Virtual power plant power coordination control method and device based on multi-source data

Abstract

Embodiments of the present disclosure disclose a virtual power plant power coordination control method and device based on multi-source data. A specific embodiment of the method comprises: based on the physical resources corresponding to the multi-source data, constructing corresponding virtual resource models in the virtual power plant; determining power adjustment range information and power change information using a preset device physical property response model; determining corresponding power response characteristic information based on the power change information; performing power power control processing on the power response characteristic information based on the power adjustment range information to obtain a set of coordination control instructions; for each coordination control instruction in the set of coordination control instructions: distributing the coordination control instruction to the corresponding edge control device to perform security verification to obtain a verification result; and executing the coordination control instruction to control the target power device. This embodiment can improve the rationality of power distribution, enhance the stability of the power grid, and reduce the risk of device overload.

Description

Technical Field

[0001] The embodiments disclosed herein relate to the field of computer technology, and more specifically to a method and apparatus for virtual power plant power collaborative control based on multi-source data. Background Technology

[0002] Currently, the proportion of distributed energy in the power distribution network is gradually increasing, and using virtual power plants to aggregate various distributed energy sources and control power has become the main approach.

[0003] The common approach to power control in virtual power plants is to use distributed algorithms to uniformly schedule aggregated resources (e.g., loads, energy storage, and distributed power sources).

[0004] However, when using the above method for power control, the following technical problems often arise:

[0005] Treating aggregated resources as a whole fails to adequately account for the differences in physical characteristics, response speeds, and regulation costs among different types of resources (e.g., industrial and commercial loads, energy storage systems, and electric vehicles). This leads to a mismatch between control commands and the actual regulation capacity of the resources, resulting in regulation deviations and resource overload.

[0006] The information disclosed in this background section is only intended to enhance the understanding of the background of the inventive concept, and therefore may contain information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0007] The summary portion of this disclosure is intended to provide a brief overview of the concepts, which will be described in detail in the detailed description portion. This summary portion is not intended to identify key or essential features of the claimed technical solutions, nor is it intended to limit the scope of the claimed technical solutions.

[0008] Some embodiments of this disclosure propose a method, apparatus, electronic device, and computer-readable medium for coordinated power control of virtual power plants based on multi-source data, in order to solve one or more of the technical problems mentioned in the background section above.

[0009] In a first aspect, some embodiments of this disclosure provide a virtual power plant power collaborative control method based on multi-source data, comprising: constructing a virtual resource model corresponding to the physical resources in a virtual power plant based on the physical resources corresponding to the multi-source data, wherein the multi-source data is data collected in real time by sensors deployed on local resources; determining power adjustment range information and power change information based on the equipment type and power operation status of the equipment corresponding to the virtual resource model, using a preset equipment physical characteristic response model; determining power response characteristic information corresponding to the virtual resource model based on the power change information; performing power control processing on the power response characteristic information based on the power adjustment range information to obtain a collaborative control instruction set; and for each collaborative control instruction in the collaborative control instruction set, performing the following steps: distributing the collaborative control instruction to the corresponding edge control device for security verification to obtain a verification result; and executing the collaborative control instruction to control the target power equipment in response to the verification result meeting a preset verification success condition.

[0010] Secondly, some embodiments of this disclosure provide a virtual power plant power collaborative control device based on multi-source data, comprising: a model building unit configured to build a virtual resource model corresponding to the physical resources in a virtual power plant based on the physical resources corresponding to the multi-source data, wherein the multi-source data is data collected in real time by sensors deployed on local resources; an information acquisition unit configured to determine power adjustment range information and power change information based on the equipment type and power operation status of the equipment corresponding to the virtual resource model, using a preset equipment physical characteristic response model; an information determination unit configured to determine power response characteristic information corresponding to the virtual resource model based on the power change information; an instruction generation unit configured to perform power control processing on the power response characteristic information based on the power adjustment range information to obtain a collaborative control instruction set; and an equipment control unit configured to perform the following steps for each collaborative control instruction in the collaborative control instruction set: distributing the collaborative control instruction to the corresponding edge control device for security verification to obtain a verification result; and executing the collaborative control instruction to control the target power equipment in response to the verification result meeting a preset verification success condition.

[0011] Thirdly, some embodiments of this disclosure provide an electronic device, including: one or more processors; and a storage device having one or more programs stored thereon, such that when the one or more programs are executed by the one or more processors, the one or more processors implement the method as described in any implementation of the first aspect.

[0012] Fourthly, some embodiments of this disclosure provide a computer-readable medium having a computer program stored thereon, wherein the program, when executed by a processor, implements the method as described in any implementation of the first aspect.

[0013] The above embodiments of this disclosure have the following beneficial effects: Through the virtual power plant power collaborative control method based on multi-source data in some embodiments of this disclosure, power equipment can be accurately controlled. By constructing a global collaborative optimization mechanism, the rationality of power allocation is improved, thereby enhancing the stability of the power grid and reducing equipment overload power anomalies. Specifically, the reason for unreasonable power allocation is that multi-source data has heterogeneity and quality differences, the source of the data is not considered, and the physical characteristics and response capabilities of aggregated resources deviate, yet a unified control strategy is still used. Based on this, the virtual power plant power collaborative control method based on multi-source data in some embodiments of this disclosure firstly constructs a virtual resource model corresponding to the physical resources in the virtual power plant based on the physical resources corresponding to the multi-source data. The multi-source data is data collected in real time by sensors deployed on local resources. This establishes a standardized resource representation system, solving the problems of inconsistent data formats and dispersed sources, and providing a data foundation for accurate modeling. Secondly, based on the equipment type and equipment power operation status corresponding to the virtual resource model, a preset equipment physical characteristic response model is used to determine the power adjustment range information and power change information. Therefore, the inherent physical constraints and dynamic response capabilities of different types of resources (e.g., energy storage, adjustable loads) are considered, avoiding misjudgments of regulation capabilities caused by ignoring individual differences during control. Next, based on the aforementioned power change information, the power response characteristics corresponding to the aforementioned virtual resource model are determined. This quantifies the dynamic response behavior of each resource (e.g., delay time, regulation rate), resolving the oscillation problem caused by timing mismatch during multi-resource collaborative response. Then, based on the aforementioned power regulation range information, power control processing is performed on the aforementioned power response characteristics to obtain a collaborative control instruction set. This achieves multi-resource collaborative optimization calculation under a global objective, integrating the original resource regulation behavior into a unified regulatory force, improving the rationality of power allocation. Finally, for each collaborative control instruction in the aforementioned collaborative control instruction set, the following steps are executed: the collaborative control instruction is distributed to the corresponding edge control device for security verification, obtaining the verification result; in response to the verification result meeting the preset verification success condition, the collaborative control instruction is executed to control the target power equipment. Therefore, through a near-end rapid verification mechanism, dangerous commands that may cause local equipment overload or voltage exceedance are identified and intercepted in advance, forming a collaborative protection between global optimization and local security. In summary, by constructing a full-process collaborative control mechanism that includes standardized data modeling, assessment of individual resources, quantification of dynamic response characteristics, and execution of edge security, power equipment is precisely controlled, improving the rationality of power allocation, enhancing grid stability, and reducing equipment overload risks. Attached Figure Description

[0014] The above and other features, advantages, and aspects of the embodiments of this disclosure will become more apparent from the accompanying drawings and the following detailed description. Throughout the drawings, the same or similar reference numerals denote the same or similar elements. It should be understood that the drawings are schematic, and elements are not necessarily drawn to scale.

