Multi-microgrid multi-element scheduling method, system and device based on virtual energy storage and medium

Abstract

This invention relates to the field of multi-microgrid management technology, and discloses a multi-microgrid multi-element scheduling method, system, device, and medium based on virtual energy storage. The method involves inputting real-time operational data of various virtual energy storage systems into a target Markov game model, and scheduling the target multi-microgrid according to the obtained scheduling strategy. The construction of the target Markov game model includes: constructing a microgrid operation benefit model and an operation cost model based on the analysis results of the operational data of various virtual energy storage systems; subsequently constructing a microgrid optimization model for each microgrid, as well as a multi-microgrid optimization model; training the initial Markov game model obtained from the microgrid optimization model and the multi-microgrid optimization model locally; and using a federated learning algorithm to iteratively update the noise gradient of each agent to obtain the target Markov game model. This method enables unified multi-element energy scheduling for multi-microgrids while ensuring data security.

Description

Technical Field

[0001] This invention relates to the field of multi-microgrid management technology, and in particular to a multi-microgrid multi-element scheduling method, system, device and medium based on virtual energy storage. Background Technology

[0002] Against the backdrop of energy transition, large-scale, highly volatile renewable energy grid integration has become a significant trend in power system development. However, the intermittent and random nature of these renewable energy sources poses a severe challenge to the stable operation of the power system. Multi-microgrids, composed of interconnected microgrids, offer an effective solution to these challenges due to their flexible operating modes and collaborative capabilities. Through energy interaction and coordinated operation among multiple microgrids, multi-microgrids can mitigate the fluctuations in renewable energy levels to a certain extent, thereby improving the stability and reliability of the power system.

[0003] To achieve energy interaction and coordinated operation among multiple microgrids, source-load coordinated scheduling of multiple microgrids is required. However, the source-load resources of each microgrid in the current multi-microgrid system are different, the proportion and capacity of distributed renewable energy configuration on the power supply side are significantly different, and the spatiotemporal characteristics and adjustability potential of various distributed controllable loads on the user side are different. Existing technologies lack a unified multi-dimensional optimization scheduling strategy, and there is a lack of effective protection for various equipment data and user-side data collected during the scheduling process, which poses a risk of data leakage.

[0004] Therefore, how to achieve multi-micronet multi-optimization scheduling and improve the data security of multi-micronets has become a technical problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0005] This invention provides a method, system, device, and medium for multi-microgrid multi-dimensional scheduling based on virtual energy storage, in order to solve the technical problems of how to achieve multi-dimensional optimized scheduling of multi-microgrids and improve the data security of multi-microgrids, thereby achieving the effect of multi-dimensional optimized scheduling of multi-microgrids and improving the data security of multi-microgrids.

[0006] In a first aspect, the present invention provides a multi-dimensional scheduling method for multi-microgrids based on virtual energy storage. The method includes: determining the virtual energy storage type of the target multi-microgrid, inputting first real-time operating data of each type of virtual energy storage and second real-time operating data of the target multi-microgrid into a target Markov game model, and scheduling the target multi-microgrid according to the obtained scheduling strategy. The construction and training process of the target Markov game model includes:

[0007] Based on the analysis results of the operation data of various virtual energy storage systems, an operation model is constructed for each microgrid within the target multi-microgrid. The operation model includes an operation benefit model and an operation cost model.

[0008] Based on the output results of the operation benefit model and the operation cost model, a microgrid optimization model is constructed for each microgrid, and the target multi-microgrid is optimized based on the microgrid optimization model of each microgrid to obtain a multi-microgrid optimization model.

[0009] Based on the microgrid optimization model and the multi-microgrid optimization model, an initial Markov game model is constructed, and the initial Markov game model is trained locally to obtain the noise gradient of each agent. The initial Markov game model defines the agent for each microgrid, and the microgrid optimization model is converted into the local model, and the multi-microgrid optimization model is converted into a global model.

[0010] The noise gradient of each agent is updated using a federated learning algorithm, and the local model is iteratively trained based on the obtained global model noise gradient until the local model meets the preset conditions, thereby obtaining the target Markov game model.

[0011] Preferably, the step of constructing an operational model for each microgrid within the target multi-microgrid based on the operational data analysis results of various types of virtual energy storage includes:

[0012] Based on the switching characteristics of electric vehicles under the guidance of time-of-use pricing, the operating data of electric vehicles are analyzed to construct a virtual energy storage relationship for electric vehicles. The virtual energy storage relationship for electric vehicles includes: a virtual capacitance value relationship for electric vehicles, a virtual state of charge relationship for electric vehicles, and a battery aging degree relationship for electric vehicles.

[0013] The process of controlling the rotor kinetic energy by adjusting the motor state of the fan is equivalent to the charging and discharging process of the fan. The fan operation data is analyzed to construct a virtual energy storage relationship for the fan. The virtual energy storage relationship for the fan includes: a virtual capacitance value relationship for the fan and a virtual state of charge relationship for the fan.

[0014] The response processes of controllable loads that can be reduced, transferred, and shifted are equivalent to the charging and discharging processes of the controllable loads. The operating data of the controllable loads are analyzed to construct virtual energy storage relationships for the controllable loads. These virtual energy storage relationships include: virtual power value relationships for reduced loads, virtual power value relationships for transferred loads, and virtual power value relationships for shifted loads.

[0015] Based on the energy conversion process of the hydrogen storage device, the operating data of the hydrogen storage device is analyzed to construct a virtual energy storage relationship for the hydrogen storage device. The virtual energy storage relationship for the hydrogen storage device includes: the hydrogen production capacity relationship of the alkaline electrolyzer, the hydrogen combustion capacity relationship of the proton exchange membrane fuel cell, the virtual state of charge relationship of the hydrogen storage tank, and the virtual capacitance value relationship of the hydrogen storage device.

[0016] Based on the virtual energy storage relationship of the electric vehicle and the virtual energy storage relationship of the hydrogen storage device, an operational efficiency model for each microgrid within the target multi-microgrid is constructed.

[0017] Based on the virtual energy storage relationship of electric vehicles, the virtual energy storage relationship of wind turbines, the virtual energy storage relationship of controllable loads, and the virtual energy storage relationship of hydrogen storage devices, an operating cost model for each microgrid within the target multi-microgrid is constructed.

[0018] Preferably, the step of constructing an operational efficiency model for each microgrid within the target multi-microgrid based on the virtual energy storage relationship of the electric vehicle and the virtual energy storage relationship of the hydrogen storage device includes:

[0019] According to the electric vehicle virtual energy storage relationship, the first data set of electric vehicle virtual energy storage participating in charging scheduling of each microgrid in the target multi-microgrid is obtained. Based on the first data set, the electric vehicle charging and discharging benefit relationship of each microgrid in the target multi-microgrid is obtained. The first data set includes at least: the total number of first electric vehicle virtual energy storage participating in charging scheduling and the charging amount of electric vehicle virtual energy storage.

[0020] The second data set of virtual energy storage of the hydrogen storage device for each microgrid is obtained according to the virtual energy storage relationship of the hydrogen storage device. Based on the second data set, the hydrogen production efficiency relationship of each microgrid is obtained. The second data set includes at least: daily hydrogen sludge mass and hydrogen production capacity of alkaline electrolyzer.

[0021] Obtain the electricity sales power of each microgrid to other microgrids, and based on the electricity sales power, obtain the electricity sales benefit relationship of each microgrid;

[0022] Based on the electric vehicle charging and discharging benefit relationship, the hydrogen production benefit relationship, and the electricity sales benefit relationship, an operational benefit model for each microgrid within the target multi-microgrid is constructed.

[0023] Preferably, the step of constructing an operating cost model for each microgrid within the target multi-microgrid based on the virtual energy storage relationship of the electric vehicle, the virtual energy storage relationship of the wind turbine, the virtual energy storage relationship of the controllable load, and the virtual energy storage relationship of the hydrogen storage device includes:

[0024] According to the electric vehicle virtual energy storage relationship, a third data set of electric vehicle virtual energy storage participating in discharge scheduling for each microgrid in the target multi-microgrid is obtained. Based on the third data set, the electric vehicle virtual energy storage scheduling cost relationship of each microgrid in the target multi-microgrid is obtained. The third data set includes at least: the total number of second electric vehicle virtual energy storage participating in discharge scheduling and the electric vehicle virtual energy storage discharge capacity.

