Virtual power plant customer portrait determination method, apparatus and device, and storage medium

Abstract

The invention discloses a virtual power plant customer portrait determination method and device, equipment and a storage medium. The method comprises the following steps: acquiring customer portrait associated information of a current virtual power plant; respectively determining a load characteristic value and a credit characteristic value according to the load resource information and the customer default information, and determining a customer willingness characteristic value by using the operation state environment information and a fuzzy membership function; and inputting the customer target information into a preset neural network model to obtain user portrait information. According to the technical scheme provided by the embodiment of the invention, accurate expression of customer willingness characteristics with relatively high subjectivity is realized, the influence of load resources, enterprise willingness and performance credit factors on a virtual power plant scheduling result is more fully considered, the customer portrait generation accuracy is improved, and the problem that the virtual power plant customer portrait is inaccurate is solved.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a method, device, equipment and storage medium for determining a customer profile of a virtual power plant. Background Art

[0002] In the context of high penetration of new energy, virtual power plants can aggregate idle consumption and power generation resources on the load side to actively support the power grid, thereby promoting power balance at the lowest economic cost. Typical load types include distributed photovoltaics, energy storage, charging stations, air conditioners, and flexible loads. The market mechanism of virtual power plants can bring more benefits to corporate customers' stock assets, and also give rise to virtual power plant operator service formats.

[0003] Since the ownership and operation rights of load-side resources belong to corporate customers, and the willingness and ability of enterprises to participate in virtual power plants will affect the actual scheduling and operation effects of virtual power plants, it is necessary for virtual power plant operators to classify, evaluate and manage corporate customers with load-side resources, give high-quality corporate customers more opportunities to participate in virtual power plants and gradually eliminate low-quality customers, thereby improving the reliability of response scheduling.

[0004] However, in actual work, the virtual power plant customer portrait often has the problem of inaccurate mathematical model of the virtual power plant customer portrait. Summary of the invention

[0005] The present invention provides a method, device, equipment and storage medium for determining a virtual power plant customer profile to solve the problem of inaccurate virtual power plant customer profile.

[0006] In a first aspect, the present invention provides a method for determining a virtual power plant customer profile, comprising:

[0007] Acquire customer profile association information of the current virtual power plant, wherein the customer profile association information includes load resource information, operating status environment information and default information of the customers of the current virtual power plant, and the operating status environment information includes the operating status of the customer and the power trading environment of the customer;

[0008] Determine a load characteristic value and a credit characteristic value according to the load resource information and the customer default information, respectively, and determine a customer willingness characteristic value by using the operation state environment information and a fuzzy membership function, wherein the customer target information includes the load characteristic value, the credit characteristic value and the customer willingness characteristic value;

[0009] The customer target information is input into a preset neural network model to obtain user portrait information.

[0010] In a second aspect, the present invention provides a virtual power plant customer profile determination device, comprising:

[0011] An information acquisition module, used to acquire customer profile-related information of the current virtual power plant, wherein the customer profile-related information includes load resource information, operating status environment information and default information of the customers of the current virtual power plant, and the operating status environment information includes the operating status of the customer and the power trading environment of the customer;

[0012] a characteristic value determination module, configured to determine a load characteristic value and a credit characteristic value according to the load resource information and the customer default information, respectively, and determine a customer willingness characteristic value by using the operation state environment information and a fuzzy membership function, wherein the customer target information includes the load characteristic value, the credit characteristic value and the customer willingness characteristic value;

[0013] The user portrait generation module is used to input the customer target information into a preset neural network model to obtain user portrait information.

[0014] In a third aspect, the present invention provides an electronic device, the electronic device comprising:

[0015] at least one processor;

[0016] and a memory communicatively coupled to the at least one processor;

[0017] Among them, the memory stores a computer program that can be executed by at least one processor, and the computer program is executed by at least one processor so that the at least one processor can execute the virtual power plant customer profile determination method of the first aspect mentioned above.

[0018] In a fourth aspect, the present invention provides a computer-readable storage medium storing computer instructions, which are used to enable a processor to implement the method for determining a virtual power plant customer profile of the first aspect when executed.

[0019] The virtual power plant customer portrait determination scheme provided by the present invention obtains the customer portrait association information of the current virtual power plant, wherein the customer portrait association information includes the load resource information, operating state environment information and default information of the customer of the current virtual power plant, the operating state environment information includes the operating state of the customer and the power trading environment of the customer, and the load characteristic value and the credit characteristic value are determined according to the load resource information and the customer default information, and the customer willingness characteristic value is determined by using the operating state environment information and the fuzzy membership function, wherein the customer target information includes the load characteristic value, the credit characteristic value and the customer willingness characteristic value, and the customer target information is input into the preset neural network model to obtain user portrait information. By adopting the above technical scheme, the customer willingness characteristic value is determined by using the operating state environment information and the fuzzy membership function, the accurate expression of the customer willingness characteristics with strong subjectivity is realized, the subjective parameters of the customer portrait are accurately characterized, the transformation from business scenario to mathematical modeling is completed, and the influence of load resources, enterprise willingness and performance credit factors on the scheduling results of the virtual power plant is more fully considered, the accuracy of customer portrait generation is improved, and the problem of inaccurate customer portrait of the virtual power plant is solved.

[0020] It should be understood that the content described in this section is not intended to identify the key or important features of the present invention, nor is it intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0022] Figure 1 is a flow chart of a method for determining a virtual power plant customer profile according to Embodiment 1 of the present invention;

[0023] Figure 2 It is a schematic diagram of a training process of a radial basis function neural network model provided according to the first embodiment of the present invention;

[0024] Figure 3 is a flow chart of a method for determining a virtual power plant customer profile according to Embodiment 2 of the present invention;

[0025] Figure 4 This is a schematic diagram of an optimization process provided according to Embodiment 2 of the present invention;

[0026] Figure 5is a structural schematic diagram of a virtual power plant customer profile determination device provided according to Embodiment 3 of the present invention;

[0027] Figure 6 It is a structural schematic diagram of an electronic device provided according to Embodiment 4 of the present invention. DETAILED DESCRIPTION

[0028] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings 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 should fall within the scope of protection of the present invention.

