deviation between the actual transaction value and the planned value between the virtual power plant operator and the prosumer cluster follows a normal distribution, as shown in the following formula:

[0099]

[0100] Where: μ b and σ b are the expected value and standard deviation of the actual delivery volume obtained by the virtual power plant operator through power purchase transactions with the prosumer cluster, μs and σ s are the expected value and standard deviation of the actual delivery volume obtained by the virtual power plant operator in the power sales transaction with the prosumer cluster, respectively.

[0101] (4-5) After the virtual power plant operator updates the local market price in round r, a certain number of prosumers are randomly selected according to the following formula and are prohibited from changing their bidding strategies in the next round of iterations:

[0102]

[0103] in: The number of prosumers that are prohibited from updating their bidding strategies; N PRS Total number of prosumers customized for participation in the virtual power plant.

[0104] Specifically, in step (5), if the local market price converges, the prosumer signs a day-ahead transaction contract with the virtual power plant operator based on the final decision of both parties to complete the customized construction of the virtual power plant. Otherwise, the process returns to step (3) to loop, specifically:

[0105] The following formula is used as the convergence criterion for local market prices

[0106]

[0107] in: is the price vector of electricity purchase and sale in the local market; π is a minimum positive value. BRIEF DESCRIPTION OF THE DRAWINGS

[0108] Figure 1 1 is a flow chart for implementing the method of the present invention; DETAILED DESCRIPTION

[0109] The present invention will be further described below in conjunction with the accompanying drawings.

[0110] like Figure 1 The figure shows a transaction-driven customized construction method of a virtual power plant. The entire implementation process is described in detail below.

[0111] Step 1: DSO conducts day-ahead market clearing based on the next-day output range and corresponding quotations reported by the virtual power plant operator, and sends the clearing results to the virtual power plant operator in the form of a day-ahead transaction contract.

[0112] Based on the next-day output forecast data reported by the prosumer cluster, the virtual power plant operator calculates the output range and corresponding quotation for each period of the next day:

[0113]

[0114]

[0115] Where: P t N is the power of the virtual power plant in the day-ahead contract, Day-ahead contract transaction electricity price; g i,t represents the power generation of prosumer i, d i,t represents the power consumption of prosumer i; the superscripts min and max represent the minimum and maximum values ​​of the corresponding items respectively; and is the quotation coefficient of the virtual power plant operator to the DSO; Ψ represents the set of production-type prosumers, Ω represents the set of consumption-type prosumers, T represents the set of trading periods, d j，t is the internal load power demand of prosumer j in the tth trading period, g j，i is the day-ahead forecast of power generation of prosumer j, and t represents the tth trading period.

[0116] Step 2: The virtual power plant operator publishes the initial local market price to the prosumer cluster.

[0117] (2-1) Initial electricity purchase price in the local market is a random number between the unified power purchase price of the power grid and the contract transaction price, as shown in the following formula

[0118]

[0119] in: Unify the electricity purchase price for the power grid.

[0120] (2-2) Initial electricity sales price in the local market is a random number between the contract transaction price and the retail market price, as shown below

[0121]

[0122] in: The retail market electricity price.

[0123] Step 3: The prosumer cluster makes the optimal bidding decision based on the local market price and submits the bidding strategy to the virtual power plant operator.

[0124] (3-1) Prosumers calculate the power generation-consumption ratio (GDR) according to the following formula: t , and thus determine their own identity - producer or consumer

[0125]

[0126] Where: g t is the electricity generated by the prosumer, d t For electricity consumption by producers and consumers; when GDR tWhen ≥1, the prosumer is a producer, otherwise it is a consumer.

[0127] (3-2) The producer and consumer uses the following quadratic utility function to calculate their own utility U when consuming different amounts of electricity t

[0128]

[0129]

[0130] Where: μ t is the user preference parameter that characterizes the user's electricity consumption behavior, ν is the parameter that each user has set, and P t con is the electricity consumption of the prosumer, and They are the upper and lower limits of electricity demand of producers and consumers respectively.

