A method and system for optimal distributed resource control considering the uncertainties of multiple factors in microgrids

Abstract

This invention relates to a method and system for optimal distributed resource regulation considering multi-factor uncertainties, belonging to the field of distributed resource regulation technology. The method and system include an uncertainty assessment model considering variable correlations. First, considering variable correlations, the conditional probability density distribution function of the error between the predicted and actual values ​​of variables is measured. Then, discrete convolution with complex correlations is used to measure the uncertainty range of the difference or sum of variables. Finally, a robust optimal distributed resource regulation strategy considering multi-factor uncertainties is proposed. Different game strategies are constructed for different distributed heterogeneous components, and individual optima are achieved through strategy exchange and updates between game supply and demand entities. A potential function ensures that the local and overall objective functions are consistent with optimality.

Description

Technical Field

[0001] This invention belongs to the field of distributed resource regulation technology, and relates to a method and system for optimal regulation of distributed resources that considers the uncertainties of multiple factors in microgrids. Background Art

[0002] In recent years, due to the extensive use of fossil fuels, environmental pollution has gradually worsened, leading to increased attention on renewable energy due to its environmentally friendly characteristics and higher economic benefits. Renewable energy, including solar power, is driving the power system towards a model based on renewable energy and supplemented by traditional fossil fuels, gradually reducing dependence on them. However, with the large-scale integration of renewable energy, its output characteristics become uncertain due to unstable weather factors (such as sunlight and wind speed), posing a major obstacle to its widespread adoption. Researchers are addressing this issue by integrating distributed heterogeneous components such as renewable energy, energy storage systems, distributed dispatchable generators, and demand response plans into the distribution network. This has led to the concept of microgrids, which provide power to local loads on a smaller-scale grid basis, achieving high penetration of renewable energy integration. Combining local renewable energy or distributed generation can reduce transmission losses and maximize output. Microgrids can operate either grid-connected as part of the traditional distribution network or off-grid. However, microgrid components have different distributed heterogeneous characteristics, and there are coupling or independent relationships between distributed heterogeneous components. In addition, the uncertainty of renewable energy output and the demand uncertainty of diverse loads need to be considered in the process of optimal regulation of distributed heterogeneous components. This has brought obstacles to the optimal regulation of distributed microgrids. How to effectively manage distributed energy, energy storage systems and loads through optimized regulation and intelligent control technologies to improve the performance, efficiency and flexibility of microgrids is the focus of microgrid research.

[0003] Game theory, with its distributed control and optimization capabilities, is widely used in microgrid optimization scheduling. It establishes strategy models for each participant and analyzes the interactions between different strategies. However, considering the multi-coupling characteristics of distributed resources, its convergence is difficult to guarantee; not every game has a Nash equilibrium, and there are conflicts between the optimal individual interests of some components and the optimal path for the overall benefit. Potential game theory, proposed by Monderer et al., provides a new approach to the optimization of distributed systems. It achieves convergence to a Nash equilibrium with finite improvement through the self-optimization of each member of the distributed game's supply and demand sides. The potential function represents the overall system function by a collection of the utility functions of the game's supply and demand sides. This approach demonstrates unique advantages in the multi-objective optimization scheduling of distributed resources. However, existing game strategies for the supply and demand sides are formulated under the assumption of stable distributed resource output and supply, neglecting the uncertainty of distributed resource supply and demand.