[0025] According to the wind turbine virtual energy storage relationship, the fourth data set of wind turbine virtual energy storage wind regulation for each microgrid is obtained. Based on the fourth data set, the wind turbine virtual energy storage scheduling cost relationship for each microgrid is obtained. The fourth data set includes at least: the total number of wind turbine virtual energy storage, the output power of wind turbine virtual energy storage before wind regulation, and the output power of wind turbine virtual energy storage after wind regulation.

[0026] According to the controllable load virtual energy storage relationship, the fifth data set for the controllable load virtual energy storage of each microgrid participating in peak shaving is obtained. Based on the fifth data set, the controllable load virtual energy storage scheduling cost relationship of each microgrid is obtained. The fifth data set includes at least: the load power value that can be reduced and scheduled, the load power value that can be transferred and scheduled, and the load power value that can be shifted and scheduled.

[0027] The sixth data set for the virtual energy storage of the hydrogen storage device in each microgrid participating in peak shaving is obtained according to the virtual energy storage relationship of the hydrogen storage device. Based on the sixth data set, the virtual energy storage scheduling cost relationship of the hydrogen storage device in each microgrid is obtained. The sixth data set includes at least: the power of the alkaline electrolyzer, the operating status of the alkaline electrolyzer, the power of the proton exchange membrane fuel cell, and the operating status of the proton exchange membrane fuel cell.

[0028] Obtain other operating cost formulas for each of the microgrids, wherein the other operating costs include at least: microgrid network loss cost formula, battery cost formula, photovoltaic operation and maintenance cost formula, and microgrid electricity purchase cost formula;

[0029] Based on the virtual energy storage scheduling cost relationship of electric vehicles, the virtual energy storage scheduling cost relationship of wind turbines, the virtual energy storage scheduling cost relationship of controllable loads, the virtual energy storage scheduling cost relationship of hydrogen storage devices, and the other operating cost relationships, an operating cost model for each microgrid within the target multi-microgrid is constructed.

[0030] Preferably, the step of constructing an initial Markov game model based on the microgrid optimization model and the multi-microgrid optimization model, and performing local model training on the initial Markov game model to obtain the noise gradient of each agent, includes:

[0031] Each microgrid within the target multi-microgrid is defined as an intelligent agent, and the state space of each microgrid is obtained for observation. The state space includes the overall state space of the microgrid and the local state space of the microgrid.

[0032] The action space of each of the aforementioned intelligent agents is obtained, and the action space includes at least: the first charging and discharging power of each type of virtual energy storage, the second charging and discharging power of the battery, and the power purchased and sold by the microgrid;

[0033] Based on the multi-micronet optimization model, a reward function for the target multi-micronet is constructed, the reward function including: optimization target achievement reward and constraint violation penalty reward;

[0034] The microgrid optimization model is converted into a local model, and the multi-microgrid optimization model is converted into a global model.

[0035] Based on the state space, the action space, the reward function, the local model, and the global model, an initial Markov game model is constructed.

[0036] A deep deterministic policy gradient reinforcement learning method is used to train the initial Markov game model locally, obtaining the latest gradient of each local model, and then Laplace noise is introduced into the latest gradient to obtain the noise gradient of each agent.

[0037] Preferably, the reward function is expressed as:

[0038]

[0039] in, Represents the reward function, The weight coefficients represent the optimization objectives of the multi-micronet optimization model. This represents the objective function of the multi-micronet optimization model. This represents the weighting coefficient for the penalty of exceeding the constraint limit. This represents the penalty cost for exceeding the limits of various constraints in the multi-micronet optimization model.

[0040] Preferably, the step of updating the noise gradient of each agent using a federated learning algorithm and iteratively training the local model based on the obtained global model noise gradient until the local model meets preset conditions to obtain the target Markov game model includes:

[0041] The noise gradient of each of the intelligent agents is uploaded to the global server to update the global model parameters, thereby obtaining the global model noise gradient.

[0042] The global model noise gradient is broadcast globally to update the noise gradient of each agent, and iterative training is performed based on the updated noise gradient until the local model meets preset conditions to obtain the target Markov game model. The preset conditions include: the local model loss curve converges and / or the local model payoff curve converges.

[0043] Secondly, the present invention also provides a multi-microgrid multi-element scheduling system based on virtual energy storage to realize the multi-microgrid multi-element scheduling method based on virtual energy storage described above. The system includes: a multi-element scheduling module.

[0044] The multi-dimensional scheduling module is used to determine the virtual energy storage type of the target multi-microgrid, and input the first real-time operation data of each type of virtual energy storage and the second real-time operation data of the target multi-microgrid into the target Markov game model, and schedule the target multi-microgrid according to the obtained scheduling strategy.

[0045] The multi-dimensional scheduling module includes: a microgrid operation model construction unit, a multi-microgrid optimization model construction unit, an initial Markov game model construction unit, and a model iteration training unit;

[0046] The microgrid operation model construction unit is used to construct an operation model for each microgrid within the target multi-microgrid based on the operation data analysis results of various types of virtual energy storage. The operation model includes an operation benefit model and an operation cost model.

[0047] The multi-microgrid optimization model construction unit is used to construct a microgrid optimization model for each microgrid based on the output results corresponding to the operation benefit model and the operation cost model, and to optimize the target multi-microgrid based on the microgrid optimization model of each microgrid to obtain the multi-microgrid optimization model.

[0048] The initial Markov game model construction unit is used to construct an initial Markov game model based on the microgrid optimization model and the multi-microgrid optimization model, and to perform local model training on the initial Markov game model to obtain the noise gradient of each agent. The initial Markov game model defines the agent for each microgrid, and converts the microgrid optimization model into the local model and the multi-microgrid optimization model into a global model.

[0049] The model iterative training unit is used to update the noise gradient of each agent using a federated learning algorithm, and to iteratively train the local model based on the obtained global model noise gradient until the local model meets the preset conditions, thereby obtaining the target Markov game model.

[0050] Thirdly, the present invention also provides a computer device, the computer device including a memory, a processor and a transceiver, which are connected to each other via a bus; the memory is used to store a set of computer program instructions and data, and to transmit the stored data to the processor, the processor executing the computer program instructions stored in the memory to execute the above-described multi-microgrid multi-element scheduling method based on virtual energy storage.

[0051] Fourthly, the present invention also provides a computer-readable storage medium storing a computer program, which, when run, implements the multi-microgrid multi-element scheduling method based on virtual energy storage described above.

[0052] This application provides a multi-microgrid multi-element scheduling method, system, device, and medium based on virtual energy storage. Compared with the prior art, the beneficial effects of the embodiments of this application are as follows:

[0053] This application discloses a multi-microgrid multi-element scheduling method based on virtual energy storage. Electric vehicles, wind turbines, controllable loads, and hydrogen storage devices are virtually represented as energy storage. Virtual capacitance values ​​are used as standardized energy regulation parameters, simplifying the operating parameters of flexible and adjustable resources in multi-microgrids, thereby improving the perception level of the energy regulation potential of multi-microgrids. Using virtual state of charge as the state criterion for virtual energy storage provides a unified quantitative benchmark for assessing the frequency regulation and peak shaving capabilities of multi-microgrids. The Markov game algorithm fully leverages the synergistic potential of multi-microgrid sources, grids, loads, and storage, perceiving the distributed renewable energy configuration on the power source side and various distributed controllable loads on the user side of each microgrid. This enables unified multi-element energy scheduling of multi-microgrids, meeting the collaborative regulation needs of multi-microgrids, significantly improving collaborative optimization benefits, achieving rational resource allocation, improving the utilization efficiency of multi-microgrid resources, and ensuring the data security of each microgrid. Attached Figure Description

[0054] Figure 1 This is a schematic diagram of the construction and training process steps of the target Markov game model provided in a preferred embodiment of the present invention;

[0055] Figure 2 This is a schematic diagram of a multi-microgrid multi-element scheduling system based on virtual energy storage provided in a preferred embodiment of the present invention;

[0056] Figure 3 This is an internal structural diagram of the computer device in an embodiment of the present invention;

[0057] Figure label:

[0058] 1-Microgrid operation model construction unit, 2-Multi-microgrid optimization model construction unit, 3-Initial Markov game model construction unit, 4-Model iterative training unit. Detailed Implementation

[0059] The embodiments of the present invention are described in detail below with reference to the accompanying drawings. The embodiments are provided for illustrative purposes only and should not be construed as limiting the invention. The accompanying drawings are for reference and illustration only and do not constitute a limitation on the scope of protection of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of this invention. In the description of this invention, the terms "first," "second," "third," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, a feature defined with "first," "second," "third," etc., may explicitly or implicitly include one or more of that feature. In the description of this invention, unless otherwise stated, "a plurality of" means two or more.

