Virtual power plant resource regulation and control method and system considering resource uncertainty

Abstract

The invention discloses a virtual power plant resource regulation and control method and system considering resource uncertainty, and relates to the technical field of power system demand response, and the method comprises the steps: obtaining virtual power plant aggregation resource data, inputting the virtual power plant aggregation resource data into a pre-established virtual power plant aggregation model, and obtaining a virtual power plant aggregation model; outputting to obtain an adjustable resource mathematical model of the virtual power plant aggregation, wherein the adjustable resource mathematical model of the virtual power plant aggregation comprises an energy storage model, an air conditioner model and an electric vehicle model; generating a scene set based on the uncertainty of the virtual power plant aggregation resource data, reducing the scene set to obtain a reduced scene set, and performing conditional value-at-risk calculation to obtain conditional value-at-risk; and based on an adjustable resource mathematical model of virtual power plant aggregation, in combination with the reduced scene set and the conditional value-at-risk, generating an operation optimization model of the virtual power plant participating in the day-ahead power market considering resource uncertainty, and performing virtual power plant resource regulation and control.

Description

Technical Field

[0001] The present invention relates to the technical field of power system demand response, and specifically to a virtual power plant resource control method and system taking resource uncertainty into consideration. Background Art

[0002] In the free and open environment of the power market, in order to avoid the volatility and dispersion of distributed renewable energy directly participating in power grid transactions, aggregating the resources on the demand side into a virtual power plant to participate in the power market has become a new idea. At present, the construction of virtual power plants in my country is still in its infancy. Under the current background, the direct call of virtual power plants on internal distributed resources and coordinated optimization of scheduling with the overall benefit optimization will still be an important scheduling mode in the future. Summary of the invention

[0003] In order to solve the deficiencies mentioned in the above background technology, the purpose of the present invention is to provide a virtual power plant resource control method and system taking resource uncertainty into consideration.

[0004] In a first aspect, the purpose of the present invention can be achieved by the following technical solution: a virtual power plant resource control method considering resource uncertainty, the method comprising the following steps:

[0005] Acquire virtual power plant aggregated resource data, input the virtual power plant aggregated resource data into a pre-established virtual power plant aggregation model, and output an adjustable resource mathematical model aggregated by the virtual power plant, wherein the adjustable resource mathematical model aggregated by the virtual power plant includes: an energy storage model, an air conditioning model, and an electric vehicle model;

[0006] Generate a scenario set based on the uncertainty of the aggregated resource data of the virtual power plant, reduce the scenario set to obtain a reduced scenario set, calculate the conditional risk value based on the uncertainty of the aggregated resource data of the virtual power plant, and obtain the conditional risk value;

[0007] Based on the mathematical model of adjustable resources aggregated by virtual power plants, combined with the reduced scenario set and conditional risk value, an operation optimization model of virtual power plants participating in the day-ahead power market considering resource uncertainty is generated. Based on the benefit results of the operation optimization model of virtual power plants participating in the day-ahead power market considering resource uncertainty, virtual power plant resources are controlled.

[0008] In combination with the first aspect, in some implementations of the first aspect, the method further includes: the energy storage model:

[0009]

[0010] Where: P i cha,max and P idis,max are the upper limits of the charging and discharging power of the i-th energy storage device, is the remaining power of the i-th energy storage device in period t under scenario k, S k,i,t is the state of charge of the i-th energy storage device in period t under scenario k, and are the lower and upper limits of the state of charge of the i-th energy storage device, is the charging and discharging efficiency factor of the i-th energy storage device, and are the starting and ending capacities of the i energy storage devices respectively;

[0011] Air conditioning model:

[0012]

[0013] In the formula, C j,a , R j,a are the equivalent heat capacity and equivalent thermal resistance of the room of air conditioner j respectively; are the indoor and outdoor temperatures of room j under air conditioner j at time t in scenario k; COP j is the energy efficiency ratio of air conditioner j; is the power consumption of air conditioner j in period t under scenario k; is the rated power of air conditioner j; T j,max 、T j,min They are the upper and lower limits of indoor temperature when air conditioner j is operating normally;

[0014] Electric car model:

[0015]

[0016] Where: are the arrival and departure times of electric vehicles, P i ev,max is the upper limit of the charging power of the ith electric vehicle, is the battery capacity of the ith electric vehicle in time period t under scenario k, S k,i,t is the battery state of charge of the ith electric vehicle at time period t under scenario k, is the maximum battery capacity of the ith electric vehicle, and are the lower and upper limits of the state of charge of the ith electric vehicle, is the charging efficiency factor of the ith electric vehicle battery, are the expected power of the owner of the i-th electric car and the power of the electric car at the end.

[0017] In combination with the first aspect, in some implementations of the first aspect, the method further includes: the uncertainty of the virtual power plant aggregated resource data includes:

[0018] The uncertainty of wind and photovoltaic generators is related to their location, altitude, and season;

[0019] The wind speed and light intensity were fitted using the probability density functions of Weibull distribution and Beta distribution;

[0020] The wind speed probability density function based on Weibull distribution is:

[0021]

[0022] Where: v is the wind speed value; k and c are shape parameters and scale parameters respectively, and k and c ≥ 0;

[0023] The probability density function of light intensity based on Beta distribution is:

[0024]

[0025] Where: w is the light intensity, and the subscript max represents its maximum value; χ is the shape parameter of Beta distribution; Γ(·) is the gamma function.

