A user resource management method and system for a virtual power plant
By building a dynamic resource quality evaluation model, using BP neural network to combine dynamic rate and contribution weighting, the real-time problem of virtual power plant user resource management is solved, the accurate and real-time management of virtual power plant user resources is achieved, and the management effect is improved.
Patent Information
- Application Number
- CN202411165492.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-23
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2044-08-23
AI Technical Summary
In the prior art, virtual power plants are difficult to realize real-time management of user resources, resulting in poor management timeliness and reducing management effect.
By obtaining the basic data, operation data and declaration data of the user resources of the virtual power plant, quantifying the dynamic rate and contribution degree, combining the BP neural network to build a dynamic resource quality evaluation model to achieve real-time evaluation of user resources.
It enhances the accuracy and sensitivity of the dynamic evaluation model of resource quality, realizes real-time and accurate grasp of virtual power plant user resource data, and improves resource management effect.
Smart Images

Figure CN118690973B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of virtual power plant management, and in particular to a user resource management method and system for a virtual power plant. Background Art
[0002] A virtual power plant is a model that uses renewable energy generation devices as the basis to purchase and sell electricity through a virtual power trading platform. Virtual power plants rely on resource scheduling by various settlement parties / users to maintain fairness, transparency and efficiency in virtual power plant operations and ensure that the interests of virtual power plants are reasonably protected. Therefore, effective management of user resources of virtual power plants can optimize power scheduling of virtual power plants and is an important part of settlement management of virtual power plants.
[0003] In the prior art, the management of user resources by virtual power plants is static management, that is, user resources cannot be managed and updated in real time according to data changes of the virtual power plant, and it is difficult to achieve real-time resource evaluation, resulting in difficulty in grasping real-time user resources during management, poor management timeliness, and reduced management effectiveness. Summary of the invention
[0004] The purpose of the present invention is to provide a user resource management method and system for a virtual power plant to solve the technical problems in the prior art that it is difficult to achieve real-time resource evaluation, resulting in difficulty in grasping real-time user resources during management, poor management timeliness, and reduced management effectiveness.
[0005] In order to solve the above technical problems, the present invention specifically provides the following technical solutions:
[0006] A user resource management method for a virtual power plant comprises the following steps:
[0007] Obtain basic data, operation data, declaration data and quality data of user resources of virtual power plants;
[0008] By quantifying the basic data, operation data, dynamic rate of declared data of user resources, and contribution to quality data;
[0009] The dynamic rate and contribution degree are combined to form the characteristic weights of basic data, operating data and declared data;
[0010] The BP neural network is used to encapsulate the mapping relationship between basic data, operating data, declared data and quality data according to the feature weights, and a resource quality dynamic evaluation model for obtaining quality data based on basic data, operating data and declared data is obtained.
[0011] As a preferred solution of the present invention, the basic data includes user name, resource account number, resource name, resource capacity, number of participations, participation duration, monthly adjustment amount, annual adjustment amount and cumulative adjustment amount;
[0012] Operational data include real-time load, baseline load, remote regulation, remote control, remote signaling, and telemetry data of resource operation;
[0013] The declaration data includes user name, resource name, contact person, declaration date, and declaration plan power;
[0014] Quality data include: baseline of selection time, plan, curve of actual adjustment amount, as well as cumulative number of participations, cumulative participation duration, cumulative participation qualification rate, cumulative participation rate, annual number of participations, annual participation qualification rate, and annual participation rate.
[0015] As a preferred solution of the present invention, the method for quantifying the dynamic rate of basic data, operation data and declaration data of user resources includes:
[0016] Quantify the similarity between the basic data, operation data, and declaration data of the user resources at the current moment and the basic data, operation data, and declaration data of the user resources at the previous moment to obtain the dynamic rate of the basic data, operation data, and declaration data of the user resources;
[0017] The dynamic rates of the basic data, operation data and declaration data of the user resources are:
[0018] ;
[0019] In the formula, K E is the dynamic rate of E, E is the character identifier of basic data, operating data, and declared data, A is the data vector of basic data, B is the data vector of operating data, C is the data vector of declared data, and E t is the current E, E t-1 is the E of the previous moment, ||E t -E t-1 ||1 is the calculation formula for L1 distance.