[0131] (3-3) Prosumers make optimal bidding decisions based on the local market electricity purchase and sales prices and submit bidding strategies to the virtual power plant operator and

[0132] 1) Prosumers as producers

[0133] Let i be the number of the production-type prosumer. When acting as a producer, the prosumer can sell electricity to the virtual power plant operator on the basis of meeting its own electricity demand. The bidding strategy submitted by prosumer i includes the price sensitivity coefficient a i,t And the power limit of electricity sales Forming a bidding strategy set

[0134] Since the prosumer has flexible loads such as central air conditioning and energy storage equipment with adjustable power, the amount of electricity available for sale is sensitive to price, as shown in the following formula.

[0135]

[0136]

[0137] in: is the electricity sales power of prosumer i in the local market; is the price elasticity coefficient of electricity sold by producer / consumer i in the local market, and the purchase price in the local market can be known. The higher it is, the more electricity prosumer i is willing to sell in the local market; represents the willingness of prosumer i to sell electricity in the local market. The premise for prosumer i to choose to sell electricity in the local market is that the purchase price in the local market is Not lower than the unified purchase price of the power grid From this we can get is the maximum power sold by producer / consumer i, which should meet the minimum power demand of internal loads Consider selling electricity to the outside world on the basis of i,t is the day-ahead forecast value of power generation of prosumer i.

[0138] The benefit Π of producer-consumer i i,t It includes the following items: the utility of its own electricity consumption, the income from selling electricity in the local market, the income from connecting the surplus electricity to the grid, and the cost of its own power generation. Prosumer i performs day-ahead optimization with the goal of maximizing utility. The objective function is shown in the following formula.

[0139]

[0140] in: is the power of the central air conditioning system of prosumer i; U i,t is the electricity consumption of prosumer i; i,t is the unit electricity generation cost of prosumer i; g i,t is the electricity generation of prosumer i; It is the power value uniformly purchased by the power grid for prosumer i.

[0141] st

[0142] a) Power balance constraint: The sum of the power consumed by producer i and the power sold in the local market should be equal to the total power generated.

[0143]

[0144]

[0145] b) Flexible load constraint. The power consumed by the prosumer i is equal to the power of the air conditioning system And uncontrollable load power Taking into account the adjustable potential of flexible load resources, the central air-conditioning power consumption of prosumer i should be between the upper and lower limits of its own power demand.

[0146]

[0147]

[0148]

[0149]

[0150] in: is the total power consumption of prosumer i; and are the minimum and maximum power of the central air-conditioning system of prosumer i respectively.

[0151] c) Price elasticity constraint. The price elasticity of producer-consumer i is constrained by the user's willingness to use and sell electricity, and should not exceed the maximum price elasticity coefficient of each user.

[0152]

[0153] 2) Prosumers as consumers

[0154] Let j be the consumer prosumer ID. When the prosumer acts as a consumer, the bidding strategy submitted includes the price sensitivity coefficient and power purchase power limit Forming a bidding strategy set

[0155] In this paper, the virtual power plant operator sells electricity to prosumers at a local market price between the retail market price and the contract price. By adjusting the local market price, the power consumption of the prosumers is changed, so that the virtual power plant can meet customized demand for external power generation. The relationship between the power consumption of the prosumers and the local power price is as follows.

[0156]

[0157]

[0158] in: is the power purchased by prosumer j in the local market; is the price elasticity coefficient of electricity purchased by producer / consumer j in the local market, and the electricity selling price in the local market can be known. The higher it is, the more electricity prosumer j is willing to sell in the local market; represents the willingness of prosumer j to purchase electricity in the local market. The premise for prosumer j to choose to purchase electricity in the local market is that the electricity price in the local market is Not higher than the retail market electricity price From this we can get is the maximum power purchase power of prosumer j, i.e., the upper limit of its power demand

[0159] The benefit of producer / consumer j j,t It includes the following items: its own electricity utility, the cost of purchasing electricity in the local market, and its own power generation cost. Prosumer j performs day-ahead optimization with the goal of maximizing utility. The objective function is shown in the following formula.

[0160]

[0161] in: is the power of the central air conditioning system of producer / consumer j; U j,t is the electricity consumption of producer / consumer j;j,t is the unit electricity generation cost of prosumer j; g j,t is the electricity generation of prosumer j.

[0162] st

[0163] a) Power balance constraint: The sum of the power purchased by producer / consumer j in the local market and its own power generation should be equal to the total power consumption.

[0164]

[0165] in: is the total power consumed by prosumer j.

[0166] b) Flexible load constraints: Constraints when the prosumer acts as a producer.