[0026] In combination with the first aspect, in some implementations of the first aspect, the method further includes: a process of generating a scenario set based on uncertainty of the virtual power plant aggregated resource data:

[0027] Assuming that the distribution of uncertainties follows the historical distribution, the uncertainty faced in the decision-making of virtual power plants is described by generating distributed resource PV, WT scenario sets and electricity price scenario sets. The WT and PV outputs and electricity price fluctuations follow a normal distribution with the predicted value as the mean.

[0028] Assume that the number of generated distributed resource scenarios is m, and the number of electricity price scenarios is u. Since the distributed resource scenarios and electricity price scenarios are independent of each other, the total number of scenarios is N. K =m×u.

[0029] In combination with the first aspect, in some implementations of the first aspect, the method further includes: the process of reducing the scene set to obtain a reduced scene set:

[0030] The Latin hypercube sampling method is used to reduce the uncertainty scenario set generated by the Monte Carlo method. The sampling steps are as follows:

[0031] Sampling: Assume there is a random variable X i , whose sampling scale is R, X i The cumulative distribution function of is:

[0032] Y i =F i(X i ),Y i ∈[0,1]

[0033] R divides Y equally i The value interval of each interval is taken as Y i The sampling value of X i The rth sample value of can be obtained by the following inverse transformation:

[0034]

[0035] Get the initial sampling matrix Z i =[X i1 ,X i2 ,…,X iR ];

[0036] Sorting: Since the distribution function is an increasing function, the initial matrix obtained by sampling has a positive correlation, and then the random processing is performed to obtain an independent normal distribution sampling matrix Z' i .

[0037] In combination with the first aspect, in some implementations of the first aspect, the method further includes: performing conditional risk value calculation based on the uncertainty of the virtual power plant aggregated resource data to obtain the conditional risk value:

[0038] The conditional value at risk (CVaR) method is used to measure the risks of distributed resources and market electricity prices faced by virtual power plants. The CVaR method evaluates the expected value of risk losses exceeding a certain probability, or the expected return value under the worst probability (1-α)×100%, under a given confidence level α. The expressions of VaR and CVaR are as follows:

[0039] Pr{V<VaR α}=1-α

[0040]

[0041] Where: V is market return, VaR α is the value at risk at the confidence level α, CVaR α is the conditional risk value under the confidence level α, X is the decision quantity combination; ξ is the uncertainty; f(X,ξ) is the return function; p(ξ) is the probability density function of ξ; E() is the calculation expectation identifier.

[0042] In combination with the first aspect, in certain implementations of the first aspect, the method further includes: a process of linearizing the conditional value at risk:

[0043] The original CVaR formula is linearized by the scenario generation method, and the constraints are added as follows:

[0044]

[0045] η k ≥ζ-R k

[0046] η k ≥0

[0047] Where: ζ and η k is an auxiliary variable introduced, ζ represents the risk value under the confidence level α, R k Represents the revenue of the virtual power plant under scenario k.

[0048] In combination with the first aspect, in some implementations of the first aspect, the method further includes: the objective function of the operation optimization model of the virtual power plant participating in the day-ahead power market considering resource uncertainty includes two parts: expected return and conditional risk value, and the objective function is as follows:

[0049] maxf=(1-β)R+β·CVaR α

[0050] Where: β∈[0,1] is the risk aversion coefficient, R is the expected return of the virtual power plant obtained based on the uncertainty scenario set, CVaR α Represents the risk-benefit of the virtual power plant at the confidence level α.

[0051] In a second aspect, in order to achieve the above-mentioned purpose, the present invention discloses a virtual power plant resource control system considering resource uncertainty, comprising:

[0052] A model processing module is used to obtain virtual power plant aggregated resource data, input the virtual power plant aggregated resource data into a pre-established virtual power plant aggregation model, and output an adjustable resource mathematical model aggregated by the virtual power plant, wherein the adjustable resource mathematical model aggregated by the virtual power plant includes: an energy storage model, an air conditioning model, and an electric vehicle model;

[0053] A resource processing module is used to generate a scenario set based on the uncertainty of the aggregated resource data of the virtual power plant, reduce the scenario set to obtain a reduced scenario set, and perform conditional risk value calculation based on the uncertainty of the aggregated resource data of the virtual power plant to obtain the conditional risk value;

[0054] The resource regulation module is used to generate an operation optimization model for virtual power plants participating in the day-ahead power market taking into account resource uncertainty based on the mathematical model of adjustable resources aggregated by virtual power plants, combined with the reduced scenario set and conditional risk value, so as to regulate the resources of virtual power plants based on the benefit results of the operation optimization model for virtual power plants participating in the day-ahead power market taking into account resource uncertainty.

[0055] In conjunction with the second aspect, in some implementations of the second aspect, the system further includes: an energy storage model of the model processing module:

[0056]

[0057]

[0058] Where: P i cha,max and P i dis,max are the upper limits of the charging and discharging power of the i-th energy storage device, is the remaining power of the i-th energy storage device in period t under scenario k, S k,i,t is the state of charge of the i-th energy storage device in period t under scenario k, and are the lower and upper limits of the state of charge of the i-th energy storage device, is the charging and discharging efficiency factor of the i-th energy storage device, and are the starting and ending capacities of the i energy storage devices respectively;

[0059] Air conditioning model:

[0060]

[0061] In the formula, C j,a , R j,a are the equivalent heat capacity and equivalent thermal resistance of the room of air conditioner j respectively; are the indoor and outdoor temperatures of room j under air conditioner j at time t in scenario k; COP j is the energy efficiency ratio of air conditioner j; is the power consumption of air conditioner j in period t under scenario k; is the rated power of air conditioner j; T j,max 、T j,min They are the upper and lower limits of indoor temperature when air conditioner j is operating normally;

[0062] Electric car model:

[0063]

[0064]