[0060] In the description of this invention, it should be noted that, unless otherwise expressly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to fixed connections, detachable connections, or integral connections; they can refer to mechanical connections or electrical connections; they can refer to direct connections or indirect connections through an intermediate medium; and they can refer to communication within two components. The terms "vertical," "horizontal," "left," "right," "upper," "lower," and similar expressions used herein are for illustrative purposes only and do not indicate or imply that the device or component referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as limiting the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items. Those skilled in the art will understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0061] In the description of this invention, it should be noted that, unless otherwise defined, all technical and scientific terms used in this invention have the same meaning as commonly understood by one of ordinary skill in the art. The terminology used in this specification is for the purpose of describing specific embodiments only and is not intended to limit the invention. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0062] In an embodiment of the present invention, a multi-microgrid multi-element scheduling method based on virtual energy storage is provided, the method comprising:

[0063] The virtual energy storage type of the target multi-microgrid is determined, and the first real-time operating data of each type of virtual energy storage and the second real-time operating data of the target multi-microgrid are input into the target Markov game model. The target multi-microgrid is then scheduled according to the obtained scheduling strategy. Figure 1 The diagram illustrates the steps involved in constructing and training a target Markov game model, including:

[0064] S1. Based on the analysis results of the operational data of various types of virtual energy storage, construct an operational model for each microgrid within the target multi-microgrid. The operational model includes an operational benefit model and an operational cost model. In a preferred embodiment of this application, virtual energy storage is introduced. Virtual energy storage is a novel energy storage concept that integrates and coordinates various dispersed, adjustable resources to simulate the functions of traditional physical energy storage. Based on the operating characteristics of the microgrid, virtual energy storage includes at least: electric vehicle virtual energy storage, wind turbine virtual energy storage, controllable load virtual energy storage, and hydrogen storage device virtual energy storage.

[0065] Based on the rechargeable and dischargeable characteristics of electric vehicles (EVs), EVs are virtualized as virtual energy storage. By analyzing the switching characteristics of EVs under time-of-use pricing, EV operating data is used to construct virtual energy storage equations. These equations include: virtual capacitance value equation, virtual state of charge equation, and battery aging degree equation. The virtual energy storage equations are expressed as follows:

[0066]

[0067] in, Indicates time period, Indicates the total time period. express Virtual capacitance value of electric vehicles during the time period express Active power of electric vehicle charging and discharging during a given period express DC voltage at the charging station during the specified time period. express Virtual state of charge of electric vehicles during a given time period. express Virtual state of charge of electric vehicles during a given time period. This represents the charging and discharging loss coefficient of an electric vehicle battery. Indicates the battery capacity of an electric vehicle. express The degree of aging of electric vehicle batteries over a period of time express The degree of aging of electric vehicle batteries over a period of time This represents the degradation coefficient of electric vehicle batteries. Indicates by Time period to The aging capacity of electric vehicle batteries over a given period of time.

[0068] A virtual energy storage model for electric vehicles is constructed based on the virtual energy storage relationship of electric vehicles, so as to perform model-based control of the virtual energy storage of electric vehicles.

[0069] Since the kinetic energy of a wind turbine is affected by its rotor speed, the process of controlling the rotor's kinetic energy by adjusting the motor state is equivalent to the charging and discharging process of the wind turbine. The wind turbine's operating data is analyzed to construct a virtual energy storage equation, which includes a virtual capacitance value equation and a virtual state of charge equation. The virtual energy storage equation is expressed as follows:

[0070]

[0071] in, express Virtual capacitance value of the wind turbine during the time period. This represents the maximum power point tracking coefficient of the wind turbine. This indicates the rated angular velocity of the fan rotor. express The angular velocity of the fan rotor during the time period, This indicates the voltage offset at the wind turbine terminals. express Virtual state of charge of wind turbines during specific time periods.

[0072] A virtual energy storage model for wind turbines is constructed based on the virtual energy storage relationship to enable model-based control of virtual energy storage in wind turbines.

[0073] Based on the adjustability of controllable loads, the loads participating in demand response can be divided into reduceable loads, transferable loads, and shiftable loads. The response processes of reduceable, transferable, and shiftable loads are equivalent to the charging and discharging processes of controllable loads. Controllable load operating data is analyzed to construct virtual energy storage equations for controllable loads. These equations include virtual power value equations for reduceable loads, transferable loads, and shiftable loads. The virtual energy storage equations for controllable loads are expressed as follows:

[0074]

[0075] in, express The virtual power value of the load can be reduced during certain periods. express The power output of the load before it participates in the dispatching process can be reduced during certain periods. express The load power that can be reduced during the time period that participates in scheduling. express Virtual power value of load transferable during the time period express Power available for load transfer before it participates in dispatching during a given period. express The transferable load power that participates in scheduling during a given time period. , and Represents 0-1 variables, express The increased power that can be transferred during a certain period of time. express The power that can be transferred during a certain period of time is reduced. express The virtual power value of the load can be shifted during the time period. express Power that can be shifted from the time period before being included in the dispatching process. express The power of shiftable loads that participate in scheduling during specific time periods. express The increased power of the load can be shifted during the time period. express The power that can be shifted during a certain period of time is reduced by the load.

[0076] A controllable load virtual energy storage model is constructed based on the controllable load virtual energy storage relationship to enable model-based control of controllable load virtual energy storage.

[0077] Based on the electro-hydrogen-electro-energy conversion process of hydrogen production via alkaline electrolyzer, hydrogen storage via compression in hydrogen storage tank, and hydrogen combustion output power in proton exchange membrane fuel cell within the hydrogen storage device, and by analyzing the operating data of the hydrogen storage device, a virtual energy storage relationship for the hydrogen storage device is constructed. This virtual energy storage relationship includes: the hydrogen production rate relationship of the alkaline electrolyzer, the hydrogen combustion rate relationship of the proton exchange membrane fuel cell, the virtual state of charge relationship of the hydrogen storage tank, and the virtual capacitance value relationship of the hydrogen storage device. The virtual energy storage relationship of the hydrogen storage device is expressed as follows:

[0078]

[0079] in, express Hydrogen production capacity of alkaline electrolyzer during a given period This indicates the conversion efficiency of the alkaline electrolyzer. express Power of the alkaline electrolyzer during the time period This indicates the conversion efficiency of the alkaline electrolytic cell converter. This indicates the operating status of the alkaline electrolytic cell. This indicates the lower heating value of hydrogen. express Hydrogen consumption per time period in proton exchange membrane fuel cells express The power of a proton exchange membrane fuel cell over a given period of time. This indicates the operating status of the proton exchange membrane fuel cell. This indicates the power generation efficiency of a proton exchange membrane fuel cell. This indicates the efficiency of the converter. express Virtual state of charge of hydrogen storage tank during a given period express Virtual state of charge of hydrogen storage tank during a given period This indicates the maximum hydrogen storage capacity of the hydrogen storage tank. express Voltage of the hydrogen storage device during a given time period.

[0080] A virtual energy storage model for the hydrogen storage device is constructed based on the virtual energy storage relationship to enable model-based control of the virtual energy storage of the hydrogen storage device.