[0167] c) Price elasticity constraint. The price elasticity of producer-consumer j is constrained by the user's own willingness to use electricity and willingness to purchase electricity, and should not be lower than the minimum price elasticity coefficient of each user.

[0168]

[0169] (3-4) The step-size control method is used to further constrain the decision-making results of the bidding coefficient to prevent it from changing significantly. The specific formula is as follows:

[0170]

[0171] Δ=|RMP t (r+1)·a i,t (r)| (24)

[0172] Where: r is the number of iterations; Δ is the step size during the iteration, a i,t (r) is the decision value of the bidding coefficient obtained in the rth round of iteration, is the optimal value of the bidding coefficient obtained in the r+1th round of iteration, RMP t ∈[0,1] is the climbing coefficient. This step is equivalent to correcting the bidding coefficient obtained by local optimization according to the step size control requirements, so as to obtain the decision value of the bidding coefficient of this round.

[0173] Step 4: The virtual power plant operator aims to maximize operating profits and is constrained by meeting the customized output requirements of the virtual power plant. It updates the local market price based on the bidding strategy of the prosumers and publishes it to each prosumer.

[0174] (4-1) Multi-scenario technology is used to simulate the uncertainty of renewable energy output, user load demand and penalty for breach of trading contracts during actual operation, and the stochastic optimization problem is transformed into a deterministic optimization problem.

[0175] (4-2) Conditional value at risk is used to measure the risk of the default part of the virtual power plant transaction contract, and to reasonably weigh the operating benefits and potential risks of the virtual power plant.

[0176] (4-3) The virtual power plant operator makes a day-ahead optimization decision based on the bidding strategy submitted by the prosumer with the goal of minimizing the customization cost of the virtual power plant. The optimization model is constructed as follows:

[0177]

[0178]

[0179] Where: VPP Customizing costs for virtual power plants; is the electricity purchase cost of the virtual power plant operator in the xth prosumer electricity sales scenario; is the electricity sales revenue of the virtual power plant operator in the yth prosumer electricity purchasing scenario; R con Revenue from signing power contracts for virtual power plant operators; x and ω y are the scenario probabilities corresponding to the xth prosumer electricity sales scenario and the yth prosumer electricity purchase scenario, respectively. There are X prosumer electricity sales scenarios and Y electricity purchase scenarios; γ is the risk aversion coefficient of the virtual power plant operator; V risk is the conditional risk value of the virtual power plant customization problem; Ξ is the decision variable set of the optimization model; ξ is the risk value of the virtual power plant customization problem; δ xyh For linearization V risk The auxiliary variables introduced.

[0180] The calculation of each part in the objective function is as follows:

[0181]

[0182] in: is the amount of electricity purchased by the virtual power plant operator from prosumer i in the xth prosumer electricity sales scenario; is the amount of electricity sold by the virtual power plant operator to prosumer j in the yth prosumer electricity purchasing scenario; Δt is the duration of the unit transaction period; α is the confidence level; ω h is the scenario probability corresponding to the h-th contract breach penalty price, and there are H contract breach penalty price scenarios in total.

[0183] st

[0184] a) Power balance constraint: The net external output of the virtual power plant should not be lower than the customized output value specified in the contract.

[0185]

[0186] b) Constraints on electricity purchase and sales prices. The electricity purchase price of the virtual power plant operator for production-type prosumers should be between the unified purchase price of the power grid and the contract transaction price, and the electricity sales price of the virtual power plant operator for consumption-type prosumers should be between the wholesale market price and the retail market price.

[0187]

[0188]

[0189] c) Risk constraints. This paper believes that if the actual power delivered by the virtual power plant operator is lower than the value specified in the contract or exceeds the value specified in the contract and the deviation is greater than ΔP max When the deviation occurs, the contract breach penalty needs to be paid for the deviation. risk The auxiliary variable δ in xyh The following constraints should be met.