[0065] Where: are the arrival and departure times of electric vehicles, P i ev,max is the upper limit of the charging power of the ith electric vehicle, is the battery capacity of the ith electric vehicle in time period t under scenario k, S k,i,tis the battery state of charge of the ith electric vehicle at time period t under scenario k, is the maximum battery capacity of the ith electric vehicle, and are the lower and upper limits of the state of charge of the ith electric vehicle, is the charging efficiency factor of the ith electric vehicle battery, are the expected power of the owner of the i-th electric car and the power of the electric car at the end;

[0066] The uncertainties of the aggregated resource data of the virtual power plant in the resource processing module include:

[0067] The uncertainty of wind and photovoltaic generators is related to their location, altitude, and season;

[0068] The wind speed and light intensity were fitted using the probability density functions of Weibull distribution and Beta distribution;

[0069] The wind speed probability density function based on Weibull distribution is:

[0070]

[0071] Where: v is the wind speed value; k and c are shape parameters and scale parameters respectively, and k and c ≥ 0;

[0072] The probability density function of light intensity based on Beta distribution is:

[0073]

[0074] Where: w is the light intensity, and the subscript max represents its maximum value; χ is the shape parameter of Beta distribution; Γ(·) is the gamma function;

[0075] The resource processing module generates a scenario set based on the uncertainty of the virtual power plant aggregated resource data:

[0076] Assuming that the distribution of uncertainties follows the historical distribution, the uncertainty faced in the decision-making of virtual power plants is described by generating distributed resource PV, WT scenario sets and electricity price scenario sets. The WT and PV outputs and electricity price fluctuations follow a normal distribution with the predicted value as the mean.

[0077] Assume that the number of generated distributed resource scenarios is m, and the number of electricity price scenarios is u. Since the distributed resource scenarios and electricity price scenarios are independent of each other, the total number of scenarios is N. K =m×u;

[0078] The resource processing module reduces the scene set to obtain the reduced scene set:

[0079] The Latin hypercube sampling method is used to reduce the uncertainty scenario set generated by the Monte Carlo method. The sampling steps are as follows:

[0080] Sampling: Assume there is a random variable X i , whose sampling scale is R, X i The cumulative distribution function of is:

[0081] Y i =F i (X i ),Y i ∈[0,1]

[0082] R divides Y equally i The value interval of each interval is taken as Y i The sampling value of X i The rth sample value of can be obtained by the following inverse transformation:

[0083]

[0084] Get the initial sampling matrix Z i =[X i1 ,X i2 ,…,X iR ];

[0085] Sorting: Since the distribution function is an increasing function, the initial matrix obtained by sampling has a positive correlation, and then the random processing is performed to obtain an independent normal distribution sampling matrix Z' i ;

[0086] The resource processing module calculates the conditional risk value based on the uncertainty of the virtual power plant aggregated resource data to obtain the process of conditional risk value:

[0087] The conditional value at risk (CVaR) method is used to measure the risks of distributed resources and market electricity prices faced by virtual power plants. The CVaR method evaluates the expected value of risk losses exceeding a certain probability, or the expected return value under the worst probability (1-α)×100%, under a given confidence level α. The expressions of VaR and CVaR are as follows:

[0088] Pr{V<VaR α}=1-α

[0089]

[0090] Where: V is market return, VaR α is the value at risk at the confidence level α, CVaR αis the conditional risk value under the confidence level α, X is the decision quantity combination; ξ is the uncertainty; f(X,ξ) is the return function; p(ξ) is the probability density function of ξ; E() is the calculation expectation identifier.

[0091] The process of linearizing the conditional risk value of the resource processing module:

[0092] The original CVaR formula is linearized by the scenario generation method, and the constraints are added as follows:

[0093]

[0094] η k ≥ζ-R k

[0095] η k ≥0

[0096] Where: ζ and η k is an auxiliary variable introduced, ζ represents the risk value under the confidence level α, R k represents the revenue of the virtual power plant under scenario k;

[0097] The objective function of the operation optimization model of the virtual power plant participating in the day-ahead power market considering resource uncertainty in the resource control module includes two parts: expected return and conditional risk value. The objective function is as follows:

[0098] maxf=(1-β)R+β·CVaR α

[0099] Where: β∈[0,1] is the risk aversion coefficient, R is the expected return of the virtual power plant obtained based on the uncertainty scenario set, CVaR α Represents the risk-benefit of the virtual power plant at the confidence level α.

[0100] Beneficial effects of the present invention:

[0101] The present invention can better enable the virtual power plant to establish a mathematical model of distributed resources based on the output characteristics of its internal units, taking into account the adjustable characteristics and uncertainties of different distributed resources. At the same time, a virtual power plant participating in the day-ahead power market optimization operation model is constructed, which is practical and reliable. The virtual power plant regulates and integrates the fragmented resources on the demand side, so that different energy sources are coordinated and complementary. BRIEF DESCRIPTION OF THE DRAWINGS

[0102] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0103] Figure 1 It is a schematic flow chart of the method of the present invention;

[0104] Figure 2 It is a schematic diagram of the system structure of the present invention. DETAILED DESCRIPTION

[0105] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0106] Embodiment 1:

[0107] The following is an introduction to the relevant terms involved in the embodiments of the present application:

[0108] A virtual power plant is a power coordination and management system that uses advanced information and communication technologies and software systems to aggregate and coordinate distributed energy resources (DER) to participate in the electricity market and power grid operations as a special power plant.