[0081] Furthermore, based on the virtual energy storage relationships of electric vehicles and hydrogen storage devices, a microgrid operation benefit model is constructed. The peak-shaving benefits of wind turbine virtual energy storage and controllable load virtual energy storage fall within the scope of the electricity market. Therefore, in the preferred embodiment of this application, for each microgrid, considering the profits from the participation of electric vehicle virtual energy storage and hydrogen storage device virtual energy storage in dispatch, electric vehicle charging and discharging benefit relationships, hydrogen production benefit relationships, and electricity sales benefit relationships are constructed.

[0082] Specifically, based on the electric vehicle virtual energy storage relationship, the first data set of electric vehicle virtual energy storage participating in charging scheduling for each microgrid within the target multi-microgrid is obtained. The first data set includes at least: the total number of electric vehicle virtual energy storage participating in charging scheduling and the charging amount of electric vehicle virtual energy storage. Based on the first data set, the electric vehicle charging and discharging benefit relationship for each microgrid within the target multi-microgrid is obtained. The electric vehicle charging and discharging benefit relationship is expressed as:

[0083]

[0084] in, Indicates the first microgrids Indicates the first Electric vehicle charging and discharging efficiency of microgrids This represents the total number of virtual energy storage units of the first electric vehicle participating in the charging schedule. Indicates the first Virtual energy storage for electric vehicles express Time period Electricity price for electric vehicle charging within a microgrid express Time period Within the first microgrid Virtual energy storage charging capacity for electric vehicles.

[0085] The second data set of the hydrogen storage device for each microgrid is obtained based on the virtual energy storage relationship of the hydrogen storage device. The second data set includes at least: the daily hydrogen sludge mass and the hydrogen production capacity of the alkaline electrolyzer. Based on the second data set, the hydrogen production benefit relationship of each microgrid is obtained, which is expressed as follows:

[0086]

[0087] in, Indicates the first Hydrogen production efficiency of microgrids This represents the byproduct revenue coefficient of a hydrogen storage device. Indicates the contents of the hydrogen storage tank Hydrogen mass during the period, Indicates the daily hydrogen sludge mass. This indicates the spot price of hydrogen.

[0088] Obtain the electricity sales power of each microgrid to other microgrids. Based on the electricity sales power, derive the electricity sales benefit formula for each microgrid. The electricity sales benefit formula is expressed as:

[0089]

[0090] in, Indicates the first Electricity sales revenue of a microgrid express Time period The electricity price sold by one microgrid to other microgrids. express Time period The amount of electricity sold by a microgrid to other microgrids.

[0091] Based on the relationships between the charging and discharging efficiency of electric vehicles, the hydrogen production efficiency, and the electricity sales efficiency, a microgrid operation efficiency model is constructed. The relationships in the microgrid operation efficiency model are expressed as follows:

[0092]

[0093] Furthermore, based on the virtual energy storage relationships for electric vehicles, wind turbines, controlled loads, and hydrogen storage devices, an operating cost model for each microgrid within the target multi-microgrid is constructed. In a preferred embodiment of this application, for each microgrid, considering the operating conditions of devices such as batteries and photovoltaics within the microgrid, virtual energy storage scheduling cost relationships for hydrogen storage devices, wind turbines, controlled loads, and hydrogen storage devices are constructed, along with microgrid network loss cost relationships, battery cost relationships, photovoltaic operation and maintenance cost relationships, and microgrid electricity purchase cost relationships. Specifically, based on the virtual energy storage relationships for electric vehicles, a third data set for the electric vehicle virtual energy storage participating in discharge scheduling for each microgrid is obtained. The third data set includes at least: the total number of second electric vehicle virtual energy storage participating in discharge scheduling and the discharge amount of electric vehicle virtual energy storage. Based on the third data set, the electric vehicle virtual energy storage scheduling cost relationship for each microgrid within the target multi-microgrid is obtained, expressed as follows:

[0094]

[0095] in, Indicates the first The cost of virtual energy storage dispatching for electric vehicles in a microgrid This represents the total number of virtual energy storage units of the second electric vehicle participating in the discharge scheduling. express Time period Electricity price compensation for virtual energy storage discharge of electric vehicles in microgrids express Time period Within the first microgrid The amount of virtual energy stored and discharged by an electric vehicle.

[0096] Based on the wind turbine virtual energy storage relationship, obtain the fourth data set for wind turbine virtual energy storage wind regulation for each microgrid. The fourth data set includes at least: the total number of wind turbine virtual energy storage units, the output power of the wind turbines before wind regulation, and the output power of the wind turbines after wind regulation. Based on the fourth data set, obtain the wind turbine virtual energy storage scheduling cost relationship for each microgrid. The wind turbine virtual energy storage scheduling cost relationship is expressed as follows:

[0097]

[0098] in, Indicates the first The cost of virtual energy storage dispatching for wind turbines in a microgrid This represents the total virtual energy storage of the wind turbines. Indicates the first Virtual energy storage for individual wind turbines Indicates the first The first microgrid Each fan's air adjustment factor express Time period The first microgrid The power output of a single wind turbine before virtual energy storage and wind regulation. express Time period The first microgrid The power output of a single wind turbine after virtual energy storage and wind regulation. Indicates the first The first microgrid The operation and maintenance cost of each wind turbine.

[0099] Based on the controllable load virtual energy storage relationship, the fifth data set for controllable load virtual energy storage participation in peak shaving for each microgrid is obtained. The fifth data set includes at least: the power values ​​of loads that can be reduced and dispatched, the power values ​​of loads that can be transferred and dispatched, and the power values ​​of loads that can be shifted and dispatched. Based on the fifth data set, the controllable load virtual energy storage dispatch cost relationship for each microgrid is obtained, and the controllable load virtual energy storage dispatch cost relationship is expressed as follows:

[0100]

[0101] in, Indicates the first The cost of virtual energy storage dispatching for controllable loads in a microgrid express Time period The ability of microgrids to participate in dispatch can reduce load power. Indicates the first The compensation coefficient for virtual energy storage that can reduce load in a microgrid. express Time period The transferable load power of a microgrid participating in dispatch Indicates the first Compensation coefficient for virtual energy storage of transferable loads in a microgrid express Time period The power of loads that can be shifted and dispatched by a microgrid Indicates the first The compensation coefficient for virtual energy storage of transferable loads in a microgrid.

[0102] Based on the virtual energy storage relationship of the hydrogen storage devices, the sixth data set for peak shaving participation of the hydrogen storage devices in each microgrid is obtained. The sixth data set includes at least: the power of the alkaline electrolyzer, the operating status of the alkaline electrolyzer, the power of the proton exchange membrane fuel cell, and the operating status of the proton exchange membrane fuel cell. Based on the sixth data set, the virtual energy storage scheduling cost relationship of the hydrogen storage devices in each microgrid is obtained, and the virtual energy storage scheduling cost relationship of the hydrogen storage devices is expressed as follows:

[0103]

[0104] in, Indicates the first Virtual energy storage dispatch cost of hydrogen storage devices in microgrids This represents the total virtual energy storage of the hydrogen storage device. Indicates the first Virtual energy storage for a hydrogen storage device Indicates the first Within the first microgrid The daily water consumption for hydrogen production by a single hydrogen storage unit. Indicates the first The unit water cost for a microgrid express Time period Within the first microgrid The power of the alkaline electrolyzer in the hydrogen storage unit, express Time period Within the first microgrid The operating status of the alkaline electrolyzer in the hydrogen storage unit. Indicates the first Within the first microgrid The operation and maintenance cost of the alkaline electrolyzer in a hydrogen storage unit. express Time period Within the first microgrid The power of a proton exchange membrane fuel cell for a hydrogen storage device, express Time period Within the first microgrid The operation status of the proton exchange membrane fuel cell in the hydrogen storage device. Indicates the first Within the first microgrid The operation and maintenance cost of a proton exchange membrane fuel cell for a hydrogen storage device.

