A dual-model driven virtual power plant power resource disposal method

Abstract

The present invention discloses a dual-model-driven method for handling power resources in a virtual power plant. This method decouples energy resource data from specific handling solutions, uses an energy resource model to determine the current real-time dispatch capability, then inputs the real-time dispatch capability and real-time operating status data into the energy handling model to output a real-time handling solution for power energy dispatch. This method avoids the black box problem of a single model, allowing for a clear understanding of changes in dispatch capability caused by changes in energy resource data. This allows the energy handling model to make judgments to ensure the feasibility of the handling solution, and allows for intuitive verification of the resource dispatch capability judgment results. Even if errors are found, the energy resource model can be individually optimized.

Description

Technical Field

[0001] The present invention relates to the field of power dispatching, and in particular to a dual-model driven virtual power plant power resource disposal method. Background Art

[0002] At present, the power resource disposal of virtual power plants usually utilizes various neural network models, and the relevant data involved include energy resource data, operating status data, etc. Although operating status data is the main basis for virtual power plants to dispose of power energy, the disposal plan is based on energy resource data. Moreover, during the model training process, the logic of generating disposal plans is also based on energy resource data. Therefore, when the energy resource data changes, it is equivalent to a change in the basic composition of the virtual power plant. At this time, it may be difficult to cope with it by relying solely on the generalization ability of the original model, and it is easy to cause unreasonable power scheduling of the disposal plan.

[0003] Therefore, if changes in energy resource data alter some of the virtual power plant's original properties, these changes may cause the model's performance to degrade, making it no longer accurately reflect the current energy resource situation. Under existing technology frameworks, the model must be retrained to maintain the expected accuracy and adaptability. Retraining requires significant computing resources and time.

[0004] In summary, when energy resource data changes, depending on the nature and magnitude of the change and the characteristics of the model itself, in order to maintain the original accuracy, it is usually necessary to retrain the model, which is costly and inefficient. Summary of the Invention

[0005] In response to the problem in the prior art that changes in the energy resource data corresponding to the virtual power plant may invalidate the original model and lead to unreasonable power resource disposal, the present invention provides a dual-model driven virtual power plant power resource disposal method, which decouples energy resource data from specific disposal plans, uses the energy resource model to determine the current real-time dispatch capability, and then inputs the real-time dispatch capability and real-time operating status data into the energy disposal model to output a real-time disposal plan for power energy dispatch. The present invention avoids the black box problem of a single model, can clearly understand the changes in dispatch capability caused by changes in energy resource data, and can be used by the energy disposal model to make judgments to ensure the feasibility of the disposal plan. At the same time, the resource dispatch capability judgment results can be intuitively verified, and even if there are errors, the energy resource model can be optimized separately.

[0006] The following are the technical solutions of the present invention.

[0007] A dual-model driven virtual power plant power resource disposal method includes the following steps:

[0008] S1: Obtain real-time operating status data and real-time energy resource data of the virtual power plant;

[0009] S2: Inputting real-time energy resource data into a machine learning-based energy resource model trained with historical energy resource data and corresponding historical dispatch capabilities to obtain the real-time dispatch capabilities of the virtual power plant;

[0010] S3: Input the real-time dispatch capability and real-time operating status data into the energy disposal model based on deep learning trained by historical operating status data, historical dispatch capability, and historical disposal plans, and output the real-time disposal plan of the virtual power plant.

[0011] In this invention, the training of the energy disposal model is decoupled from energy resource data, eliminating the need for a direct correlation between the final disposal solution and the energy resource data. Instead, the energy resource model receives the energy resource data and converts it into dispatch capabilities readable by the energy disposal model. This eliminates the direct impact of changes in energy resource data on the energy disposal model, instead buffering and converting it. This allows the energy disposal model to focus more on the connection between dispatch capabilities and disposal solutions, rather than focusing too much on energy resource data, which can lead to problems with existing technologies.

[0012] Therefore, the present invention uses the energy resource model to output real-time scheduling capabilities based on real-time energy resource data, which can intuitively show the changes in the scheduling capabilities of the virtual power plant, and also serves as one of the input conditions of the energy disposal model. Since the energy disposal model does not directly use energy resource data during training, but uses historical operating status data, historical scheduling capabilities, and historical disposal plans, the impact of energy resource data is isolated in the energy resource model. As long as the output of the energy resource model is reliable, the output of the disposal plan can be guaranteed to be unaffected.