[0109] The core concepts of virtual power plants are "communication" and "aggregation". Through intelligent scheduling, real-time monitoring and in-depth analysis, it can formulate the best adjustment plan, thereby achieving multi-energy complementarity on the power supply side, two-way interaction on the load side, and flexible adjustment on the energy storage side. Virtual power plants can aggregate various distributed resources, such as charging piles, air conditioners, new energy storage, data centers, etc., and through digital and intelligent management, these scattered resources are brought together in the cloud to form a new energy management system.

[0110] like Figure 1 As shown, a virtual power plant resource control method considering resource uncertainty includes the following steps:

[0111] S101: acquiring virtual power plant aggregated resource data, inputting the virtual power plant aggregated resource data into a pre-established virtual power plant aggregation model, and outputting an adjustable resource mathematical model aggregated by the virtual power plant, wherein the adjustable resource mathematical model aggregated by the virtual power plant includes: an energy storage model, an air conditioning model, and an electric vehicle model;

[0112] Among them, energy storage modeling: During the operation of the energy storage equipment, in order to extend its service life, the charge state of each time period must be maintained within a certain range. In addition, in order to facilitate scheduling the next day, the charge state of the energy storage equipment at the end of the day should be equal to the state at the beginning of the day.

[0113]

[0114] Where: P i cha,max and P i dis,max are the upper limits of the charging and discharging power of the i-th energy storage device, is the remaining power of the i-th energy storage device in period t under scenario k, S k,i,t is the state of charge of the i-th energy storage device in period t under scenario k, and are the lower and upper limits of the state of charge of the i-th energy storage device, is the charging and discharging efficiency factor of the i-th energy storage device, and are the starting and ending capacities of the i energy storage devices respectively.

[0115] Air Conditioning Modeling:

[0116]

[0117] In the formula, C j,a , R j,a are the equivalent heat capacity and equivalent thermal resistance of the room of air conditioner j respectively; are the indoor and outdoor temperatures of room j under air conditioner j at time t in scenario k; COP j is the energy efficiency ratio of air conditioner j; is the power consumption of air conditioner j in period t under scenario k; is the rated power of air conditioner j; T j,max 、T j,min They are respectively the upper and lower limits of indoor temperature when air conditioner j is operating normally and meet the user's comfort requirements.

[0118] Electric vehicle modeling: Optimizing the charging behavior of electric vehicles based on a model of their arrival and departure patterns. The focus is on using charging piles to control the charging behavior of electric vehicles.

[0119]

[0120] Where: are the arrival and departure times of electric vehicles, P i ev,max is the upper limit of the charging power of the ith electric vehicle, is the battery capacity of the ith electric vehicle in time period t under scenario k, S k,i,t is the battery state of charge of the ith electric vehicle at time period t under scenario k, is the maximum battery capacity of the ith electric vehicle, and are the lower and upper limits of the state of charge of the ith electric vehicle, is the charging efficiency factor of the ith electric vehicle battery, are the expected power of the owner of the i-th electric car and the power of the electric car at the end.

[0121] S102: generating a scenario set based on the uncertainty of the aggregated resource data of the virtual power plant, reducing the scenario set to obtain a reduced scenario set, and performing conditional risk value calculation based on the uncertainty of the aggregated resource data of the virtual power plant to obtain a conditional risk value;

[0122] Among them, the uncertainty of wind power and photovoltaic power generation units is related to factors such as location, altitude, season, etc., and is an uncontrollable variable, but the virtual power plant can predict the output of distributed renewable energy for the next day based on long-term historical data statistics and forecast information. A large number of studies have shown that wind speed and light intensity can be well fitted with the probability density functions of Weibull distribution and Beta distribution, which are widely applicable.

[0123] The uncertainties in the aggregated resource data of the virtual power plant include:

[0124] The uncertainty of wind and photovoltaic generators is related to their location, altitude, and season;

[0125] The wind speed and light intensity were fitted using the probability density functions of Weibull distribution and Beta distribution;

[0126] The wind speed probability density function based on Weibull distribution is:

[0127]

[0128] Where: v is the wind speed value; k and c are shape parameters and scale parameters respectively, and k and c ≥ 0;

[0129] The probability density function of light intensity based on Beta distribution is:

[0130]

[0131] Where: w is the light intensity, and the subscript max represents its maximum value; χ is the shape parameter of Beta distribution; Γ(·) is the gamma function.

[0132] Assuming that the distribution of uncertainties follows the historical distribution, the uncertainty faced in the virtual power plant decision-making is described by generating a set of distributed resource (PV, WT) scenarios and a set of electricity price scenarios. The WT and PV output and electricity price fluctuations follow a normal distribution with the predicted value as the mean.

[0133] Assume that the number of generated distributed resource scenarios is m, and the number of electricity price scenarios is u. Since the distributed resource scenarios and electricity price scenarios are independent of each other, the total number of scenarios is N. K =m×u.

[0134] The Latin hypercube sampling method is used to reduce the uncertainty scenario set generated by the Monte Carlo method, improving efficiency while maintaining calculation accuracy.

[0135] Latin hypercube sampling is essentially a stratified sampling method that can cover the same size of sample space with fewer samples while not ignoring the sample values ​​at the tail. The sampling steps are as follows:

[0136] (1) Sampling: Assume there is a random variable X i , whose sampling scale is R, X i The cumulative distribution function of is:

[0137] Y i =F i (X i ),Y i ∈[0,1]

[0138] R divides Y equally i The value interval of each interval is taken as Y i The sampling value of X i The rth sample value of can be obtained by the following inverse transformation:

[0139]

[0140] Then we can get the initial sampling matrix Z i =[X i1 ,X i2 ,…,X iR ].

[0141] (2) Sorting: Since the distribution function is an increasing function, the initial matrix obtained by the above sampling has a positive correlation. After random processing, an independent normal distribution sampling matrix Z' is obtained. i .