[0004] Copula theory, as an effective method for modeling nonlinear dependencies among various uncertainties, has been introduced into stochastic dependency modeling in uncertainty analysis. Copula theory transforms random variables into a common uniform domain through an invertible cumulative distribution function, and then uses the Copula function within this domain to model multivariate stochastic dependencies. The Gaussian Copula function has been applied in modeling the impact of wind power grid connection on system load in the Netherlands, laying the foundation for stochastic correlation modeling in power systems. Summary of the Invention

[0005] In view of this, the technical problem to be solved by this invention is: In the optimal control of distributed resources, this invention designs a method and system for optimal control of distributed resources that considers the uncertainties of multiple factors in a microgrid. To address the impact of the complex uncertainties inherent in distributed resources, firstly, considering correlation, the conditional probability distribution of the error between the predicted and actual values ​​is measured. Then, a discrete convolution method that measures correlation is used to measure the range of uncertainty of the variables. Furthermore, considering that in a distributed environment, the objective functions formulated by different devices may contradict each other and the overall system, leading to inconsistencies between individual optimization paths and the overall system optimization, and that the different characteristics of distributed resources may prevent the system from reaching a convergent optimal state, a distributed optimal resource control method considering correlated uncertainties is proposed. In a distributed environment, different game strategy sets are constructed for different distributed heterogeneous resources, and individual optima are achieved through strategy exchange and updating optimization between game supply and demand entities. The potential function ensures consistency between individual and overall microgrid optima.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] A method for optimal control of distributed resources considering the uncertainties of multiple factors in microgrids, the method comprising the following steps:

[0008] S1: Use the Copula function to measure the relationship between the predicted and actual values ​​of uncertain variables, and calculate the range of prediction error for uncertain variables;

[0009] S2: Using the Copula function and the discrete convolution method, the problem of requiring variables to be independent in discrete convolution is solved by adding the Copula function, which measures the correlation between different variables, into the discrete convolution calculation. The uncertainty range of variables with different coupling relationships is measured through the above steps.

[0010] S3: Using the potential game method with distributed characteristics and mapping relationships, the rationality of individual interests is constructed through utility functions and payoff functions, and the overall importance is expressed through the potential function;

[0011] S4: Based on the constructed game strategy set and the set priority sequence, the algorithm completes the strategy iteration and optimization by utilizing the local information of the game supply and demand members and the strategies of their neighbors.

[0012] Furthermore, in step S1, the step of using the Copula function to measure the relationship between the predicted and actual values ​​of the uncertain variable, and calculating the range of the prediction error of the uncertain variable, specifically includes:

[0013] S11: The output of a distributed system is affected by many factors. Let X be the measured value of the uncertain variable and Y be the predicted value. Solve for the invertible cumulative distribution function F of the measured value and the predicted value. X (x) = P(X < x) and F Y (y) = P(Y < y), where if F X (x) and F Y If (y) is considered a random variable, then they are all considered to be uniformly distributed:

[0014] S12: Use the Copula function to construct a multivariate cumulative distribution function F from the uniform marginal distribution of the measured and predicted values ​​of the uncertainty variable. X,Y (x,y)=C(F X (x),F Y (y)), and rewrite the multivariate cumulative distribution function of the uncertainty variable in Copula form.

[0015] S13: Use the Gaussian Copula function to obtain the joint probability density function f of the uncertain variables from the given point prediction data. X,Y(x,y), the joint probability density function can be used to evaluate the performance of the prediction model and reveal the dependency between the actual and predicted values ​​of the variables;

[0016] S14: Given a point prediction value, obtain the invertible cumulative distribution function of the actual output, and transform it to obtain the conditional probability density function of the prediction error value of the uncertain variable. Calculate the conditional probability density function of the prediction error of the uncertain variable using the actual output distribution and the Copula density function, where the Copula function includes the correlation between the predicted and actual values. This equation provides a method for quantifying the conditional prediction error of variables using point prediction values.

[0017] Furthermore, step S2 specifically includes:

[0018] S21: Using the conditional probability function of the prediction variable error as input, calculate the joint probability distribution function f of the variables and their sum. X,Y (x,y) is then rewritten as a discrete convolution of variables;

[0019] S22: Considering the complex coupling relationship between variables, the Copula function is introduced to measure the complex correlation between the conditional probability density functions of variables, and it is introduced into the convolution calculation of the difference or sum of variables to solve the problem that convolution requires the variables to be independent; through Copula theory, it is rewritten into a related convolution form with the Copula function;

[0020] S23: Discretizes the correlated convolution form with variable correlation into a discrete convolution form, which has a similar structure to the convolution formula, but adds a Copula function to build the dependency relationship between variables compared to the previous one;

[0021] S24: Based on the sum or difference relationship between variables, apply the expansion of the relevant discrete convolution described in S23 to calculate the variable prediction error distribution to measure the uncertainty of the complex coupling relationship between different uncertain variables.