[0013] In the present invention, the changes in energy resource data are judged by the energy resource model, so the intermediate process that has the greatest impact on the disposal plan is no longer a black box. According to the results, it can be clearly understood whether there is a problem in the link of judging the scheduling capability or in the link of generating the disposal plan. Therefore, even if an error occurs, it is no longer necessary to retrain all of them, thereby improving execution efficiency and accuracy.

[0014] In addition, the machine learning algorithm in the present invention is more suitable for the first half of the disposal solution generation process, that is, the energy resource model of this application, due to its strong interpretability, strong ability to process discrete data, high computational efficiency, and easy tuning. The deep learning algorithm, on the other hand, is more suitable for the second half of the disposal solution generation process, that is, the energy disposal model of this application, due to its strong nonlinear modeling ability, strong self-learning ability, strong generalization ability, and ability to capture long-term dependencies. This combination can fully leverage the advantages of each algorithm and improve the efficiency and accuracy of the virtual power plant's power resource disposal.

[0015] Preferably, in S2 and S3, the steps of acquiring training data required for the energy resource model and the energy disposal model include: A1: acquiring the operation records of the virtual power plant, extracting historical disposal plans from the operation records, and associating historical energy resource data and historical operation status data corresponding to each historical disposal plan;

[0016] A2: Determine the historical dispatch capacity corresponding to each energy resource based on historical disposal plans and historical energy resource data.

[0017] Preferably, the historical energy resource data includes:

[0018] The number, types, and operating parameters of distributed power generation facilities; the number, types, and operating parameters of distributed energy storage systems; the number, types, and operating parameters of controllable loads.

[0019] Preferably, A1: obtaining the operation records of the virtual power plant, extracting historical disposal plans from the operation records, and associating historical energy resource data and historical operation status data corresponding to each historical disposal plan, includes:

[0020] Obtain operation records from the virtual power plant management system using API interfaces or data capture;

[0021] Use data parsing algorithms to extract historical disposal plans from acquired operation records;

[0022] According to the time of the historical disposal plan, the historical energy resource data and historical operation status data corresponding to the time are determined, and associated with the historical disposal plan. The associated data is verified and cleaned, and the associated data is stored in a relational database.

[0023] Preferably, A2: determining the historical dispatch capacity corresponding to each energy resource based on historical disposal plans and historical energy resource data, includes:

[0024] Statistics and calculations are performed on the scheduling performance of each energy resource under different historical disposal plans to obtain the historical scheduling priority and historical scheduling value of each energy resource;

[0025] Statistics and calculations are made on the dispatch performance of different energy resources in different historical disposal plans to obtain the complementarity and substitutability between different energy resources and establish the complementary or substitutable relationship between different energy resources;

[0026] Based on the historical scheduling priority and historical scheduling value of each energy resource, as well as the complementary or substitution relationship between different energy resources, a structured scheduling information database representing the historical scheduling capabilities is constructed.

[0027] In this invention, by statistically analyzing and calculating the performance of each energy resource under different historical disposal scenarios, we can determine the frequency and value of the energy resource's historical scheduling, and then determine the historical scheduling priority and historical scheduling value, providing data support for subsequent resource allocation and scheduling strategies. At the same time, by statistically analyzing and calculating the performance of different energy resources under different historical disposal scenarios, we can determine the complementarity and substitutability between them. Complementarity refers to the fact that certain energy resources can enhance each other during scheduling, while substitutability refers to the fact that in certain circumstances, one energy resource can replace another to meet energy demand.

[0028] Preferably, the calculation method of the historical scheduling priority includes:

[0029] Based on the statistical results of the scheduling performance of each energy resource under different historical disposal plans, the historical scheduling times n of each energy resource are obtained. According to the number of historical disposal plans m, Y = n / m is calculated to obtain the historical scheduling rate Y, where the higher the historical scheduling rate Y, the higher the scheduling priority.

[0030] Preferably, the training process of the energy resource model includes:

[0031] A primary energy resource model is established based on a decision tree or random forest algorithm, and the primary energy resource model is trained using historical energy resource data and corresponding historical scheduling capabilities to obtain a trained energy resource model.