[0015] Figure 1 This is a flowchart of some embodiments of the virtual power plant power collaborative control method based on multi-source data according to the present disclosure;

[0016] Figure 2 This is a schematic diagram of the structure of some embodiments of the virtual power plant power coordination control device based on multi-source data according to the present disclosure;

[0017] Figure 3 This is a schematic diagram of the structure of an electronic device suitable for implementing some embodiments of the present disclosure. Detailed Implementation

[0018] Embodiments of this disclosure will now be described in more detail with reference to the accompanying drawings. While some embodiments of this disclosure are shown in the drawings, it should be understood that this disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of this disclosure. It should be understood that the accompanying drawings and embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of protection of this disclosure.

[0019] It should also be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings. Unless otherwise specified, the embodiments and features described in this disclosure can be combined with each other.

[0020] It should be noted that the concepts of "first" and "second" mentioned in this disclosure are used only to distinguish different devices, modules or units, and are not used to limit the order of functions performed by these devices, modules or units or their interdependencies.

[0021] It should be noted that the terms "a" and "a plurality of" used in this disclosure are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0022] The names of messages or information exchanged between multiple devices in the embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of such messages or information.

[0023] This disclosure will now be described in detail with reference to the accompanying drawings and embodiments.

[0024] refer to Figure 1The flowchart 100 illustrates some embodiments of the virtual power plant power coordination control method based on multi-source data according to this disclosure. The virtual power plant power coordination control method based on multi-source data includes the following steps:

[0025] Step 101: Based on the physical resources corresponding to the multi-source data, construct a virtual resource model corresponding to the aforementioned physical resources in the virtual power plant.

[0026] In some embodiments, the executor (e.g., an electronic device) of the above-described virtual power plant power collaborative control method based on multi-source data can be hardware or software. When the computing device is hardware, it can be implemented as a distributed cluster composed of multiple servers or terminal devices, or as a single server or a single terminal device. When the computing device is software, it can be installed in the hardware devices listed above. It can be implemented as multiple software programs or software modules to provide distributed services, or as a single software program or software module. No specific limitations are made here.

[0027] In some embodiments, the aforementioned executing entity can construct a virtual resource model corresponding to the physical resources in the virtual power plant based on the physical resources corresponding to the multi-source data. The multi-source data refers to data collected in real-time by sensors deployed on local resources. This multi-source data can be a heterogeneous data set used to describe various attributes of the same physical resource, obtained from different sources and through different methods within the virtual power plant environment. The physical resources can be physical devices or systems aggregated by the virtual power plant that truly exist in the physical world and possess the capacity for electricity production, consumption, or storage. The virtual resource model can be a standardized digital mapping model within the information space of the virtual power plant.

[0028] In some optional implementations of certain embodiments, the aforementioned execution entity can construct a virtual resource model corresponding to the physical resources in the virtual power plant based on the physical resources corresponding to the multi-source data, which may include the following steps:

[0029] The first step is to determine the resource types of the aforementioned multi-source data. These resource types include: industrial and commercial load types, energy storage system types, new energy charging station types, and distributed photovoltaic (PV) types. These resource types can be standardized classifications based on the physical characteristics of the resources and their functional type within the power grid. The industrial and commercial load types can refer to the types corresponding to adjustable power consumption equipment covering both commercial (e.g., supermarkets) and industrial (e.g., manufacturing industries). The energy storage system types can refer to the types corresponding to equipment with energy storage capabilities (e.g., battery energy storage). The new energy charging station types can refer to the types corresponding to charging facilities designed for vehicles to connect to the grid. The distributed PV types can include both industrial and commercial PV and residential PV.

[0030] The second step is to create an initial virtual resource model based on the aforementioned multi-source data and resource types. This initial virtual resource model can be a standardized mathematical model of the resource's basic attributes (e.g., capacity, location, grid connection point). In practice, firstly, the corresponding initial model can be called based on the resource type. Then, information corresponding to the basic attributes in the multi-source data is extracted and populated into the initial model to obtain the initial virtual resource model.

[0031] The third step involves configuring the measured index information into the initial virtual resource model, in response to the real-time data collection from local sensors deployed via the IoT acquisition module, thus obtaining the virtual resource model. The IoT acquisition module can be a module (e.g., an API interface) used to collect data transmitted from sensors deployed on the local resources. For example, the local sensors could be smart meters or voltage / current transformers. The measured index information can be data collected in real-time by sensors deployed on the local resources. In practice, a virtual power plant acquisition module can be used to associate the measured indicators (e.g., current, power, voltage) with the interfaces defined in the initial virtual resource model to obtain the virtual resource model.

[0032] The fourth step, in response to the fact that the aforementioned physical resources are collected through a third-party platform, involves configuring the third-party platform data into the initial virtual resource model to obtain the virtual resource model. This third-party platform data can be resource data aggregated by equipment manufacturers, energy service providers, or other management systems. In practice, firstly, data can be exchanged with the third-party platform via a standardized API (Application Programming Interface) to obtain its resource operation data. Then, this data is parsed and populated into the corresponding fields of the initial virtual resource model to obtain the virtual resource model. For example, the aforementioned physical resource could be a distributed photovoltaic power station. Its power generation and inverter status data can then be obtained through a third-party cloud platform.

[0033] The fifth step involves responding to the aforementioned physical resources by autonomously collecting information and configuring this autonomous information into the initial virtual resource model, thus obtaining a virtual resource model. This autonomous information is collected through the user's own electrification production management system. This autonomous information can be data proactively provided by the resource owner or obtained from its internal management system (e.g., Enterprise Resource Planning (ERP) system, Manufacturing Execution System (MES)). In practice, firstly, electrification production information (e.g., equipment power consumption, user production plans, equipment maintenance schedules) can be obtained through provided data interfaces or system integration methods to obtain autonomous information. Then, this autonomous information is populated into the corresponding fields of the initial virtual resource model to obtain the virtual resource model. For example, the aforementioned physical resource could be a factory with uninterrupted production. The factory's production scheduling information for the next 24 hours would then be configured into the model, allowing the virtual power plant to consider the factory's production constraints when formulating power control strategies.

[0034] Step 102: Based on the equipment type and power operation status corresponding to the virtual resource model, determine the power adjustment range information and power change information using the preset equipment physical characteristic response model.

[0035] In some embodiments, the aforementioned execution entity can determine power regulation range information and power change information based on the device type and power operation status corresponding to the aforementioned virtual resource model, using a preset device physical characteristic response model. The aforementioned device power operation status can refer to the real-time power condition (e.g., power consumption and power off) and power capacity parameters (e.g., voltage) of the device corresponding to the physical resource (e.g., charging pile) at a specific time. The aforementioned preset device physical characteristic response model can be a neural network model based on a Long Short-Term Memory (LSTM) network. The aforementioned neural network model can be trained based on device basic data and historical device operation data. The aforementioned device basic data can be experimental data from the device manufacturer. The aforementioned neural network model can include a physical feature encoding layer (converting static attributes such as device type and rated parameters into machine-recognizable features, which can use a multilayer perceptron), a temporal feature extraction layer (extracting temporal dependent features from real-time device operation data, which can use an LSTM structure), and an output layer (generating power regulation range information and power change information, which can use a branch network structure).

[0036] The aforementioned power regulation range information can be the boundary range within which the equipment can safely regulate power relative to its current operating state over a certain period of time. The aforementioned power change information can be power value information generated in response to grid demand (e.g., frequency regulation).

[0037] In some optional implementations of certain embodiments, the execution entity can determine power adjustment range information and power change information based on the device type and power operation status corresponding to the virtual resource model, using a preset device physical characteristic response model. This may include the following steps:

[0038] The first step, based on the aforementioned operational status, is to acquire real-time power data corresponding to the virtual resource model. This real-time power data includes power measurement index data and electrification production data. The power measurement index data can be electrical operating parameters acquired through an IoT acquisition module. For example, these electrical operating parameters may include active power, reactive power, voltage, current, and frequency. The electrification production data can be business operation information acquired through an enterprise resource management system. For example, this electrification production data includes production plans, equipment maintenance schedules, and power resource energy consumption costs.