[0020] As a preferred solution of the present invention, the method for quantifying the contribution of basic data, operation data, and declaration data of user resources to quality data includes:
[0021] Quantify the mutual information between the basic data, operation data, declaration data and quality data of user resources to obtain the contribution of the basic data, operation data and declaration data of user resources to the quality data;
[0022] The contribution of the basic data, operation data and declaration data of the user resources to the quality data is:
[0023] ;
[0024] Where, L Eis the contribution of E to the quality data, E is the character identifier of basic data, operational data and declared data, A is the data vector of basic data, B is the data vector of operational data, C is the data vector of declared data, F is the data vector of quality data, p(E,F) is the joint probability distribution function of E and F, p(E) is the marginal probability distribution function of E, and p(F) is the marginal probability distribution function of F.
[0025] As a preferred solution of the present invention, the method of combining the dynamic rate and contribution degree to form the characteristic weights of basic data, operating data, and declared data includes:
[0026] The dynamic rate and the contribution are combined by power and normalized to obtain a normalized combination value of the dynamic rate and the contribution. The normalized combination value of the dynamic rate and the contribution is: ;
[0027] By using the normalized combined value of the dynamic rate and contribution, the characteristic weights of the basic data, operating data, and declared data that are positively correlated with the dynamic rate and contribution are obtained. The characteristic weights are:
[0028] ;
[0029] Where, X E is the feature weight of E, K E is the dynamic rate of E, L E is the contribution of E to the quality data, E is the character identifier of basic data, operational data and declared data, A is the data vector of basic data, B is the data vector of operational data, C is the data vector of declared data, and F is the data vector of quality data.
[0030] As a preferred solution of the present invention, the method for constructing the resource quality dynamic evaluation model includes:
[0031] The basic data, operation data and declaration data with characteristic weights are used as input items of BP neural network, and the quality data is used as output item of BP neural network;
[0032] The BP neural network is used to map input items and output items to obtain a dynamic resource quality assessment model;
[0033] The resource quality dynamic assessment model is:
[0034] ;
[0035] In the formula, F is the quality data, BP is the BP neural network, X Eis the feature weight of E, E is the character identifier of basic data, operating data and declared data, A is the data vector of basic data, B is the data vector of operating data, and C is the data vector of declared data.
[0036] As a preferred solution of the present invention, the training loss of the BP neural network is the mean square error between the predicted value and the true value.
[0037] As a preferred solution of the present invention, it also includes using a resource quality dynamic evaluation model to perform real-time evaluation of user resource quality, wherein the method of using the resource quality dynamic evaluation model to perform real-time evaluation of user resource quality includes:
[0038] Obtain the basic data, operation data, and declaration data of the virtual power plant at the current moment;
[0039] Calculating the dynamic rate and contribution at the current moment;
[0040] The basic data, operation data, declaration data at the current moment and the dynamic rate and contribution degree at the current moment are input into the resource quality dynamic evaluation model, and the quality data of the user resources at the current moment is output.
[0041] As a preferred solution of the present invention, the present invention provides a user resource management system for a virtual power plant, which is applied to the user resource management method for a virtual power plant, and the system includes:
[0042] A data collection unit is used to obtain basic data, operation data, declaration data and quality data of user resources of the virtual power plant;
[0043] Data preprocessing unit, used to quantify the basic data, operation data, dynamic rate of declared data of user resources, and contribution to quality data;
[0044] The dynamic rate and contribution degree are combined to form the characteristic weights of basic data, operating data and declared data;
[0045] Using the BP neural network to encapsulate the mapping relationship between the basic data, the operating data, the declared data and the quality data according to the feature weights, a resource quality dynamic evaluation model for obtaining the quality data according to the basic data, the operating data and the declared data is obtained;
[0046] The quality display unit is used to visualize the quality data obtained by the resource quality dynamic assessment model in charts.
[0047] As a preferred solution of the present invention, the quality display unit uses a resource quality dynamic evaluation model to perform real-time evaluation of user resource quality to obtain the quality data, wherein the method of using the resource quality dynamic evaluation model to perform real-time evaluation of user resource quality includes:
[0048] Obtain the basic data, operation data, and declaration data of the virtual power plant at the current moment;
[0049] Calculating the dynamic rate and contribution at the current moment;
[0050] The basic data, operation data, declaration data at the current moment and the dynamic rate and contribution degree at the current moment are input into the resource quality dynamic evaluation model, and the quality data of the user resources at the current moment is output.