[0105] The relationship between microgrid network loss costs is constructed and expressed as follows:

[0106]

[0107] in, Indicates the first Microgrid loss cost of a microgrid express Time period The grid loss of a microgrid Indicates the first The network loss cost coefficient of a microgrid.

[0108] The battery cost relationship is constructed as follows:

[0109]

[0110] in, Indicates the first The cost of batteries for a microgrid Indicates the first The total number of batteries in each microgrid Indicates the first One battery, express Time period Within the first microgrid The charging and discharging power of each battery. Indicates the first Within the first microgrid The maintenance cost of a single battery Indicates the first Within the first microgrid The investment cost of a single battery Indicates the first Within the first microgrid The number of charge / discharge cycles of a battery. Indicates the first Within the first microgrid The depth of discharge of a battery. Indicates the first Within the first microgrid The rated capacity of each battery.

[0111] The relationship between photovoltaic (PV) operation and maintenance costs is constructed and expressed as follows:

[0112]

[0113] in, Indicates the first The cost of photovoltaic operation and maintenance within a microgrid Indicates the first The total number of photovoltaic cells in each microgrid Indicates the first A photovoltaic, express Time period The first microgrid The output of a photovoltaic power plant Indicates the first The first microgrid The operation and maintenance cost coefficient of a photovoltaic system.

[0114] The relationship between microgrid electricity purchase costs is constructed and expressed as follows:

[0115]

[0116] in, Indicates the first The cost of electricity purchase for a microgrid express Time period Electricity purchases for individual microgrids express Time period The electricity purchase price for each microgrid.

[0117] Furthermore, based on the cost relationships for virtual energy storage scheduling of electric vehicles, wind turbines, controllable loads, hydrogen storage devices, microgrid network loss, battery costs, photovoltaic operation and maintenance, and microgrid electricity purchase, a microgrid operation cost model is constructed for each microgrid. The relationships in the microgrid operation cost model are expressed as follows:

[0118]

[0119] S2. Construct a microgrid optimization model for each microgrid based on the output results of the operation benefit model and the operation cost model, and optimize the target multi-microgrid based on the microgrid optimization model for each microgrid to obtain a multi-microgrid optimization model; In a preferred embodiment of this application, a microgrid optimization model is constructed for each microgrid. The microgrid optimization model is determined by the relationship between the microgrid operation benefit model and the microgrid operation cost model. The relationship of the microgrid optimization model is expressed as follows:

[0120]

[0121] in, Indicates the first Optimization benefits of individual microgrids.

[0122] Furthermore, taking the minimization of the difference between the operating benefits and operating costs of the multi-microgrid as the optimization objective, an optimization objective function is constructed. Based on this objective function, a multi-microgrid optimization model is built. The objective function expression of the multi-microgrid optimization model is as follows:

[0123]

[0124] in, This represents the objective function of the multi-micronet optimization model. This represents the collection of microgrids contained in a multi-microgrid.

[0125] The various constraints of the multi-micronet optimization model are expressed as follows:

[0126]

[0127] in, and These represent the upper and lower limits of the virtual energy storage state of charge of the electric vehicle, respectively. and These represent the upper and lower limits of the aging degree of electric vehicle batteries, respectively. and These represent the upper and lower limits of the virtual energy storage state of charge of the wind turbine, respectively. This indicates the upper limit of the load that can be reduced. and These represent the upper and lower limits of transferable load, respectively. and These represent the upper and lower limits of the transferable load, respectively. and These represent the upper and lower limits of the virtual energy storage state of charge of the hydrogen storage device, respectively. express The state of charge of the battery during a given period. and These represent the upper and lower limits of the battery's state of charge, respectively. Indicates line power. Indicates the maximum transmission power of the line. and They represent Time period The charging and discharging power of batteries within a microgrid express Time period The power of the inherent load of a microgrid.

[0128] The multi-microgrid optimization model constructed in this application virtualizes electric vehicles, wind turbines, controllable loads, and hydrogen storage devices as energy storage components. Using virtual capacitance values ​​as standardized energy regulation parameters, it simplifies the operating parameters of flexible and adjustable resources in multi-microgrids, thereby improving the perception level of the energy regulation potential of multi-microgrids. Using virtual state of charge as the state criterion for virtual energy storage provides a unified quantitative benchmark for evaluating the frequency regulation and peak shaving capabilities of multi-microgrids.

[0129] S3. Based on the microgrid optimization model and the multi-microgrid optimization model, an initial Markov game model is constructed, and local model training is performed on the initial Markov game model to obtain the noise gradient of each agent. The initial Markov game model defines the agent for each microgrid, and the microgrid optimization model is converted into the local model, while the multi-microgrid optimization model is converted into a global model. In a preferred embodiment of this application, a Markov game algorithm is used for multi-dimensional scheduling of multi-microgrids, each microgrid is defined as an agent, a local model is constructed based on the microgrid optimization model of each microgrid, and a global model is constructed based on the multi-microgrid optimization model. The Markov game algorithm is a mathematical model used to describe the interactive decision-making of multiple agents in a dynamic environment. The basic elements of the Markov game algorithm include state space, action space, transition probability, and reward function. The state space contains all possible states of the game objects. In the preferred embodiment of this application, the overall observation information of each microgrid is the same, while the local observation information varies. A state space is constructed from the microgrid observation information at both the local and global levels. The global state space focuses on the unit operation and maintenance costs and time of each device, including the spot hydrogen price, unit water cost, compensation coefficients for virtual energy storage that can reduce load, compensation coefficients for virtual energy storage that can transfer load, compensation coefficients for virtual energy storage that can be shifted, electricity purchase price, and time period. The local state space focuses on the electricity sales price and electric vehicle charging and discharging price of each microgrid, photovoltaic and wind turbine output, virtual energy storage state of charge, and inherent load, including the electricity sales price to other microgrids, electric vehicle charging price, electric vehicle virtual energy storage discharge compensation price, photovoltaic output, battery state of charge, electric vehicle virtual energy storage state of charge, electric vehicle aging level, wind turbine virtual energy storage state of charge, and hydrogen storage device virtual energy storage state of charge.

[0130] The state space is represented as:

[0131]

[0132] in, Indicates the first The local state space of a microgrid Indicates the first The global state space of a microgrid Indicates the first The state space of a microgrid.

[0133] Furthermore, the action space of each microgrid is constructed. Based on the optimizable variables of various devices within each microgrid, namely the charging and discharging power of various virtual energy storage devices, the charging and discharging power of batteries, and the power purchased and sold by the microgrid, the action space of multiple microgrids is established. The action space is represented as follows:

[0134]

[0135] Furthermore, a reward function for the multi-micronet is constructed. In a preferred embodiment of this application, the reward function includes an optimization objective achievement reward and a constraint violation penalty reward. Taking full account of the optimization objective function and constraints of the multi-micronet, the expression of the reward function is as follows:

[0136]

[0137] in, Represents the reward function, The weight coefficients represent the optimization objective of the multi-micronet optimization model. To constrain the weighting coefficient of the penalty for exceeding the limit, This represents the penalty cost for exceeding the limits of various constraints in the multi-micronet optimization model.

[0138] Based on the state space, action space, reward function, microgrid optimization model, and multi-microgrid optimization model, an initial Markov game model is constructed. This initial Markov game model includes a local model built using the microgrid optimization model and a global model built using the multi-microgrid optimization model. The state of the initial Markov game model at the next moment depends only on the current state and the current action taken, and is independent of past historical states. This allows the model to effectively handle decision-making problems in dynamic environments, simplifying the problem's complexity. Considering the interactions between multiple agents, each agent's decision not only affects its own reward but also the rewards of other agents and the state transitions of the entire multi-microgrid. The Markov game model effectively describes the dynamic changes of the multi-microgrid over time and how the actions of a particular microgrid affect the state transitions of the multi-microgrid. It evaluates and compares different scheduling strategies, obtains the optimal scheduling strategy for each microgrid under different states, achieves rational resource allocation, and improves the utilization efficiency of multi-microgrid resources.