[0039] The second step is to preprocess the aforementioned real-time power data to obtain preprocessed real-time power data. In practice, this preprocessing may include data cleaning, outlier handling, and standardization steps.

[0040] The third step involves inputting the preprocessed real-time power data into the preset equipment physical characteristic response model based on the type of the aforementioned virtual resource model, thereby obtaining the power gain information corresponding to the virtual resource model. This power gain information can be used to quantify the cost to the resource owner per unit of power regulation. The power gain information can include: power resource cost and marginal benefit. For example, for interruptible loads, a positive power gain information indicates cost savings from reducing unit power consumption. For energy storage systems, a negative power gain information indicates equipment loss costs from increasing unit discharge.

[0041] The fourth step is to determine the initial power adjustment range information corresponding to the virtual resource model based on the aforementioned power gain information. In practice, the adjustment ranges can be prioritized according to the power gain information to obtain the corresponding power adjustment range information, which serves as the initial power adjustment range information. Resources with higher power gain information (i.e., greater adjustment benefits or lower resource consumption costs) have a larger adjustable power range.

[0042] The fifth step involves revising the initial power regulation range information based on the grid power gain information of the virtual power plant, thereby obtaining the power regulation range information and power change information. The grid power gain information can include real-time grid power supply and demand, remaining power resources, nodal prices, and regulation demand. In practice, this revision process can be optimized with the goal of maximizing the overall benefit of the virtual power plant to obtain the power regulation range information and power change information. For example, in a grid regulation scenario, when the grid experiences a power deficit requiring increased power, resources with lower power gain information (i.e., lower regulation costs) can be prioritized, while limiting the regulation range of resources with excessively high power gain information to ensure overall regulation stability.

[0043] In addressing the technical challenges of differentiated resource control in virtual power plants, the following technical solutions are employed to address the specific application scenario: Virtual power plants need to simultaneously coordinate various heterogeneous resources such as energy storage systems, industrial and commercial loads, and electric vehicles to participate in grid regulation. This often presents the following technical challenges: Differentiated data processing and modeling methods must be adopted for different types of resources to avoid a uniform control strategy that would prevent proper resource regulation of equipment. Considering the following requirements for this application scenario—namely, the need to consider resource specificity, the need to establish a unified management framework, and the need to account for the unique characteristics of various resources—we have decided to adopt the following solution:

[0044] In some optional implementations of certain embodiments, the execution entity can determine power adjustment range information and power change information based on the device type and power operation status corresponding to the virtual resource model, using a preset device physical characteristic response model. This may include the following steps:

[0045] The first step involves generating a global power control function and power boundary condition information based on the grid operation status and power value attribute information of the virtual power plant. The power boundary condition information includes: the total power regulation demand information of the virtual power plant, the system regulation rate threshold information, and voltage safety constraint information. The global power control function can be an optimization function guiding the overall operation of the virtual power plant to achieve its objective. This objective can be minimizing the operating cost of the virtual power plant. The power boundary condition information represents the constraints that the virtual power plant must satisfy during optimization calculations. The total power regulation demand information represents the total power regulation amount completed by the virtual power plant within a specific time period. The system regulation rate threshold information represents the maximum allowable power change rate limit when the grid regulation demand is met. The voltage safety constraint information represents the voltage fluctuation range of each node in the grid for safe operation. In practice, under grid regulation scenarios, firstly, an optimization function with the objective of minimizing the total operating cost of the virtual power plant can be constructed to obtain the global power control function. Then, the regulation costs of various resources and the grid's electricity purchase cost are used as constraints to obtain the power boundary condition information. For example, the optimization function described above can control the total power of the virtual power plant by minimizing the sum of generation cost, regulation cost, and power deviation penalty term.

[0046] The second step involves sending the aforementioned global power control function and power boundary condition information to the edge control device corresponding to the virtual resource model. This edge control device can be an intelligent device deployed at a physical resource location, possessing computing, storage, and communication capabilities. This intelligent device can be used to perform local computing tasks and control functions. In practice, a message queue can be used to distribute the global power control function and power boundary condition information to the edge control device.

[0047] Third, based on the aforementioned edge control device, perform the following steps:

[0048] The first sub-step involves acquiring power measurement data from local sensors and electrification production data from the production management system, based on the operational status corresponding to the aforementioned virtual resource model. The power measurement data can be physical quantity measurements directly collected by sensors. The electrification production data can be information related to electrification production obtained from the enterprise management system. For example, in a grid regulation scenario, real-time power data can be read from smart meters. State of charge (SOC) data can be obtained from the battery management system. Production plan data can be obtained from the enterprise ERP system. The virtual resource model can be a factory. The power measurement data may include: current power consumption of 800kW and battery SOC of 65%. The electrification production data may include electrification production scheduling information for the next week.

[0049] The second sub-step involves preprocessing the aforementioned power measurement index data to obtain preprocessed power measurement index data. This preprocessing may include data cleaning, outlier handling, and standardization steps.

[0050] The third sub-step involves combining the preprocessed power measurement index data and the electrification production data into a feature vector. This feature vector can be a numerical vector formed by arranging multiple feature parameters in a specific order. In practice, the preprocessed power measurement index data and electrification production data can be normalized and combined into a vector form as the feature vector.

[0051] Fourth, based on the aforementioned global power control function and power boundary condition information, the aforementioned edge control device calls the aforementioned preset device physical characteristic response model to perform the following information determination steps:

[0052] The first sub-step involves embedding the aforementioned power boundary condition information into the preset equipment physical characteristic response model to obtain the physical characteristic response model. This physical characteristic response model can be a model with constraint information added. In practice, power constraints and voltage constraints can be embedded into a neural network model using the Lagrange multiplier method to form a constrained optimization model.

[0053] The second sub-step involves inputting the aforementioned feature vector into the aforementioned physical characteristic response model to obtain initial power parameters. These initial power parameters can be preliminary power adjustment recommendations obtained from the model.

[0054] The fifth step involves refining the initial power parameters over time to obtain power adjustment range information and power change information. In practice, firstly, the initial power parameters can be expanded along the time dimension. Then, for each time scale, a corresponding power setpoint is generated, and the corresponding adjustment rate is obtained, thus yielding power adjustment range information and power change information. For example, the initial power parameters (e.g., total adjustment) could be 200kW. Secondly, this is decomposed into 10 time steps, with each step adjusting 20kW to form a complete power adjustment curve. The power adjustment curve is then processed to obtain power adjustment range information and power change information.

[0055] The above-described operational steps, as an inventive point of this disclosure, solve the technical problem mentioned in the background art: "treating aggregated resources as a whole fails to fully consider the differences in physical characteristics, response speed, and regulation costs of different types of resources." In practice, traditional unified modeling methods cannot accurately reflect the actual operating characteristics of various resources. For example, using the same response model for energy storage systems as for adjustable loads may overestimate the load's regulation speed or underestimate the regulation accuracy of energy storage, leading to control effects deviating from expectations. This disclosure designs a scheme for data differentiation processing and response modeling based on resource types. For different types of resources, feature vectors and response models are constructed separately, ensuring that each type of resource can participate in grid regulation according to its own equipment state. Therefore, it maintains unified management of the virtual power plant and the individual differences of various resources, enabling the energy storage system to respond quickly and perform power regulation, allowing industrial and commercial loads to participate in regulation within the constraints of their electrified production, and allowing electric vehicles to provide flexibility within their available time windows, thereby significantly improving the accuracy of virtual power plant control and the utilization rate of power resources.

