Virtual power plant scheduling method and device based on perceptual information model, equipment and medium

By optimizing the energy dispatch of virtual power plants using a perception information model and reinforcement learning algorithms, the stability and economic issues of the power system under uncertainties are solved, and efficient dispatch and stable operation of virtual power plants are achieved.

CN118982196APending Publication Date: 2024-11-19INFORMATION & COMM BRANCH OF STATE GRID JIANGSU ELECTRIC POWER +3
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202411069025.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-06
Publication Date
2024-11-19

AI Technical Summary

Technical Problem

How to maximize the economic benefits of the power system while reducing grid fluctuations and ensuring the stability of power supply, especially in the face of complex environments with uncertainties such as market price fluctuations, the intermittency of renewable energy and changes in user demand.

Method used

A virtual power plant scheduling method based on a perception information model is adopted. By acquiring equipment status data and historical power data, automatic time series prediction network is used for prediction, and reinforcement learning algorithm is combined to optimize energy scheduling strategy, determine target power generation data, and realize the economic benefits and operational stability of the virtual power plant.

Benefits of technology

This improves the economic efficiency and operational stability of virtual power plants, enabling them to better cope with uncertainties and ensure the stability and flexibility of power supply.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118982196A_ABST
    Figure CN118982196A_ABST
Patent Text Reader

Abstract

The embodiment of the invention discloses a virtual power plant scheduling method and device based on a perceptual information model, equipment and a medium. The method is applied to a virtual power plant dispatching system, and the virtual power plant dispatching system comprises a dispatching management module, a dispatching control module and at least one virtual power plant. The method comprises the following steps: acquiring equipment state data of the virtual power plant and historical power data of a set time length; inputting the historical power data into a set prediction network to obtain predicted power data; and determining an energy scheduling strategy of the virtual power plant based on the predicted power data and the equipment state data, and determining target power generation data of the virtual power plant through the energy scheduling strategy. According to the technical scheme, the economic benefit and the operation stability of the virtual power plant are improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of power system dispatching, and in particular to a virtual power plant dispatching method, device, equipment and medium based on a perception information model. Background Art

[0002] In recent years, virtual power plant (VPP) technology has been regarded as an emerging energy management strategy, which achieves energy complementarity and comprehensive utilization by connecting and integrating different energy resources.

[0003] However, with the increasing complexity of energy systems, the uncertainties faced by all parties involved in the power system are also increasing. These uncertainties come from factors such as market price fluctuations, the intermittent and unpredictable nature of renewable energy, and changes in user demand. How to minimize grid fluctuations while maximizing the economic benefits of the power system and ensure the stability of power supply has become an urgent problem that needs to be solved in the operation and planning of the power system. Summary of the invention

[0004] The present invention provides a virtual power plant scheduling method, device, equipment and medium based on a perception information model, which improves the economic benefits and operational stability of the virtual power plant.

[0005] According to one aspect of the present invention, a virtual power plant scheduling method based on a perception information model is provided, the method being applied to a virtual power plant scheduling system, the virtual power plant scheduling system comprising a scheduling management module, a scheduling control module and at least one virtual power plant; the method comprising:

[0006] Obtaining equipment status data of the virtual power plant and historical power data of a set period;

[0007] Inputting the historical power data into a set prediction network to obtain predicted power data;

[0008] An energy dispatching strategy of the virtual power plant is determined based on the predicted power data and the equipment status data, and target power generation data of the virtual power plant is determined through the energy dispatching strategy.

[0009] Optionally, before obtaining the device status data of the virtual power plant and the historical power data of a set period, the method further includes:

[0010] Obtaining basic attributes and parameter information of each device in the virtual power plant;

[0011] A perception information model is constructed according to the basic attributes and the parameter information, so that the perception information model shares the status data of each device.

[0012] Optionally, the historical power data includes historical load data, historical output data and historical electricity price data; the set prediction network is an automatic time series prediction network;

[0013] Inputting the historical power data into a set prediction network to obtain predicted power data includes:

[0014] The historical load data, historical output data and historical electricity price data of a set time period are input into the automatic time series prediction network to obtain the predicted load data, predicted output data and predicted electricity price data.

[0015] Optionally, determining an energy dispatch strategy of the virtual power plant based on the predicted power data and the device status data includes:

[0016] Cost of obtaining electricity;

[0017] Determine the equipment operation cost respectively according to the equipment status data and the predicted power data;

[0018] constructing an objective function based on the electricity cost and the equipment operation cost;

[0019] The energy dispatch strategy of the virtual power plant is determined according to the objective function and the constraints.

[0020] Optionally, the equipment operating costs include operating costs of power generation equipment and energy storage equipment; the power generation equipment includes fossil power generation, photovoltaic power generation and wind power generation;

[0021] Determining the equipment operation cost according to the equipment status data and the predicted power data respectively includes:

[0022] Determine the operating costs of fossil power generation, photovoltaic power generation, wind power generation and energy storage equipment respectively according to the equipment status data and the predicted power data;

[0023] The operating costs of the fossil power generation, the photovoltaic power generation, the wind power generation and the energy storage device are integrated to obtain the equipment operating cost.

[0024] Optionally, determining the target power generation data of the virtual power plant through the energy scheduling strategy includes:

[0025] Modeling the energy scheduling strategy as a reinforcement learning agent;

[0026] The reinforcement learning agent is processed by a reinforcement learning algorithm to obtain target power generation data of the virtual power plant.

[0027] Optionally, the reinforcement learning agent is processed by a reinforcement learning algorithm to obtain target power generation data of the virtual power plant, including:

[0028] Taking the system energy state at the current moment as a state set, and taking the scheduling actions of the power generation equipment and the energy storage equipment as an action behavior set;

[0029] Determine the transition probability function of the system energy state at the next moment according to the state set and the action behavior set;

[0030] Determine a reward function for the system energy state based on the objective function and the transition probability function;

[0031] The target power generation data of the virtual power plant is determined according to the reward function.

[0032] According to another aspect of the present invention, a virtual power plant scheduling device based on a perception information model is provided, wherein the device is configured in a virtual power plant scheduling system, wherein the virtual power plant scheduling system comprises a scheduling management module, a scheduling control module and at least one virtual power plant; the device comprises:

[0033] A data acquisition module, used to acquire equipment status data of the virtual power plant and historical power data of a set period;

[0034] A data prediction module, used for inputting the historical power data into a set prediction network to obtain predicted power data;

[0035] An energy scheduling strategy determination module is used to determine the energy scheduling strategy of the virtual power plant based on the predicted power data and the equipment status data, and determine the target power generation data of the virtual power plant through the energy scheduling strategy.