[0032] Preferably, the training process of the energy disposal model includes:

[0033] A primary energy disposal model is established based on the long short-term memory network algorithm, and the primary energy disposal model is trained using historical operating status data, historical scheduling capabilities, and historical disposal plans to obtain a trained energy disposal model.

[0034] Preferably, the S2 further includes:

[0035] According to the data dimension of the real-time scheduling capability, multi-dimensional vector parameters are set, and the real-time scheduling capability is recorded in the form of multi-dimensional vector parameters;

[0036] The cosine similarity of the multidimensional vector parameters before and after is calculated. If the cosine similarity is less than the preset similarity, the energy resource model is updated.

[0037] The present invention also provides a dual-model driven virtual power plant power resource disposal system. When the dual-model driven virtual power plant power resource disposal system is in operation, the dual-model driven virtual power plant power resource disposal method described above is executed.

[0038] The present invention also provides an electronic device comprising a memory and a processor, wherein the memory stores a computer program, and when the processor calls the computer program in the memory, it implements the steps of the above-mentioned dual-model driven virtual power plant power resource disposal method.

[0039] The present invention also provides a storage medium, in which computer executable instructions are stored. When the computer executable instructions are loaded and executed by a processor, the steps of the above-mentioned dual-model driven virtual power plant power resource disposal method are implemented.

[0040] The substantial effects of the present invention include:

[0041] By introducing a dual-model architecture of an energy resource model and an energy disposal model, the present invention achieves the decoupling of energy resource data from specific disposal solutions. This design avoids the black box problem of a single model, making the impact of changes in energy resource data on the disposal solution controllable and predictable, thereby improving the stability and reliability of the system. The machine learning algorithm in the present invention is more suitable for the first half of the disposal solution generation process, i.e., the energy resource model of the present application, due to its strong interpretability, strong ability to process discrete data, high computational efficiency, and easy optimization. The deep learning algorithm is more suitable for the second half of the disposal solution generation process, i.e., the energy disposal model of the present application, due to its advantages such as strong nonlinear modeling ability, strong self-learning ability, strong generalization ability, and the ability to capture long-term dependencies. This combination can give full play to the advantages of each algorithm and improve the efficiency and accuracy of the virtual power plant's power resource disposal.

[0042] This invention utilizes an energy resource model to process real-time energy resource data and output real-time dispatch capabilities, providing accurate and reliable input for the energy disposal model. Because the energy disposal model doesn't directly utilize energy resource data during training, but instead uses historical operating status data, historical dispatch capabilities, and historical disposal plans, it effectively isolates the direct impact of energy resource data on the disposal model, allowing the disposal model to focus more on the connection between dispatch capabilities and disposal plans.

[0043] In addition, by counting and calculating the performance of each energy resource under different historical disposal schemes, the present invention obtains the historical scheduling priority and historical scheduling value of the energy resource, providing strong data support for subsequent resource allocation and scheduling strategies. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 is a flow chart of an embodiment of the present invention. DETAILED DESCRIPTION

[0045] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions will be clearly and completely described below in conjunction with the embodiments. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0046] It should be understood that in various embodiments of the present invention, the size of the sequence number of each process does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0047] It should be understood that in the present invention, "include" and "have" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or apparatus that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to these processes, methods, products or apparatuses.

[0048] It should be understood that in the present invention, "multiple" refers to two or more. "And / or" is only a description of the association relationship of associated objects, indicating that three relationships can 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 previous and next associated objects are in an "or" relationship. "Contains A, B and C", "Contains A, B, C" means that A, B, and C are all included, "Contains A, B or C" means that one of A, B, and C is included, and "Contains A, B and / or C" means that any one, any two, or any three of A, B, and C are included.

[0049] The technical solution of the present invention is described in detail below with reference to specific embodiments. The embodiments may be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0050] Example 1:

[0051] A dual-model driven virtual power plant power resource disposal method, such as Figure 1 As shown, the following steps are included:

[0052] S1: Obtain real-time operating status data and real-time energy resource data of the virtual power plant;

[0053] S2: Inputting real-time energy resource data into a machine learning-based energy resource model trained with historical energy resource data and corresponding historical dispatch capabilities to obtain the real-time dispatch capabilities of the virtual power plant;

[0054] S3: Input the real-time dispatch capability and real-time operating status data into the energy disposal model based on deep learning trained by historical operating status data, historical dispatch capability, and historical disposal plans, and output the real-time disposal plan of the virtual power plant.