[0051] Compared with the prior art, the present invention has the following beneficial effects:
[0052] The present invention constructs a dynamic assessment model for resource quality, and realizes the acquisition of quality data based on basic data, operating data, and declared data. Moreover, the basic data, operating data, and declared data are weighted in the dynamic assessment model for resource quality by weights composed of dynamic rates and contribution degrees, thereby enhancing the accuracy and sensitivity of the dynamic assessment model for resource quality, realizing real-time and accurate grasp of resource data of virtual power plant users, and improving the effect of resource management. BRIEF DESCRIPTION OF THE DRAWINGS
[0053] In order to more clearly illustrate the implementation methods of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for the implementation methods or the description of the prior art. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other implementation drawings can be derived from the provided drawings without creative work.
[0054] Figure 1 A flow chart of a user resource management method for a virtual power plant provided by an embodiment of the present invention;
[0055] Figure 2 A block diagram of a user resource management system for a virtual power plant provided in an embodiment of the present invention. DETAILED DESCRIPTION
[0056] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0057] like Figure 1 As shown, the present invention provides a user resource management method for a virtual power plant, comprising the following steps:
[0058] Obtain basic data, operation data, declaration data and quality data of user resources of virtual power plants;
[0059] By quantifying the basic data, operation data, dynamic rate of declared data of user resources, and contribution to quality data;
[0060] The dynamic rate and contribution are combined to form the characteristic weights of basic data, operating data, and declared data;
[0061] The BP neural network is used to encapsulate the mapping relationship between basic data, operating data, declared data and quality data according to feature weights, and a dynamic resource quality assessment model for obtaining quality data based on basic data, operating data and declared data is obtained.
[0062] The present invention encapsulates the mapping relationship between basic data, operation data, declaration data and quality data to construct a dynamic resource quality assessment model, which can evaluate the quality of user resources based on basic data, operation data and declaration data and automatically measure them.
[0063] Furthermore, in order to improve the accuracy and interpretability of user resource evaluation of the constructed dynamic resource quality evaluation model, the present invention quantifies the contribution of the basic data, operating data, and declared data of user resources to the quality data, and uses the contribution as a part of the weight structure of the feature data in the dynamic resource quality evaluation model. Since the contribution is measured by mutual information, the higher the mutual information value between the basic data, operating data, declared data and quality data, the higher the correlation between the basic data, operating data, declared data and quality data, and the greater the contribution to solving the quality data. The data with a greater contribution is assigned a higher feature weight, so that the dynamic resource quality evaluation model can better utilize those features that have a greater impact on the prediction results during training, thereby improving the accuracy of the prediction.
[0064] And by looking at the weights of each feature, you can understand which features have a greater impact on the model's prediction results, which features are relatively small, and may even be irrelevant. This transparency helps users better understand the model's decision-making process, thereby enhancing trust in the model's prediction results, making the model's prediction results more explainable.
[0065] Similarly, in order to improve the temporal sensitivity of user resource evaluation of the constructed dynamic resource quality evaluation model, the present invention quantifies the dynamic rate of the basic data, operating data, and declared data of user resources, and uses the dynamic rate as a part of the weight composition of the characteristic data in the dynamic resource quality evaluation model. Since the dynamic rate is measured by the changes between the previous and next moments, the higher the degree of change of the basic data, operating data, and declared data, the higher the weight is assigned. This makes it easier for the constructed dynamic resource quality evaluation model to pay attention to data with a higher degree of change when evaluating the quality of user resources, and the data perception sensitivity is higher. That is, when the basic data, operating data, and declared data of user resources change, the dynamic resource quality evaluation model will be triggered to perform quality evaluation on the user resources, thereby realizing the dynamic real-time nature of the user resource quality evaluation.
[0066] The dynamic rate and contribution degree are combined to form the characteristic weight of the data in the dynamic evaluation model of resource quality. The contribution degree is taken as the power of the dynamic rate. The dynamic sensitivity of the dynamic evaluation model of resource quality is improved mainly, and the accuracy and interpretability of the dynamic evaluation model of resource quality is improved as an auxiliary. This can maximize the dynamic real-time evaluation on the basis of accurate evaluation of user resources of the dynamic evaluation model of resource quality.