[0036] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0037] at least one processor; and

[0038] a memory communicatively connected to the at least one processor; wherein,

[0039] The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the virtual power plant scheduling method based on the perception information model described in any embodiment of the present invention.

[0040] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the virtual power plant scheduling method based on the perception information model described in any embodiment of the present invention when executed.

[0041] The technical solution of the embodiment of the present invention obtains the equipment status data of the virtual power plant and the historical power data of a set time; inputs the historical power data into a set prediction network to obtain predicted power data; determines the energy dispatch strategy of the virtual power plant based on the predicted power data and the equipment status data, and determines the target power generation data of the virtual power plant through the energy dispatch strategy. This technical solution improves the economic benefits and operational stability of the virtual power plant.

[0042] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0044] Figure 1 is a flow chart of a virtual power plant scheduling method based on a perception information model provided according to Embodiment 1 of the present invention;

[0045] Figure 2 is an example diagram of the architecture of a perception information model provided according to the first embodiment of the present invention;

[0046] Figure 3 is a flow chart of a virtual power plant scheduling method based on a perception information model provided according to Embodiment 2 of the present invention;

[0047] Figure 4 This is a diagram of an energy management architecture for virtual scheduling provided according to Embodiment 2 of the present invention;

[0048] Figure 5 is a structural schematic diagram of a virtual power plant scheduling device based on a perception information model provided according to Embodiment 3 of the present invention;

[0049] Figure 6 It is a schematic diagram of the structure of an electronic device provided according to Embodiment 4 of the present invention. DETAILED DESCRIPTION

[0050] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0051] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0052] Embodiment 1

[0053] Figure 1 This is a flowchart of a virtual power plant scheduling method based on a perception information model provided according to the first embodiment of the present invention. This embodiment is applicable to the case of dynamic scheduling of virtual power plant data. The method can be executed by a virtual power plant scheduling device. The virtual power plant scheduling device can be implemented in the form of hardware and / or software. The virtual power plant scheduling device can be configured in an electronic device with data processing capabilities. Figure 1 As shown, the method includes:

[0054] The technical solution of this embodiment is applied to a virtual power plant dispatching system, which includes a dispatching management module, a dispatching control module and at least one virtual power plant. The dispatching management module can be used to manage the dispatching operations of various devices in the virtual power plant. The dispatching control module can be used to control various devices based on the dispatching management module. In the energy Internet application scenario, the virtual power plant VPP technology further enhances the flexibility and reliability of the energy system and can better cope with various uncertainties. The virtual power plant VPP optimizes the allocation and use of power resources, reduces the fluctuations of the power system, and improves economic benefits by integrating distributed energy, energy storage equipment and intelligent control systems.

[0055] In this embodiment, an innovative VPP dual architecture is proposed, which can be divided into a commercial virtual power plant CVPP and a technical virtual power plant TVPP. The two work together to ensure the maximization of technical efficiency and economic benefits. Specifically, the commercial virtual power plant CVPP is mainly responsible for integrating real-time market electricity prices, predicted load information and predicted output of renewable energy to formulate an electric energy dispatching strategy. Subsequently, the commercial virtual power plant CVPP passes the corresponding electric energy dispatching strategy to the technical virtual power plant TVPP to optimize the power generation of distributed power generation units, and feeds back the optimization results to the commercial virtual power plant CVPP. In this embodiment, the power dynamic dispatching mechanism of the virtual power plant dual architecture can be adopted. The mutual inductance of the virtual power plant can be enhanced through the detachable modules, and the energy distribution and power dispatching are optimized in combination with the timing prediction algorithm and the dispatching mechanism, so as to improve the economic benefits and operation stability of the virtual power plant dispatching system and realize efficient dispatching in a complex energy environment.

[0056] S110. Acquire equipment status data of the virtual power plant and historical power data of a set period based on the perception information model.

[0057] Among them, the device status data may include static attribute data and dynamic attribute data of each device in the virtual power plant. Specifically, the static attributes may include descriptive information of each device, such as device identifier, device name, type, physical location, manufacturer, and operating status. The equipment of the virtual power plant may include distributed energy equipment, energy storage equipment, and power consumption equipment. Dynamic attribute data may include the power generation capacity, energy type, and grid connection point identification of distributed energy equipment; it may also include operating data such as the storage capacity, maximum charging power, maximum discharge power, and current power percentage of energy storage equipment; it may also include data such as device model, device power consumption, and grid connection point identification. The perception information model may be pre-built and used to share data such as the status of each device in the virtual power plant in real time. The perception information model in this embodiment defines and standardizes the parameters and information interaction formats of various types of equipment to achieve seamless communication and integration between different devices and systems, improves the intelligent management and operation efficiency of the distributed new energy acquisition and control system, and makes the model more applicable.

[0058] The set duration may be one day, one week, one month, one year, etc., and may be set according to actual needs. The historical power data may include the historical power data of the set duration in the virtual power plant. In this embodiment, the historical power data may include historical load data, historical output data, and historical electricity price data. Specifically, the historical load data may refer to historical baseline load data; the historical output data may include historical wind power output data and historical solar power output data; and the historical electricity price data may refer to historical time-of-use electricity price data.

[0059] In this embodiment, static attribute data and dynamic attribute data of each device in the virtual power plant, as well as historical load data, historical output data and historical electricity price data for a set period of time can be obtained.

[0060] In this embodiment, optionally, before obtaining the equipment status data of the virtual power plant and the historical power data of a set period of time, it also includes: obtaining basic attributes and parameter information of each device in the virtual power plant; constructing a perception information model based on the basic attributes and parameter information, so that the perception information model shares the status data of each device.

[0061] Among them, basic attributes may include basic information, static attributes, and dynamic attributes. Parameter information may include input parameters and output parameter information of the device. The perception information model may be constructed based on data such as basic information, static attributes, and dynamic attributes of each device in the virtual power plant, as well as input parameters and output parameter information. In this embodiment, the perception information model can be used to share the status data of each device in real time, realize seamless interoperability between different devices and systems, and enhance the overall performance and reliability of the system.