[0055] In this embodiment, the training of the energy disposal model is decoupled from the energy resource data, eliminating the need for a direct correlation between the final disposal plan and the energy resource data. Instead, the energy resource model receives the energy resource data and converts it into dispatch capabilities readable by the energy disposal model. This eliminates the direct impact of changes in energy resource data on the energy disposal model, instead buffering and converting the data. This allows the energy disposal model to focus more on the connection between dispatch capabilities and disposal plans, rather than focusing too much on energy resource data, which can lead to problems with existing technologies.

[0056] Therefore, this embodiment uses the energy resource model to output real-time scheduling capabilities based on real-time energy resource data, so that changes in the scheduling capabilities of the virtual power plant can be intuitively seen, and it also serves as one of the input conditions of the energy disposal model. Since the energy disposal model does not directly use energy resource data during training, but uses historical operating status data, historical scheduling capabilities, and historical disposal plans, the impact of energy resource data is isolated in the energy resource model. As long as the output of the energy resource model is reliable, the output of the disposal plan can be guaranteed to be unaffected.

[0057] In this embodiment, the changes in energy resource data are judged by the energy resource model. Therefore, the intermediate process that has the greatest impact on the disposal plan is no longer a black box. According to the results, it can be clearly understood whether there is a problem in the link of judging the scheduling capability or in the link of generating the disposal plan. Therefore, even if an error occurs, it is no longer necessary to retrain all of them, thereby improving execution efficiency and accuracy.

[0058] The energy resource model and energy disposal model mentioned in S2 and S3 are important aspects of implementing this embodiment. The steps for acquiring training data for them include:

[0059] A1: Obtain the operation records of the virtual power plant, extract the historical disposal plans in the operation records, and associate the historical energy resource data and historical operation status data corresponding to each historical disposal plan, including:

[0060] Obtain operation records from the virtual power plant management system using an API or data capture method. For example, obtain the virtual power plant's operation records for the past year in real time through the API, including daily power generation, energy storage, and load data.

[0061] Utilize data parsing algorithms to extract historical disposal plans from acquired operation records. For example, operation records may contain multiple daily or emergency dispatch plans that were actually implemented. Data parsing algorithms can identify and extract these emergency dispatch plans as historical disposal plans.

[0062] Based on the timing of historical resolution plans, the corresponding historical energy resource data and historical operating status data are determined and associated with the historical resolution plans. The associated data is then verified and cleaned, and stored in a relational database. For example, an emergency dispatch plan implemented due to a sudden drop in wind power generation requires associating relevant information such as wind power generation data at that point in time, the energy storage system's charge and discharge status data, and grid load data.

[0063] In this embodiment, historical energy resource data includes: the number, type, and operating parameters of distributed generation facilities; the number, type, and operating parameters of distributed energy storage systems; and the number, type, and operating parameters of controllable loads. For example, a virtual power plant might include 10 wind turbines, 5 photovoltaic power generation facilities, 5 energy storage stations, and 11 industrial production lines. The number and type (e.g., wind or photovoltaic) and operating parameters (e.g., rated power, generation efficiency, etc.) of each facility are recorded in detail.

[0064] A2: Determine the historical dispatch capacity corresponding to each energy resource based on historical disposal plans and historical energy resource data, including:

[0065] Statistics and calculations are performed on the scheduling performance of each energy resource under different historical disposal plans to obtain the historical scheduling priority and historical scheduling value of each energy resource;

[0066] Statistics and calculations are made on the dispatch performance of different energy resources in different historical disposal plans to obtain the complementarity and substitutability between different energy resources and establish the complementary or substitutable relationship between different energy resources;

[0067] Based on the historical scheduling priority and historical scheduling value of each energy resource, as well as the complementary or substitution relationship between different energy resources, a structured scheduling information database representing the historical scheduling capabilities is constructed.

[0068] The calculation method of historical scheduling priority includes:

[0069] Based on the statistical results of the scheduling performance of each energy resource under different historical disposal plans, the historical scheduling times n of each energy resource are obtained. According to the number of historical disposal plans m, Y = n / m is calculated to obtain the historical scheduling rate Y, where the higher the historical scheduling rate Y, the higher the scheduling priority.