[0022] Furthermore, step S3 specifically includes the following steps:

[0023] S31: Identify the device members involved in resource regulation and collect the input and output variables of different members within the system;

[0024] S32: Define the basic types of supply and demand members in the system. The individuals participating in the game are not limited to the basic individuals listed below, but can be any combination of these basic components.

[0025] S33: Each member of the supply and demand side participating in the distributed resource regulation game has a corresponding payoff function, that is, the payoff generated by its strategy, as follows:

[0026]

[0027] S34: Based on different members, the various constraints of the members participating in distributed resource regulation are transformed into functions in the game strategy set. The strategies of the supply and demand entities participating in the game mainly refer to the output and input demand range of each basic component, where P i The strategy of the participants is represented by their strategy set as follows:

[0028] S35: Determine Γ=<N,{Y i} i∈N ,{U i} i∈N > is a strategy game with a finite number of participants, where N = {1, 2, ..., n} is the number of participants, and Y i For the strategy set of participating member i, U i The benefit / utility function representing member i;

[0029] S36: The basis of the game theory model is to satisfy the utility and payoff of each player under the premise of satisfying the supply and demand balance of distributed resource regulation, and to set balance constraints.

[0030] S37: Each member of the supply and demand side in the game should be responsible for the supply and demand balance constraints of distributed resource regulation. However, the strategy set objective of each member of the supply and demand side is only to maximize their own utility function. A penalty function is introduced to handle the global constraints of the supply and demand side members:

[0031] Furthermore, step S4 specifically includes:

[0032] S41: Determine the member constraint function, revenue function, and utility function for optimal regulation of distributed resources;

[0033] S42: Initialize the optimization algorithm parameters, construct the individual game model of the supply and demand members of the game according to the different constraints described in S34, and set the priority queue in the game according to the importance.

[0034] S43: Send a request to the previous player in the priority queue; evaluate the player's own strategy and select the best strategy; the player updates their strategy, calculates the rate of change of strategy, and sets the current convergence status;

[0035] S44: Through strategy iteration and penalty system constraints, all players in the game determine the current game state by communicating with their neighbors.

[0036] S45: Determine whether the supply and demand constraint balance is satisfied. If not, optimize by iterating and increasing the penalty coefficient. If satisfied, all players in the supply and demand sectors retain their current strategies.

[0037] S46: Output the game strategy and optimization results ({P) i,t}i=1,2,...,N; t=1,2,...,T).

[0038] The present invention also provides a distributed resource optimal control system that takes into account the uncertainties of multiple factors in microgrids.

[0039] The beneficial effects of the present invention are:

[0040] This invention designs an optimal distributed resource control method considering the multi-factor uncertainties of microgrids. Addressing the issue of insufficient consideration of the complex coupling uncertainties among different distributed resources in a distributed environment, this method fully considers the variables themselves and the complex coupling relationships between different resources, measures the range of uncertainty of distributed resource variables, and constructs a conditional probability density function for prediction errors. Furthermore, considering the multi-coupling characteristics of distributed resources, its convergence is difficult to guarantee, not every game has a Nash equilibrium, and there is a conflict between the optimal individual interests of some components and the optimal path for global interests. This invention employs a distributed potential game approach, treating all heterogeneous distributed resource components as independent game supply and demand members. The rationality of individual interests is constructed through utility and payoff functions, and the overall importance is expressed through a potential function. Finally, based on the constructed game strategy set and the set priority sequence, the method utilizes the local information of the game supply and demand members and the strategies of their neighbors to complete the strategy iteration and optimization.