[0056] In addressing the aforementioned technical issues of insufficient real-time performance in models, and considering the application scenario—the virtual power plant—which needs to adapt to the integration of new resources and changes in the operational characteristics of existing resources, often presents the following technical challenge: traditional pre-trained models struggle to adapt to changes in resource characteristics, leading to a decline in prediction accuracy over time. Given the specific requirements of this application scenario—namely, the need to support rapid model iteration and personalized customization of resource entities to ensure the model always reflects the latest operational characteristics of the resources—we have decided to adopt the following solution:

[0057] In some optional implementations of certain embodiments, the aforementioned preset device physical characteristic response model is a neural network model; and

[0058] The training steps for the aforementioned preset device physical characteristic response model may include the following steps:

[0059] The first step is to extract measurement index information and electrification production information from the above multi-source data.

[0060] The second step involves anomaly detection of the aforementioned measurement indicators and electrification production information to identify power data anomalies. These anomalies can be data points that deviate from the normal range. In practice, anomaly detection methods based on statistical distributions (e.g., the Z-score algorithm) can be used to identify these deviations. For example, at a certain moment, the energy storage power data might show an anomaly of -9999.

[0061] The third step, assuming the aforementioned power data anomaly is a single, isolated anomaly, involves using a preset interpolation algorithm to repair it and obtain standard power data. In practice, this preset interpolation algorithm can be a linear interpolation algorithm. The standard power data can be the data obtained after anomaly detection and repair. In practice, firstly, the interpolation can be obtained using adjacent normal data before and after the anomaly. Then, the interpolated data is used to replace the power data anomaly to obtain standard power data.

[0062] The fourth step involves addressing the continuous abnormality of the aforementioned power data anomalies using a pre-defined replacement algorithm to repair them and obtain standard power data. This pre-defined replacement algorithm can be a historical average algorithm. In practice, firstly, normal data from the same time points over a past period can be obtained. Then, the average value corresponding to the normal data is determined, and this average value is used to replace the power data anomalies to obtain standard power data.

[0063] The fifth step involves resampling the aforementioned standard power data to obtain standardized sampled data. This standardized sampled data can be resampled data with a uniform time granularity. In practice, firstly, data from different sampling frequencies in the standard power data are unified to a standard time granularity. Then, interpolation methods are used to increase data density (and averaging methods are used to decrease data density) to obtain standardized sampled data. For example, the original data (second-level) can be resampled into 5-minute average data.

[0064] The sixth step is to perform window smoothing filtering and noise reduction on the standardized sampled data to obtain the power dataset. In practice, the sliding window averaging method can be used to filter and reduce noise on the standardized sampled data to obtain the power dataset.

[0065] Step 7: Based on the type of the virtual resource model described above, select the parameter information of the corresponding neural network model from the preset model library. This parameter information can be the structural parameters and training parameters of the neural network model. The parameter information can include: the number of network layers, the number of nodes, and the learning rate. For example, if the virtual resource model corresponds to an energy storage system, then an LSTM network is selected. If the virtual resource model corresponds to an adjustable load type, then a CNN network is selected.

[0066] Step 8: Receive the above parameter information through a graphical configuration interface to configure the initial neural network model. This initial neural network model can be an instance model that has not yet begun training. In practice, operators can configure the model structure through a graphical interface (e.g., a web interface). For example, configuration can be done by dragging and dropping components and filling in parameter values.

[0067] The ninth step involves training the initial neural network model using the aforementioned power dataset to obtain the preset equipment physical characteristic response model. In practice, firstly, the power dataset can be divided into training, validation, and test sets. Secondly, the backpropagation algorithm can be used to train the neural network. Then, the model parameters are optimized using a loss function to obtain the preset equipment physical characteristic response model.

[0068] The above-described operational steps, combined with step 105, constitute an inventive point of this disclosure, solving the technical problem mentioned in the background art: "treating aggregated resources as a whole fails to fully consider the differences in physical characteristics, response speed, and adjustment costs of different types of resources." In practice, traditional pre-trained models use fixed historical datasets, which cannot adapt to the dynamic changes in resource operating characteristics. For example, energy storage systems experience performance degradation with use, and industrial and commercial loads change their electricity consumption characteristics due to production plan adjustments. Using outdated training data leads to increased model prediction bias. This disclosure designs a model training scheme based on real-time power data updates and graphical customization. By continuously collecting the latest operating data and using a graphical interface to quickly configure the corresponding model for each virtual resource, real-time synchronization between the model and resource characteristics is achieved. Therefore, it ensures that each virtual resource model is trained and optimized using the latest operating data, keeping the prediction model always up-to-date, improving the accuracy and adaptability of power prediction for different types of resources, and avoiding efficiency deviations in control equipment caused by outdated models.

[0069] Step 103: Based on the power change information, determine the power response characteristic information corresponding to the virtual resource model.

[0070] In some embodiments, the execution entity may determine the power response characteristic information corresponding to the virtual resource model based on the power change information. The power response characteristic information may include key performance indicators such as response delay time, regulation rate, and regulation accuracy.

[0071] As an example, firstly, information can be extracted from power change information to obtain feature information related to the response. Then, the feature information is processed using corresponding processing methods to obtain the aforementioned power response characteristic information. For example, the aforementioned virtual resource model can be an adjustable industrial conformity model. The response characteristics can then include: power regulation rate and power regulation accuracy. The processing method for the power regulation rate can include the following steps: First, differentiate the power change curve to obtain the instantaneous rate of change at each moment. Then, statistically analyze its maximum and typical values ​​to obtain the response characteristic information. The processing method for the power regulation accuracy can include the following steps: First, establish a deviation sequence between the target power and the actual power. Second, determine the corresponding power mean, power variance, and power extreme values, and generate an accuracy assessment report to obtain the power response characteristic information.

[0072] In some optional implementations of certain embodiments, the execution entity may determine the power response characteristic information corresponding to the virtual resource model based on the power change information, which may include the following steps:

[0073] The first step is to extract the power characteristic parameters from the aforementioned power change information. These characteristic parameters can include parameters such as the rate of change of power, power overshoot, settling time, and power fluctuation frequency. In practice, signal processing techniques (e.g., Fourier transform) can be used to extract the corresponding characteristic parameters from the power change curve.

[0074] The second step involves querying the response parameters of similar devices from a pre-defined response characteristic knowledge base based on the device type in the aforementioned virtual resource model. This pre-defined response characteristic knowledge base can be a knowledge base storing response characteristic data for various resources. The knowledge in this knowledge base can originate from the technical specifications of the device manufacturer, historical operating statistics, or laboratory test results. The response parameters can be theoretical parameters. These parameters may include: theoretical response delay, rated adjustment rate, and standard adjustment accuracy.

[0075] The third step is to determine the degree of deviation between the aforementioned power characteristic parameters and power response parameters, thereby obtaining deviation information. This deviation information can be the deviation between actual and theoretical parameter information. In practice, the degree of deviation between the actual and theoretical parameters can be determined by calculating the Euclidean distance.

[0076] The fourth step involves inputting the aforementioned deviation information into a pre-trained response characteristic regression model to obtain quantified values ​​of the power response features. These quantified values ​​can be specific response parameter values. These response parameter values ​​may include: response delay time, actual regulation rate, and reliability score. The aforementioned response characteristic regression model can be trained based on a machine learning algorithm (e.g., support vector machine) and is capable of predicting actual response characteristics based on the input feature deviations.

[0077] The fifth step involves correcting the quantified values ​​of the power response characteristics based on the real-time operating parameters of the aforementioned virtual resource model, thereby obtaining power response characteristic information. These real-time operating parameters can be the actual performance parameters of the equipment during operation. These performance parameters may include equipment temperature and operating time. The correction process can consider the impact of equipment performance factors on the equipment response characteristics. For example, for energy storage systems operating in high-temperature environments, the quantified values ​​output by the response characteristic regression model are better. Therefore, the influence of temperature on battery performance can be considered, and the scores for adjustment rate and duration can be appropriately reduced to obtain power response characteristic information that better reflects the actual situation.