[0070] For example, a wind power facility was dispatched five times under different historical dispatch plans, a photovoltaic power facility was dispatched three times, and a battery energy storage facility was dispatched twice to supplement power. If the total number of historical dispatch plans, m, is 20, then the historical dispatch rate for the wind power facility, Y, is 5 / 20 = 0.25, the historical dispatch rate for the photovoltaic power facility, Y, is 3 / 20 = 0.15, and the historical dispatch rate for the battery energy storage facility, Y, is 2 / 20 = 0.1. Clearly, the wind power facility has the highest historical dispatch priority. The historical dispatch value can be understood as the total dispatch capacity of each facility, for example, 60 MWh.

[0071] For example, when statistics and calculations were performed on the dispatch performance of different energy resources under different historical disposal plans, it was found that when wind power output was insufficient, battery energy storage was dispatched more frequently to meet demand, demonstrating the complementarity between wind power and battery energy storage. PV power output was higher during the day, while wind power output was higher at night, complementing each other at different times, also demonstrating complementarity. Furthermore, to meet electricity demand, a certain diesel generator set was dispatched more frequently to replace photovoltaic and wind power facilities in the area, filling the power gap caused by insufficient photovoltaic and wind power generation. At this point, the diesel generator set and other power generation facilities became substitutable.

[0072] The completed structured dispatch information database includes at least parameters representing the regulation capabilities of different power generation facilities, such as maximum output (MW), average output (MW), response time (seconds), and regulation accuracy (%). It also identifies the complementarity and substitutability of different facilities. Parameters like maximum output can often be directly derived from energy resource data, while other parameters require calculations based on large amounts of data using energy resource models based on the actual operating conditions of different facilities.

[0073] In this embodiment, by statistically analyzing and calculating the performance of each energy resource under different historical disposal scenarios, we can determine the frequency and value of the energy resource's historical scheduling, thereby determining its historical scheduling priority and historical scheduling value, providing data support for subsequent resource allocation and scheduling strategies. Furthermore, by statistically analyzing and calculating the performance of different energy resources under different historical disposal scenarios, we can determine their complementarity and substitutability. Complementarity refers to the fact that certain energy resources can enhance each other during scheduling, while substitutability refers to the fact that, in certain circumstances, one energy resource can replace another to meet energy demand.

[0074] On this basis, the training process of the energy resource model includes:

[0075] A primary energy resource model is established based on a decision tree or random forest algorithm, and the primary energy resource model is trained using historical energy resource data and corresponding historical scheduling capabilities to obtain a trained energy resource model.

[0076] The decision tree model builds a model by recursively partitioning the data feature space. Each node represents a decision point for a feature variable, and the leaf node represents the prediction result. The random forest model improves the accuracy and stability of the model by constructing multiple decision trees and combining their prediction results.

[0077] The decision tree algorithm uses a tree structure to represent the decision-making process. Each internal node represents a judgment on an attribute, each branch represents the output of a judgment result, and each leaf node represents a category. This structure makes the relationship between energy resource data and dispatch capability very intuitive, easy to understand and verify. Among decision tree algorithms, commonly used algorithms include ID3, C4.5, CART, etc. In the field of virtual power plants, since continuous data and missing value processing may be involved, C4.5 or CART algorithms can generally be used first. The decision tree model has good interpretability. The structure of the decision tree can be viewed through visualization tools to understand the decision-making process of the model. For example, by viewing the decision tree, you can clearly understand how the type and number of a power station affect the dispatch capability.

[0078] Then, characteristic variables are extracted from historical energy resource data. These characteristic variables should be able to reflect the output characteristics and availability of energy resources. Characteristic variables may include the type, quantity, connectivity, average output, maximum output, minimum output, output volatility, etc. of energy resources.

[0079] During training, the model learns the mapping between characteristic variables and historical dispatch capabilities—in other words, how to predict dispatch capabilities based on energy resource characteristics. The trained energy resource model is then evaluated using methods such as cross-validation to check its predictive performance. Based on the evaluation results, the model is fine-tuned, such as adjusting parameters like the depth of the decision tree and the number of trees in the random forest, to improve model accuracy and generalization.