[0029] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In the description of the present invention, unless otherwise specified, "multiple" refers to two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist, for example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the objects associated before and after are an "or" relationship. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0030] When deciding which corporate customers to invite to participate in the virtual power plant dispatch, virtual power plant operators need to consider the subjective willingness and performance credit of corporate customers in addition to being constrained by load-side resources. The subjective willingness of corporate customers will change dynamically due to many factors such as the company's own operating conditions, and it is not static. In addition to objective conditions such as load failures and energy safety, performance credit may also be subject to subjective intentional breach of contract. At present, it is difficult to mathematically model customer behavior that contains subjective factors.

[0031] Embodiment 1

[0032] Figure 1A flowchart of a method for determining a customer profile of a virtual power plant is provided for the first embodiment of the present invention. This embodiment is applicable to the case of determining a customer profile of a virtual power plant. The method can be executed by a virtual power plant customer profile determination device. The virtual power plant customer profile determination device can be implemented in the form of hardware and / or software. The virtual power plant customer profile determination device can be configured in an electronic device. The electronic device can be composed of two or more physical entities or one physical entity.

[0033] like Figure 1 As shown, the first embodiment of the present invention provides a method for determining a virtual power plant customer profile, which specifically includes the following steps:

[0034] S101. Obtain customer portrait-related information of the current virtual power plant, wherein the customer portrait-related information includes load resource information, operating status environment information and default information of the customers of the current virtual power plant, and the operating status environment information includes the operating status of the customer and the power trading environment of the customer.

[0035] In this embodiment, in order to mathematically model the customer portrait, the customer portrait-related information of the current virtual power plant can be obtained in advance. Load resource information includes load types, such as energy storage, photovoltaics, charging stations, air conditioning, flexible loads, smart buildings, smart homes, heat storage boilers, and ground source heat pumps, as well as related indicators of each load type, such as adjustable power (MW) and adjustable capacity (MWh). Operation status and environmental information include the customer's operating status, such as revenue, and the customer's electricity trading environment, such as the trading market environment. Default information includes subjective default behavior information and objective default behavior information, such as the number of defaults.

[0036] S102. Determine a load characteristic value and a credit characteristic value according to the load resource information and the customer default information respectively, and determine a customer willingness characteristic value using the operating status environment information and a fuzzy membership function, wherein the customer target information includes the load characteristic value, the credit characteristic value and the customer willingness characteristic value.

[0037] In this embodiment, the load characteristic value and the credit characteristic value can be determined respectively by using the load resource information and the customer default information. For example, the load characteristic value and the credit characteristic value can be determined respectively by calculating the indicators in the load resource information and the customer default information respectively. The customer willingness characteristic value can be determined by calculating the indicators in the operating status environment information or the indicators associated with the operating status environment information using a fuzzy membership function.

[0038] S103: Input the customer target information into a preset neural network model to obtain user portrait information.

[0039] In this embodiment, the load characteristic value, credit characteristic value and customer willingness characteristic value can be input into a pre-trained preset neural network model, which can output user portrait information, such as outputting user classification labels (such as key customers, potential customers or low-quality customers, etc.).

[0040] The method for determining the customer portrait of a virtual power plant provided in an embodiment of the present invention obtains the customer portrait association information of the current virtual power plant, wherein the customer portrait association information includes the load resource information, operating state environment information and default information of the customer of the current virtual power plant, the operating state environment information includes the operating state of the customer and the power trading environment of the customer, and the load characteristic value and the credit characteristic value are determined according to the load resource information and the customer default information, and the customer willingness characteristic value is determined by using the operating state environment information and the fuzzy membership function, wherein the customer target information includes the load characteristic value, the credit characteristic value and the customer willingness characteristic value, and the customer target information is input into a preset neural network model to obtain user portrait information. The technical solution of the embodiment of the present invention determines the customer willingness characteristic value by using the operating state environment information and the fuzzy membership function, realizes the accurate expression of the customer willingness characteristics with strong subjectivity, accurately characterizes the subjective parameters of the customer portrait, completes the transformation from business scenario to mathematical modeling, and more fully considers the influence of load resources, enterprise willingness and performance credit factors on the scheduling results of the virtual power plant, improves the accuracy of customer portrait generation, and solves the problem of inaccurate customer portrait of the virtual power plant.

[0041] Optionally, determining the load characteristic value based on the load resource information includes: for each customer of the current virtual power plant, determining a first proportion of the current customer's adjustable total power to the total adjustable power of all customers in the current virtual power plant, and determining a second proportion of the current customer's adjustable total capacity to the total adjustable capacity of all customers, wherein the adjustable total power is the sum of the adjustable powers of the load energies owned by the customers, and the adjustable total capacity is the sum of the adjustable capacities of the load energies owned by the customers; for each customer of the current virtual power plant, determining the sum of the first proportion and the second proportion corresponding to the current customer as the load characteristic value of the current customer.

[0042] Exemplarily, the load characteristic value U is determined by:

[0043]

[0044] Among them, P i is the total adjustable power of the i-th customer in the current virtual power plant, ∑P i is the total adjustable power of all customers in the current virtual power plant, is the first ratio, Qi is the total adjustable capacity of the i-th customer in the current virtual power plant, ∑Q i is the total adjustable amount of all customers in the current virtual power plant, If the first ratio or the second ratio is higher, it means that the customer's load resources are relatively sufficient for the virtual power plant operator, otherwise, it means that they are not sufficient.

[0045] Optionally, determining a credit characteristic value based on the customer default information includes: presetting a first deduction point for the customer's subjective default behavior and a second deduction point for the objective default behavior, wherein the objective default behavior includes a default caused by force majeure, and the first deduction point is different from the second deduction point; determining a credit characteristic value using the first deduction point, the second deduction point and the customer default information, wherein the customer default information includes the number of default behaviors, and the number of default behaviors includes the number of subjective default behaviors and the number of objective default behaviors.

[0046] Exemplarily, the credit characteristic value R is determined by:

[0047] R=(100-5*r1-2*r2) / 100

[0048] The above method means: if there is one subjective (malicious) breach of contract by the customer, the credit score will be deducted by 5 points (i.e. the first deduction = 5), and the cumulative number of times is r1. If there is one objective (force majeure) breach of contract, the credit score will be deducted by 2 points (i.e. the second deduction = 2), and the cumulative number of times is r2.

[0049] Optionally, the method of determining the customer willingness characteristic value by using the operating status environment information and the fuzzy membership function includes: according to the operating status environment information, using multiple different fuzzy membership functions to respectively generate corresponding fuzzy matrices, and using the fuzzy matrices to determine the customer willingness characteristic value matrix, wherein the fuzzy matrix includes a matrix containing fuzzy impact values ​​of the customer's load resources on the customer's production and operation, a matrix containing fuzzy sensitivity to benefits of the customer, and a matrix containing fuzzy impact values ​​of the power trading environment on the customer's willingness.