[0062] Specifically, in this embodiment, the specific method of constructing a perception information model based on basic attributes and parameter information is to define the objects of the information model, and to model the devices included in each object, as well as the basic attributes and parameter information of the devices. In addition, standardized device models, monitoring messages, control services, topology models, and communication protocols are also defined. A single device model can be aggregated to form a regional-level information perception model, so that the perception information model shares the status data of each device.

[0063] For example, the architecture of the perception information model constructed in this embodiment is shown in FIG. Figure 2As shown in Figure 1. The model objects in the perception information model can include basic information, static attributes, dynamic attributes, messages, and service objects. Specifically, the basic information part contains the model identifier and model description, which are used to uniquely identify and describe the detailed information of the model. The model identifier provides a unique identification code for each model entity to ensure uniqueness in the entire information system. It is a globally unique ID and can be a serial number, database primary key, or any other form of unique identification. The model description provides detailed information about the model to help understand the function, purpose, and role of the model in the system. It includes the purpose, design concept, application scenario, and related business logic description of the model. The static attribute part covers the identifier, name, type, location, and physical characteristics of the device. By standardizing the device information, it ensures that different types of devices can be managed and interacted within a unified framework. The dynamic attribute part focuses on the specific attributes of distributed power sources, energy storage devices, power consumption devices, and other power grid devices, such as installed capacity, energy type, key equipment parameters, etc. The message part defines the interaction information and content between different devices, pays attention to the operating status and mode of various types of equipment, and communicates between different devices through input and output parameters. For example, different types of messages such as equipment operating status, equipment alarms, voltage and power over-limit alarms are used for real-time monitoring and early warning to ensure the safe and stable operation of the system. Finally, the service part includes load forecasting, power forecasting, economic optimization scheduling and general services. Load forecasting and power forecasting provide information such as daily forecast values, short-term forecast values ​​and real-time values ​​for predicting photovoltaic power generation. The economic optimization scheduling part contains optimization goals and equipment information for optimizing energy scheduling and resource allocation. General services provide functions such as equipment status monitoring and equipment event monitoring to ensure the safe and efficient operation of equipment.

[0064] For example, the specific content of static attributes is shown in Table 1. Static attributes define the basic attributes of a device, including unique identifier (mRID), name, type, physical location, manufacturer, and status. The static attribute objects in the constructed perception information model provide comprehensive description and management information for each device, ensuring that each device can be uniquely identified and effectively managed in the system.

[0065] Table 1 Static properties

[0066]

[0067] In this embodiment, the dynamic attributes specifically include the dynamic attributes of distributed energy, the dynamic attributes of energy storage equipment, and the dynamic attributes of power consumption equipment. Among them, the dynamic attributes of distributed energy can be as shown in Table 2. The dynamic attributes of distributed energy include information such as power generation capacity (capacity), energy type (energyType), and connection point identification (connectionPoint). The information included in the dynamic attributes of distributed energy in this embodiment provides a comprehensive description and management for each distributed energy device, ensuring its uniqueness and efficient management in the system.

[0068] Table 2 Dynamic properties of distributed power sources

[0069]

[0070]

[0071] The dynamic properties of the energy storage device can be shown in Table 3, which shows the measurement model of the energy storage device, including information such as storage capacity (capacity), maximum charging power (chargeRate), maximum discharge power (dischargeRate) and current power percentage (stateOfCharge). The information included in the dynamic properties of the energy storage device in this embodiment provides detailed operating parameters and status monitoring for each energy storage device to ensure its efficient operation and management in the system.

[0072] Table 3 Dynamic properties of energy storage devices

[0073]

[0074] The dynamic attributes of the power consumption equipment can be shown in Table 4, which shows the measurement model of the power consumption equipment, including equipment model (model), equipment power consumption (loadAmount), grid connection point identification (connectionPoint) and other information. The information included in the dynamic attributes of the power consumption equipment in this embodiment provides complete usage records and management for each power consumption equipment, ensuring its reasonable configuration and control in the system.

[0075] Table 4 Dynamic properties of power consumption equipment

[0076]

[0077] In this embodiment, the messages contained in the model object in the perception information model specifically include messages of distributed energy, messages of energy storage equipment, and messages of power consumption equipment. Among them, Table 5 gives the messages of distributed power sources, specifically including the input parameters contained in Table 5 (a) and the output parameters contained in Table 5 (b). The information of input parameters and output parameters provides detailed data for the operation monitoring and status management of distributed power sources, ensuring their efficient operation and stability in the system.

[0078] Table 5(a) Input parameters of distributed power generation

[0079]

[0080] Table 5(b) Output parameters of distributed power supply

[0081]

[0082] Among them, Table 6 gives the information of the energy storage device, including Table 6 (a) the input parameters of the energy storage device and Table 6 (b) the output parameters of the energy storage device. The information of input parameters and output parameters provides an important reference for the power management and operation status of the energy storage device, ensuring its effectiveness and reliability in the system.

[0083] Table 6 (a) Input parameters of energy storage equipment

[0084]

[0085] Table 6(b) Output parameters of energy storage equipment

[0086]

[0087] Among them, Table 7 gives the information of the power consumption equipment, including Table 7 (a) the input parameters of the power consumption equipment and Table 7 (b) the output parameters of the power consumption equipment. The information of input parameters and output parameters provides key data for power consumption monitoring and status management of the power consumption equipment, ensuring its reasonable use and maintenance in the system.

[0088] Table 7(a) Input parameters of power consumption equipment

[0089]

[0090] Table 7(b) Output parameters of power consumption equipment

[0091]

[0092] Among them, Table 8 shows the general message, specifically including the input parameters of the general message in Table 8 (a) and the output parameters of the general message in Table 8 (b). In this embodiment, the input parameters and output parameters of these general messages provide comprehensive data support for the electrical parameter monitoring and safety management of the equipment, ensuring the safety and stability of the system.

[0093] Table 8 (a) Input parameters of general message

[0094]

[0095] Table 8(b) Output parameters of general message

[0096]

[0097] The services included in the model object in the perception information model in this embodiment specifically include the content of load forecasting service, the content of power forecasting service, the content of economic optimization dispatching service and the content of general service. Among them, Table 9 shows the content of load forecasting service, which provides comprehensive environmental and meteorological data support for load forecasting, ensuring the accuracy and reliability of the forecast results.