[0080] In this embodiment, the machine algorithm can provide a more intuitive decision-making path, making the relationship between energy resource data and scheduling capabilities more transparent, easy to understand and verify. For discrete data such as the number of power stations and the type of energy storage stations, the machine learning algorithm can better capture the laws behind them, thereby accurately judging the scheduling capabilities. Compared with complex algorithms such as deep learning, machine learning algorithms usually have higher computing efficiency during training and prediction, and are suitable for scenarios with high real-time requirements. In addition, the machine learning algorithm has relatively few parameters and the tuning process is relatively simple, and the model performance can be quickly adjusted according to actual needs. Therefore, the machine learning algorithm is more suitable for the first half of the disposal plan generation process, that is, the energy resource model of this application.

[0081] In addition, the training process of the energy disposal model includes:

[0082] A primary energy disposal model is established based on the long short-term memory network algorithm, and the primary energy disposal model is trained using historical operating status data, historical scheduling capabilities, and historical disposal plans to obtain a trained energy disposal model.

[0083] Before training, the LSTM model structure must be designed, including an input layer, hidden layers (one or more LSTM layers), and an output layer. The input layer receives preprocessed feature data, the hidden layer captures long-term dependencies in time series data through LSTM units, and the output layer generates predictions. Parameter initialization is then performed: the weights and biases of the LSTM model are randomly initialized to prepare for the training process. An appropriate loss function (such as mean squared error (MSE)) is selected to evaluate the difference between the model's predictions and the actual values, and an appropriate optimizer (such as Adam) is selected to update the model parameters to minimize the loss function. The LSTM model is then iteratively trained using the training set data. In each iteration, the feature data is input into the model, the loss between the predictions and the actual values ​​is calculated, and the model parameters are then updated using the backpropagation algorithm. Performance on the validation set is monitored to assess the model's generalization ability and avoid overfitting.

[0084] Unlike the machine learning algorithms used in energy resource models, deep learning algorithms are good at processing complex nonlinear relationships and can capture the deep connections between operating status data, scheduling capacity data, etc., so as to formulate more accurate disposal plans. Through a large amount of training data, deep learning algorithms can automatically learn and optimize internal parameters and adapt to different operating scenarios without human intervention. The fully trained deep learning model has a strong generalization ability and can maintain stable performance output when facing new operating states. For time series data (such as load change trends), deep learning algorithms such as LSTM can capture the long-term dependencies between data and provide strong support for the formulation of long-term and effective disposal plans. Therefore, deep learning algorithms are more suitable for the second half of the disposal plan generation process, that is, the energy disposal model of this application.

[0085] Specifically, the LSTM model can capture the complex nonlinear relationships between operating status data, dispatch capacity data, and other data, enabling more precise response plans. It also automatically adjusts network weights through backpropagation and gradient descent, achieving self-learning. During training, the LSTM continuously optimizes its internal parameters to adapt to different operating scenarios and data distributions. Therefore, a fully trained LSTM model has strong generalization capabilities and can maintain stable performance output when faced with new operating conditions. This enables it to maintain reliable response capabilities in the complex and changing virtual power plant environment. Furthermore, for time series data (such as load trends and the charge and discharge cycles of energy storage devices), the LSTM can capture long-term dependencies between data. This is crucial for developing response plans that consider both historical data and future trends. For example, when predicting future load changes, the LSTM can comprehensively consider data from past days or even weeks to make more accurate predictions.

[0086] In other words, machine learning algorithms are suitable for building energy resource models due to their strong interpretability, ability to process discrete data, high computational efficiency, and ease of tuning. Deep learning algorithms, on the other hand, are more suitable for building energy disposal models due to their strong nonlinear modeling capabilities, self-learning capabilities, strong generalization capabilities, and ability to capture long-term dependencies. This combination leverages the strengths of each algorithm to improve the efficiency and accuracy of virtual power plant power resource disposal.

[0087] As a supplement to Example 1, S2 may further include:

[0088] According to the data dimension of the real-time scheduling capability, multi-dimensional vector parameters are set, and the real-time scheduling capability is recorded in the form of multi-dimensional vector parameters;

[0089] The cosine similarity of the multidimensional vector parameters before and after is calculated. If the cosine similarity is less than the preset similarity, the energy resource model is updated.