[0067] Basic data includes user name, resource account number, resource name, resource capacity, number of participations, participation duration, monthly adjustment amount, annual adjustment amount, and cumulative adjustment amount;
[0068] Operational data include real-time load, baseline load, remote regulation, remote control, remote signaling, and telemetry data of resource operation;
[0069] The declaration data includes user name, resource name, contact person, declaration date, and declaration plan power;
[0070] Quality data include: baseline of selection time, plan, curve of actual adjustment amount, as well as cumulative number of participations, cumulative participation duration, cumulative participation qualification rate, cumulative participation rate, annual number of participations, annual participation qualification rate, and annual participation rate.
[0071] Methods for quantifying the dynamic rates of basic data, operational data, and declared data of user resources include:
[0072] Quantify the similarity between the basic data, operation data, and declaration data of the user resources at the current moment and the basic data, operation data, and declaration data of the user resources at the previous moment to obtain the dynamic rate of the basic data, operation data, and declaration data of the user resources;
[0073] The dynamic rates of basic data, operation data and declaration data of user resources are:
[0074] ;
[0075] In the formula, K E is the dynamic rate of E, E is the character identifier of basic data, operating data, and declared data, A is the data vector of basic data, B is the data vector of operating data, C is the data vector of declared data, and E t is the current E, E t-1 is the E of the previous moment, ||E t -E t-1 ||1 is the calculation formula for Manhattan distance / L1 distance.
[0076] In order to improve the accuracy and interpretability of user resource evaluation of the constructed dynamic resource quality evaluation model, the present invention quantifies the contribution of basic data, operation data, and declaration data of user resources to quality data, and uses the contribution as a part of the weight composition of feature data in the dynamic resource quality evaluation model. Since the contribution is measured by mutual information, the higher the mutual information value between the basic data, operation data, declaration data and quality data, the higher the correlation between the basic data, operation data, declaration data and quality data, and the greater the contribution to solving the quality data, the higher the feature weight is assigned to the data with greater contribution, so that the dynamic resource quality evaluation model can better utilize those features that have a greater impact on the prediction results during training, thereby improving the accuracy of the prediction.
[0077] And by looking at the weights of each feature, you can understand which features have a greater impact on the model's prediction results, which features are relatively small, and may even be irrelevant. This transparency helps users better understand the model's decision-making process, thereby enhancing trust in the model's prediction results, making the model's prediction results more explainable.
[0078] Methods for quantifying the contribution of basic data, operational data, and declared data of user resources to quality data include:
[0079] Quantify the mutual information between the basic data, operation data, declaration data and quality data of user resources to obtain the contribution of the basic data, operation data and declaration data of user resources to the quality data;
[0080] The contribution of user resource basic data, operation data, and declaration data to quality data is as follows:
[0081] ;
[0082] Where, L Eis the contribution of E to the quality data, E is the character identifier of basic data, operational data and declared data, A is the data vector of basic data, B is the data vector of operational data, C is the data vector of declared data, F is the data vector of quality data, p(E,F) is the joint probability distribution function of E and F, p(E) is the marginal probability distribution function of E, and p(F) is the marginal probability distribution function of F.
[0083] In order to improve the temporal sensitivity of user resource evaluation of the constructed dynamic evaluation model of resource quality, the present invention quantifies the dynamic rate of basic data, operation data and declaration data of user resources, and uses the dynamic rate as a part of the weight composition of characteristic data in the dynamic evaluation model of resource quality. Since the dynamic rate is measured by the change between the previous and next moments, the higher the degree of change of the basic data, operation data and declaration data, the higher the weight is assigned. This makes it easier for the constructed dynamic evaluation model of resource quality to pay attention to data with a higher degree of change when evaluating the quality of user resources, and the sensitivity of data perception is higher. That is, when the basic data, operation data and declaration data of user resources change, the dynamic evaluation model of resource quality will be triggered to perform quality evaluation on user resources, thereby realizing the dynamic real-time performance of user resource quality evaluation.
[0084] Methods for combining dynamic rate and contribution to form characteristic weights of basic data, operating data, and declared data include:
[0085] The dynamic rate and contribution are combined by power and normalized to obtain the normalized combination value of the dynamic rate and contribution. The normalized combination value of the dynamic rate and contribution is: ;
[0086] Using the normalized combined value of the dynamic rate and contribution, the characteristic weights of the basic data, operating data, and declared data that are positively correlated with the dynamic rate and contribution are obtained. The characteristic weights are:
[0087] ;
[0088] Where, X E is the feature weight of E, K E is the dynamic rate of E, L E is the contribution of E to the quality data, E is the character identifier of basic data, operational data and declared data, A is the data vector of basic data, B is the data vector of operational data, C is the data vector of declared data, and F is the data vector of quality data.