[0098] Table 9 Load forecasting service

[0099]

[0100] Among them, Table 10 gives the content of the power forecast service. This information provides detailed power market and environmental data support for power forecasting, ensuring the accuracy and effectiveness of the forecast results.

[0101] Table 10 Power prediction service

[0102]

[0103] Among them, Table 11 gives the content of economic optimization scheduling services. This information provides basic data and conditions for achieving economic optimization scheduling and ensuring the optimal allocation and use of energy resources.

[0104] Table 11 Economic optimization dispatch service

[0105]

[0106] Among them, Table 12 gives the content of general services, which provide necessary monitoring and control means for the daily management and operation of the equipment to ensure the stability and security of the system.

[0107] Table 12 General Services

[0108]

[0109] In this embodiment, environmental and meteorological data such as power generation, load demand, irradiance, temperature, humidity and wind speed are used to provide load forecasting and power forecasting services. At the same time, the operating status of the photovoltaic inverter, the status of the energy storage system, the charging pile information, and the charging and discharging status of the energy storage equipment in the control service are used to perform economic optimization scheduling and decision-making control. According to the provided perception information model, these data and services ensure the efficient allocation and use of energy resources. In this embodiment, the perception information model is constructed by the device modeling method for the balanced consumption of distributed new energy, which can ensure effective communication and control between edge intelligent devices and distributed new energy devices. In this embodiment, the perception information model can aggregate a single device model to form a regional aggregation model to achieve seamless interoperability between different devices and systems to enhance the overall performance and reliability of the system.

[0110] S120, inputting historical power data into a set prediction network to obtain predicted power data.

[0111] The set prediction network may be a trained network. It may be applied to predict power data. Exemplarily, in this embodiment, the set prediction network may be an automatic time series prediction network. The predicted power data may be predicted data outputted by the set prediction network. Specifically, the predicted power data may include predicted load data, predicted output data, and predicted electricity price data.

[0112] In this embodiment, historical load data, historical output data and historical electricity price data of a set time period can be input into the Auto-NHiTs prediction network to obtain predicted load data, predicted output data and predicted electricity price data.

[0113] S130. Determine an energy dispatching strategy for the virtual power plant based on the predicted power data and the equipment status data, and determine the target power generation data of the virtual power plant through the energy dispatching strategy.

[0114] Among them, the energy dispatching strategy of the virtual power plant can be understood as the energy supply and demand situation in the virtual power plant and the energy allocation strategy for each device. In this embodiment, the method of determining the energy dispatching strategy of the virtual power plant based on the predicted power data and the equipment status data can be to establish an objective function based on the predicted power data and the equipment status data in combination with the set conditions, so as to schedule the energy of each device based on the objective function and the constraints, which can be used as an energy dispatching strategy. The target power generation data can be understood as coal-fired power generation data. Exemplarily, the target power generation data can be coal-fired power generation power or other data. In this embodiment, the coal-fired power generation power of the virtual power plant can be determined according to the determined energy dispatching strategy.

[0115] The technical solution of the embodiment of the present invention obtains the equipment status data of the virtual power plant and the historical power data of a set time; inputs the historical power data into the set prediction network to obtain the predicted power data; determines the energy dispatch strategy of the virtual power plant based on the predicted power data and the equipment status data, and determines the target power generation data of the virtual power plant through the energy dispatch strategy. This technical solution improves the economic benefits and operational stability of the virtual power plant.

[0116] Embodiment 2

[0117] Figure 3 This is a flowchart of a virtual power plant scheduling method based on a perception information model provided in accordance with the second embodiment of the present invention. This embodiment is optimized based on the above embodiment. The specific optimization is: inputting historical power data into a set prediction network to obtain predicted power data, including: inputting historical load data, historical output data, and historical electricity price data of a set time into an automatic time series prediction network to obtain predicted load data, predicted output data, and predicted electricity price data. Figure 3 As shown, the method includes:

[0118] S310. Acquire equipment status data of the virtual power plant and historical power data of a set period based on the perception information model.

[0119] Among them, the historical power data includes historical load data, historical output data and historical electricity price data; the prediction network is set to an automatic time series prediction network.

[0120] S320, inputting historical load data, historical output data and historical electricity price data of a set time period into an automatic time series prediction network to obtain predicted load data, predicted output data and predicted electricity price data.

[0121] Among them, the automatic time series prediction network can be AutoN-HiTS. Exemplarily, the automatic time series prediction network can be an Auto-NHiTs prediction network based on automatic parameter tuning, or other prediction networks, which can be set according to actual conditions, and are not limited here. In this embodiment, AutoN-HiTS is an automated time series prediction network, which is improved based on N-HiTS, and can automatically select and optimize model parameters to adapt to different types of data and prediction requirements. AutoN-HiTS further improves the accuracy and robustness of the prediction through automated parameter adjustment and model selection. The predicted load data can be the load data predicted within a set time range. The predicted output data is the wind output data and / or solar output data predicted within a set time range. The predicted electricity price data can be the electricity price predicted within a set time range.

[0122] In this embodiment, the historical load data, historical output data and historical electricity price data of the set time are input into the automatic time series prediction network to obtain the predicted load data, predicted output data and predicted electricity price data. It can be understood that in this embodiment, the longer the set time of the historical data input is, the higher the accuracy of the obtained prediction data.

[0123] In this embodiment, in the prediction part, the historical power data of a certain place in the first nine months is divided into a training set, and the data of the last three months is divided into a verification set, and a 7-day short-term prediction of power generation, power consumption and market electricity price is performed. The advanced N-HiTS time series prediction model and AutoN-HiTS time series prediction model are used to provide reliable data support for scheduling. Through the prediction data analysis, the true value (True) is compared with the prediction results of the two prediction methods (N-HiTS and AutoN-HiTS) using a graphical comparison method to obtain the comparison result AutoN-HiTS is improved based on N-HiTS, automatically selects and optimizes model parameters, and can adapt to different types of data and prediction requirements; through automated parameter adjustment and model selection, the accuracy and robustness of the prediction are further improved. Based on the comparison results of the photovoltaic power generation prediction results, it can be seen that the AutoN-HiTS algorithm has higher accuracy and robustness when processing data with strong periodicity such as photovoltaic power generation. Based on the comparison of wind power generation forecast results, it can be seen that the AutoN-HiTS algorithm has higher accuracy when processing complex and volatile wind power generation data; based on the comparison of load forecast results, it can be seen that both the AutoN-HiTS algorithm and the N-HiTS algorithm can better capture the trend of load changes, but the prediction results of the AutoN-HiTS algorithm in each time period are smoother and closer to the true value, with smaller errors, reflecting its superiority in load forecasting. Based on the comparison of electricity price forecast results, it can be seen that the AutoN-HiTS algorithm has higher robustness and accuracy in electricity price forecasting. From various data, the error of the AutoN-HiTs model is smaller than that of the N-HiTs model in all prediction parameters, indicating that the AutoN-HiTs model is more accurate in prediction.