[0078] Step 104: Based on the power regulation range information, perform power power control processing on the power response characteristic information to obtain a coordinated control instruction set.

[0079] In some embodiments, the aforementioned executing entity can perform power control processing on the aforementioned power response characteristic information based on the aforementioned power regulation range information to obtain a cooperative control instruction set. The aforementioned power control processing can be a process for handling a multi-objective optimization problem with optimization objectives and constraints. The aforementioned optimization objectives may include: power resource utilization rate, accuracy of the grid command model, and power resource consumption cost. The aforementioned constraints may include: equipment power regulation range, response characteristics, and grid security information.

[0080] As an example, the power regulation range of each resource can be used as the boundary constraint of the decision variable. Power response characteristic information (e.g., delay time and regulation rate) is used as dynamic constraint, and grid characteristic information (e.g., regulation demand and nodal prices) is comprehensively considered to solve the problem and obtain the corresponding coordinated control command.

[0081] In addressing the technical challenges of power coordination control in virtual power plants, the following technical issues arise in the application scenario: virtual power plants participate in grid peak shaving and frequency regulation, serving ancillary service markets (e.g., controlling power consumption and reducing grid costs and increasing efficiency). However, this often presents the following challenges: how to coordinate global optimization objectives (overall grid frequency) with local constraints (individual device power consumption) while considering resource heterogeneity, avoiding control conflicts and system oscillations caused by differences in resource regulation characteristics. Considering the following requirements for this application scenario—precise and rapid control, and the need to balance the economic benefits of various resource entities (e.g., energy storage systems, factory industrial loads)—we have decided to adopt the following solution:

[0082] In some optional implementations of certain embodiments, the execution entity may perform power control processing on the power response characteristic information based on the power adjustment range information to obtain a coordinated control instruction set, which may include the following steps:

[0083] The first step is to construct an initial control function with the objective of minimizing the total regulation cost of the virtual power plant. This initial control function incorporates information about the power regulation range, power response characteristics, and power gain of the virtual resource model. Specifically, this initial control function can be a function that minimizes the weighted absolute value sum of the regulation cost and marginal benefit of the virtual resource model. The regulation cost can be the cost incurred when each resource regulates its power. The power gain information can be the economic benefit brought by unit power regulation.

[0084] The second step involves adding constraints to the initial control function to obtain the power control function. These constraints include physical operation constraints of the virtual resource model and grid safety operation constraints. The physical operation constraints include upper and lower power limits and ramp rate limits. The grid safety operation constraints include node voltage deviations and line capacity limits. The power control function can be a function with added constraints. The upper and lower power limits can be the minimum and maximum power values ​​corresponding to the virtual resource model. The ramp rate limit can be the maximum allowable power change per unit time. The node voltage deviation can be the allowable fluctuation range of voltage at each node in the grid. The line capacity limit can be the maximum allowable power transmission value of the power line.

[0085] The third step is to determine the power control information corresponding to the aforementioned power control function. The preset distributed optimization algorithm can be a parallel optimization algorithm. The optimization algorithm can be the alternating direction multiplier method. The optimization result can be the result information after solving the power control function.

[0086] The fourth step involves generating power adjustment instructions corresponding to the virtual resource model based on the aforementioned power control information. These instructions include a power setpoint, adjustment time, and priority. The power adjustment instructions can be commands executed to adjust the power output of the equipment. The power setpoint can be the specific power value that the resource needs to achieve. The adjustment time can be the start time and duration of the instruction execution. The priority can be the execution order information in case of conflicting instructions. For example, in a grid regulation scenario, the virtual resource model can be an energy storage system. The power adjustment instructions could include: a power setpoint of 150 kilowatts, a start time of 2 PM, an end time of 3 PM, and a priority of level one.

[0087] The fifth step is to standardize the aforementioned power regulation commands to obtain a coordinated control command set. This standardization process can be a process of unifying the command format. In practice, this process may include adding timestamps and checksums. Specifically, this can be achieved by encapsulating the commands using a particular data format (e.g., command version number, timestamp, and checksum) to obtain the coordinated control commands.

[0088] The above-described operational steps, combined with step 105, constitute an inventive point of this disclosure, solving the technical problem mentioned in the background art: "treating aggregated resources as a whole fails to fully consider the differences in physical characteristics, response speed, and regulation costs of different types of resources." In practice, traditional centralized optimization methods are ill-suited to the dynamic and heterogeneous nature of resources in virtual power plants. Distributed algorithms, lacking a unified coordination framework, may lead to conflicting autonomous decisions among resources, hindering the formation of effective regulatory synergy and even causing power oscillations in local power grids. This disclosure designs a collaborative control scheme based on hierarchical distributed optimization. By constructing a two-layer architecture of global power control and local device condition constraints, power control calculations are performed at each edge node based on local resource characteristics. Finally, collaborative execution is achieved through a standardized instruction set. Therefore, the differentiated characteristics of individual resources are considered, ensuring the synergy and economy of the virtual power plant as a whole participating in grid regulation, effectively avoiding regulation deviations and resource overload.

[0089] Step 105: For each cooperative control instruction in the cooperative control instruction set, perform the following steps:

[0090] Step 1051: Distribute the collaborative control command to the corresponding edge control device for security verification and obtain the verification result.

[0091] In some embodiments, the executing entity may distribute the aforementioned collaborative control instructions to corresponding edge control devices for security verification and obtain verification results. The edge control devices may be intelligent devices with computing, storage, and communication capabilities deployed at physical resource locations. For example, for an energy storage power station, the edge control device may be an industrial computer deployed within a substation. The verification results may include comprehensive judgment information across multiple dimensions. These multiple dimensions may include: instruction legality dimension, local security dimension, and grid impact dimension. For example, the instruction legality dimension may verify whether the instruction format, signature, and timing conform to a security protocol. The local security dimension may check whether the power value required by the instruction exceeds the device's current safe operating range. The grid impact dimension may determine whether executing the instruction will lead to voltage over-limit and line overload.

[0092] In some optional implementations of certain embodiments, the aforementioned executing entity may distribute the aforementioned collaborative control instructions to the corresponding edge control device for security verification and obtain the verification result, which may include the following steps:

[0093] The first step is to extract the instruction format information and power information from the control commands based on the aforementioned edge control device. The instruction format information may include information corresponding to the instruction's format specification (e.g., communication protocol type and data packet structure). The power information may be control parameters such as the power setpoint, power change rate, and execution time.

[0094] The second step involves executing the aforementioned coordinated control commands within a preset power grid model, assuming the command format information meets a preset communication protocol and the power information is within the safe operating range of the corresponding physical resources. This yields the operating results of the power equipment. The preset communication protocol can be a pre-defined protocol that ensures the reliability of the command source and the integrity of data transmission. The safe operating range can be the safe operating interval determined by the physical characteristics of the equipment. The preset power grid model can be a digital model of the topology, line parameters, and load distribution of the physical resources within the local power grid. The operating results can be the information output by the equipment after executing the commands.

[0095] The third step, based on the above operational results, is to determine the power anomaly indicators after executing the aforementioned coordinated control commands, thereby obtaining power anomaly indicator information. This information includes voltage deviation information and voltage frequency information. The power anomaly indicator information can be complete information quantified from various power anomaly indicators. These power anomaly indicators can be parameters that may pose a threat to the stable operation of power grid equipment (or system) after the execution of the control commands. For example, voltage deviation indicators and voltage frequency indicators. The voltage deviation information can be information on the degree to which node voltage deviates from its rated value. The voltage frequency information can be information on voltage frequency fluctuations.

[0096] As an example, in the above results, the node voltage dropped to 0.87 per unit, and the voltage frequency fluctuated to 50.1 Hz. This quantified data can then be used as a corresponding power anomaly indicator to obtain power anomaly information.