[0142] The conditional value at risk (CVaR) method is used to measure the risks of distributed resources and market electricity prices faced by virtual power plants. The CVaR method evaluates the expected value of risk losses exceeding a certain probability, or the expected return value under the worst probability (1-α)×100%, with a given confidence level α. The specific expressions of VaR and CVaR are as follows:

[0143] Pr{V<VaR α}=1-α

[0144]

[0145] Where: V is market return, VaR α CVaR is the value at risk at the confidence level α. α is the conditional risk value under the confidence level α, X is the decision quantity combination; ξ is the uncertainty; f(X,ξ) is the return function; p(ξ) is the probability density function of ξ; E() is the calculation expectation identifier.

[0146] The original CVaR formula is linearized by using the scenario generation method, and constraints are added:

[0147]

[0148] η k ≥ζ-R k

[0149] η k ≥0

[0150] Where: ζ and η k is an auxiliary variable introduced, ζ represents the risk value under the confidence level α, R k Represents the revenue of the virtual power plant under scenario k.

[0151] S103: Based on the mathematical model of adjustable resources aggregated by virtual power plants, combined with the reduced scenario set and conditional risk value, an operation optimization model of virtual power plants participating in the day-ahead power market taking into account resource uncertainty is generated, thereby performing virtual power plant resource regulation based on the benefit results of the operation optimization model of virtual power plants participating in the day-ahead power market taking into account resource uncertainty.

[0152] The objective function of the operation optimization model of the virtual power plant participating in the day-ahead power market considering resource uncertainty includes two parts: expected return and conditional risk value. The objective function is as follows:

[0153] maxf=(1-β)R+β·CVaR α

[0154] Where: β∈[0,1] is the risk aversion coefficient, R is the expected return of the virtual power plant obtained based on the uncertainty scenario set, CVaR α Represents the risk-benefit of the virtual power plant at the confidence level α.

[0155] Embodiment 2: In the second aspect, as Figure 2 As shown, in order to achieve the above-mentioned purpose, the present invention discloses a virtual power plant resource control system considering resource uncertainty, including:

[0156] The model processing module 11 is used to obtain the aggregated resource data of the virtual power plant, input the aggregated resource data of the virtual power plant into a pre-established virtual power plant aggregation model, and output an adjustable resource mathematical model of the virtual power plant aggregation, wherein the adjustable resource mathematical model of the virtual power plant aggregation includes: an energy storage model, an air conditioning model, and an electric vehicle model;

[0157] The resource processing module 12 is used to generate a scenario set based on the uncertainty of the aggregated resource data of the virtual power plant, reduce the scenario set to obtain a reduced scenario set, and perform conditional risk value calculation based on the uncertainty of the aggregated resource data of the virtual power plant to obtain the conditional risk value;

[0158] The resource control module 13 is used to generate an operation optimization model for the virtual power plant participating in the day-ahead power market taking into account resource uncertainty based on the mathematical model of adjustable resources aggregated by the virtual power plant, combined with the reduced scenario set and conditional risk value, so as to control the resources of the virtual power plant based on the benefit results of the operation optimization model for the virtual power plant participating in the day-ahead power market taking into account resource uncertainty.

[0159] In conjunction with the second aspect, in some implementations of the second aspect, the system further includes: an energy storage model of the model processing module 11:

[0160]

[0161]

[0162] Where: P i cha,max and P i dis,max are the upper limits of the charging and discharging power of the i-th energy storage device, is the remaining power of the i-th energy storage device in period t under scenario k, S k,i,t is the state of charge of the i-th energy storage device in period t under scenario k, and are the lower and upper limits of the state of charge of the i-th energy storage device, is the charging and discharging efficiency factor of the i-th energy storage device, and are the starting and ending capacities of the i energy storage devices respectively;

[0163] Air conditioning model:

[0164]

[0165] In the formula, C j,a , R j,a are the equivalent heat capacity and equivalent thermal resistance of the room of air conditioner j respectively; are the indoor and outdoor temperatures of room j under air conditioner j at time t in scenario k; COP j is the energy efficiency ratio of air conditioner j; is the power consumption of air conditioner j in period t under scenario k; is the rated power of air conditioner j; T j,max 、T j,min They are the upper and lower limits of indoor temperature when air conditioner j is operating normally;

[0166] Electric car model:

[0167]

[0168]

[0169] Where: are the arrival and departure times of electric vehicles, P i ev,max is the upper limit of the charging power of the ith electric vehicle, is the battery capacity of the ith electric vehicle in time period t under scenario k, S k,i,t is the battery state of charge of the ith electric vehicle at time period t under scenario k, is the maximum battery capacity of the ith electric vehicle, and are the lower and upper limits of the state of charge of the ith electric vehicle, is the charging efficiency factor of the ith electric vehicle battery, are the expected power of the owner of the i-th electric car and the power of the electric car at the end;

[0170] The uncertainties of the aggregated resource data of the virtual power plant in the resource processing module 12 include:

[0171] The uncertainty of wind and photovoltaic generators is related to their location, altitude, and season;

[0172] The wind speed and light intensity were fitted using the probability density functions of Weibull distribution and Beta distribution;

[0173] The wind speed probability density function based on Weibull distribution is:

[0174]

[0175] Where: v is the wind speed value; k and c are shape parameters and scale parameters respectively, and k and c ≥ 0;

[0176] The probability density function of light intensity based on Beta distribution is:

[0177]

[0178] Where: w is the light intensity, and the subscript max represents its maximum value; χ is the shape parameter of Beta distribution; Γ(·) is the gamma function;

[0179] The resource processing module 12 generates a scenario set based on the uncertainty of the virtual power plant aggregated resource data:

[0180] Assuming that the distribution of uncertainties follows the historical distribution, the uncertainty faced in the decision-making of virtual power plants is described by generating distributed resource PV, WT scenario sets and electricity price scenario sets. The WT and PV outputs and electricity price fluctuations follow a normal distribution with the predicted value as the mean.