[0190]

[0191]

[0192]

[0193]

[0194]

[0195] in: The contract breach penalty for the virtual power plant operator under the xth prosumer electricity sales scenario, the yth prosumer electricity purchase scenario, and the hth contract breach penalty price scenario; and They are the actual power purchase deviation ratio and actual power sales deviation ratio of virtual power plant operators respectively; and are the h-th contract breach penalty prices in the case of insufficient and excessive delivery, respectively; ΔP max Indicates the maximum deviation allowed for overdelivery; [v] + means taking the larger value between v and 0, where v refers to and

[0196] (4-4) The deviation between the actual transaction value and the planned value between the virtual power plant operator and the prosumer cluster follows a normal distribution, as shown in the following formula:

[0197]

[0198] Where: μ b and σ bare the expected value and standard deviation of the actual delivery volume obtained by the virtual power plant operator through power purchase transactions with the prosumer cluster, μ s and σ s are the expected value and standard deviation of the actual delivery volume obtained by the virtual power plant operator in the power sales transaction with the prosumer cluster, respectively.

[0199] (4-5) After the virtual power plant operator updates the local market price in round r, a certain number of prosumers are randomly selected according to the following formula and are prohibited from changing their bidding strategies in the next round of iterations:

[0200]

[0201] in: The number of prosumers that are prohibited from updating their bidding strategies; N PRS Total number of prosumers customized for participation in the virtual power plant.

[0202] Step 5: If the local market prices converge, the prosumer signs a day-ahead transaction contract with the virtual power plant operator based on the final decision of both parties to complete the customized construction of the virtual power plant. Otherwise, return to step 3 to repeat the cycle.

[0203] The following formula is used as the convergence criterion for local market prices

[0204]

[0205] in: is the price vector of electricity purchase and sale in the local market; ε π is a minimum positive value.