[0124] In this embodiment, the AutoN-HiTS model based on the automated tuning framework participates in each stage of source-load-grid-storage, performs load, energy output and electricity price forecasts, and provides solid data support for the diversified energy management of the virtual power plant VPP. It provides effective guidance for scheduling decisions with powerful forecasting capabilities, improves the grid's ability to adapt to market and environmental changes, and ensures the stable operation of the grid.

[0125] S330: Obtain electricity cost.

[0126] The power cost can be understood as the cost of power trading between the virtual power plant and the main grid. In this embodiment, the power time price within the time period can be obtained to determine the power cost of the virtual power plant and the main grid for capacity trading.

[0127] Specifically, in this embodiment, the power cost is obtained The method can be:

[0128]

[0129] in, Describes the cost of energy trading between VPP and the main grid, ρ t It can be expressed as the predicted electricity price data for time period t, which can enable the virtual power plant VPP to sell electricity according to ρ t Get benefits, and when purchasing electricity, t Bear the cost, where β represents the premium coefficient of the electricity purchase cost; is the power exchanged with the main grid during time period t.

[0130] In this embodiment, the electricity cost can be determined by obtaining the transaction electricity volume and predicted electricity price data in a certain period of time.

[0131] S340: Determine the equipment operation cost according to the equipment status data and the predicted power data.

[0132] Among them, the equipment operating cost can be understood as the total operating cost of the equipment within a set time range. Specifically, the equipment operating cost in this embodiment may include the total operating cost of both the power generation equipment and the energy storage equipment. In this embodiment, the method of determining the equipment operating cost according to the equipment status data and the predicted power data can be to determine the operating cost of the power generation equipment and the energy storage equipment according to the status data of the power generation equipment and the energy storage equipment and the predicted load data and output data, and then superimpose the operating cost of the power generation equipment and the operating cost of the energy storage equipment to obtain the equipment operating cost.

[0133] In this embodiment, optionally, the equipment operating cost includes the operating cost of power generation equipment and energy storage equipment; the power generation equipment includes fossil power generation, photovoltaic power generation and wind power generation; the equipment operating cost is determined respectively according to the equipment status data and the predicted power data, including: determining the operating costs of fossil power generation, photovoltaic power generation, wind power generation and energy storage equipment respectively according to the equipment status data and the predicted power data; the operating costs of fossil power generation, photovoltaic power generation, wind power generation and energy storage equipment are integrated to obtain the equipment operating cost.

[0134] Among them, fossil power generation can also be called fossil fuel power generation, which refers to a power generation method that uses fossil fuels such as coal, oil, and natural gas to burn to generate heat energy, which is then converted into electrical energy. Photovoltaic power generation can be understood as a power generation method that uses the photoelectric effect to convert solar energy into electrical energy. Wind power generation can be understood as a power generation method that uses wind energy to convert into electrical energy. In this embodiment, wind power generation can be generated by wind turbines.

[0135] In this embodiment, the operating costs of fossil power generation, photovoltaic power generation, wind power generation and energy storage equipment are determined according to the equipment status data and predicted power data. It can be understood as the factors affecting the operation of fossil power generation, photovoltaic power generation, wind power generation and energy storage equipment according to the output parameters of the corresponding equipment and the predicted power data, and combined with other set cost coefficients of the equipment to determine the operating costs of fossil power generation, photovoltaic power generation, wind power generation and energy storage equipment. Further, in this embodiment, the operating cost of fossil power generation can reflect the use of traditional energy, while photovoltaic power generation and wind turbine power generation can be used to represent the operating cost of renewable energy, and the operating cost of energy storage equipment can represent the efficiency loss during energy storage and release.

[0136] In this embodiment, the specific method of determining the operating costs of fossil power generation, photovoltaic power generation, wind power generation and energy storage equipment respectively according to the equipment status data and the predicted power data is as follows:

[0137]

[0138]

[0139] In formula (3-2), the total operating cost of power generation equipment and power storage equipment can be composed of the operating costs of fossil power generation, photovoltaic power generation, wind power generation and energy storage equipment. Formula (3-3) represents the operating cost of fossil power generation units, where: Characterizes the output power at time T. m refers to the specific type of pollutant, and M can represent the total number of pollutant types. Parameter β m It can represent the cost of treating each unit of pollutant m, α m It can represent the amount of pollutants m emitted by coal-fired power units when producing each unit of electricity. Formula (3-4) represents the operating cost of photovoltaic power generation, where K v is the operating cost coefficient of the PV generator, which reflects the cost incurred by the operation and maintenance activities of the PV generator during its life cycle; It is represented by the power generated by the ith photovoltaic generator at time t. Formula (3-5) represents the operating cost of wind turbines for wind power generation, where K wIt can be expressed as the operating cost factor of the wind turbine, which reflects the cost incurred by the wind turbine due to operation and maintenance activities during its life cycle; It is represented by the power generated by the i-th wind turbine at time t. Equation (3-6) represents the operating cost of the energy storage system (ESS), Kb 1 It can represent the operating cost coefficient of energy storage equipment ESS, Kb 2 It can be deduced to represent the maintenance cost coefficient of the energy storage equipment ESS, E b Indicates the rated power of the energy storage device ESS; It is represented as the charging and discharging power of the energy storage device. Further, formula (3-5) defines the power generation limit of the coal power unit, while (3-6) and (3-7) enforce the rising and falling constraints of the coal power generation unit respectively.

[0140] S350: construct an objective function according to electricity cost and equipment operation cost.