[0181] Assume that the number of generated distributed resource scenarios is m, and the number of electricity price scenarios is u. Since the distributed resource scenarios and electricity price scenarios are independent of each other, the total number of scenarios is N. K =m×u;

[0182] The resource processing module 12 reduces the scene set to obtain a reduced scene set:

[0183] The Latin hypercube sampling method is used to reduce the uncertainty scenario set generated by the Monte Carlo method. The sampling steps are as follows:

[0184] Sampling: Assume there is a random variable X i , whose sampling scale is R, X i The cumulative distribution function of is:

[0185] Y i =F i (X i ),Y i ∈[0,1]

[0186] R divides Y equally i The value interval of each interval is taken as Y i The sampling value of X i The rth sample value of can be obtained by the following inverse transformation:

[0187]

[0188] Get the initial sampling matrix Z i =[X i1 ,X i2 ,…,X iR ];

[0189] Sorting: Since the distribution function is an increasing function, the initial matrix obtained by sampling has a positive correlation, and then the random processing is performed to obtain an independent normal distribution sampling matrix Z' i ;

[0190] The resource processing module 12 calculates the conditional risk value based on the uncertainty of the virtual power plant aggregated resource data to obtain the process of the conditional risk value:

[0191] The conditional value at risk (CVaR) method is used to measure the risks of distributed resources and market electricity prices faced by virtual power plants. The CVaR method evaluates the expected value of risk losses exceeding a certain probability, or the expected return value under the worst probability (1-α)×100%, under a given confidence level α. The expressions of VaR and CVaR are as follows:

[0192] Pr{V<VaR α}=1-α

[0193]

[0194] Where: V is market return, VaR α is the value at risk at the confidence level α, CVaR α is the conditional risk value under the confidence level α, X is the decision quantity combination; ξ is the uncertainty; f(X,ξ) is the return function; p(ξ) is the probability density function of ξ; E() is the calculation expectation identifier.

[0195] Resource Processing Module 12 Conditional Risk Value Linearization Process:

[0196] The original CVaR formula is linearized by the scenario generation method, and the constraints are added as follows:

[0197]

[0198] η k ≥ζ-R k

[0199] η k ≥0

[0200] Where: ζ and η k is an auxiliary variable introduced, ζ represents the risk value under the confidence level α, R k represents the revenue of the virtual power plant under scenario k;

[0201] The objective function of the operation optimization model of the virtual power plant participating in the day-ahead power market considering resource uncertainty in the resource regulation module 13 includes two parts: expected return and conditional risk value. The objective function is as follows:

[0202] maxf=(1-β)R+β·CVaR α

[0203] Where: β∈[0,1] is the risk aversion coefficient, R is the expected return of the virtual power plant obtained based on the uncertainty scenario set, CVaRα Represents the risk-benefit of the virtual power plant at the confidence level α.

[0204] Based on the same inventive concept, the present invention also provides a computer device, which includes: one or more processors, and a memory for storing one or more computer programs; the program includes program instructions, and the processor is used to execute the program instructions stored in the memory. The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, which is used to implement one or more instructions, specifically for loading and executing one or more instructions in a computer storage medium to implement the above method.

[0205] It needs to be further explained that, based on the same inventive concept, the present invention also provides a computer storage medium, on which a computer program is stored, and the computer program is executed by a processor to execute the above method. The storage medium can adopt any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electrical, magnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples (non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, a computer-readable storage medium can be any tangible medium containing or storing a program, which can be used by an instruction execution system, device or device or used in combination with it.

[0206] In the description of this specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present disclosure. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0207] The above shows and describes the basic principles, main features and advantages of the present disclosure. Those skilled in the art should understand that the present disclosure is not limited by the above embodiments, and the above embodiments and descriptions are only for explaining the principles of the present disclosure. Without departing from the spirit and scope of the present disclosure, the present disclosure may have various changes and improvements, and these changes and improvements fall within the scope of the present disclosure to be protected.

Claims

1. A virtual power plant resource control method considering resource uncertainty, characterized in that: The method comprises the following steps: Acquire virtual power plant aggregated resource data, input the virtual power plant aggregated resource data into a pre-established virtual power plant aggregation model, and output an adjustable resource mathematical model aggregated by the virtual power plant, wherein the adjustable resource mathematical model aggregated by the virtual power plant includes: an energy storage model, an air conditioning model, and an electric vehicle model; Generate a scenario set based on the uncertainty of the aggregated resource data of the virtual power plant, reduce the scenario set to obtain a reduced scenario set, calculate the conditional risk value based on the uncertainty of the aggregated resource data of the virtual power plant, and obtain the conditional risk value; Based on the mathematical model of adjustable resources aggregated by virtual power plants, combined with the reduced scenario set and conditional risk value, an operation optimization model of virtual power plants participating in the day-ahead power market considering resource uncertainty is generated. Based on the benefit results of the operation optimization model of virtual power plants participating in the day-ahead power market considering resource uncertainty, virtual power plant resources are controlled.