[0206] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A transaction-driven customized construction method for a virtual power plant, characterized by: The steps include: (1) DSO conducts day-ahead market clearing based on the next-day output range and corresponding quotations reported by the virtual power plant operator, and issues the clearing results as day-ahead transaction contracts to the virtual power plant operator. Specifically: Based on the next-day output forecast data reported by the prosumer cluster, the virtual power plant operator calculates the output range and corresponding quotation for each period of the next day: Where: P t N is the power of the virtual power plant in the day-ahead contract, Day-ahead contract transaction electricity price; g i,t represents the power generation of prosumer i, d i,t represents the power consumption of prosumer i; the superscripts min and max represent the minimum and maximum values ​​of the corresponding items respectively; and is the quotation coefficient of the virtual power plant operator to the DSO; Ψ represents the set of production-type prosumers, Ω represents the set of consumption-type prosumers, and T represents the set of trading periods; d j，t is the internal load power demand of prosumer j in the tth trading period, g j，t is the day-ahead forecast of the power generation of prosumer j, and t represents the tth trading period; (2) The virtual power plant operator publishes the initial local market price to the prosumer cluster, which is: (2-1) Initial electricity purchase price in the local market is a random number between the unified power purchase price of the power grid and the contract transaction price, as shown in the following formula in: Unify the purchase price of electricity for the power grid; (2-2) Initial electricity sales price in the local market is a random number between the contract transaction price and the retail market price, as shown below in: for retail market electricity prices; (3) The prosumer cluster makes the best bidding decision based on the local market price and submits a bidding strategy to the virtual power plant operator. Specifically: (3-1) Prosumers calculate the power generation-consumption ratio (GDR) according to the following formula: t , and thus determine their own identity - producer or consumer Where: g t is the electricity generated by the prosumer, d t For electricity consumption by producers and consumers; when GDR t When ≥1, the prosumer is a producer, otherwise it is a consumer; (3-2) The producer and consumer uses the following quadratic utility function to calculate their own utility U when consuming different amounts of electricity t Where: μ t is the user preference parameter that characterizes the user's electricity consumption behavior, ν is the parameter that each user has set, and P t con is the electricity consumption of the prosumer, and They are the upper and lower limits of electricity demand of prosumers; (3-3) Prosumers make optimal bidding decisions based on the local market electricity purchase and sales prices and submit bidding strategies to the virtual power plant operator and 1) Prosumers as producers Let i be the number of the production-type prosumer. When acting as a producer, the prosumer can sell electricity to the virtual power plant operator on the basis of meeting its own electricity demand. The bidding strategy submitted by prosumer i includes the price sensitivity coefficient a i,t And the power limit of electricity sales Forming a bidding strategy set Since the prosumer has flexible loads such as central air conditioning and energy storage equipment with adjustable power, the amount of electricity available for sale is sensitive to price, as shown in the following formula: in: is the electricity sales power of prosumer i in the local market; is the price elasticity coefficient of electricity sold by producer / consumer i in the local market, and the purchase price in the local market can be known. The higher it is, the more electricity prosumer i is willing to sell in the local market; represents the willingness of prosumer i to sell electricity in the local market. The premise for prosumer i to choose to sell electricity in the local market is that the purchase price in the local market is Not lower than the unified purchase price of the power grid From this we can get is the maximum power sold by producer / consumer i, which should meet the minimum power demand of internal loads Consider selling electricity to the outside world on the basis of i,t is the day-ahead forecast value of power generation of prosumer i; The benefit Π of producer-consumer i i,t It includes the following items: the utility of its own electricity consumption, the income from selling electricity in the local market, the income from connecting the surplus electricity to the grid, and the cost of its own power generation. The producer and consumer i performs day-ahead optimization with the goal of maximizing utility. The objective function is shown in the following formula: in: is the power of the central air conditioning system of prosumer i; U i,t is the electricity consumption of prosumer i; i,t is the unit electricity generation cost of prosumer i; g i,t is the electricity generation of prosumer i; is the power value uniformly purchased by the power grid for producer and consumer i; a) Power balance constraint: the sum of the power consumed by producer i and the power sold in the local market should be equal to the total power generated: b) Flexible load constraint, the power consumed by the prosumer i is equal to the power of the air conditioning system And uncontrollable load power The sum of the power consumption of the central air conditioner of the producer and consumer i should be between the upper and lower limits of its own power demand, considering the adjustable potential of the flexible load resources; in: is the total power consumption of prosumer i; and are the minimum and maximum power of the central air-conditioning system of prosumer i, respectively; c) Price elasticity constraint: The price elasticity of producer-consumer i is subject to the user's willingness to consume and sell electricity, and should not exceed the maximum price elasticity coefficient of each user: 2) Prosumers as consumers Let j be the consumer prosumer ID. When the prosumer acts as a consumer, the bidding strategy submitted includes the price sensitivity coefficient and power purchase power limit Forming a bidding strategy set The virtual power plant operator sells electricity to prosumers at a local market price between the retail market price and the contract price. By adjusting the local market price, the power consumption of prosumers is changed, so that the virtual power plant can meet customized demand for external power generation. The relationship between the power consumption of prosumers and the local price is as follows: in: is the power purchased by prosumer j in the local market; is the price elasticity coefficient of electricity purchased by producer / consumer j in the local market, and the electricity selling price in the local market can be known. The higher it is, the more electricity prosumer j is willing to sell in the local market; represents the willingness of prosumer j to purchase electricity in the local market. The premise for prosumer j to choose to purchase electricity in the local market is that the electricity price in the local market is Not higher than the retail market electricity price From this we can get is the maximum power purchase power of prosumer j, i.e., the upper limit of its power demand