[0141] Among them, the objective function can be a function constructed based on the power cost and the equipment operating cost. In this embodiment, the total operating cost formed by the power cost and the equipment operating cost is used to construct the objective function by minimizing the total operating cost of the entire energy scheduling system. In this embodiment, the objective function constructed based on the power cost and the equipment operating cost is specifically:

[0142]

[0143] in, It can be expressed as the operating cost of the power generation equipment and energy storage equipment corresponding to the generator set. The electricity cost of the virtual power plant for electricity trading with the main grid can be used. In this embodiment, the objective function formula (3-1) aims to minimize the total operating cost within a predetermined time range T, including the operating costs of the power generation equipment and energy storage equipment of the generator set in (3-2) and the electricity cost (3-7) of the cost of buying and selling electricity with the main grid.

[0144] S360. Determine the energy dispatch strategy of the virtual power plant based on the objective function and constraints.

[0145] The constraint conditions may be pre-set. The constraint conditions in this embodiment may be constraint conditions for establishing the objective function. The energy scheduling strategy may be understood as an energy adjustment strategy for minimizing the total operating cost of the virtual power plant within a set time range based on the objective function and the constraint conditions.

[0146] Specifically, the constraint conditions in this embodiment may be:

[0147]

[0148] Furthermore, in this embodiment, the problem of the optimal energy scheduling strategy is characterized and solved by constructing nonlinear programming model formulas (3-1) to (3-15). The energy storage device ESS is modeled using (3-11) to (3-13). In this model, the energy storage device ESS is allowed to schedule its discharge and charging power in advance. Expression (3-11) defines the charge and discharge power limit, expression (3-12) models the state of charge SOC (State of Charge) as a function of the charge and discharge power, and expression (3-13) limits the energy stored in the energy storage device ESS to avoid the effects of overcharging and over-discharging. This paper also models the main network import and export power limit through the expression in formula (3-14). For the entire VPP system model, one of the key factors is to ensure power balance, that is, the total power consumption of each distributed energy DG unit must match the total power demand of the user, as shown in formula (3-15).

[0149] In this embodiment, an energy adjustment strategy that minimizes the total operating cost of the virtual power plant within a set time range can be determined based on the objective function and set constraints.

[0150] S370. Determine the target power generation data of the virtual power plant through energy dispatch strategy.

[0151] Among them, the target power generation data can be understood as coal-fired power generation data. Specifically, the target power generation data in this embodiment can be coal-fired power generation power. In this embodiment, the specific way to determine the target power generation data of the virtual power plant through the energy scheduling strategy is that the energy scheduling strategy can be expressed as a Markov decision process (Markov Decision Process, MDP), and the Markov decision process is processed by the reinforcement learning algorithm to obtain the coal-fired power generation data of the virtual power plant. Among them, the Markov decision process can be composed of a five-tuple (S, A, P, R, γ), where S represents the state set of the system, A represents the action behavior set, P represents the state transition probability function, R represents the reward function, and γ represents the discount factor. In this embodiment, a deep deterministic policy gradient DDPG (Deep Deterministic Policy Gradient) algorithm based on the WU (Watt Utility) activation function is designed in the Markov decision process MDP to adapt to the nonlinear characteristics and complex environmental dynamics in the energy system. This activation function enhances the algorithm's ability to explore the problem space by simulating the uncertainty of energy markets and environmental conditions and their impact on virtual power plant (VPP) scheduling decisions, thereby improving adaptability and decision-making capabilities in a changing environment and ensuring the real-time and economic performance of the scheduling plan.

[0152] In this embodiment, optionally, the target power generation data of the virtual power plant is determined by an energy scheduling strategy, including: modeling the energy scheduling strategy as a reinforcement learning agent; processing the reinforcement learning agent through a reinforcement learning algorithm to obtain the target power generation data of the virtual power plant.

[0153] Among them, the reinforcement learning agent can be understood as an RL agent, that is, a reinforcement learning (RL) agent, which is a machine learning entity that learns the optimal strategy by interacting with the environment. The reinforcement learning algorithm can be an algorithm that can process the reinforcement learning agent. In this embodiment, the reinforcement learning algorithm can be DRL (Deep Reinforcement Learning). In this embodiment, the DRL algorithm is a model-free method that can learn this dynamic from the interaction and improve the efficiency and performance of the algorithm. In this embodiment, the energy scheduling strategy is modeled as an RL agent, and the reinforcement learning agent is processed by the DRL algorithm to finally determine the target power generation data of the virtual power plant. The state information of the system provides an important basis for the operation scheduling unit. In this embodiment, the processing is carried out by the reinforcement learning method, and the algorithm can continuously optimize the scheduling strategy and gradually improve the overall performance and efficiency of the system.

[0154] In this embodiment, optionally, the reinforcement learning agent is processed by a reinforcement learning algorithm to obtain the target power generation data of the virtual power plant, including: taking the system energy state at the current moment as the state set, and taking the scheduling actions of the power generation equipment and the energy storage equipment as the action behavior set; determining the transition probability function of the system energy state at the next moment according to the state set and the action behavior set; determining the reward function of the system energy state based on the objective function and the transition probability function; and determining the target power generation data of the virtual power plant according to the reward function.

[0155] Among them, the current moment may be moment t. In this embodiment, the system energy state at moment t may be taken as a state set, and the scheduling actions of the power generation equipment and the energy storage equipment may be taken as an action behavior set. Among them, the next moment may be understood as the next moment of the current moment. Exemplarily, if the current moment is moment t, the next moment may be moment t+1. In this embodiment, the transition probability function of the system energy state at the next moment may be predicted and determined based on the state set and action behavior set at the current moment.

[0156] Specifically, in this embodiment, a system state at time t can be defined as a state set And define the dispatching actions of distributed energy DG of power generation equipment and energy storage equipment ESS as action behavior set In this energy dispatch strategy, the agent does not directly control the affairs between the energy system of the virtual power plant and the main grid. Instead, after performing any operation, it will export / import power from the main grid to maintain power balance in accordance with the maximum in / out power capacity limit. In this embodiment, the specific method of determining the transition probability function of the system energy state at the next moment according to the state set and action behavior set is as follows: the state set s at a given time t t and action set a t , determine the energy system transfer to the next state s t+1 The transition probability function is:

[0157]

[0158] This part of the example simulates the dynamics of the energy system. In model-based algorithms, uncertainty is predicted by a fixed value or sampling from a prior probability distribution. In contrast, the DRL algorithm is a model-free method that can learn this dynamics from interactions. To guide learning, the environment must provide a reward r t , so that the agent can quantify the reward or penalty for the action taken, thereby guiding the indicator of the policy update direction. In the optimal energy scheduling strategy of the energy system of the virtual power plant, the reward function should guide the agent to take actions that minimize the operating cost while enforcing the power balance constraint, which can be achieved by using the reward function of formula (3-17). In this embodiment, the specific way to determine the reward function of the system energy state based on the objective function and the transition probability function is:

[0159]

[0160] where ΔP corresponds to the power imbalance at time step t and is defined as

[0161]

[0162] Among them, σ 1 and σ 2 Used to control the magnitude and trade-off between minimizing the operating cost and the penalty incurred in case of power imbalance.