[0089] Among them, the feature weight is positively correlated with the dynamic rate and contribution, that is, the larger the dynamic rate and contribution, the greater the feature weight. Since the contribution is measured by mutual information, the higher the mutual information value of basic data, operating data, declared data and quality data, the higher the correlation between basic data, operating data, declared data and quality data, and the greater the contribution to solving quality data. The data with greater contribution is assigned a higher feature weight. Since the dynamic rate is measured by the changes between the previous and next moments, the higher the degree of change in basic data, operating data and declared data, the higher the weight is assigned.
[0090] The dynamic rate and contribution degree are combined to form the characteristic weight of the data in the dynamic evaluation model of resource quality. The contribution degree is taken as the power of the dynamic rate. The dynamic sensitivity of the dynamic evaluation model of resource quality is improved mainly, and the accuracy and interpretability of the dynamic evaluation model of resource quality is improved as an auxiliary. This can maximize the dynamic real-time evaluation on the basis of accurate evaluation of user resources of the dynamic evaluation model of resource quality.
[0091] The present invention encapsulates the mapping relationship between basic data, operation data, declaration data and quality data, and constructs a dynamic resource quality assessment model, which can evaluate the quality of user resources based on basic data, operation data and declaration data, and automatically calculate them, as follows:
[0092] The construction method of the resource quality dynamic assessment model includes:
[0093] The basic data, operation data and declaration data with characteristic weights are used as input items of BP neural network, and the quality data is used as output item of BP neural network;
[0094] The BP neural network is used to map input items and output items to obtain a dynamic resource quality assessment model;
[0095] The dynamic evaluation model of resource quality is:
[0096] ;
[0097] In the formula, F is the quality data, BP is the BP neural network, X E is the feature weight of E, E is the character identifier of basic data, operating data and declared data, A is the data vector of basic data, B is the data vector of operating data, and C is the data vector of declared data.
[0098] The training loss of the BP neural network is the mean square error between the predicted value and the true value.
[0099] The method also includes using a dynamic resource quality assessment model to perform real-time assessment of user resource quality, wherein the method of using the dynamic resource quality assessment model to perform real-time assessment of user resource quality includes:
[0100] Obtain the basic data, operation data, and declaration data of the virtual power plant at the current moment;
[0101] Calculate the dynamic rate and contribution at the current moment;
[0102] The basic data, operation data, declaration data, dynamic rate and contribution degree at the current moment are input into the dynamic evaluation model of resource quality, and the quality data of user resources at the current moment is output.
[0103] like Figure 2 As shown, the present invention provides a user resource management system for a virtual power plant, which is applied to a user resource management method for a virtual power plant, and the system includes:
[0104] A data collection unit is used to obtain basic data, operation data, declaration data and quality data of user resources of the virtual power plant;
[0105] Data preprocessing unit, used to quantify the basic data, operation data, dynamic rate of declared data of user resources, and contribution to quality data;
[0106] The dynamic rate and contribution are combined to form the characteristic weights of basic data, operating data, and declared data;
[0107] The BP neural network is used to encapsulate the mapping relationship between basic data, operating data, declared data and quality data according to the feature weights, and a resource quality dynamic evaluation model for obtaining quality data based on basic data, operating data and declared data is obtained;
[0108] The quality display unit is used to visualize the quality data obtained by the resource quality dynamic assessment model in charts.
[0109] The basic data in the data collection unit (user name, resource account number, resource name, resource capacity, number of participations, participation duration, monthly adjustment amount, annual adjustment amount, and cumulative adjustment amount) are pre-imported by the management of the virtual power plant;
[0110] The declaration data (user name, resource name, contact person, declaration date and declared planned power) in the data collection unit is imported in real time by the users of the virtual power plant.
[0111] The operating data in the data acquisition unit (real-time load, baseline load, remote regulation, remote control, remote signaling, and telemetry data of resource operation) is obtained and imported for monitoring the virtual power plant.