[0050] Specifically, the first target fuzzy membership function can be used to process the indicators in the operating state environment information that characterize the impact of load resources on the production and operation of customers, and / or the indicators associated with the impact of load resources on the production and operation of customers, so as to obtain a matrix representing the fuzzy impact value of the customer's load resources on the production and operation of customers. The second target fuzzy membership function can be used to process the indicators in the operating state environment information that characterize the customer's sensitivity to benefits, and / or the indicators associated with the customer's sensitivity to benefits, so as to obtain a matrix representing the customer's fuzzy sensitivity to benefits. The third target fuzzy membership function can be used to process the indicators in the operating state environment information that characterize the impact of the power trading environment on customers, and / or the indicators associated with the impact of the power trading environment on customers, so as to obtain a matrix representing the fuzzy impact value of the power trading environment on the customer's willingness.

[0051] Optionally, the step of inputting the customer target information into a preset neural network model to obtain user portrait information includes: inputting the customer target information into a radial basis function neural network model to obtain user portrait information, wherein the user portrait information includes a user label. The advantage of such a setting is that the RBF (Radial Basis Function) neural network model can generate a user portrait based on the customer target information, and for customer target information that changes dynamically, the model can ensure convergence speed while ensuring approximation accuracy.

[0052] Specifically, Figure 2 Figure 1 is a schematic diagram of the training process of a radial basis function neural network model. Figure 2 As shown, the neural network parameters of the radial basis function neural network model can be initialized first, such as setting the learning factor η = 0.1, the model approximation accuracy (that is, the iteration termination accuracy) ε = 0.1, and adjusting the weight α = 0.1. Then calculate the root mean square error (RMS), and determine whether the RMS is less than or equal to ε. If so, the training ends and the parameters and number of iterations are saved. Otherwise, the iterative training continues. The convergence speed of the RBF neural network algorithm is related to the value of the window function σ of the Gaussian function. The larger σ is, the larger the output of the Gaussian kernel function of the hidden layer in the radial basis function neural network model will be for the local response of the input signal. Although this can improve the local approximation accuracy of the network, it will also increase the iterative training time of the algorithm. Because customer behavior changes dynamically, it is necessary to optimize the training time while ensuring the approximation accuracy. You can first assign σ = 0.1 as the initial value, and then perform iterative training each time with a step size of 0.1, and stop when σ increases to the preset value, such as Figure 2 The cutoff time is σ=10. Figure 2As shown, when σ is greater than or equal to 10, the σ corresponding to the minimum number of iterations is taken as the convergence speed optimization result, and the corresponding radial basis function neural network model parameters are output, such as the adjustment weights, center and width parameters of the neural network parameters, so as to obtain the trained radial basis function neural network model. When the radial basis function neural network model is actually used, there is no need to repeat the convergence speed optimization calculation, and the result can be directly reused. The advantage of the above iterative training is that the convergence speed of the algorithm is improved by adaptively adjusting the shape of the Gaussian kernel function.

[0053] Among them, the structure of the radial basis function neural network model can be designed as:

[0054] The number of input layers in the first layer of the neural network is 3, corresponding to the load characteristic value, credit characteristic value and customer willingness characteristic value respectively. The number of hidden layers in the second layer of the neural network is 5, and the kernel function of the hidden layer neuron is a Gaussian function to achieve nonlinear spatial mapping transformation. The number of output layers in the third layer of the neural network is 3, corresponding to the three label values ​​of customer classification in the user portrait information. Different label values ​​correspond to different customer classifications. The adjustment weights, center and width parameters between the input layer and the hidden layer, and between the hidden layer and the output layer of the radial basis function neural network model can be obtained by gradient descent method and iterative training.

[0055] Embodiment 2

[0056] Figure 3 A flowchart of a method for determining a customer profile of a virtual power plant is provided in Embodiment 2 of the present invention. The technical solution of the embodiment of the present invention is further optimized on the basis of the above-mentioned optional technical solutions, and a specific method for determining the customer profile of a virtual power plant is given.

[0057] Optionally, according to the operating status environment information, corresponding fuzzy matrices are generated respectively using multiple different fuzzy membership functions, and the customer willingness eigenvalue matrix is ​​determined using the fuzzy matrix, including: generating a first weight vector using a first preset weight correspondence relationship and the customer's load resources, and generating a first membership function value using a first fuzzy membership function and the first weight vector to obtain a first fuzzy matrix containing the first weight vector and the first membership function value, wherein the first preset weight correspondence relationship includes a correspondence between the customer's load resources and a first preset weight, and the first membership function is used to characterize the fuzzy impact value of the customer's load resources on the customer's production and operation; generating a second weight vector using a second preset weight correspondence relationship and the customer's operating status, and generating a second membership function value using a second fuzzy membership function and the second weight vector to obtain a second fuzzy matrix containing the second weight vector and the second membership function value, wherein the second preset weight correspondence relationship includes The method comprises the following steps: generating a correspondence between the operating status of the customer and a second preset weight, wherein the operating status includes revenue status, profit status and industry trend, and the second membership function is used to characterize the fuzzy sensitivity of the customer to revenue; generating a third weight vector using a third preset weight correspondence and the customer's load resources, and generating a third membership function value using a third fuzzy membership function and the third weight vector to obtain a third fuzzy matrix including the third weight vector and the third membership function value, wherein the third preset weight correspondence includes a correspondence between the customer's power trading environment and a third preset weight, wherein the power trading environment includes the degree of mechanism perfection, the degree of market activity and the trading category, and the third membership function is used to characterize the fuzzy influence value of the power trading environment on the customer's willingness; performing an intersection operation on the first fuzzy matrix, the second fuzzy matrix and the third fuzzy matrix to obtain a customer willingness eigenvalue matrix; wherein the first membership function, the second membership function and the third membership function are all different. The advantage of this setting is that different membership functions are used to generate corresponding fuzzy matrices for different customer willingness characteristics, avoiding the interference of human subjective judgment on the fuzzy representation results, so that the final customer willingness characteristic values ​​are more in line with the actual situation.

[0058] like Figure 3 As shown, a method for determining a virtual power plant customer profile provided by Embodiment 2 of the present invention specifically includes the following steps:

[0059] S201. Obtain customer profile related information of the current virtual power plant.