[0163] The goal of the RL algorithm in this embodiment is to find an optimal scheduling strategy π * (random or deterministic) to maximize the total expected discount reward of the formulated MDP. In this embodiment, the optimal energy scheduling strategy can be determined according to the reward function, and the coal-fired power generation capacity of the virtual power plant can be determined according to the optimal energy scheduling strategy; wherein, the optimal scheduling strategy can be implemented as follows:

[0164]

[0165] In this embodiment, a penalty mechanism is used to implement power balance constraints, and its application in the dynamic scheduling problem of virtual power plants (VPPs) is explored. The DRL algorithm obtains reward signals by interacting with the environment, and uses these rewards to update the parameters of the Critic deep neural network based on the time difference theory. According to the policy gradient theory, the update of these parameters can further affect the DNN parameters of the Actor. Therefore, in this embodiment, a deep reinforcement learning framework with continuous decision-making capabilities is used to solve the dynamic scheduling problem of VPPs.

[0166] In addition, in this embodiment, an optimization scheduling algorithm based on deep reinforcement learning is obtained through simulation testing, which is combined with the Watt Utility function and the deep deterministic policy gradient algorithm to achieve efficient and economic optimization scheduling of the microgrid; among them, the WU-DDPG algorithm performs more stably and efficiently under the condition of electricity price fluctuations, and can better adapt to market changes and maximize economic benefits.

[0167] The technical solution of the embodiment of the present invention obtains the equipment status data of the virtual power plant and the historical power data of a set time; inputs the historical load data, historical output data and historical electricity price data of the set time into the automatic time series prediction network to obtain the predicted load data, predicted output data and predicted electricity price data. Obtain the electricity cost; determine the equipment operation cost according to the equipment status data and the predicted power data respectively; construct the objective function according to the electricity cost and the equipment operation cost; determine the energy dispatching strategy of the virtual power plant according to the objective function and the constraints; determine the target power generation data of the virtual power plant through the energy dispatching strategy. This technical solution, based on the automatic time series prediction network, performs load, energy output and electricity price forecasts by participating in various stages of source, load, network and storage. Through accurate forecasting, the power grid can more flexibly respond to peak and valley load changes, ensure the stable operation of the power grid and the reliable power supply of users, and improve the economic benefits and operational stability of the virtual power plant.

[0168] A common virtual power plant VPP includes energy conversion equipment, energy storage equipment, energy transmission network and load, etc. For example, the energy management architecture diagram of the virtual dispatching system in the virtual power plant in this embodiment is as follows: Figure 4As shown. In this architecture, energy conversion equipment, i.e., distributed power sources, include solar photovoltaic power generation PV (Photovoltaics), wind turbine WT (Wind Turbine) and fossil fuel power generation, and other equipment such as energy storage system ESS (Energy Storage System), load EL (Electric Load), grid equipment, etc.; loads may include general loads and sensitive loads. Regarding rectifiers, area energy routers, local coordination controllers, and intelligent edge devices, which are conventional equipment in virtual power plants, specific settings can be made according to actual needs. This embodiment proposes an innovative VPP dual architecture, which is divided into a commercial virtual power plant CVPP and a technical virtual power plant TVPP. The two work together to ensure the maximization of technical efficiency and economic benefits. CVPP is mainly responsible for formulating power supply strategies based on the comprehensive real-time market electricity price, predicted load information, and predicted output of renewable energy. Subsequently, the strategy is passed to TVPP to optimize the power generation of distributed power generation units and calculate the power generation cost, and the optimization results are fed back to CVPP. Through the two-way dynamic interaction between CVPP and TVPP, VPP can maximize economic benefits while ensuring technical efficiency. Virtual power plants (VPPs) can interact with the main power grid and use coordinated optimization control methods to import or export electricity from the main grid. By utilizing multiple energy sources in the system, they can solve the problem of surplus or shortage of demand, reduce energy processing costs, and improve energy utilization.

[0169] Embodiment 3

[0170] Figure 5 is a structural diagram of a virtual power plant dispatching device based on a perception information model provided according to Embodiment 3 of the present invention. The device is configured in a virtual power plant dispatching system, and the virtual power plant dispatching system includes a dispatching management module, a dispatching control module and at least one virtual power plant; Figure 5 As shown, the device comprises:

[0171] A data acquisition module 510, for acquiring device status data of the virtual power plant and historical power data of a set period based on the perception information model;

[0172] The data prediction module 520 is used to input the historical power data into the set prediction network to obtain the predicted power data;

[0173] The energy scheduling strategy determination module 530 is used to determine the energy scheduling strategy of the virtual power plant based on the predicted power data and the equipment status data, and determine the target power generation data of the virtual power plant through the energy scheduling strategy.

[0174] Optionally, the device also includes a perception model construction module, which is used to obtain basic attributes and parameter information of each device in the virtual power plant before obtaining the equipment status data of the virtual power plant and historical power data of a set period of time; and construct a perception information model based on the basic attributes and parameter information so that the perception information model shares the status data of each device.

[0175] Optionally, the historical power data includes historical load data, historical output data and historical electricity price data; the prediction network is set to be an automatic time series prediction network;

[0176] The data prediction module 520 is specifically used for:

[0177] The historical load data, historical output data and historical electricity price data of a set time period are input into the automatic time series prediction network to obtain the predicted load data, predicted output data and predicted electricity price data.

[0178] Optionally, the energy scheduling strategy determination module 530 includes:

[0179] an acquisition unit, for acquiring electricity cost;

[0180] an equipment cost determination unit, for determining equipment operating costs according to equipment status data and predicted power data;

[0181] An objective function building unit, used for building an objective function according to electricity cost and equipment operation cost;

[0182] The scheduling strategy determination unit is used to determine the energy scheduling strategy of the virtual power plant according to the objective function and constraint conditions.