[0097] The fourth step involves generating a verification failure message and triggering an alarm mechanism in response to the aforementioned power anomaly indicators exceeding preset threshold values. The preset threshold values ​​can be safety boundary values ​​set by safe operation standards. The alarm mechanism can be a local audible and visual warning and a message notification to the virtual power plant's maintenance personnel. For example, in a grid regulation scenario, when the voltage deviation is within 5% or the frequency deviation is within 0.2Hz, the edge control device will generate a verification failure message and simultaneously notify the maintenance personnel via a platform message.

[0098] Fifth, in response to the above-mentioned power anomaly indicator information not exceeding the preset threshold, a verification result is generated. This verification result can be complete information after successful verification. This complete information may include: a verification pass flag, specific verification data, and timestamp information.

[0099] In addressing the technical challenges of virtual power plant security control, the following technical solutions were adopted for the application scenario: Virtual power plants participate in real-time grid regulation, requiring rapid response to dispatch commands and ensuring safe execution. However, this often presents the following technical issues: In a distributed verification environment, inconsistent verification standards across edge control devices can lead to overall control failures due to local device decisions, resulting in grid frequency changes. Considering the specific requirements of this application scenario—namely, the need for coordinated central control and autonomous edge decision-making to ensure control timeliness and device execution security—we decided to adopt the following solution:

[0100] In some optional implementations of certain embodiments, the aforementioned executing entity may distribute the aforementioned collaborative control instructions to the corresponding edge control device for security verification and obtain the verification result, which may include the following steps:

[0101] The first step involves generating an instruction data packet corresponding to the aforementioned coordinated control instructions based on the central control terminal of the virtual power plant. This instruction data packet includes a digital signature and a timestamp. The instruction data packet can be a data unit generated by structuring the data of each part of the instruction. The digital signature can be an identity identifier generated using an asymmetric encryption algorithm.

[0102] In practice, firstly, the central control unit can use the RSA-2048 algorithm to hash the instruction content and encrypt it to generate a digital signature. Secondly, the timestamp can be obtained using the ISO 8601 standard format. Finally, the data in the instruction (power setpoint, power change rate, and control parameters for execution time) is structured to obtain the instruction data packet.

[0103] The second step involves distributing the aforementioned instruction data packets to the corresponding edge control devices via a secure communication link. This secure communication link can be an end-to-end encrypted communication channel established using a preset protocol. In practice, after the central control unit verifies the identity of the edge devices using digital certificates, it transmits the instruction data packets to the edge control devices over the established secure channel.

[0104] Third, based on the aforementioned edge control device, perform the following verification steps:

[0105] The first sub-step verifies the validity of the digital signature and the timeliness of the timestamp, obtaining a verification result. This verification result can be used to distinguish whether an instruction has expired. In practice, a pre-set central control terminal public key can be used to verify the digital signature and check whether the difference between the timestamp and the current time is within an allowed time window (e.g., within 30 seconds) to obtain a verification result.

[0106] The second sub-step, in response to the verification result meeting the preset success condition, extracts the power setting value, adjustment time, and priority parameters from the instruction data packet. The preset success condition can be that the current instruction has not expired. This "not expired" condition can be that the digital signature has been verified and the timestamp is within the allowed time window. In practice, JSON format data in the instruction data packet can be parsed to extract the control parameters, obtaining the power setting value, adjustment time, and priority parameters. For example, the power setting value could be 500kW. The adjustment time could be 14:00:00-15:00:00. The priority parameter could be level 1 (the highest level).

[0107] The third sub-step involves verifying whether the aforementioned power setting value is within the device's permissible operating range, based on the device's locally stored safe operating parameters. These safe operating parameters can be information indicating that the device can operate safely. For example, for an energy storage system with a rated capacity of 600kW, the safe operating parameters could be between 0 and 600kW. In practice, the safe operating parameters can be obtained by querying the local device's parameter database, and then it can be determined whether the power setting value is within these parameters.

[0108] The fourth sub-step, in response to the aforementioned power setpoint being within the allowable operating range of the aforementioned equipment, invokes the local power grid topology model to simulate the execution of the aforementioned coordinated control commands, obtaining the power impact results. These power impact results include: voltage impact information, frequency impact information, and line load impact information. Specifically, the power impact results can be data showing changes in the power grid operating state after the equipment executes the control commands. The voltage impact information can be information on changes in the voltage amplitude at various nodes in the power grid. The frequency impact information can be information on the dynamic changes in the power grid system frequency. The line load impact information can be information on changes in the power transmission volume of each power line.

[0109] The fifth sub-step involves determining power anomaly indicator information based on the aforementioned power impact results. This power anomaly indicator information includes: voltage deviation rate, frequency fluctuation rate, and line load rate. These power anomaly indicators can be used to assess the safety of the power grid. The voltage deviation rate can be the percentage deviation between the actual voltage value and the rated voltage value. The frequency fluctuation rate can be the percentage deviation between the actual frequency value and the rated frequency value. The line load rate can be the percentage ratio of the actual transmitted power of the line to its rated capacity. In practice, power anomaly indicator information can be obtained using corresponding algorithms.

[0110] The sixth sub-step involves generating a verification failure message and triggering a local alarm device in response to the aforementioned power anomaly indicators exceeding the corresponding safety thresholds. The aforementioned safety thresholds can be thresholds affecting power grid fluctuations but falling within the safe range of the power grid. The aforementioned verification failure message can be information used to identify power grid anomalies. In practice, when a voltage deviation rate exceeding 5%, a frequency fluctuation rate exceeding 0.5%, or a line load rate exceeding 95% is detected, a verification failure message is generated, an audible and visual alarm is activated, and the verification failure message is sent to the central control terminal of the virtual power plant.

[0111] The seventh sub-step generates a verification pass message in response to the power anomaly indicator information being within the safe threshold range.

[0112] The fourth step involves sending the verification pass information to the central control terminal of the virtual power plant to obtain the verification result. This result includes: command validity information, local operation information, and grid impact information. The command validity information can be the result of verifying the compliance and security of the control command. The local operation information can be information assessing the feasibility and security of the target equipment executing the control command in its current operating state. The grid impact information can be a quantitative assessment of the impact of the control command execution on the grid's operating state.

[0113] The above-described operational steps, combined with step 1052, constitute an inventive point of this disclosure, solving the technical problem mentioned in the background art: "treating aggregated resources as a whole fails to fully consider the differences in physical characteristics, response speed, and regulation costs of different types of resources." In practice, traditional centralized security verification methods are ill-suited to the heterogeneity and widespread distribution of resources in virtual power plants. A single central verification mechanism may cause verification results to deviate from actual operating conditions due to communication delays or incomplete local information, leading to equipment overload and grid fluctuations. This disclosure designs a distributed security control scheme based on hierarchical verification. By implementing command validity verification at the central control terminal and equipment security and grid impact verification at the edge control devices, a multi-layered protection mechanism is established. Therefore, it ensures the rapid execution of control commands, prevents control deviations and equipment overloads caused by differences in resource characteristics, and improves the security of virtual power plants participating in grid regulation and the controllability of grid frequency.

[0114] Step 1052: In response to the verification result meeting the preset verification success condition, execute the cooperative control command to control the target power equipment.

[0115] In some embodiments, the execution entity may execute the cooperative control command to control the target power equipment in response to the verification result meeting the preset verification success condition. The preset verification success condition may be a pre-set condition used to determine whether the verification result meets the safety execution standard. The target power equipment may be a physical device that needs to receive power control commands from various distributed resources aggregated by the virtual power plant. For example, the preset verification success condition may be a format verification condition, a safety range condition, a grid safety condition, and a device status condition. The format verification condition may be that the command format conforms to a preset communication protocol specification, the data packet is complete, and it passes encryption verification. The safety range condition may be that the power setting value is within the rated operating range of the device and does not exceed the instantaneous withstand capacity. The grid safety condition may be that the predicted voltage deviation does not exceed 5%, and the frequency deviation is within 0.2Hz. The device status condition may be that the target device is in a ready state, has no fault alarms, and the communication connection is normal. The physical equipment may be power generation equipment (e.g., wind power generation), energy storage equipment (e.g., bidirectional converters), and load equipment (e.g., switches for regulating industrial loads, frequency converters for air conditioning systems).