[0060] S202. For each customer of the current virtual power plant, determine a first proportion of the current customer's adjustable total power to the total adjustable power of all customers in the current virtual power plant, and determine a second proportion of the current customer's adjustable total capacity to the total adjustable capacity of all customers; for each customer of the current virtual power plant, determine the sum of the first proportion and the second proportion corresponding to the current customer as the load characteristic value of the current customer.

[0061] S203, presetting a first deduction point for the customer's subjective breach of contract behavior and a second deduction point for the objective breach of contract behavior; and determining a credit characteristic value using the first deduction point, the second deduction point and the customer breach of contract information.

[0062] S204. Generate a first weight vector using a first preset weight correspondence relationship and the customer's load resources, and generate a first membership function value using a first fuzzy membership function and the first weight vector to obtain a first fuzzy matrix including the first weight vector and the first membership function value.

[0063] Among them, the first preset weight correspondence relationship includes the correspondence relationship between the customer's load resources and the first preset weight, and the first membership function is used to characterize the fuzzy impact value of the customer's load resources on the customer's production and operation.

[0064] Optionally, the first membership function includes an S-type membership function, the second membership function includes a Gaussian membership function, and the third membership function includes a generalized bell-shaped membership function.

[0065] For example, energy storage, photovoltaics, and charging stations are relatively decoupled from the customer's production and operation, and have little impact, while heat pumps and flexible loads, as part of the production process, have a greater impact on the customer's production and operation. Therefore, when evaluating the impact of a customer's load resources on production and operation, the load type can be referenced, so the first preset weight correspondence relationship can be pre-set. The correspondence between the customer's load resources and the first preset weight included in the first preset weight correspondence relationship is shown in the first preset weight correspondence table in Table 1 below:

[0066] Table 1 First preset weight correspondence table

[0067]

[0068] The weight of each customer's load type can be accumulated to obtain a value range greater than or equal to 0 and less than or equal to 6, with a total of 7 possible accumulated weight values, thereby obtaining a 7-dimensional first weight vector (m1, m2, ..., m7).

[0069] Generally speaking, the impact of load resources on production and operation presents a two-level differentiation feature, which conforms to the S-type distribution. Therefore, the S-type membership function can be used to generate the first membership function value to obtain the first fuzzy matrix including the first weight vector and the first membership function value. Specifically, the determination method of the first membership function value f1 includes:

[0070]

[0071] Among them, the positive or negative value of a determines the opening direction of the S-type membership function, the absolute value of a determines the steepness of the S-type membership function, and c determines whether the initial value of the horizontal translation of the S-type membership function is 0, i = 1, 2, ..., 7. The first fuzzy matrix F1 can be expressed as:

[0072] F1={(m1,f1(m1,a,0)),(m2,f1(m2,a,0)),…,(m7,f1(m7,a,0))}

[0073] S205. Generate a second weight vector using the second preset weight correspondence relationship and the operating status of the customer, and generate a second membership function value using a second fuzzy membership function and the second weight vector to obtain a second fuzzy matrix including the second weight vector and the second membership function value.

[0074] Among them, the second preset weight correspondence includes the correspondence between the customer's operating status and the second preset weight, the operating status includes revenue status, profit status and industry trend, and the second membership function is used to characterize the customer's fuzzy sensitivity to revenue.

[0075] For example, different customers have different sensitivities to the benefits of virtual power plants. Generally speaking, customers with less operating pressure and good profitability do not care about the benefits of virtual power plants, so they are not very motivated to participate in transactions. On the contrary, customers who have operational difficulties attach more importance to the benefits of participating in virtual power plants. Through expert experience, industry trend analysis, and browsing related websites, the customer's revenue growth, profit growth and industry trend information can be obtained to preset the correspondence between the customer's operating status and the second preset weight. The correspondence between the customer's operating status and the second preset weight is shown in Table 2, the second preset weight correspondence table:

[0076] Table 2 Second preset weight correspondence table

[0077] Serial number Dimensions Feature description and second preset weight 1 Revenue Increase (1), Maintain (0), Decline (-1) 2 profit Increase (1), Maintain (0), Decline (-1) 3 Industry Trends Increase (1), Maintain (0), Decline (-1)

[0078] Among them, by weight accumulation, a value range of greater than or equal to -3 and less than or equal to 3 can be obtained, and there are 7 possible accumulated weight values ​​in total. Thus, a 7-dimensional second weight vector (n1, n2, ..., n7) can be obtained.

[0079] Generally speaking, customers with particularly good business conditions are not sensitive to compensation benefits, while companies with poor business conditions pay more attention to the benefits of virtual power plants. Therefore, customers' sensitivity to benefits basically conforms to the Gaussian distribution. Therefore, the Gaussian membership function can be used to generate the second membership function value to obtain a second fuzzy matrix containing the second weight vector and the second membership function value. Specifically, the second membership function value f2 is determined by:

[0080]

[0081] Wherein, σ is the standard deviation of the normal distribution, which indicates the degree of dispersion of the data. μ is the mean value in the normal distribution, which can be 3. The second fuzzy matrix F2 can be expressed as:

[0082] F2={(n1,f2(n1,σ,3)),(n2,f2(n2,σ,3)),…,(n7,f2(n7,σ,3))}

[0083] S206. Generate a third weight vector using a third preset weight correspondence relationship and the customer's load resources, and generate a third membership function value using a third fuzzy membership function and the third weight vector to obtain a third fuzzy matrix including the third weight vector and the third membership function value.

[0084] Among them, the third preset weight correspondence includes the correspondence between the customer's electricity trading environment and the third preset weight, the electricity trading environment includes the degree of mechanism perfection, market activity and trading categories, and the third membership function is used to characterize the fuzzy impact value of the electricity trading environment on the customer's willingness.

[0085] For example, in the current organization of the construction of virtual power plants, there are still objective problems such as imperfect policy mechanisms and the need to improve trading platforms. At the same time, there are differences in the types of transactions that have been carried out in different trading markets. Generally speaking, they are divided into demand response, ancillary services and spot markets. Therefore, it is necessary to evaluate the impact of the policy market environment in the customer's area and their subjective trading category preferences on customer willingness. Information such as the degree of perfection of policy mechanisms, market activity, and trading categories can be obtained through expert experience, industry trend analysis, and browsing relevant websites. In this way, the correspondence between the customer's electricity trading environment and the third preset weight can be preset. The correspondence between the customer's electricity trading environment and the third preset weight is shown in Table 3 below:

[0086] Table 3 The third preset weight correspondence table

[0087]

[0088] Among them, by weight accumulation, a value range of greater than or equal to 3 and less than or equal to 9 can be obtained, and there are 7 possible accumulated weight values ​​in total. Thus, a 7-dimensional third weight vector (t1, t2, ..., t7) can be obtained.