[0183] Optionally, the equipment operating cost includes the operating cost of power generation equipment and energy storage equipment; the power generation equipment includes fossil power generation, photovoltaic power generation and wind power generation;

[0184] The equipment cost determination unit is specifically used to determine the operating costs of fossil power generation, photovoltaic power generation, wind power generation and energy storage equipment respectively according to the equipment status data and the predicted power data; the operating costs of fossil power generation, photovoltaic power generation, wind power generation and energy storage equipment are integrated to obtain the equipment operating costs.

[0185] Optionally, the energy scheduling strategy determination module 530 includes:

[0186] A modeling unit for modeling energy scheduling strategies as reinforcement learning agents;

[0187] The processing unit is used to process the reinforcement learning agent through the reinforcement learning algorithm to obtain the target power generation data of the virtual power plant.

[0188] Optionally, the processing unit is specifically used to take the system energy state at the current moment as a state set, and the scheduling actions of the power generation equipment and the energy storage equipment as an action behavior set; determine the transition probability function of the system energy state at the next moment according to the state set and the action behavior set; determine the reward function of the system energy state based on the objective function and the transition probability function; and determine the target power generation data of the virtual power plant according to the reward function.

[0189] A virtual power plant scheduling device based on a perception information model provided in an embodiment of the present invention can execute a virtual power plant scheduling method based on a perception information model provided in any embodiment of the present invention, and has functional modules and beneficial effects corresponding to the execution method.

[0190] Embodiment 4

[0191] Figure 6 1 is a schematic diagram of the structure of an electronic device provided according to Embodiment 4 of the present invention. The electronic device 10 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0192] like Figure 6 As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0193] A number of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0194] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The processor 11 executes the various methods and processes described above, such as a virtual power plant scheduling method based on a perception information model.

[0195] In some embodiments, the virtual power plant scheduling method based on the perception information model can be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the virtual power plant scheduling method based on the perception information model described above can be performed. Alternatively, in other embodiments, the processor 11 can be configured to execute the virtual power plant scheduling method based on the perception information model in any other appropriate manner (for example, by means of firmware).

[0196] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0197] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0198] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0199] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0200] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0201] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0202] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0203] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A virtual power plant scheduling method based on perception information model, It is characterized in that The method is applied to a virtual power plant dispatching system, which includes a dispatching management module, a dispatching control module and at least one virtual power plant; the method includes: Acquire equipment status data of the virtual power plant and historical power data of a set period based on a perception information model; Inputting the historical power data into a set prediction network to obtain predicted power data; An energy dispatching strategy of the virtual power plant is determined based on the predicted power data and the equipment status data, and target power generation data of the virtual power plant is determined through the energy dispatching strategy.

2. The method according to claim 1, It is characterized in that Before obtaining the equipment status data of the virtual power plant and the historical power data of a set period, the method further includes: Obtaining basic attributes and parameter information of each device in the virtual power plant; A perception information model is constructed according to the basic attributes and the parameter information, so that the perception information model shares the status data of each device.

3. The method according to claim 1, It is characterized in that The historical power data includes historical load data, historical output data and historical electricity price data; the set prediction network is an automatic time series prediction network; Inputting the historical power data into a set prediction network to obtain predicted power data includes: The historical load data, historical output data and historical electricity price data of a set time period are input into the automatic time series prediction network to obtain the predicted load data, predicted output data and predicted electricity price data.

4. The method according to claim 3, It is characterized in that Determining an energy dispatch strategy of the virtual power plant based on the predicted power data and the device status data includes: Cost of obtaining electricity; Determine the equipment operation cost respectively according to the equipment status data and the predicted power data; constructing an objective function based on the electricity cost and the equipment operation cost; The energy dispatch strategy of the virtual power plant is determined according to the objective function and the constraints.

5. The method according to claim 4, It is characterized in that The equipment operating costs include the operating costs of power generation equipment and energy storage equipment; power generation equipment includes fossil power generation, photovoltaic power generation and wind power generation; Determining the equipment operation cost according to the equipment status data and the predicted power data respectively includes: Determine the operating costs of fossil power generation, photovoltaic power generation, wind power generation and energy storage equipment respectively according to the equipment status data and the predicted power data; The operating costs of the fossil power generation, the photovoltaic power generation, the wind power generation and the energy storage device are integrated to obtain the equipment operating cost.

6. The method according to claim 5, It is characterized in that Determining target power generation data of the virtual power plant through the energy dispatch strategy includes: Modeling the energy scheduling strategy as a reinforcement learning agent; The reinforcement learning agent is processed by a reinforcement learning algorithm to obtain target power generation data of the virtual power plant.

7. The method according to claim 6, It is characterized in that The reinforcement learning agent is processed by a reinforcement learning algorithm to obtain target power generation data of the virtual power plant, including: Taking the system energy state at the current moment as a state set, and taking the scheduling actions of the power generation equipment and the energy storage equipment as an action behavior set; Determine the transition probability function of the system energy state at the next moment according to the state set and the action behavior set; Determine a reward function for the system energy state based on the objective function and the transition probability function; The target power generation data of the virtual power plant is determined according to the reward function.

8. A virtual power plant dispatching device based on perception information model, It is characterized in that The device is configured in a virtual power plant dispatching system, and the virtual power plant dispatching system includes a dispatching management module, a dispatching control module and at least one virtual power plant; the device includes: A data acquisition module, used to acquire equipment status data of the virtual power plant and historical power data of a set period; A data prediction module, used for inputting the historical power data into a set prediction network to obtain predicted power data; An energy scheduling strategy determination module is used to determine the energy scheduling strategy of the virtual power plant based on the predicted power data and the equipment status data, and determine the target power generation data of the virtual power plant through the energy scheduling strategy.

9. An electronic device, It is characterized in that The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the virtual power plant scheduling method based on the perception information model as described in any one of claims 1-7.

10. A computer-readable storage medium, It is characterized in that The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the virtual power plant scheduling method based on a perception information model as described in any one of claims 1 to 7 when executed.

Citation Information

Cited By

  • Virtual power plant optimal scheduling method, system and equipment combined with vehicle network interaction and medium

    CN119831243A