[0116] The above embodiments of this disclosure have the following beneficial effects: Through the virtual power plant power collaborative control method based on multi-source data in some embodiments of this disclosure, power equipment can be accurately controlled. By constructing a global collaborative optimization mechanism, the rationality of power allocation is improved, thereby enhancing the stability of the power grid and reducing equipment overload power anomalies. Specifically, the reason for unreasonable power allocation is that multi-source data has heterogeneity and quality differences, the source of the data is not considered, and the physical characteristics and response capabilities of aggregated resources deviate, yet a unified control strategy is still used. Based on this, the virtual power plant power collaborative control method based on multi-source data in some embodiments of this disclosure firstly constructs a virtual resource model corresponding to the physical resources in the virtual power plant based on the physical resources corresponding to the multi-source data. The multi-source data is data collected in real time by sensors deployed on local resources. This establishes a standardized resource representation system, solving the problems of inconsistent data formats and dispersed sources, and providing a data foundation for accurate modeling. Secondly, based on the equipment type and equipment power operation status corresponding to the virtual resource model, a preset equipment physical characteristic response model is used to determine the power adjustment range information and power change information. Therefore, the inherent physical constraints and dynamic response capabilities of different types of resources (e.g., energy storage, adjustable loads) are considered, avoiding misjudgments of regulation capabilities caused by ignoring individual differences during control. Next, based on the aforementioned power change information, the power response characteristics corresponding to the aforementioned virtual resource model are determined. This quantifies the dynamic response behavior of each resource (e.g., delay time, regulation rate), resolving the oscillation problem caused by timing mismatch during multi-resource collaborative response. Then, based on the aforementioned power regulation range information, power control processing is performed on the aforementioned power response characteristics to obtain a collaborative control instruction set. This achieves multi-resource collaborative optimization calculation under a global objective, integrating the original resource regulation behavior into a unified regulatory force, improving the rationality of power allocation. Finally, for each collaborative control instruction in the aforementioned collaborative control instruction set, the following steps are executed: the collaborative control instruction is distributed to the corresponding edge control device for security verification, obtaining the verification result; in response to the verification result meeting the preset verification success condition, the collaborative control instruction is executed to control the target power equipment. Therefore, through a near-end rapid verification mechanism, dangerous commands that may cause local equipment overload or voltage exceedance are identified and intercepted in advance, forming a collaborative protection between global optimization and local security. In summary, by constructing a full-process collaborative control mechanism that includes standardized data modeling, assessment of individual resources, quantification of dynamic response characteristics, and execution of edge security, power equipment is precisely controlled, improving the rationality of power allocation, enhancing grid stability, and reducing equipment overload risks.

[0117] Further reference Figure 2As an implementation of the methods shown in the above figures, this disclosure provides some embodiments of a virtual power plant power coordination control device based on multi-source data. These device embodiments are similar to... Figure 1 Corresponding to the method embodiments shown, this virtual power plant power coordination control device based on multi-source data can be specifically applied to various electronic devices.

[0118] like Figure 2 As shown, a virtual power plant power collaborative control device 200 based on multi-source data includes: a model building unit 201, an information acquisition unit 202, an information determination unit 203, an instruction generation unit 204, and an equipment control unit 205. The model building unit 201 is configured to: construct a virtual resource model corresponding to the physical resources in the virtual power plant based on the physical resources corresponding to the multi-source data, wherein the multi-source data is data collected in real time by sensors deployed on the local resources. The information acquisition unit 202 is configured to: determine power regulation range information and power change information based on the equipment type and power operation status corresponding to the virtual resource model, using a preset equipment physical characteristic response model. The information determination unit 203 is configured to: determine the power response characteristic information corresponding to the virtual resource model based on the power change information. The instruction generation unit 204 is configured to: perform power control processing on the power response characteristic information based on the power regulation range information to obtain a collaborative control instruction set. The device control unit 205 is configured to perform the following steps for each collaborative control instruction in the above collaborative control instruction set: distribute the above collaborative control instruction to the corresponding edge control device for security verification and obtain the verification result; in response to the above verification result satisfying the preset verification success condition, execute the above collaborative control instruction to control the target power equipment.

[0119] It is understandable that the units described in the virtual power plant power coordination control device 200 based on multi-source data are related to the reference... Figure 1 The steps in the described method correspond accordingly. Therefore, the operations, features, and beneficial effects described above for the method also apply to the virtual power plant power coordination control device 200 based on multi-source data and the units contained therein, and will not be repeated here.

[0120] The following is for reference. Figure 3 It shows a schematic diagram of the structure of an electronic device (e.g., an electronic device) 300 suitable for implementing some embodiments of the present disclosure. Figure 3 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments of this disclosure.

[0121] like Figure 3As shown, the electronic device 300 may include a processing unit (e.g., a central processing unit, a graphics processing unit, etc.) 301, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 302 or a program loaded from a storage device 308 into a random access memory (RAM) 303. The RAM 303 also stores various programs and data required for the operation of the electronic device 300. The processing unit 301, ROM 302, and RAM 303 are interconnected via a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.

[0122] Typically, the following devices can be connected to I / O interface 305: input devices 306 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 307 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 308 including, for example, magnetic tapes, hard disks, etc.; and communication devices 309. Communication device 309 allows electronic device 300 to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 3 An electronic device 300 with various devices is shown; however, it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed alternatively. Figure 3 Each box shown can represent a device or multiple devices as needed.

[0123] In particular, according to some embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, some embodiments of this disclosure include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication device 309, or installed from storage device 308, or installed from ROM 302. When the computer program is executed by processing device 301, it performs the functions defined in the methods of some embodiments of this disclosure.

[0124] It should be noted that, in some embodiments of this disclosure, the computer-readable medium described above may be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In some embodiments of this disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In some embodiments of this disclosure, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0125] In some implementations, clients and servers can communicate using any currently known or future-developed network protocol such as HTTP (Hypertext Transfer Protocol) and can interconnect with digital data communication (e.g., communication networks) of any form or medium. Examples of communication networks include local area networks (“LANs”), wide area networks (“WANs”), the Internet (e.g., the Internet of Things), and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any currently known or future-developed networks.

[0126] The aforementioned computer-readable medium may be included in the aforementioned electronic device; or it may exist independently and not assembled into the electronic device. The aforementioned computer-readable medium carries one or more programs that, when executed by the electronic device, cause the electronic device to: construct a virtual resource model corresponding to the physical resources in a virtual power plant based on the physical resources corresponding to the multi-source data, wherein the multi-source data is data collected in real time by sensors deployed on local resources; determine power regulation range information and power change information based on the equipment type and power operation status corresponding to the virtual resource model, using a preset equipment physical characteristic response model; determine power response characteristic information corresponding to the virtual resource model based on the power change information; perform power control processing on the power response characteristic information based on the power regulation range information to obtain a set of cooperative control instructions; for each cooperative control instruction in the set of cooperative control instructions, perform the following steps: distribute the cooperative control instruction to the corresponding edge control device for security verification to obtain a verification result; in response to the verification result satisfying a preset verification success condition, execute the cooperative control instruction to control the target power equipment.