[0139] Furthermore, a deep deterministic policy gradient reinforcement learning method is employed for local model training. Specifically, the parameters of the Actor network and Critic network for each agent are initialized. The Actor network is responsible for generating action policies, and the Critic network evaluates the value of the action policies generated by the Actor network, obtaining reward feedback, which is then fed back to the Actor network for policy optimization. Each agent obtains its own state space from the environment as input. The Actor network executes a set of actions to the Critic network, obtaining corresponding reward feedback and post-action state. The state space, actions, reward feedback, and post-action state are combined as experience samples and placed into an experience buffer. Each agent draws small batches of experience samples from the experience buffer for training and model updates, and each agent calculates the latest gradient of the model.

[0140] In a preferred embodiment of this application, to protect the privacy of the local model training results, Laplace noise is introduced into the latest gradient according to the Laplace mechanism in differential privacy to obtain the noise gradient of each agent. The noise gradient is represented as follows:

[0141]

[0142] in, Indicates the first Noise gradient of an agent, Indicates the first [number] after introducing noise. The network parameters of each agent (Actor). Indicates the number of noise introduced. Critic network parameters for each agent Indicates the first The initial Actor network parameters for each agent, Indicates the first Initial Critic network parameters for each agent Indicates the first Laplace noise of an agent.

[0143] S4. The noise gradient of each agent is updated using a federated learning algorithm, and the local model is iteratively trained based on the obtained global model noise gradient until the local model meets the preset conditions, thus obtaining the target Markov game model. In a preferred embodiment of this application, a federated learning algorithm is used for global model aggregation. According to the training framework of federated learning, the noise gradient of each agent is uploaded to the global server for global model parameter update, thus obtaining the global model noise gradient. The global model noise gradient is broadcast globally to update the noise gradient of the local model of each agent, and iterative training is performed based on the updated noise gradient until the local model loss curve converges and / or the local model reward curve converges, thus obtaining the target Markov game model.

[0144] After obtaining the target Markov game model, the first real-time operating data of virtual energy storage and the second real-time operating data of the target multi-microgrid are input into the target Markov game model to obtain an optimized scheduling strategy. The target multi-microgrid is then scheduled according to the optimized scheduling strategy. The first real-time operating data includes at least the parameters involved in constructing the virtual energy storage relationships for electric vehicles, wind turbines, controllable loads, and hydrogen storage devices. The second real-time operating data includes at least the parameters involved in constructing the microgrid network loss cost relationship, battery cost relationship, photovoltaic operation and maintenance cost relationship, and microgrid electricity purchase cost relationship. This fully leverages the synergistic potential of multi-microgrid generation, grid, load, and storage, sensing the distributed renewable energy configuration on the power source side and various distributed controllable loads on the user side of each microgrid. This enables unified multi-dimensional energy scheduling of the multi-microgrid, meeting the needs of multi-microgrid coordinated regulation, significantly improving synergistic optimization benefits, achieving rational resource allocation, improving the utilization efficiency of multi-microgrid resources, and ensuring the data security of each microgrid.

[0145] In a preferred embodiment of the present invention, the virtual energy storage type of the target multi-microgrid is determined, and the first real-time operation data of each type of virtual energy storage and the second real-time operation data of the target multi-microgrid are input into the target Markov game model. The target multi-microgrid is then scheduled according to the obtained scheduling strategy. The construction and training process of the target Markov game model includes: constructing an operation model for each microgrid within the target multi-microgrid based on the analysis results of the operation data of each type of virtual energy storage; the operation model includes an operation benefit model and an operation cost model; constructing a microgrid optimization model for each microgrid based on the output results corresponding to the operation benefit model and the operation cost model; and optimizing the microgrid model based on the microgrid optimization results of each microgrid. The model optimizes the target multi-microgrid to obtain a multi-microgrid optimization model. Based on the microgrid optimization model and the multi-microgrid optimization model, an initial Markov game model is constructed, and local model training is performed on the initial Markov game model to obtain the noise gradient of each agent. The initial Markov game model defines agents for each microgrid, and the microgrid optimization model is converted into a local model, while the multi-microgrid optimization model is converted into a global model. A federated learning algorithm is used to update the noise gradient of each agent, and the local model is iteratively trained based on the obtained global model noise gradient until the local model meets the preset conditions, thus obtaining the target Markov game model. The multi-microgrid multi-element scheduling method based on virtual energy storage disclosed in this application treats electric vehicles, wind turbines, controllable loads, and hydrogen storage devices as virtual energy storage, and uses virtual capacitance values ​​as standardized energy regulation parameters to simplify the operating parameters of flexible and adjustable resources in multi-microgrids, thereby improving the perception level of the energy regulation potential of multi-microgrids. Using virtual state of charge as the state criterion for virtual energy storage provides a unified quantitative benchmark for evaluating the frequency regulation and peak shaving capabilities of multi-microgrids. Markov game theory algorithms fully leverage the collaborative potential of multiple microgrids' sources, grids, loads, and storage. They sense the distributed renewable energy configuration on the power source side and various distributed controllable loads on the user side of each microgrid, enabling unified multi-dimensional energy dispatching for multiple microgrids. This meets the collaborative regulation and control needs of multiple microgrids, significantly improves collaborative optimization benefits, achieves rational resource allocation, enhances the utilization efficiency of multi-microgrid resources, and ensures the data security of each microgrid.

[0146] Accordingly, such as Figure 2 The diagram shows a multi-microgrid multi-electrode scheduling system based on virtual energy storage. Based on a multi-microgrid multi-electrode scheduling method based on virtual energy storage, this embodiment of the invention also provides a multi-microgrid multi-electrode scheduling system based on virtual energy storage, which implements the multi-microgrid multi-electrode scheduling method based on virtual energy storage disclosed in this embodiment of the invention. The system includes: a multi-electrode scheduling module.

[0147] The multi-dimensional scheduling module is used to determine the virtual energy storage type of the target multi-microgrid, and input the first real-time operation data of each type of virtual energy storage and the second real-time operation data of the target multi-microgrid into the target Markov game model, and schedule the target multi-microgrid according to the obtained scheduling strategy.

[0148] The multi-dimensional scheduling module includes: a microgrid operation model construction unit 1, a multi-microgrid optimization model construction unit 2, an initial Markov game model construction unit 3, and a model iteration training unit 4;

[0149] The microgrid operation model construction unit 1 is used to construct the operation model of each microgrid in the target multi-microgrid based on the operation data analysis results of various types of virtual energy storage. The operation model includes an operation benefit model and an operation cost model.

[0150] The multi-microgrid optimization model construction unit 2 is used to construct a microgrid optimization model for each microgrid based on the output results corresponding to the operation benefit model and the operation cost model, and to optimize the target multi-microgrid based on the microgrid optimization model of each microgrid to obtain the multi-microgrid optimization model.

[0151] The initial Markov game model construction unit 3 is used to construct an initial Markov game model based on the microgrid optimization model and the multi-microgrid optimization model, and to perform local model training on the initial Markov game model to obtain the noise gradient of each agent. The initial Markov game model defines the agent for each microgrid, and converts the microgrid optimization model into the local model and the multi-microgrid optimization model into a global model.

[0152] The model iterative training unit 4 is used to update the noise gradient of each agent using a federated learning algorithm, and to iteratively train the local model according to the obtained global model noise gradient until the local model meets the preset conditions, thereby obtaining the target Markov game model.

[0153] For specific limitations regarding a multi-microgrid multi-element scheduling system based on virtual energy storage, please refer to the above-described limitations regarding a multi-microgrid multi-element scheduling method based on virtual energy storage, which will not be repeated here. Those skilled in the art will recognize that the various modules and steps described in conjunction with the embodiments disclosed in this invention can be implemented in hardware, software, or a combination of both. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this invention.

[0154] like Figure 3As shown, an embodiment of the present invention provides a computer device including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the steps as described in the above embodiment of multi-microgrid multi-element scheduling based on virtual energy storage, for example... Figure 1 Steps S1 to S4 as described above.

[0155] Those skilled in the art will understand that the illustrations Figure 3 This is merely an example of a computer device and does not constitute a limitation on the computer device. It may include more or fewer components than shown, or combine certain components, or different components. For example, the computer device may also include input / output devices, network access devices, buses, etc.