[0112] The quality display unit uses the resource quality dynamic evaluation model to perform real-time evaluation of user resource quality to obtain quality data (selecting the baseline of time, plan, curve of actual adjustment amount, and cumulative number of participations, cumulative participation duration, cumulative participation qualification rate, cumulative participation rate, annual number of participations, annual participation qualification rate, and annual participation rate). The method of using the resource quality dynamic evaluation model to perform real-time evaluation of user resource quality includes:
[0113] Obtain the basic data, operation data, and declaration data of the virtual power plant at the current moment;
[0114] Calculate the dynamic rate and contribution at the current moment;
[0115] The basic data, operation data, declaration data, dynamic rate and contribution degree at the current moment are input into the dynamic evaluation model of resource quality, and the quality data of user resources at the current moment is output.
[0116] BP (back propagation) neural network is a concept proposed by scientists led by Rumelhart and McClelland. It is a multi-layer feedforward neural network trained according to the error back propagation algorithm and is one of the most widely used neural network models.
[0117] BP neural network has the ability to classify patterns of any complexity and excellent multi-dimensional function mapping ability. It can learn certain rules through its own training and obtain the result closest to the expected output value when the input value is given, that is, determine the mapping relationship between input and output. The BP algorithm includes two processes: forward propagation of signals and reverse propagation of errors. That is, when calculating the error output, it is done from input to output, while adjusting the weights and thresholds from output to input. During forward propagation, the input signal acts on the output node through the hidden layer, and after nonlinear transformation, the output signal is generated. If the actual output does not match the expected output, the error reverse propagation process is entered. Error back propagation is to propagate the output error through the hidden layer to the input layer layer by layer, and distribute the error to all units in each layer, and use the error signal obtained from each layer as the basis for adjusting the weights of each unit. By adjusting the connection strength between the input node and the hidden layer node, the connection strength between the hidden layer node and the output node, and the threshold, the error is reduced along the gradient direction. After repeated learning and training, the network parameters (weights and thresholds) corresponding to the minimum error are determined, and the training stops. At this point, the trained neural network can process the input information of similar samples and obtain the output information by itself after nonlinear transformation with the smallest output error.
[0118] By utilizing the function mapping property of the BP neural network, the present invention applies the BP neural network to the quality assessment process and trains the BP neural network to learn the mapping relationship between quality data and basic data, operation data, and declaration data.
[0119] The present invention constructs a dynamic assessment model for resource quality, and realizes the acquisition of quality data based on basic data, operating data, and declared data. Moreover, the basic data, operating data, and declared data are weighted in the dynamic assessment model for resource quality by weights composed of dynamic rates and contribution degrees, thereby enhancing the accuracy and sensitivity of the dynamic assessment model for resource quality, realizing real-time and accurate grasp of resource data of virtual power plant users, and improving the effect of resource management.
[0120] The above embodiments are only exemplary embodiments of the present application and are not intended to limit the present application. The protection scope of the present application is defined by the claims. Those skilled in the art may make various modifications or equivalent substitutions to the present application within the essence and protection scope of the present application, and such modifications or equivalent substitutions shall also be deemed to fall within the protection scope of the present application.
Claims
1. A user resource management method for a virtual power plant, It is characterized in that The following steps are involved: Obtain basic data, operation data, declaration data and quality data of user resources of virtual power plants; By quantifying the basic data, operation data, dynamic rate of declared data of user resources, and contribution to quality data; The dynamic rate and contribution degree are combined to form the characteristic weights of basic data, operating data and declared data; Using the BP neural network to encapsulate the mapping relationship between the basic data, the operating data, the declared data and the quality data according to the feature weights, a resource quality dynamic evaluation model for obtaining the quality data according to the basic data, the operating data and the declared data is obtained; Methods for quantifying the dynamic rates of basic data, operational data, and declared data of user resources include: Quantify the similarity between the basic data, operation data, and declaration data of the user resources at the current moment and the basic data, operation data, and declaration data of the user resources at the previous moment to obtain the dynamic rate of the basic data, operation data, and declaration data of the user resources; The dynamic rates of basic data, operation data and declaration data of user resources are: ; In the formula, K E is the dynamic rate of E, E is the character identifier of basic data, operating data, and declared data, A is the data vector of basic data, B is the data vector of operating data, C is the data vector of declared data, and E t is the current E, E t-1 is the E of the previous moment, ||E t -E t-1 || 1 is the calculation formula of L1 distance; Methods for quantifying the contribution of basic data, operational data, and declared data of user resources to quality data include: Quantify