[0089] Generally speaking, the impact of the power trading environment on customer willingness has a time lag feature, and customers need to go through a research and learning cycle before accepting the concept of virtual power plants and participating in operations. Therefore, in terms of distribution characteristics, the impact of the power trading environment on customer willingness basically conforms to the bell-shaped characteristics. Therefore, the generalized bell-shaped membership function can be used to generate the third membership function value to obtain the third fuzzy matrix containing the third weight vector and the third membership function value. Specifically, the method for determining the third membership function value f3 includes:

[0090]

[0091] Among them, b is usually positive, e is located at the center of the curve and can be 0, and d can be 1. The third fuzzy matrix F3 can be expressed as:

[0092] F3={(t1,f3(t1,b,1,0)),(t2,f3(t2,b,1,0)),…,(t7,f3(t7,b,1,0))}

[0093] S207: Perform an intersection operation on the first fuzzy matrix, the second fuzzy matrix and the third fuzzy matrix to obtain a customer willingness eigenvalue matrix.

[0094] The first membership function, the second membership function and the third membership function are all different.

[0095] Specifically, the intersection operation can be understood as the customer's customer willingness characteristics are the result of the multiple effects of these three influencing factors. Through the above steps, the fuzzy modeling of the three main factors affecting the company's willingness can be achieved. Among them, the elements in the customer willingness eigenvalue matrix are the customer willingness eigenvalues. The customer willingness eigenvalue matrix F can be expressed as:

[0096] F(a,σ,b)=F1∩F2∩F3

[0097] Optionally, after performing an intersection operation on the first fuzzy matrix, the second fuzzy matrix and the third fuzzy matrix to obtain a customer willingness eigenvalue matrix, it also includes: determining a probability distribution function that the customer willingness eigenvalue obeys, and determining a function value of the probability distribution function using the number of rows of the customer willingness eigenvalue matrix; determining optimal membership parameters using the function values ​​and the variance of the target customer willingness eigenvalue vector, wherein the element values ​​in the target customer willingness eigenvalue vector are consistent with the number of rows corresponding to the function values, and the optimal membership parameters include parameters in the first membership function, the second membership function and the third membership function; using the optimal membership parameters to update the first membership function, the second membership function and the third membership function to obtain an updated membership function, and performing an intersection operation on the updated membership function to obtain an updated customer willingness eigenvalue matrix.

[0098] Specifically, we can first determine the probability distribution function that the customer willingness eigenvalues ​​in the customer willingness eigenvalue matrix obey, such as Poisson distribution, normal distribution, gamma distribution, F distribution, and binomial distribution. For example, if it obeys Poisson distribution, it means that most customers are on the sidelines about participating in virtual power plant transactions, and only a few are highly motivated or particularly opposed. The second column of the customer willingness eigenvalue matrix F can represent the willingness of corporate customers (close to 0 means weak willingness, close to 1 means strong willingness). If the distribution of corporate customer willingness obeys Poisson distribution:

[0099]

[0100] Among them, γ can be 2, k represents the number of rows of the second column of matrix F, k = 1, 2, ..., 7. It can be seen from this that in order to achieve relatively objective and accurate fuzzy expression, it is also necessary to perform global traversal optimization for a in the S-type membership function, σ in the Gaussian membership function, and b in the generalized bell-shaped membership function. The idea of ​​this traversal optimization is to find a set of optimal a, σ, and b so that the matrix F can basically satisfy the Poisson distribution. Figure 4 The process of global traversal optimization is as follows: Figure 4 As shown, we can first initialize the values ​​of a, σ and b corresponding to the S-type membership function, Gaussian membership function and generalized bell-shaped membership function, and then traverse the Poisson probability values ​​p(k) corresponding to different values ​​of a, σ and b, and calculate the variance of the second column data of the matrix F and p(k). After the traversal is completed, the a, σ and b corresponding to the minimum variance can be output to obtain the optimal a, σ and b. The advantage of this setting is that the gradient of the fuzzy representation can be adjusted by adaptively adjusting the parameters of the fuzzy membership function, thereby more accurately representing the willingness characteristic parameters of the corporate customer portrait.

[0101] S208. Input the customer target information into the radial basis function neural network model to obtain user portrait information.

[0102] Among them, the user portrait information includes user tags.

[0103] Specifically, user tags include key customers (tag value can be defined as 1), potential customers (tag value can be defined as 0) and poor customers (tag value can be defined as -1). Key customers can be defined as the 20% customer base that accounts for 80% of the revenue of the current virtual power plant business. Key customers can be given priority in participating in transactions during virtual power plant scheduling. Potential customers can be defined as customers in non-key customer groups who have the resource conditions to participate in virtual power plants, but have not yet participated in transactions in large quantities due to insufficient cognition or ability. It is necessary to strengthen the training and guidance of potential customers in normal times. They may be converted into key customers later. Poor-quality customers can be defined as customers in non-key customer groups who have not yet formed a transaction volume due to insufficient performance capabilities or weak subjective willingness.

[0104] The method for determining the customer profile of a virtual power plant provided in an embodiment of the present invention utilizes different membership functions to generate corresponding fuzzy matrices for different customer willingness characteristics, thereby avoiding the interference of human subjective judgment on the fuzzy representation results, so that the final customer willingness characteristic values ​​are more in line with the actual situation. This method improves the customer classification accuracy and has dynamic adaptability, which can help operators of virtual power plants improve their operating levels.

[0105] Embodiment 3

[0106] Figure 5 This is a schematic diagram of the structure of a virtual power plant customer profile determination device provided in Embodiment 3 of the present invention. Figure 5 As shown, the device includes: an information acquisition module 301, a feature value determination module 302 and a user portrait generation module 303, wherein:

[0107] An information acquisition module, used to acquire customer profile-related information of the current virtual power plant, wherein the customer profile-related information includes load resource information, operating status environment information and default information of the customers of the current virtual power plant, and the operating status environment information includes the operating status of the customer and the power trading environment of the customer;

[0108] a characteristic value determination module, configured to determine a load characteristic value and a credit characteristic value according to the load resource information and the customer default information, respectively, and determine a customer willingness characteristic value by using the operation state environment information and a fuzzy membership function, wherein the customer target information includes the load characteristic value, the credit characteristic value and the customer willingness characteristic value;

[0109] The user portrait generation module is used to input the customer target information into a preset neural network model to obtain user portrait information.