[0090] In this embodiment, the energy resource model is updated using incremental learning based on knowledge distillation, which includes the following steps: using the original source resource model as the first teacher model, constructing and training a second teacher model using incremental energy resource data and corresponding scheduling capabilities;

[0091] Use the first teacher model to predict the old data, and use the second teacher model to predict the new data, and generate corresponding soft labels respectively;

[0092] Build a student model and set a distillation loss function to measure the difference between the student model output and the soft labels of the two teacher models. For example, the soft labels of the two teacher models are weighted averaged as the target that the student model should imitate.

[0093] Based on the distillation loss function, the soft labels of the two teacher models are used to guide the learning direction of the student model, and the parameters of the student model are updated to obtain the updated energy resource model.

[0094] This approach preserves the parameters of the old model, allowing for replacement or rollback when necessary. Compared to direct retraining, it offers greater controllability and transparency, making it easier to identify and adjust existing issues.

[0095] Example 2:

[0096] A dual-model driven virtual power plant power resource handling system, when the dual-model driven virtual power plant power resource handling system is running, executes a dual-model driven virtual power plant power resource handling method of embodiment 1.

[0097] Example 3:

[0098] An electronic device includes a memory and a processor, wherein a computer program is stored in the memory, and when the processor calls the computer program in the memory, the steps of the above-mentioned dual-model driven virtual power plant power resource disposal method are implemented.

[0099] Example 4:

[0100] A storage medium stores computer-executable instructions, which, when loaded and executed by a processor, implement the steps of the above-mentioned dual-model driven virtual power plant power resource disposal method.

[0101] In summary, the substantial effects of the above embodiments include:

[0102] By introducing a dual-model architecture consisting of an energy resource model and an energy disposal model, this embodiment decouples energy resource data from specific disposal solutions. This design avoids the black box problem inherent in a single model, making the impact of changes in energy resource data on disposal solutions controllable and predictable, thereby improving system stability and reliability.

[0103] This embodiment utilizes an energy resource model to process real-time energy resource data and output real-time dispatch capabilities, providing accurate and reliable input for the energy disposal model. Because the energy disposal model doesn't directly utilize energy resource data during training, but instead uses historical operating status data, historical dispatch capabilities, and historical disposal solutions, it effectively isolates the direct impact of energy resource data on the disposal model, allowing the disposal model to focus more on the connection between dispatch capabilities and disposal solutions.

[0104] Furthermore, this embodiment provides a rich set of training data acquisition steps and model training processes, ensuring that the energy resource model and energy disposal model accurately reflect the actual operation and dispatch capabilities of the virtual power plant. By statistically analyzing and calculating the performance of each energy resource under different historical disposal scenarios, this embodiment derives the historical dispatch priority and historical dispatch value of each energy resource, providing strong data support for subsequent resource allocation and scheduling strategies.

[0105] Through the description of the above implementation methods, technical personnel in the relevant field can understand that for the convenience and simplicity of description, only the division of the above-mentioned functional modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional modules as needed, that is, the internal structure of the specific device can be divided into different functional modules to complete all or part of the functions described above.

[0106] In the embodiments provided in this application, it should be understood that the disclosed structures and methods can be implemented in other ways. For example, the embodiments of the structure described above are merely illustrative. For example, the division of modules or units is only a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another structure, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interface, structure or unit, which can be electrical, mechanical or other forms.

[0107] Units described as separate components may or may not be physically separate, and components shown as units may be one physical unit or multiple physical units, that is, they may be located in one place or distributed in multiple places. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0108] In addition, the functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated units may be implemented in the form of hardware or software functional units.