the mutual information between the basic data, operation data, declaration data and quality data of user resources to obtain the contribution of the basic data, operation data and declaration data of user resources to the quality data; The contribution of basic data, operation data, and declaration data of user resources to quality data is as follows: ; Where, L E is the contribution of E to the quality data, E is the character identifier of basic data, operation data and declaration data, A is the data vector of basic data, B is the data vector of operation data, C is the data vector of declaration data, F is the data vector of quality data, p(E,F) is the joint probability distribution function of E and F, p(E) is the marginal probability distribution function of E, and p(F) is the marginal probability distribution function of F; Methods for combining dynamic rate and contribution to form characteristic weights of basic data, operating data, and declared data include: The dynamic rate and contribution are combined by power and normalized to obtain the normalized combination value of the dynamic rate and contribution. The normalized combination value of the dynamic rate and contribution is: ; Using the normalized combined value of the dynamic rate and contribution, the characteristic weights of the basic data, operating data, and declared data that are positively correlated with the dynamic rate and contribution are obtained. The characteristic weights are: ; Where, X E is the feature weight of E, K E is the dynamic rate of E, L E is the contribution of E to the quality data, E is the character identifier of basic data, operation data and declaration data, A is the data vector of basic data, B is the data vector of operation data, C is the data vector of declaration data, and F is the data vector of quality data; The basic data includes user name, resource account number, resource name, resource capacity, number of participations, participation duration, monthly adjustment amount, annual adjustment amount and cumulative adjustment amount; Operational data include real-time load, baseline load, remote regulation, remote control, remote signaling, and telemetry data of resource operation; The declaration data includes user name, resource name, contact person, declaration date, and declaration plan power; Quality data include: baseline, plan, curve of actual adjustment amount, as well as cumulative number of participations, cumulative participation duration, cumulative participation qualification rate, cumulative participation rate, annual number of participations, annual participation qualification rate, and annual participation rate; The method for constructing the resource quality dynamic assessment model includes: The basic data, operation data and declaration data with characteristic weights are used as input items of BP neural network, and the quality data is used as output item of BP neural network; The BP neural network is used to map input items and output items to obtain a dynamic resource quality assessment model; The resource quality dynamic assessment model is: ; In the formula, F is the quality data, BP is the BP neural network, X E is the feature weight of E, E is the character identifier of basic data, operating data and declared data, A is the data vector of basic data, B is the data vector of operating data, and C is the data vector of declared data.
2. A user resource management method for a virtual power plant according to claim 1, Features: The training loss of the BP neural network is the mean square error between the predicted value and the true value.
3. A user resource management method for a virtual power plant according to claim 2, Features: The method also includes using a dynamic resource quality assessment model to perform real-time assessment of user resource quality, wherein the method of using the dynamic resource quality assessment model to perform real-time assessment of user resource quality includes: Obtain the basic data, operation data, and declaration data of the virtual power plant at the current moment; Calculating the dynamic rate and contribution at the current moment; The basic data, operation data, declaration data at the current moment and the dynamic rate and contribution degree at the current moment are input into the resource quality dynamic evaluation model, and the quality data of the user resources at the current moment is output.
4. A user resource management system for a virtual power plant, It is characterized in that A user resource management method for a virtual power plant according to any one of claims 1 to 3, the system comprising: A data collection unit is used to obtain basic data, operation data, declaration data and quality data of user resources of the virtual power plant; Data preprocessing unit, used to quantify the basic data, operation data, dynamic rate of declared data of user resources, and contribution to quality data; The dynamic rate and contribution degree are combined to form the characteristic weights of basic data, operating data and declared data; Using the BP neural network to encapsulate the mapping relationship between the basic data, the operating data, the declared data and the quality data according to the feature weights, a resource quality dynamic evaluation model for obtaining the quality data according to the basic data, the operating data and the declared data is obtained; The quality display unit is used to visualize the quality data obtained by the resource quality dynamic assessment model in charts.
5. A user resource management system for a virtual power plant according to claim 4, Features: The quality display unit uses the resource quality dynamic evaluation model to perform real-time evaluation of user resource quality to obtain the quality data, wherein the method of using the resource quality dynamic evaluation model to perform real-time evaluation of user resource quality includes: Obtain the basic data, operation data, and declaration data of the virtual power plant at the current moment; Calculating the dynamic rate and contribution at the current moment; The basic data, operation data, declaration data at the current moment and the dynamic rate and contribution degree at the current moment are input into the resource quality dynamic evaluation model, and the quality data of the user resources at the current moment is output.