[0110] The virtual power plant customer portrait determination device provided in the embodiment of the present invention determines the customer willingness characteristic value by using the operating status environment information and the fuzzy membership function, thereby achieving the accurate expression of the customer willingness characteristics with strong subjectivity, accurately characterizing the subjective parameters of the customer portrait, completing the transformation from the business scenario to the mathematical modeling, and more fully considering the influence of load resources, enterprise willingness and performance credit factors on the virtual power plant scheduling results, thereby improving the accuracy of customer portrait generation and solving the problem of inaccurate customer portrait of the virtual power plant.

[0111] Optionally, the feature value determination module includes:

[0112] a ratio determination unit, for determining, for each customer of the current virtual power plant, a first ratio of the current customer's total adjustable power to the sum of the total adjustable powers of all customers in the current virtual power plant, and determining a second ratio of the current customer's total adjustable capacity to the sum of the total adjustable capacities of all customers, wherein the total adjustable power is the sum of the adjustable powers of the load energy sources owned by the customer, and the total adjustable capacity is the sum of the adjustable capacities of the load energy sources owned by the customer;

[0113] The load characteristic value determination unit is used to determine, for each customer of the current virtual power plant, the sum of the first ratio and the second ratio corresponding to the current customer as the load characteristic value of the current customer.

[0114] Optionally, the feature value determination module includes:

[0115] a deduction point preset unit, used to preset a first deduction point for a subjective breach of contract and a second deduction point for an objective breach of contract of the customer, wherein the objective breach of contract includes a breach of contract caused by force majeure, and the first deduction point is different from the second deduction point;

[0116] A credit characteristic value determination unit is used to determine a credit characteristic value by using the first deduction point, the second deduction point and the customer default information, wherein the customer default information includes the number of default behaviors, and the number of default behaviors includes the number of subjective default behaviors and the number of objective default behaviors.

[0117] Optionally, the eigenvalue determination module is specifically used to generate corresponding fuzzy matrices according to the operating status environment information using multiple different fuzzy membership functions, and use the fuzzy matrices to determine the customer willingness eigenvalue matrix, wherein the fuzzy matrix includes a matrix containing fuzzy impact values ​​of the customer's load resources on the customer's production and operation, a matrix containing fuzzy sensitivity to benefits of the customer, and a matrix containing fuzzy impact values ​​of the power trading environment on the customer's willingness.

[0118] Optionally, according to the operating status environment information, corresponding fuzzy matrices are generated respectively using multiple different fuzzy membership functions, and the customer willingness eigenvalue matrix is ​​determined using the fuzzy matrix, including: generating a first weight vector using a first preset weight correspondence relationship and the customer's load resources, and generating a first membership function value using a first fuzzy membership function and the first weight vector to obtain a first fuzzy matrix containing the first weight vector and the first membership function value, wherein the first preset weight correspondence relationship includes a correspondence between the customer's load resources and a first preset weight, and the first membership function is used to characterize the fuzzy impact value of the customer's load resources on the customer's production and operation; generating a second weight vector using a second preset weight correspondence relationship and the customer's operating status, and generating a second membership function value using a second fuzzy membership function and the second weight vector to obtain a second fuzzy matrix containing the second weight vector and the second membership function value, wherein the second preset weight correspondence relationship includes The method comprises the following steps: generating a correspondence between the operating status of the customer and a second preset weight, wherein the operating status includes revenue status, profit status and industry trend, and the second membership function is used to characterize the fuzzy sensitivity of the customer to revenue; generating a third weight vector using a third preset weight correspondence and the customer's load resources, and generating a third membership function value using a third fuzzy membership function and the third weight vector to obtain a third fuzzy matrix including the third weight vector and the third membership function value, wherein the third preset weight correspondence includes a correspondence between the customer's power trading environment and a third preset weight, wherein the power trading environment includes the degree of mechanism perfection, the degree of market activity and the trading category, and the third membership function is used to characterize the fuzzy influence value of the power trading environment on the customer's willingness; performing an intersection operation on the first fuzzy matrix, the second fuzzy matrix and the third fuzzy matrix to obtain a customer willingness eigenvalue matrix; wherein the first membership function, the second membership function and the third membership function are all different.

[0119] Optionally, the feature value determination module further includes:

[0120] a function value determination unit, configured to determine the probability distribution function obeyed by the customer willingness eigenvalues ​​after performing an intersection operation on the first fuzzy matrix, the second fuzzy matrix and the third fuzzy matrix to obtain the customer willingness eigenvalue matrix, and determine the function value of the probability distribution function using the number of rows of the customer willingness eigenvalue matrix;

[0121] A membership optimal parameter determination unit, used to determine the membership optimal parameter by using the variance of the function value and the target customer willingness characteristic value vector, wherein the element value in the target customer willingness characteristic value vector is consistent with the number of rows corresponding to the function value, and the membership optimal parameter includes the parameters in the first membership function, the second membership function and the third membership function;

[0122] A matrix updating unit is used to update the first membership function, the second membership function and the third membership function using the optimal membership parameter to obtain an updated membership function, and to obtain an updated customer willingness eigenvalue matrix by performing an intersection operation on the updated membership function.

[0123] Optionally, the first membership function includes an S-type membership function, the second membership function includes a Gaussian membership function, and the third membership function includes a generalized bell-shaped membership function.

[0124] Optionally, the user portrait generation module is specifically used to input the customer target information into a radial basis function neural network model to obtain user portrait information, wherein the user portrait information includes a user tag.

[0125] The virtual power plant customer profile determination device provided in the embodiment of the present invention can execute the virtual power plant customer profile determination method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0126] Embodiment 4

[0127] Figure 6 A schematic diagram of the structure of an electronic device 40 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0128] like Figure 6 As shown, the electronic device 40 includes at least one processor 41, and a memory connected to the at least one processor 41, such as a read-only memory (ROM) 42, a random access memory (RAM) 43, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 41 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 42 or the computer program loaded from the storage unit 48 to the random access memory (RAM) 43. In the RAM 43, various programs and data required for the operation of the electronic device 40 can also be stored. The processor 41, the ROM 42, and the RAM 43 are connected to each other through a bus 44. An input / output (I / O) interface 45 is also connected to the bus 44.