2. A virtual power plant resource control method considering resource uncertainty according to claim 1, characterized in that: The energy storage model: Where: and are the upper limits of the charging and discharging power of the i-th energy storage device, is the remaining power of the i-th energy storage device in period t under scenario k, S k,i,t is the state of charge of the i-th energy storage device in period t under scenario k, and are the lower and upper limits of the state of charge of the i-th energy storage device, is the charging and discharging efficiency factor of the i-th energy storage device, and are the starting and ending capacities of the i energy storage devices respectively; Air conditioning model: In the formula, C j,a , R j,a are the equivalent heat capacity and equivalent thermal resistance of the room of air conditioner j respectively; are the indoor and outdoor temperatures of room j under air conditioner j at time t in scenario k; COP j is the energy efficiency ratio of air conditioner j; is the power consumption of air conditioner j in period t under scenario k; is the rated power of air conditioner j; T j,max , T j,min They are the upper and lower limits of indoor temperature when air conditioner j is operating normally; Electric car model: Where: are the arrival and departure times of electric vehicles, is the upper limit of the charging power of the ith electric vehicle, is the battery capacity of the ith electric vehicle in time period t under scenario k, S k,i,t is the battery state of charge of the ith electric vehicle at time period t under scenario k, is the maximum battery capacity of the ith electric vehicle, and are the lower and upper limits of the state of charge of the ith electric vehicle, is the charging efficiency factor of the ith electric vehicle battery, are the expected power of the owner of the i-th electric car and the power of the electric car at the end.

3. A virtual power plant resource control method considering resource uncertainty according to claim 1, characterized in that: The uncertainties in the aggregated resource data of the virtual power plant include: The uncertainty of wind and photovoltaic generators is related to their location, altitude, and season; The wind speed and light intensity were fitted using the probability density functions of Weibull distribution and Beta distribution; The wind speed probability density function based on Weibull distribution is: Where: v is the wind speed value; k and c are shape parameters and scale parameters respectively, and k and c ≥ 0; The probability density function of light intensity based on Beta distribution is: Where: w is the light intensity, and the subscript max represents its maximum value; are the shape parameters of Beta distribution; Γ(·) is the gamma function.

4. A virtual power plant resource control method considering resource uncertainty according to claim 1, characterized in that: The process of generating a scenario set based on uncertainty of virtual power plant aggregated resource data: Assuming that the distribution of uncertainties follows the historical distribution, the uncertainty faced in the decision-making of virtual power plants is described by generating distributed resource PV, WT scenario sets and electricity price scenario sets. The WT and PV outputs and electricity price fluctuations follow a normal distribution with the predicted value as the mean. Assume that the number of generated distributed resource scenarios is m, and the number of electricity price scenarios is u. Since the distributed resource scenarios and electricity price scenarios are independent of each other, the total number of scenarios is N. K =m×u.

5. The method for controlling resources of a virtual power plant considering resource uncertainty according to claim 1, characterized in that: The process of reducing the scene set to obtain the reduced scene set: The Latin hypercube sampling method is used to reduce the uncertainty scenario set generated by the Monte Carlo method. The sampling steps are as follows: Sampling: Assume there is a random variable X i , whose sampling scale is R, X i The cumulative distribution function of is: Y i =F i (X i ),Y i ∈[0,1] R divides Y equally i The value interval of each interval is taken as Y i The sampling value of X i The rth sample value of can be obtained by the following inverse transformation: Get the initial sampling matrix Z i =[X i1 ,X i2 ,…,X iR ]; Sorting: Since the distribution function is an increasing function, the initial matrix obtained by sampling has a positive correlation, and then the random processing is performed to obtain an independent normal distribution sampling matrix Z' i .

6. A virtual power plant resource control method considering resource uncertainty according to claim 1, characterized in that: The process of calculating the conditional risk value based on the uncertainty of the virtual power plant aggregated resource data to obtain the conditional risk value is as follows: The conditional value at risk (CVaR) method is used to measure the risks of distributed resources and market electricity prices faced by virtual power plants. The CVaR method evaluates the expected value of risk losses exceeding a certain probability, or the expected return value under the worst probability (1-α)×100%, under a given confidence level α. The expressions of VaR and CVaR are as follows: Pr{V<VaR α }=1-α Where: V is market return, VaR α is the value at risk at the confidence level α, CVaR α is the conditional risk value under the confidence level α, X is the decision quantity combination; ξ is the uncertainty; f(X,ξ) is the return function; p(ξ) is the probability density function of ξ; E() is the calculation expectation identifier.

7. A virtual power plant resource control method considering resource uncertainty according to claim 6, characterized in that: The process of linearizing the conditional VaR is as follows: The original CVaR formula is linearized by the scenario generation method, and the constraints are added as follows: or k ≥ζ-R k or k ≥0 Where: ζ and η k is an auxiliary variable introduced, ζ represents the risk value under the confidence level α, R k Represents the revenue of the virtual power plant under scenario k.

8. The method for controlling resources of a virtual power plant considering resource uncertainty according to claim 1, characterized in that: The objective function of the operation optimization model of the virtual power plant participating in the day-ahead power market considering resource uncertainty includes two parts: expected return and conditional risk value. The objective function is as follows: maxf=(1-β)R+β·CVaR α Where: β∈[0,1] is the risk aversion coefficient, R is the expected return of the virtual power plant obtained based on the uncertainty scenario set, CVaR α Represents the risk-benefit of the virtual power plant at the confidence level α.