[0041] Other advantages, objectives, and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination, or may be learned from practice of the invention. The objectives and other advantages of the invention can be realized and obtained through the following description. Attached Figure Description

[0042] To make the objectives, technical solutions, and advantages of the present invention clearer, the preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings, wherein:

[0043] Figure 1 This is a diagram illustrating a joint uncertainty assessment method considering variable correlations as described in this invention.

[0044] Figure 2 A diagram illustrating a distributed optimal resource regulation method considering multi-factor uncertainty designed for this invention;

[0045] Figure 3A flowchart of a distributed optimal resource regulation method considering multiple factors and uncertainties is provided for an embodiment of the present invention. Detailed Implementation

[0046] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0047] The accompanying drawings are for illustrative purposes only and are schematic diagrams, not actual pictures. They should not be construed as limiting the invention. To better illustrate the embodiments of the invention, some parts in the drawings may be omitted, enlarged, or reduced, and do not represent the actual product dimensions. It is understandable to those skilled in the art that some well-known structures and their descriptions may be omitted in the drawings.

[0048] In the accompanying drawings of the embodiments of the present invention, the same or similar reference numerals correspond to the same or similar components. In the description of the present invention, it should be understood that if terms such as "upper," "lower," "left," "right," "front," and "rear" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, they are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, the terms used to describe positional relationships in the drawings are only for illustrative purposes and should not be construed as limiting the present invention. For those skilled in the art, the specific meaning of the above terms can be understood according to the specific circumstances.

[0049] like Figure 1 The diagram shows a method for measuring uncertainty based on multiple factors.

[0050] in Figure 2 This illustrates a distributed resource optimal control method that considers the uncertainties of multiple factors.

[0051] Figure 3 The diagram shows a process for optimal control of distributed resources considering multiple uncertainties, including the following steps:

[0052] Step S1: Use the Copula function to measure the relationship between the predicted and actual values ​​of the uncertain variable, and calculate the range of the prediction error of the uncertain variable.

[0053] Step S2: Using the Copula function and the discrete convolution method, the problem of requiring variables to be independent in discrete convolution is solved by adding the Copula function, which measures the correlation between different variables, into the discrete convolution calculation. The uncertainty range of variables with different coupling relationships is measured through the above steps.

[0054] Step S3: Using the range of uncertainty of variables measured in Steps S1 and S2, as well as the utility function and payoff function of distributed resources, construct different game strategies. Using the potential game method with distributed characteristics and mapping relationships, construct the rationality of individual interests through the utility function and payoff function, and express the overall importance through the potential function.

[0055] Step S4: Based on the constructed game strategy set and the set priority sequence, use the local information of the game supply and demand members and the policies of the neighbors to complete the strategy iteration and optimization algorithm.

[0056] Step S11: Let X and Y be random variables representing the measured values ​​and predicted point values, respectively. Then the CDF of the distributions of the measured values ​​and predicted point values ​​can be expressed as: F X (x) = P(X < x) and F Y (y) = P(Y < y), where if F X (x) and F Y If (y) is considered a random variable, then they are all considered to be uniformly distributed:

[0057]

[0058] Where u X It follows a uniform distribution in the interval [0,1].

[0059] Step S12: The Copula function C can construct a multivariate cumulative distribution function F from the uniform marginal distribution of the measured values ​​and the predicted values ​​of the variable: X,Y (x,y)=C(F X (x),F Y (y)). Where F X,Y (x, y) is the multivariate cumulative distribution function of X and Y. It can be rewritten as...

[0060] Step S13: Use the Gaussian Copula function to predict the joint probability density function from the given point data: f X,Y The joint probability density function (x, y) can be used to evaluate the performance of the prediction model, reveal the dependency between actual and predicted values, and provide important information for optimizing energy management strategies. Specifically:

[0061]

[0062] Where c is the Copula density function of C, f X (x) and f Y (y) represents the marginal probability density functions of X and Y, respectively.