[0156] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the computer device, connecting various parts of the computer device via various interfaces and lines.

[0157] The memory can be used to store the computer programs and / or modules. The processor implements various functions of the computer device by running or executing the computer programs and / or modules stored in the memory and by calling data stored in the memory. The memory may mainly include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the mobile phone (such as audio data, phonebook, etc.). In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0158] If the modules integrated into the computer device are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc.

[0159] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0160] Accordingly, embodiments of the present invention provide a computer-readable storage medium, the computer-readable storage medium including a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to perform steps in the multi-microgrid multi-element scheduling based on virtual energy storage as described in the above embodiments, for example... Figure 1 Steps S1 to S4 as described above.

[0161] In summary, the embodiments of this application provide a method, system, device, and medium for multi-microgrid multi-dimensional scheduling based on virtual energy storage, which solves the technical problems of how to achieve multi-dimensional optimized scheduling of multi-microgrids and improve the data security of multi-microgrids. The method includes: determining the virtual energy storage type of the target multi-microgrid, and inputting the first real-time operating data of various virtual energy storage types and the second real-time operating data of the target multi-microgrid into a target Markov game model; scheduling the target multi-microgrid according to the obtained scheduling strategy; the construction and training process of the target Markov game model includes: constructing an operating model for each microgrid within the target multi-microgrid based on the analysis results of the operating data of various virtual energy storage types, the operating model including an operating benefit model and an operating cost model; and scheduling the target multi-microgrid according to the operating benefit model and the operating cost model. The corresponding output results are used to construct a microgrid optimization model for each microgrid, and the target multi-microgrid is optimized based on the microgrid optimization model for each microgrid to obtain a multi-microgrid optimization model. Based on the microgrid optimization model and the multi-microgrid optimization model, an initial Markov game model is constructed, and local model training is performed on the initial Markov game model to obtain the noise gradient of each agent. The initial Markov game model defines agents for each microgrid, and the microgrid optimization model is converted into a local model, while the multi-microgrid optimization model is converted into a global model. A federated learning algorithm is used to update the noise gradient of each agent, and the local model is iteratively trained based on the obtained global model noise gradient until the local model meets preset conditions, thus obtaining the target Markov game model. The multi-microgrid multi-element scheduling method based on virtual energy storage disclosed in this application virtualizes electric vehicles, wind turbines, controllable loads, and hydrogen storage devices as energy storage, using virtual capacitance values ​​as standardized energy regulation parameters, simplifying the operating parameters of flexible and adjustable resources in the multi-microgrid, thereby improving the perception level of the energy regulation potential of the multi-microgrid. Using the virtual state of charge (VBC) as the state criterion for virtual energy storage provides a unified quantitative benchmark for assessing the frequency regulation and peak shaving capabilities of multi-microgrids. The Markov game algorithm fully leverages the synergistic potential of multi-microgrid sources, grids, loads, and storage, sensing the distributed renewable energy configuration on the power source side and various distributed controllable loads on the user side of each microgrid. This enables unified multi-dimensional energy dispatching across multiple microgrids, meeting the needs of coordinated regulation and control, significantly improving synergistic optimization benefits, achieving rational resource allocation, increasing the utilization efficiency of multi-microgrid resources, and ensuring data security for each microgrid.

[0162] The various embodiments in this specification are described in a progressive manner. For directly identical or similar parts of the embodiments, refer to each other. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments. It should be noted that the technical features of the above embodiments can be combined arbitrarily. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as the combination of these technical features does not contradict each other, it should be considered within the scope of this specification.

[0163] The embodiments described above are merely preferred embodiments of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the patent application. It should be noted that those skilled in the art can make various improvements and substitutions without departing from the technical principles of this application, and these improvements and substitutions should also be considered within the scope of protection of this application. Therefore, the scope of protection of this patent application should be determined by the scope of the claims.

Claims

1. A multi-microgrid multi-element scheduling method based on virtual energy storage, characterized in that, The method includes: The virtual energy storage type of the target microgrid is determined, and the first real-time operating data of each type of virtual energy storage and the second real-time operating data of the target microgrid are input into the target Markov game model. The target microgrid is then scheduled according to the obtained scheduling strategy. The construction and training process of the target Markov game model includes: Based on the analysis results of the operation data of various virtual energy storage systems, an operation model is constructed for each microgrid within the target multi-microgrid. The operation model includes an operation benefit model and an operation cost model. Based on the output results of the operation benefit model and the operation cost model, a microgrid optimization model is constructed for each microgrid, and the target multi-microgrid is optimized based on the microgrid optimization model of each microgrid to obtain a multi-microgrid optimization model. Based on the microgrid optimization model and the multi-microgrid optimization model, an initial Markov game model is constructed, and the initial Markov game model is trained locally to obtain the noise gradient of each agent. The initial Markov game model defines the agent for each microgrid, and the microgrid optimization model is converted into the local model, and the multi-microgrid optimization model is converted into a global model. The noise gradient of each agent is updated using a federated learning algorithm, and the local model is iteratively trained based on the obtained global model noise gradient until the local model meets the preset conditions, thereby obtaining the target Markov game model. The step of training the initial Markov game model locally to obtain the noise gradient for each agent includes: A deep deterministic policy gradient reinforcement learning method is used to train the initial Markov game model locally, obtaining the latest gradient of each local model, and then Laplace noise is introduced into the latest gradient to obtain the noise gradient of each agent.

2. The multi-microgrid multi-element scheduling method based on virtual energy storage as described in claim 1, characterized in that, The step of constructing an operational model for each microgrid within the target multi-microgrid based on the operational data analysis results of various types of virtual energy storage includes: Based on the switching characteristics of electric vehicles under the guidance of time-of-use pricing, the operating data of electric vehicles are analyzed to construct a virtual energy storage relationship for electric vehicles. The virtual energy storage relationship for electric vehicles includes: a virtual capacitance value relationship for electric vehicles, a virtual state of charge relationship for electric vehicles, and a battery aging degree relationship for electric vehicles. The process of controlling the rotor kinetic energy by adjusting the motor state of the fan is equivalent to the charging and discharging process of the fan. The fan operation data is analyzed to construct a virtual energy storage relationship for the fan. The virtual energy storage relationship for the fan includes: a virtual capacitance value relationship for the fan and a virtual state of charge relationship for the fan. The response processes of controllable loads that can be reduced, transferred, and shifted are equivalent to the charging and discharging processes of the controllable loads. The operating data of the controllable loads are analyzed to construct virtual energy storage relationships for the controllable loads. These virtual energy storage relationships include: virtual power value relationships for reduced loads, virtual power value relationships for transferred loads, and virtual power value relationships for shifted loads. Based on the energy conversion process of the hydrogen storage device, the operating data of the hydrogen storage device is analyzed to construct a virtual energy storage relationship for the hydrogen storage device. The virtual energy storage relationship for the hydrogen storage device includes: the hydrogen production capacity relationship of the alkaline electrolyzer, the hydrogen combustion capacity relationship of the proton exchange membrane fuel cell, the virtual state of charge relationship of the hydrogen storage tank, and the virtual capacitance value relationship of the hydrogen storage device. Based on the virtual energy storage relationship of the electric vehicle and the virtual energy storage relationship of the hydrogen storage device, an operational efficiency model for each microgrid within the target multi-microgrid is constructed. Based on the virtual energy storage relationship of electric vehicles, the virtual energy storage relationship of wind turbines, the virtual energy storage relationship of controllable loads, and the virtual energy storage relationship of hydrogen storage devices, an operating cost model for each microgrid within the target multi-microgrid is constructed.