[0129] A number of components in the electronic device 40 are connected to the I / O interface 45, including: an input unit 46, such as a keyboard, a mouse, etc.; an output unit 47, such as various types of displays, speakers, etc.; a storage unit 48, such as a disk, an optical disk, etc.; and a communication unit 49, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 49 allows the electronic device 40 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0130] The processor 41 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 41 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The processor 41 executes the various methods and processes described above, such as a method for determining a virtual power plant customer profile.

[0131] In some embodiments, the method for determining a virtual power plant customer profile may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 48. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 40 via the ROM 42 and / or the communication unit 49. When the computer program is loaded into the RAM 43 and executed by the processor 41, one or more steps of the method for determining a virtual power plant customer profile described above may be performed. Alternatively, in other embodiments, the processor 41 may be configured to execute the method for determining a virtual power plant customer profile in any other appropriate manner (e.g., by means of firmware).

[0132] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0133] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0134] The computer device provided above can be used to execute the virtual power plant customer profile determination method provided in any of the above embodiments, and has corresponding functions and beneficial effects.

[0135] Embodiment 5

[0136] In the context of the present invention, a computer-readable storage medium may be a tangible medium, and the computer-executable instructions, when executed by a computer processor, are used to perform a method for determining a virtual power plant customer profile, the method comprising:

[0137] Acquire customer profile association information of the current virtual power plant, wherein the customer profile association information includes load resource information, operating status environment information and default information of the customers of the current virtual power plant, and the operating status environment information includes the operating status of the customer and the power trading environment of the customer;

[0138] Determine a load characteristic value and a credit characteristic value according to the load resource information and the customer default information, respectively, and determine a customer willingness characteristic value by using the operation state environment information and a fuzzy membership function, wherein the customer target information includes the load characteristic value, the credit characteristic value and the customer willingness characteristic value;

[0139] The customer target information is input into a preset neural network model to obtain user portrait information.

[0140] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by an instruction execution system, device or equipment or used with an instruction execution system, device or equipment or used in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, 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 foregoing.

[0141] The computer device provided above can be used to execute the virtual power plant customer profile determination method provided in any of the above embodiments, and has corresponding functions and beneficial effects.

[0142] It is worth noting that in the embodiment of the above-mentioned virtual power plant customer portrait determination device, the various units and modules included are only divided according to functional logic, but are not limited to the above-mentioned division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0143] Note that the above are only preferred embodiments of the present invention and the technical principles used. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A method for determining a virtual power plant customer profile, characterized in that: include: Acquire customer profile association information of the current virtual power plant, wherein the customer profile association information includes load resource information, operating status environment information and default information of the customers of the current virtual power plant, and the operating status environment information includes the operating status of the customer and the power trading environment of the customer; Determine a load characteristic value and a credit characteristic value according to the load resource information and the customer default information, respectively, and determine a customer willingness characteristic value by using the operation state environment information and a fuzzy membership function, wherein the customer target information includes the load characteristic value, the credit characteristic value and the customer willingness characteristic value; The customer target information is input into a preset neural network model to obtain user portrait information.

2. The method according to claim 1, characterized in that Determining a load characteristic value according to the load resource information includes: For each customer of the current virtual power plant, determine a first proportion of the current customer's total adjustable power to the sum of the total adjustable powers of all customers in the current virtual power plant, and determine a second proportion of the current customer's total adjustable capacity to the sum of the total adjustable capacities of all customers, wherein the total adjustable power is the sum of the adjustable powers of the load energy sources owned by the customer, and the total adjustable capacity is the sum of the adjustable capacities of the load energy sources owned by the customer; For each customer of the current virtual power plant, the sum of the first ratio corresponding to the current customer and the second ratio corresponding to the current customer is determined as the load characteristic value of the current customer.

3. The method according to claim 1, characterized in that Determining a credit characteristic value according to the customer default information includes: Presetting a first deduction point for a subjective breach of contract and a second deduction point for an objective breach of contract of the customer, wherein the objective breach of contract includes a breach of contract caused by force majeure, and the first deduction point is different from the second deduction point; The credit characteristic value is determined by using the first deduction points, the second deduction points and the customer default information, wherein the customer default information includes the number of default behaviors, and the number of default behaviors includes the number of subjective default behaviors and the number of objective default behaviors.

4. The method according to any one of claims 1 to 3, characterized in that The determining of the customer willingness characteristic value by using the operation status environment information and the fuzzy membership function includes: According to the operating status environment information, corresponding fuzzy matrices are generated respectively using multiple different fuzzy membership functions, and the fuzzy matrices are used to determine the customer willingness eigenvalue matrix, wherein the fuzzy matrix includes a matrix containing fuzzy impact values ​​for characterizing the customer's load resources on the customer's production and operation, a matrix containing fuzzy sensitivity to benefits for the customer, and a matrix containing fuzzy impact values ​​for characterizing the power trading environment on the customer's willingness.

5. The method according to claim 4, characterized in that The method of generating corresponding fuzzy matrices using a plurality of different fuzzy membership functions according to the operating status environment information, and determining a customer willingness eigenvalue matrix using the fuzzy matrices includes: A first weight vector is generated by using a first preset weight correspondence relationship and the customer's load resources, and a first membership function value is generated by using a first fuzzy membership function and the first weight vector, so as to obtain a first fuzzy matrix including the first weight vector and the first membership function value, wherein the first preset weight correspondence relationship includes a correspondence relationship between the customer's load resources and the first preset weight, and the first membership function is used to characterize the fuzzy influence value of the customer's load resources on the customer's production and operation; A second weight vector is generated by using a second preset weight correspondence relationship and the operating status of the customer, and a second membership function value is generated by using a second fuzzy membership function and the second weight vector, so as to obtain a second fuzzy matrix including the second weight vector and the second membership function value, wherein the second preset weight correspondence relationship includes a correspondence relationship between the operating status of the customer and the second preset weight, the operating status includes a revenue status, a profit status and an industry trend, and the second membership function is used to characterize the fuzzy sensitivity of the customer to revenue; A third weight vector is generated by using a third preset weight correspondence relationship and the customer's load resources, and a third membership function value is generated by using a third fuzzy membership function and the third weight vector, so as to obtain a third fuzzy matrix including the third weight vector and the third membership function value, wherein the third preset weight correspondence relationship includes a correspondence relationship between the customer's power trading environment and the third preset weight, the power trading environment includes the degree of mechanism perfection, the degree of market activity and the transaction category, and the third membership function is used to characterize the fuzzy influence value of the power trading environment on the customer's willingness; Performing an intersection operation on the first fuzzy matrix, the second fuzzy matrix and the third fuzzy matrix to obtain a customer willingness eigenvalue matrix; The first membership function, the second membership function and the third membership function are all different.