9. A virtual power plant resource control system considering resource uncertainty, characterized in that: include: A model processing module is used to obtain virtual power plant aggregated resource data, input the virtual power plant aggregated resource data into a pre-established virtual power plant aggregation model, and output an adjustable resource mathematical model aggregated by the virtual power plant, wherein the adjustable resource mathematical model aggregated by the virtual power plant includes: an energy storage model, an air conditioning model, and an electric vehicle model; A resource processing module is used to generate a scenario set based on the uncertainty of the aggregated resource data of the virtual power plant, reduce the scenario set to obtain a reduced scenario set, and perform conditional risk value calculation based on the uncertainty of the aggregated resource data of the virtual power plant to obtain the conditional risk value; The resource regulation module is used to generate an operation optimization model for virtual power plants participating in the day-ahead power market taking into account resource uncertainty based on the mathematical model of adjustable resources aggregated by virtual power plants, combined with the reduced scenario set and conditional risk value, so as to regulate the resources of virtual power plants based on the benefit results of the operation optimization model for virtual power plants participating in the day-ahead power market taking into account resource uncertainty.

10. A virtual power plant resource control system considering resource uncertainty according to claim 9, characterized in that: Energy storage model of the model processing module: Where: and are the upper limits of the charging and discharging power of the i-th energy storage device, is the remaining power of the i-th energy storage device in period t under scenario k, S k,i,t is the state of charge of the i-th energy storage device in period t under scenario k, and are the lower and upper limits of the state of charge of the i-th energy storage device, is the charging and discharging efficiency factor of the i-th energy storage device, and are the starting and ending capacities of the i energy storage devices respectively; Air conditioning model: In the formula, C j,a , R j,a are the equivalent heat capacity and equivalent thermal resistance of the room of air conditioner j respectively; are the indoor and outdoor temperatures of room j under air conditioner j at time t in scenario k; COP j is the energy efficiency ratio of air conditioner j; is the power consumption of air conditioner j in period t under scenario k; is the rated power of air conditioner j; T j,max , T j,min They are the upper and lower limits of indoor temperature when air conditioner j is operating normally; Electric car model: Where: are the arrival and departure times of electric vehicles, is the upper limit of the charging power of the ith electric vehicle, is the battery capacity of the ith electric vehicle in time period t under scenario k, S k,i,t is the battery state of charge of the ith electric vehicle at time period t under scenario k, is the maximum battery capacity of the ith electric vehicle, and are the lower and upper limits of the state of charge of the ith electric vehicle, is the charging efficiency factor of the ith electric vehicle battery, are the expected power of the owner of the i-th electric car and the power of the electric car at the end; The uncertainties of the aggregated resource data of the virtual power plant in the resource processing module include: The uncertainty of wind and photovoltaic generators is related to their location, altitude, and season; The wind speed and light intensity were fitted using the probability density functions of Weibull distribution and Beta distribution; The wind speed probability density function based on Weibull distribution is: Where: v is the wind speed value; k and c are shape parameters and scale parameters respectively, and k and c ≥ 0; The probability density function of light intensity based on Beta distribution is: Where: w is the light intensity, and the subscript max represents its maximum value; are the shape parameters of Beta distribution respectively; Γ(·) is the gamma function; The resource processing module generates a scenario set based on the uncertainty of the virtual power plant aggregated resource data: Assuming that the distribution of uncertainties follows the historical distribution, the uncertainty faced in the decision-making of virtual power plants is described by generating distributed resource PV, WT scenario sets and electricity price scenario sets. The WT and PV outputs and electricity price fluctuations follow a normal distribution with the predicted value as the mean. Assume that the number of generated distributed resource scenarios is m, and the number of electricity price scenarios is u. Since the distributed resource scenarios and electricity price scenarios are independent of each other, the total number of scenarios is N. K =m×u; The resource processing module reduces the scene set to obtain the reduced scene set: The Latin hypercube sampling method is used to reduce the uncertainty scenario set generated by the Monte Carlo method. The sampling steps are as follows: Sampling: Assume there is a random variable X i , whose sampling scale is R, X i The cumulative distribution function of is: Y i =F i (X i ),Y i ∈[0,1] R divides Y equally i The value interval of each interval is taken as Y i The sampling value of X i The rth sample value of can be obtained by the following inverse transformation: Get the initial sampling matrix Z i =[X i1 ,X i2 ,…,X iR ]; Sorting: Since the distribution function is an increasing function, the initial matrix obtained by sampling has a positive correlation, and then the random processing is performed to obtain an independent normal distribution sampling matrix Z' i ; The resource processing module calculates the conditional risk value based on the uncertainty of the virtual power plant aggregated resource data to obtain the process of conditional risk value: The conditional value at risk (CVaR) method is used to measure the risks of distributed resources and market electricity prices faced by virtual power plants. The CVaR method evaluates the expected value of risk losses exceeding a certain probability, or the expected return value under the worst probability (1-α)×100%, under a given confidence level α. The expressions of VaR and CVaR are as follows: Pr{V<VaR α }=1-α Where: V is market return, VaR α is the value at risk at the confidence level α, CVaR α is the conditional risk value under the confidence level α, X is the decision quantity combination; ξ is the uncertainty; f(X,ξ) is the return function; p(ξ) is the probability density function of ξ; E() is the calculation expectation identifier; The process of linearizing the conditional risk value of the resource processing module: The original CVaR formula is linearized by the scenario generation method, and the constraints are added as follows: or k ≥ζ-R k or k ≥0 Where: ζ and η k is an auxiliary variable introduced, ζ represents the risk value under the confidence level α, R k represents the revenue of the virtual power plant under scenario k; The objective function of the operation optimization model of the virtual power plant participating in the day-ahead power market considering resource uncertainty in the resource control module includes two parts: expected return and conditional risk value. The objective function is as follows: max f=(1-β)R+β·CVaR α Where: β∈[0,1] is the risk aversion coefficient, R is the expected return of the virtual power plant obtained based on the uncertainty scenario set, CVaR α Represents the risk-benefit of the virtual power plant at the confidence level α.

Images