[0063]

[0064] Where N is N∈[-1,1] 2×2 The correlation matrix is ​​I, where I is the identity matrix.

[0065] Step S14: Given the predicted value y = r, the conditional probability density function f of the actual output can be obtained. X|Y That is, the conditional probability density function of the actual output value given the predicted value:

[0066]

[0067] The conditional probability density function for the prediction error value can be obtained by transforming the above formula:

[0068]

[0069] Where E = XY, e = xy, e ∈ [-r, max(x) - r]. The above equation shows that the distribution of prediction error can be calculated using the actual power output distribution and the Copula density function, where c(·) includes the correlation between variables. This equation provides a method for quantifying the conditional prediction error of variable b using point prediction values.

[0070] Step S2: Utilizing the Copula function and discrete convolution method, the problem of variable independence in discrete convolution is addressed by incorporating the Copula function, which measures the correlation between different variables, into the discrete convolution calculation. This step measures the uncertainty range of variables with different coupling relationships. Specifically, it includes the following steps:

[0071] Step S21: Assume X and Y are different sub-resources with marginal distribution functions f. X (x) and f Y The joint probability distribution function f of (y) X,Y The conditional probability density function of (x, y) and X+Y can be expressed as f. X+Y (x,y) can be written as:

[0072]

[0073] Step S22: Its discrete form of convolution is:

[0074]

[0075] Step S23: Considering the complex coupling relationships between variables, this invention introduces the Copula function to measure the complex correlation between the conditional probability density function of sub-resource prediction error and the conditional probability density function of sub-resource prediction error, and introduces it into the convolution calculation of the sum or difference of sub-resources to solve the problem of the prerequisite that convolution requires independent variables. The discrete correlation convolution is as follows:

[0076]

[0077] Step S24: Based on the modeling in steps S21 and S22, the probability distribution of the prediction error of the sum or difference of probabilistic sub-resources is calculated by applying the extension of the relevant discrete convolution. This distribution is represented as follows:

[0078]

[0079] Step S3: Using the range of uncertainty of variables measured in Steps S1 and S2, as well as the utility function and payoff function of distributed resources, construct different game strategies. Using the potential game method with distributed characteristics and mapping relationships, construct the rationality of individual interests through the utility function and payoff function, and express the overall importance through the potential function.

[0080] S31: Identify the device members involved in resource regulation and collect the input and output variables of different members in the system.

[0081] S32: Define the basic types of supply and demand participants in the system. The participants are not limited to the following basic individuals, but can be any combination of these basic components: N = {X1, X2, ... X} n1 ,Y1,Y2,...,Y n2 ,Z1,Z2,...,Z n3 ···}.

[0082] S33: Each member of the supply and demand side participating in the distributed resource regulation game has a corresponding payoff function, that is, the payoff generated by its strategy, as follows:

[0083]

[0084] S34: Based on different members, the various constraints of the members participating in distributed resource regulation are transformed into functions in the game strategy set. The strategies of the supply and demand entities participating in the game mainly refer to the output and input demand range of each basic component, where P i The strategy of the participants is represented by their strategy set as follows:

[0085] S35: Determine Γ=<N,{Y i} i∈N ,{Ui} i∈N > is a strategy game with a finite number of participants, where N = {1, 2, ..., n} is the number of participants, and Y i For the strategy set of participating member i, U i The benefit / utility function representing member i.

[0086] S36: The basis of the game theory model is to satisfy the utility and payoff of each player under the premise of satisfying the supply and demand balance of distributed resource regulation, and to set balance constraints:

[0087] S37: Each member of the supply and demand side in the game should be responsible for the supply and demand balance constraints of distributed resource regulation. However, the strategy set objective of each member of the supply and demand side is only to maximize their own utility function. A penalty function is introduced to handle the global constraints of the supply and demand side members:

[0088] Step S4: Based on the constructed game strategy set and the set priority sequence, utilize the local information of the game's supply and demand members and the policies of their neighbors to complete the strategy iteration and optimization algorithm. Specifically, this includes the following steps:

[0089] S41: Determine the member constraint function, revenue function, and utility function for optimal regulation of distributed resources.