6. The method according to claim 5, characterized in that After performing the intersection operation on the first fuzzy matrix, the second fuzzy matrix and the third fuzzy matrix to obtain the customer willingness eigenvalue matrix, the method further includes: Determine a probability distribution function that the customer willingness eigenvalue obeys, and determine a function value of the probability distribution function using the number of rows of the customer willingness eigenvalue matrix; Determine the optimal membership parameters by using the variance of the function values ​​and the target customer willingness feature value vector, wherein the element value in the target customer willingness feature value vector is consistent with the number of rows corresponding to the function value, and the optimal membership parameters include the parameters in the first membership function, the second membership function and the third membership function; The first membership function, the second membership function and the third membership function are updated using the optimal membership parameter to obtain an updated membership function, and an updated customer willingness eigenvalue matrix is ​​obtained by performing an intersection operation on the updated membership function.

7. The method according to claim 5, characterized in that The first membership function includes an S-type membership function, the second membership function includes a Gaussian membership function, and the third membership function includes a generalized bell-shaped membership function.

8. The method according to claim 1, characterized in that The step of inputting the customer target information into a preset neural network model to obtain user portrait information includes: The customer target information is input into a radial basis function neural network model to obtain user portrait information, wherein the user portrait information includes a user label.

9. A virtual power plant customer profile determination device, characterized in that: include: An information acquisition module, used to acquire customer profile-related information of the current virtual power plant, wherein the customer profile-related information includes load resource information, operating status environment information and default information of the customers of the current virtual power plant, and the operating status environment information includes the operating status of the customer and the power trading environment of the customer; a characteristic value determination module, configured to determine a load characteristic value and a credit characteristic value according to the load resource information and the customer default information, respectively, and determine a customer willingness characteristic value by using the operation state environment information and a fuzzy membership function, wherein the customer target information includes the load characteristic value, the credit characteristic value and the customer willingness characteristic value; The user portrait generation module is used to input the customer target information into a preset neural network model to obtain user portrait information.

10. The device according to claim 9, characterized in that The characteristic value determination module comprises: a ratio determination unit, for determining, for each customer of the current virtual power plant, a first ratio of the current customer's total adjustable power to the sum of the total adjustable powers of all customers in the current virtual power plant, and determining a second ratio of the current customer's total adjustable capacity to the sum of the total adjustable capacities of all customers, wherein the total adjustable power is the sum of the adjustable powers of the load energy sources owned by the customer, and the total adjustable capacity is the sum of the adjustable capacities of the load energy sources owned by the customer; The load characteristic value determination unit is used to determine, for each customer of the current virtual power plant, the sum of the first ratio and the second ratio corresponding to the current customer as the load characteristic value of the current customer.

11. The device according to claim 9, characterized in that The characteristic value determination module comprises: a deduction point preset unit, used to preset a first deduction point for a subjective breach of contract and a second deduction point for an objective breach of contract of the customer, wherein the objective breach of contract includes a breach of contract caused by force majeure, and the first deduction point is different from the second deduction point; A credit characteristic value determination unit is used to determine a credit characteristic value by using the first deduction point, the second deduction point and the customer default information, wherein the customer default information includes the number of default behaviors, and the number of default behaviors includes the number of subjective default behaviors and the number of objective default behaviors.

12. The device according to any one of claims 9 to 11, characterized in that The eigenvalue determination module is specifically used to generate corresponding fuzzy matrices using multiple different fuzzy membership functions according to the operating status environment information, and use the fuzzy matrices to determine the customer willingness eigenvalue matrix, wherein the fuzzy matrix includes a matrix containing fuzzy impact values ​​of the customer's load resources on the customer's production and operation, a matrix containing fuzzy sensitivity to benefits of the customer, and a matrix containing fuzzy impact values ​​of the power trading environment on the customer's willingness.

13. The device according to claim 12, characterized in that The method of generating corresponding fuzzy matrices using a plurality of different fuzzy membership functions according to the operating status environment information, and determining a customer willingness eigenvalue matrix using the fuzzy matrices includes: A first weight vector is generated by using a first preset weight correspondence relationship and the customer's load resources, and a first membership function value is generated by using a first fuzzy membership function and the first weight vector, so as to obtain a first fuzzy matrix including the first weight vector and the first membership function value, wherein the first preset weight correspondence relationship includes a correspondence relationship between the customer's load resources and the first preset weight, and the first membership function is used to characterize the fuzzy influence value of the customer's load resources on the customer's production and operation; A second weight vector is generated by using a second preset weight correspondence relationship and the operating status of the customer, and a second membership function value is generated by using a second fuzzy membership function and the second weight vector, so as to obtain a second fuzzy matrix including the second weight vector and the second membership function value, wherein the second preset weight correspondence relationship includes a correspondence relationship between the operating status of the customer and the second preset weight, the operating status includes a revenue status, a profit status and an industry trend, and the second membership function is used to characterize the fuzzy sensitivity of the customer to revenue; A third weight vector is generated by using a third preset weight correspondence relationship and the customer's load resources, and a third membership function value is generated by using a third fuzzy membership function and the third weight vector, so as to obtain a third fuzzy matrix including the third weight vector and the third membership function value, wherein the third preset weight correspondence relationship includes a correspondence relationship between the customer's power trading environment and the third preset weight, the power trading environment includes the degree of mechanism perfection, the degree of market activity and the transaction category, and the third membership function is used to characterize the fuzzy influence value of the power trading environment on the customer's willingness; Performing an intersection operation on the first fuzzy matrix, the second fuzzy matrix and the third fuzzy matrix to obtain a customer willingness eigenvalue matrix; The first membership function, the second membership function and the third membership function are all different.

14. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the virtual power plant customer profile determination method described in any one of claims 1-8.

15. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the virtual power plant customer profile determination method described in any one of claims 1-8 when executed.

Images