[0109] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a readable storage medium. Based on this understanding, the technical solution of the embodiment of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for enabling a device (which can be a single-chip microcomputer, chip, etc.) or a processor (processor) to execute all or part of the steps of the various embodiments of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0110] The above content is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. A dual-model driven virtual power plant power resource disposal method, characterized in that: The following steps are involved: S1: Obtain real-time operating status data and real-time energy resource data of the virtual power plant; S2: Inputting real-time energy resource data into a machine learning-based energy resource model trained with historical energy resource data and corresponding historical dispatch capabilities to obtain the real-time dispatch capabilities of the virtual power plant; The historical scheduling capability is represented by a structured scheduling information database, which is constructed based on the historical scheduling priority and historical scheduling value of each energy resource, as well as the complementary or substitution relationship between different energy resources. The energy resource model is updated using incremental learning based on knowledge distillation. S2 also includes: According to the data dimension of the real-time scheduling capability, multi-dimensional vector parameters are set, and the real-time scheduling capability is recorded in the form of multi-dimensional vector parameters; Calculating the cosine similarity of the multidimensional vector parameters at the previous and next moments. If the cosine similarity is less than a preset similarity, updating the energy resource model. The energy resource model training process includes: establishing a primary energy resource model based on a decision tree or random forest algorithm, and training the primary energy resource model using historical energy resource data and corresponding historical scheduling capabilities to obtain a trained energy resource model. S3: Input the real-time dispatching capability and real-time operating status data into the deep learning-based energy disposal model trained by historical operating status data, historical dispatching capability, and historical disposal plans, and output the real-time disposal plan of the virtual power plant; the training process of the energy disposal model includes: establishing a primary energy disposal model based on the long short-term memory network algorithm, and training the primary energy disposal model using historical operating status data, historical dispatching capability, and historical disposal plans to obtain a trained energy disposal model.

2. A dual-model driven virtual power plant power resource disposal method according to claim 1, characterized in that: In S2 and S3, the steps for obtaining training data required for the energy resource model and energy disposal model include: A1: Obtain the operation records of the virtual power plant, extract the historical disposal plans in the operation records, and associate the historical energy resource data and historical operation status data corresponding to each historical disposal plan; A2: Determine the historical dispatch capacity corresponding to each energy resource based on historical disposal plans and historical energy resource data.

3. The dual-model driven virtual power plant power resource disposal method according to claim 1, characterized in that: The historical energy resource data includes: The number, types, and operating parameters of distributed power generation facilities; the number, types, and operating parameters of distributed energy storage systems; the number, types, and operating parameters of controllable loads.

4. The dual-model driven virtual power plant power resource disposal method according to claim 2, characterized in that: A1: Obtain the operation records of the virtual power plant, extract the historical disposal plans in the operation records, and associate the historical energy resource data and historical operation status data corresponding to each historical disposal plan, including: Obtain operation records from the virtual power plant management system using API interfaces or data capture; Use data parsing algorithms to extract historical disposal plans from acquired operation records; According to the time of the historical disposal plan, the historical energy resource data and historical operation status data corresponding to the time are determined, and associated with the historical disposal plan. The associated data is verified and cleaned, and the associated data is stored in a relational database.

5. The dual-model driven virtual power plant power resource disposal method according to claim 2, characterized in that: A2: Determine the historical dispatch capacity corresponding to each energy resource based on historical disposal plans and historical energy resource data, including: Statistics and calculations are performed on the scheduling performance of each energy resource under different historical disposal plans to obtain the historical scheduling priority and historical scheduling value of each energy resource; Statistics and calculations are made on the dispatch performance of different energy resources in different historical disposal plans to obtain the complementarity and substitutability between different energy resources and establish the complementary or substitutable relationship between different energy resources; Based on the historical scheduling priority and historical scheduling value of each energy resource, as well as the complementary or substitution relationship between different energy resources, a structured scheduling information database representing the historical scheduling capabilities is constructed.

6. A dual-model driven virtual power plant power resource disposal method according to claim 5, characterized in that: The calculation method of the historical scheduling priority includes: Based on the statistical results of the scheduling performance of each energy resource under different historical disposal plans, the historical scheduling number n of each energy resource is obtained. According to the number of historical disposal plans m, the historical scheduling rate Y is calculated as Y=n / m. Among them, the higher the historical scheduling rate Y, the higher the scheduling priority.

7. An electronic device, characterized in that: It includes a memory and a processor, wherein a computer program is stored in the memory, and when the processor calls the computer program in the memory, it implements the steps of a dual-model driven virtual power plant power resource disposal method as described in any one of claims 1 to 6.

8. A storage medium, characterized in that: The storage medium stores computer-executable instructions, which, when loaded and executed by the processor, implement the steps of a dual-model driven virtual power plant power resource disposal method as described in any one of claims 1 to 6.

Images