[0090] S42: Initialize the optimization algorithm parameters, construct the individual game model of the supply and demand members according to the different constraints described in S34, and set the priority queue in the game according to importance.

[0091] S43: Send a request to the preceding player in the priority queue. Evaluate the player's own strategy and select the best strategy. Game players update their strategies, calculate the rate of change of strategies, and set the current convergence status:

[0092] S44: Through strategy iteration and penalty system constraints, all players in the game determine the current game state by communicating with their neighbors.

[0093] S45: Determine whether the supply and demand constraints are met: ( For power generation, (For electricity consumption), if the requirements are not met, optimization can be achieved by iteratively increasing the penalty coefficient. If the conditions are met, all members of the supply and demand sides in the game retain their current strategies.

[0094] S46: Output the game strategy and optimization results ({P) i,t}i=1,2,...,N; t=1,2,...,T).

[0095] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications can be made to the technical solutions of the present invention without departing from the spirit and scope of the present invention, and all such modifications should be covered within the scope of the claims of the present invention.

Claims

1. A method for optimal control of distributed resources considering the uncertainties of multiple factors in microgrids, characterized in that: The method includes the following steps: S1: Use the Copula function to measure the relationship between the predicted and actual values ​​of uncertain variables, and calculate the range of prediction error for uncertain variables; S2: Using the Copula function and the discrete convolution method, the problem of requiring variables to be independent in discrete convolution is solved by adding the Copula function, which measures the correlation between different variables, into the discrete convolution calculation. The uncertainty range of variables with different coupling relationships is measured through the above steps. S3: Using the potential game method with distributed characteristics and mapping relationships, the rationality of individual interests is constructed through utility functions and payoff functions, and the overall importance is expressed through the potential function; S4: Based on the constructed game strategy set and the set priority sequence, the algorithm uses the local information of the game supply and demand members and the strategies of the neighbors to complete the strategy iteration and optimization. Step S3 specifically includes the following steps: S31: Identify the device members involved in resource regulation and collect the input and output variables of different members within the system; S32: Define the basic types of supply and demand participants in the system. The participants are not limited to the basic individuals listed below, but can be any combination of these basic components: N={X1,X2,...X n1 ,Y1,Y2,…,Y n2 ,Z1,Z2,…,Z n3 ···}? S33: Each member of the supply and demand side participating in the distributed resource regulation game has a corresponding payoff function, that is, the payoff generated by its strategy, as follows: Among them U i Let F be the utility function of distributed heterogeneous component i participating in a distributed potential game. i Let i be the payoff function for a distributed heterogeneous component i participating in a distributed potential game. Let X be the payoff function of the J-th distributed heterogeneous component X participating in the distributed potential game; S34: Based on different members, the various constraints of the members participating in distributed resource regulation are transformed into functions in the game strategy set. The strategies of the supply and demand entities participating in the game include the output and input demand range of each basic component, where P i The strategy of the participants is represented by their strategy set as follows: S35: Determine Γ=<N,{Y i } i∈N ,{U i } i∈N > is a strategy game with a finite number of participants, where N = {1, 2, ..., n} is the number of participants, and Y i For the strategy set of participating member i, U i The utility function representing member i; S36: The basis of the game theory model is to satisfy the utility and payoff of each player under the premise of satisfying the supply and demand balance of distributed resource regulation, and to set balance constraints. S37: In a game, each player in the supply and demand dynamics is responsible for the supply and demand balance constraints of distributed resource regulation. Each player's strategy set objective is solely to maximize their own utility function, thus introducing a penalty function. To handle the global constraints of the supply and demand participants in the game: Where P Z This represents the difference between the supply and demand of distributed resources.