The benefit of producer / consumer j j,t It includes the following items: own electricity utility, the cost of purchasing electricity in the local market, and the cost of own power generation. The producer and consumer j performs day-ahead optimization with the goal of maximizing utility. The objective function is shown in the following formula: in: is the power of the central air conditioning system of producer / consumer j; U j,t is the electricity consumption of producer / consumer j; j,t is the unit electricity generation cost of prosumer j; g j,t is the electricity generation of prosumer j; a) Power balance constraint: the sum of the power purchased by producer / consumer j in the local market and its own power generation should be equal to the total power consumption: in: is the total power consumption of prosumer j; b) Flexible load constraints, which are constraints imposed when the prosumer acts as a producer; c) Price elasticity constraint: the price elasticity of producer / consumer j is subject to the user's own willingness to consume and purchase electricity, and should not be lower than the minimum price elasticity coefficient of each user. (3-4) The step-size control method is used to further constrain the decision-making results of the bidding coefficient to prevent it from changing significantly. The specific formula is as follows: Δ=|RMP t (r+1)·a i,t (r)| (24) Where: r is the number of iterations; Δ is the step size during the iteration, a i,t (r) is the decision value of the bidding coefficient obtained in the rth round of iteration, is the optimal value of the bidding coefficient obtained in the r+1th round of iteration, RMP t ∈[0,1] is the climbing coefficient. This step is equivalent to correcting the bidding coefficient obtained by local optimization according to the step size control requirements, so as to obtain the decision value of the bidding coefficient of this round; (4) The virtual power plant operator aims to maximize operating profits and is constrained by meeting the customized output requirements of the virtual power plant. It updates the local market price based on the bidding strategy of the prosumers and publishes it to each prosumer. Specifically, (4-1) Use multi-scenario technology to simulate the uncertainty of renewable energy output, user load demand and penalty for breach of trading contract during actual operation, and transform the stochastic optimization problem into a deterministic optimization problem; (4-2) Use conditional value at risk to measure the risk of the default part of the virtual power plant transaction contract and reasonably weigh the operating benefits and potential risks of the virtual power plant; (4-3) The virtual power plant operator makes a day-ahead optimization decision based on the bidding strategy submitted by the prosumer with the goal of minimizing the customization cost of the virtual power plant. The optimization model is constructed as follows: Where: VPP Customizing costs for virtual power plants; is the electricity purchase cost of the virtual power plant operator in the xth prosumer electricity sales scenario; is the electricity sales revenue of the virtual power plant operator in the yth prosumer electricity purchasing scenario; R con Revenue from signing power contracts for virtual power plant operators; x and ω y are the scenario probabilities corresponding to the xth prosumer electricity sales scenario and the yth prosumer electricity purchase scenario, respectively. There are X prosumer electricity sales scenarios and Y electricity purchase scenarios; γ is the risk aversion coefficient of the virtual power plant operator; V risk is the conditional risk value of the virtual power plant customization problem; Ξ is the decision variable set of the optimization model; ξ is the risk value of the virtual power plant customization problem; δ xyh For linearization V risk And the auxiliary variables introduced; The calculation of each part in the objective function is as follows: in: is the amount of electricity purchased by the virtual power plant operator from prosumer i in the xth prosumer electricity sales scenario; is the amount of electricity sold by the virtual power plant operator to prosumer j in the yth prosumer electricity purchasing scenario; Δt is the duration of the unit transaction period; α is the confidence level; ω h is the probability of the scenario corresponding to the h-th contract breach penalty price, and there are H contract breach penalty price scenarios; a) Power balance constraint: the net external output of the virtual power plant should not be less than the customized output value specified in the contract: b) Constraints on electricity purchase and sales prices: the electricity purchase price of the virtual power plant operator for production-type prosumers should be between the unified purchase price of the power grid and the contract transaction price, and the electricity sales price of the virtual power plant operator for consumption-type prosumers should be between the wholesale market price and the retail market price: c) Risk constraint: If the actual power delivered by the virtual power plant operator is lower than the value specified in the contract or exceeds the value specified in the contract and the deviation is greater than ΔP max When the contract breach penalty is paid for the deviation of electricity, the conditional risk value V risk The auxiliary variable δ in xyh The following constraints should be met: in: The contract breach penalty for the virtual power plant operator under the xth prosumer electricity sales scenario, the yth prosumer electricity purchase scenario, and the hth contract breach penalty price scenario; and They are the actual power purchase deviation ratio and actual power sales deviation ratio of virtual power plant operators respectively; and are the h-th contract breach penalty prices in the case of insufficient and excessive delivery, respectively; ΔP max Indicates the maximum deviation allowed for overdelivery; [v] + means taking the larger value between v and 0, where v refers to and (4-4) The deviation between the actual transaction value and the planned value between the virtual power plant operator and the prosumer cluster follows a normal distribution, as shown in the following formula: Where: μ b and σ b are the expected value and standard deviation of the actual delivery volume obtained by the virtual power plant operator through power purchase transactions with the prosumer cluster, μ s and σ s are the expected value and standard deviation of the actual delivery volume obtained by the virtual power plant operator in the power sales transaction with the prosumer cluster, respectively; (4-5) After the virtual power plant operator updates the local market price in round r, a certain number of prosumers are randomly selected according to the following formula and are prohibited from changing their bidding strategies in the next round of iterations: in: The number of prosumers that are prohibited from updating their bidding strategies; N PRS The total number of prosumers customized for participation in the VPP; (5) If the local market price converges, the prosumer signs a day-ahead transaction contract with the virtual power plant operator based on the final decision of both parties to complete the customized construction of the virtual power plant; otherwise, it returns to (3) and loops as follows: The following formula is used as the convergence criterion of local market prices in: is the price vector of electricity purchase and sale in the local market; ε π is a minimum positive value.

Images