[0127] Computer program code for performing operations of some embodiments of this disclosure can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0128] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0129] The units described in some embodiments of this disclosure can be implemented in software or hardware. The described units can also be housed in a processor; for example, a processor may be described as including a model building unit, an information acquisition unit, an information determination unit, an instruction generation unit, and a device control unit. The names of these units do not necessarily limit the specific unit itself. For instance, the information acquisition unit may be described as "a unit that, based on the device type and power operating status corresponding to the aforementioned virtual resource model, determines power adjustment range information and power change information using a preset device physical characteristic response model."

[0130] The functions described above in this document can be performed at least in part by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), system-on-a-chip (SoCs), complex programmable logic devices (CPLDs), and so on.

[0131] The above description is merely a selection of preferred embodiments of this disclosure and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in the embodiments of this disclosure is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described inventive concept. For example, technical solutions formed by substituting the above-described features with (but not limited to) technical features with similar functions disclosed in the embodiments of this disclosure.

Claims

1. A virtual power plant power collaborative control method based on multi-source data, comprising: Based on the physical resources and resource types corresponding to the multi-source data, a virtual resource model corresponding to the physical resources is constructed in the virtual power plant. The multi-source data is data collected in real time by sensors deployed on local resources. The multi-source data is a heterogeneous data set obtained from different sources and through different methods in the virtual power plant environment, used to describe various attributes of the same physical resource. Based on the equipment type and power operation status corresponding to the virtual resource model, and using a preset equipment physical characteristic response model, power regulation range information and power change information are determined, including: generating a global power control function and power boundary condition information based on the grid operation status information and power value attribute information of the virtual power plant, and sending them to the edge control device corresponding to the virtual resource model; based on the edge control device, the following steps are performed: based on the operation status corresponding to the virtual resource model, acquiring power measurement index data from local sensors and electrification production data from the production management system; preprocessing the power measurement index data to obtain preprocessed power measurement index data; combining the preprocessed power measurement index data and the electrification production data into a feature vector; based on the global power control function and the power boundary condition information, calling the preset equipment physical characteristic response model on the edge control device, and performing the following information determination steps: embedding the power boundary condition information into the preset equipment physical characteristic response model to obtain a physical characteristic response model; inputting the feature vector into the physical characteristic response model to obtain initial power parameters; performing time-series refinement processing on the initial power parameters to obtain power regulation range information and power change information; Based on the power change information, determine the power response characteristic information corresponding to the virtual resource model; Based on the power regulation range information, the power response characteristic information is processed for power control to obtain a set of coordinated control instructions; For each collaborative control instruction in the collaborative control instruction set, the following steps are performed: the collaborative control instruction is distributed to the corresponding edge control device for security verification to obtain a verification result; in response to the verification result satisfying a preset verification success condition, the collaborative control instruction is executed to control the target power equipment.

2. The method according to claim 1, wherein, The construction of a virtual resource model corresponding to the physical resources in the virtual power plant based on the physical resources corresponding to multi-source data includes: The resource types of the multi-source data are determined, including: industrial and commercial load types, energy storage system types, new energy charging station types, and distributed photovoltaic types. Based on the multi-source data and the resource types, an initial virtual resource model is created; In response to the physical resources being collected in real time by local sensors deployed through the IoT acquisition module, the measurement index information is configured into the initial virtual resource model to obtain the virtual resource model; In response to the physical resources being collected through a third-party platform, the data from the third-party platform is configured into the initial virtual resource model to obtain the virtual resource model; In response to the physical resources, the autonomous information is collected and configured into the initial virtual resource model to obtain a virtual resource model, wherein the autonomous information is collected through the user's own electrification production management system.

3. The method according to claim 1, wherein, The step of determining the power response characteristic information corresponding to the virtual resource model based on the power change information includes: Extract the power characteristic parameters from the power change information; Based on the device type of the virtual resource model, query the power response parameters of similar devices from the preset response characteristic knowledge base; Determine the degree of deviation between the power characteristic parameters and the power response parameters to obtain deviation information; The deviation information is input into a pre-trained response characteristic regression model to obtain the quantified value of the power response feature; Based on the real-time operating parameters of the virtual resource model, the quantified values ​​of the power response characteristics are corrected to obtain power response characteristic information.

4. The method according to claim 1, wherein, The step of distributing the collaborative control command to the corresponding edge control device for security verification and obtaining the verification result includes: Based on the edge control device, extract the instruction format information and power information from the control instructions; In response to the instruction format information satisfying the preset communication protocol and the power information being within the safe operating range of the corresponding physical resources, the cooperative control instruction is executed within the preset power grid model to obtain the operating results of the power equipment; Based on the operation results, the power anomaly indicators after executing the coordinated control command are determined, and power anomaly indicator information is obtained, wherein the power anomaly indicator information includes: voltage deviation information and voltage frequency information; In response to the power anomaly indicator information exceeding the preset branch threshold, a verification failure message is generated, and an alarm mechanism is triggered; In response to the power anomaly indicator information not exceeding the preset threshold, a verification result is generated.

5. The method according to claim 1, wherein, Based on the equipment type and power operation status corresponding to the virtual resource model, and using a preset equipment physical characteristic response model, the power adjustment range information and power change information are determined, including: Based on the operating status, real-time power data corresponding to the virtual resource model is obtained, wherein the real-time power data includes: power measurement index data and electrification production data; The real-time power data is preprocessed to obtain preprocessed real-time power data. Based on the type corresponding to the virtual resource model, the preprocessed real-time power data is input into the preset equipment physical characteristic response model to obtain the power gain information corresponding to the virtual resource model. Based on the power gain information, the initial power adjustment range information corresponding to the virtual resource model is determined; Based on the grid power gain information of the virtual power plant, the initial power regulation range information is corrected to obtain power regulation range information and power change information.

6. A virtual power plant power coordination control device based on multi-source data, comprising: The model building unit is configured to build a virtual resource model corresponding to the physical resources in the virtual power plant based on the physical resources and resource types of the multi-source data. The multi-source data is data collected in real time by sensors deployed on the local resources. The multi-source data is a heterogeneous data set obtained from different sources and through different methods in the virtual power plant environment, used to describe various attributes of the same physical resource. The information acquisition unit is configured to determine power regulation range information and power change information based on the equipment type and power operation status corresponding to the virtual resource model, using a preset equipment physical characteristic response model. This includes: generating a global power control function and power boundary condition information based on the grid operation status information and power value attribute information of the virtual power plant, and sending them to the edge control device corresponding to the virtual resource model; and, based on the edge control device, performing the following steps: acquiring power measurement index data from local sensors and electrification production data from the production management system based on the operation status corresponding to the virtual resource model; and processing the power measurement index data... Preprocessing is performed to obtain preprocessed power measurement index data; the preprocessed power measurement index data and the electrification production data are combined into a feature vector; based on the global power control function and the power boundary condition information, the edge control device calls the preset equipment physical characteristic response model and performs the following information determination steps: embedding the power boundary condition information into the preset equipment physical characteristic response model to obtain the physical characteristic response model; inputting the feature vector into the physical characteristic response model to obtain initial power parameters; performing time-series refinement processing on the initial power parameters to obtain power adjustment range information and power change information; The information determination unit is configured to determine the power response characteristic information corresponding to the virtual resource model based on the power change information; The instruction generation unit is configured to perform power control processing on the power response characteristic information based on the power adjustment range information to obtain a set of coordinated control instructions; The device control unit is configured to perform the following steps for each cooperative control instruction in the cooperative control instruction set: The collaborative control command is distributed to the corresponding edge control device for security verification, and the verification result is obtained. In response to the verification result meeting the preset verification success condition, the cooperative control command is executed to control the target power equipment.

7. An electronic device, comprising: One or more processors; Storage device, on which one or more programs are stored, When the one or more programs are executed by the one or more processors, the one or more processors implement the method as described in any one of claims 1-5.

8. A computer-readable medium having a computer program stored thereon, wherein, When the program is executed by the processor, it implements the method as described in any one of claims 1-5.

Images