2. The method for optimal distributed resource regulation considering multi-factor uncertainties in microgrids according to claim 1, characterized in that: In step S1, the use of the Copula function to measure the relationship between the predicted and actual values ​​of the uncertain variable, and to calculate the range of the prediction error of the uncertain variable, specifically includes: S11: The output of a distributed system is affected by many factors. Let X be the measured value of the uncertain variable and Y be the predicted value. Solve for the invertible cumulative distribution function F of the measured value and the predicted value. X (x) = P(X < x) and F Y (y) = P(Y < y), where if F X (x) and F Y If (y) is considered a random variable, then they are all considered to be uniformly distributed: Where u X and u Y It follows a uniform distribution in the interval [0,1]. S12: Use the Copula function C(·) to construct a multivariate cumulative distribution function F from the uniform marginal distribution of the measured and predicted values ​​of the uncertainty variable. X,Y (x,y)=C(F X (x),F Y (y)), and rewrite the multivariate cumulative distribution function of the uncertainty variable in Copula form. S13: Using the Gaussian Copula function C(·), obtain the joint probability density function f of the uncertain variables from the given point prediction data. X,Y (x,y), the joint probability density function can be used to evaluate the performance of the prediction model and reveal the dependency between the actual and predicted values ​​of the variables; S14: Given the predicted value, obtain the invertible cumulative distribution function of the actual output, and transform it to obtain the conditional probability density function of the prediction error value of the uncertain variable. Calculate the conditional probability density function of the prediction error of the uncertain variable through the actual output distribution and the Copula density function, where the Copula function contains the correlation between the predicted value and the actual value.

3. The method for optimal distributed resource control considering the multi-factor uncertainty of microgrids according to claim 2, characterized in that: Step S2 specifically includes: S21: Using the conditional probability function of the prediction variable error as input, calculate the joint probability distribution function f of the variables and their sum. X,Y (x,y) is then rewritten as a discrete convolution of variables; S22: Considering the complex coupling relationship between variables, the Copula function is introduced to measure the complex correlation between the conditional probability density functions of variables, and it is introduced into the convolution calculation of the difference or sum of variables to solve the problem that convolution requires the variables to be independent; through Copula theory, it is rewritten into a related convolution form with the Copula function; S23: Discretizes the correlated convolution form with variable correlation into a discrete convolution form, which has a similar structure to the convolution formula, but adds a Copula function to build the dependency relationship between variables compared to the previous one; S24: Based on the sum or difference relationship between variables, apply the expansion of the relevant discrete convolution described in S23 to calculate the variable prediction error distribution to measure the uncertainty of the complex coupling relationship between different uncertain variables.

4. The method for optimal distributed resource control considering multi-factor uncertainties in microgrids according to claim 3, characterized in that: Step S4 specifically includes: S41: Determine the member constraint function, revenue function, and utility function for optimal regulation of distributed resources; S42: Initialize the optimization algorithm parameters, construct the individual game model of the supply and demand members of the game according to the different constraints described in S34, and set the priority queue in the game according to the importance. S43: Send a request to the preceding player in the priority queue; evaluate the player's own strategy and select the best strategy. Game players update their strategies, calculate the rate of change of strategy, and set the current convergence status: S44: Through strategy iteration and penalty system constraints, all players in the game determine the current game state by communicating with their neighbors. S45: Determine whether the supply and demand constraints are met: For power generation, For electricity consumption; if the requirements are not met, optimize by iteratively increasing the penalty coefficient: If the conditions are met, all members of the supply and demand sides in the game retain their current strategies; S46: Output the game strategy and optimization results ({P) i,t }i=1,2,...,N; t=1,2,...,T).

5. A distributed resource optimal control system considering the uncertainties of multiple factors in a microgrid, characterized in that: The system employs the method described in any one of claims 1 to 4.

Images