3. The multi-microgrid multi-element scheduling method based on virtual energy storage as described in claim 2, characterized in that, The step of constructing an operational efficiency model for each microgrid within the target multi-microgrid based on the virtual energy storage relationship of the electric vehicle and the virtual energy storage relationship of the hydrogen storage device includes: According to the electric vehicle virtual energy storage relationship, the first data set of electric vehicle virtual energy storage participating in charging scheduling of each microgrid in the target multi-microgrid is obtained. Based on the first data set, the electric vehicle charging and discharging benefit relationship of each microgrid in the target multi-microgrid is obtained. The first data set includes at least: the total number of first electric vehicle virtual energy storage participating in charging scheduling and the charging amount of electric vehicle virtual energy storage. The second data set of virtual energy storage of the hydrogen storage device for each microgrid is obtained according to the virtual energy storage relationship of the hydrogen storage device. Based on the second data set, the hydrogen production efficiency relationship of each microgrid is obtained. The second data set includes at least: daily hydrogen sludge mass and hydrogen production capacity of alkaline electrolyzer. Obtain the electricity sales power of each microgrid to other microgrids, and based on the electricity sales power, obtain the electricity sales benefit relationship of each microgrid; Based on the electric vehicle charging and discharging benefit relationship, the hydrogen production benefit relationship, and the electricity sales benefit relationship, an operational benefit model for each microgrid within the target multi-microgrid is constructed.

4. The multi-microgrid multi-element scheduling method based on virtual energy storage as described in claim 2, characterized in that, The step of constructing an operating cost model for each microgrid within the target multi-microgrid based on the virtual energy storage relationships of electric vehicles, wind turbines, controllable loads, and hydrogen storage devices includes: According to the electric vehicle virtual energy storage relationship, a third data set of electric vehicle virtual energy storage participating in discharge scheduling for each microgrid in the target multi-microgrid is obtained. Based on the third data set, the electric vehicle virtual energy storage scheduling cost relationship of each microgrid in the target multi-microgrid is obtained. The third data set includes at least: the total number of second electric vehicle virtual energy storage participating in discharge scheduling and the electric vehicle virtual energy storage discharge capacity. According to the wind turbine virtual energy storage relationship, the fourth data set of wind turbine virtual energy storage wind regulation for each microgrid is obtained. Based on the fourth data set, the wind turbine virtual energy storage scheduling cost relationship for each microgrid is obtained. The fourth data set includes at least: the total number of wind turbine virtual energy storage, the output power of wind turbine virtual energy storage before wind regulation, and the output power of wind turbine virtual energy storage after wind regulation. According to the controllable load virtual energy storage relationship, the fifth data set for the controllable load virtual energy storage of each microgrid participating in peak shaving is obtained. Based on the fifth data set, the controllable load virtual energy storage scheduling cost relationship of each microgrid is obtained. The fifth data set includes at least: the load power value that can be reduced and scheduled, the load power value that can be transferred and scheduled, and the load power value that can be shifted and scheduled. The sixth data set for the virtual energy storage of the hydrogen storage device in each microgrid participating in peak shaving is obtained according to the virtual energy storage relationship of the hydrogen storage device. Based on the sixth data set, the virtual energy storage scheduling cost relationship of the hydrogen storage device in each microgrid is obtained. The sixth data set includes at least: the power of the alkaline electrolyzer, the operating status of the alkaline electrolyzer, the power of the proton exchange membrane fuel cell, and the operating status of the proton exchange membrane fuel cell. Obtain other operating cost formulas for each of the microgrids, wherein the other operating costs include at least: microgrid network loss cost formula, battery cost formula, photovoltaic operation and maintenance cost formula, and microgrid electricity purchase cost formula; Based on the virtual energy storage scheduling cost relationship of electric vehicles, the virtual energy storage scheduling cost relationship of wind turbines, the virtual energy storage scheduling cost relationship of controllable loads, the virtual energy storage scheduling cost relationship of hydrogen storage devices, and the other operating cost relationships, an operating cost model for each microgrid within the target multi-microgrid is constructed.

5. The multi-microgrid multi-element scheduling method based on virtual energy storage as described in claim 1, characterized in that, The step of constructing an initial Markov game model based on the microgrid optimization model and the multi-microgrid optimization model includes: Each microgrid within the target multi-microgrid is defined as an intelligent agent, and the state space of each microgrid is obtained for observation. The state space includes the overall state space of the microgrid and the local state space of the microgrid. The action space of each of the aforementioned intelligent agents is obtained, and the action space includes at least: the first charging and discharging power of each type of virtual energy storage, the second charging and discharging power of the battery, and the power purchased and sold by the microgrid; Based on the multi-micronet optimization model, a reward function for the target multi-micronet is constructed, the reward function including: optimization target achievement reward and constraint violation penalty reward; The microgrid optimization model is converted into a local model, and the multi-microgrid optimization model is converted into a global model. An initial Markov game model is constructed based on the state space, the action space, the reward function, the local model, and the global model.

6. The multi-microgrid multi-element scheduling method based on virtual energy storage as described in claim 5, characterized in that, The reward function is expressed as follows: in, Represents the reward function, The weight coefficients represent the optimization objectives of the multi-micronet optimization model. This represents the objective function of the multi-micronet optimization model. This represents the weighting coefficient for the penalty of exceeding the constraint limit. This represents the penalty cost for exceeding the limits of various constraints in the multi-micronet optimization model.

7. The multi-microgrid multi-element scheduling method based on virtual energy storage as described in claim 1, characterized in that, The method involves updating the noise gradient of each agent using a federated learning algorithm, and iteratively training the local model based on the obtained global model noise gradient until the local model meets preset conditions, thereby obtaining the target Markov game model, including: The noise gradient of each of the intelligent agents is uploaded to the global server to update the global model parameters, thereby obtaining the global model noise gradient. The global model noise gradient is broadcast globally to update the noise gradient of each agent, and iterative training is performed based on the updated noise gradient until the local model meets preset conditions to obtain the target Markov game model. The preset conditions include: the local model loss curve converges and / or the local model payoff curve converges.

8. A multi-microgrid multi-element scheduling system based on virtual energy storage, used to implement the multi-microgrid multi-element scheduling method based on virtual energy storage as described in any one of claims 1-7, characterized in that, The system includes: a multi-dimensional scheduling module; The multi-dimensional scheduling module is used to determine the virtual energy storage type of the target multi-microgrid, and input the first real-time operation data of each type of virtual energy storage and the second real-time operation data of the target multi-microgrid into the target Markov game model, and schedule the target multi-microgrid according to the obtained scheduling strategy. The multi-dimensional scheduling module includes: a microgrid operation model construction unit, a multi-microgrid optimization model construction unit, an initial Markov game model construction unit, and a model iteration training unit; The microgrid operation model construction unit is used to construct an operation model for each microgrid within the target multi-microgrid based on the operation data analysis results of various types of virtual energy storage. The operation model includes an operation benefit model and an operation cost model. The multi-microgrid optimization model construction unit is used to construct a microgrid optimization model for each microgrid based on the output results corresponding to the operation benefit model and the operation cost model, and to optimize the target multi-microgrid based on the microgrid optimization model of each microgrid to obtain the multi-microgrid optimization model. The initial Markov game model construction unit is used to construct an initial Markov game model based on the microgrid optimization model and the multi-microgrid optimization model, and to perform local model training on the initial Markov game model to obtain the noise gradient of each agent. The initial Markov game model defines the agent for each microgrid, and converts the microgrid optimization model into the local model and the multi-microgrid optimization model into a global model. The model iterative training unit is used to update the noise gradient of each agent using a federated learning algorithm, and to iteratively train the local model according to the obtained global model noise gradient until the local model meets the preset conditions to obtain the target Markov game model. The step of training the initial Markov game model locally to obtain the noise gradient for each agent includes: A deep deterministic policy gradient reinforcement learning method is used to train the initial Markov game model locally, obtaining the latest gradient of each local model, and then Laplace noise is introduced into the latest gradient to obtain the noise gradient of each agent.

9. A computer device, characterized in that: The computer device includes a memory, a processor, and a transceiver, which are connected to each other via a bus; the memory is used to store a set of computer program instructions and data, and to transmit the stored data to the processor, and the processor executes the computer program instructions stored in the memory to perform the multi-microgrid multi-element scheduling method based on virtual energy storage as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, which, when run, implements the multi-microgrid multi-element scheduling method based on virtual energy storage as described in any one of claims 1 to 7.

Images