Energy storage system energy scheduling optimization method based on demand response

Abstract

The invention discloses an energy storage system energy dispatching optimization method based on demand response, and relates to the technical field related to power dispatching optimization, and the method comprises the steps: carrying out the source load prediction of a target region, determining the energy supply and demand data, and introducing an energy storage element; executing a distribution decision based on the energy supply and demand data, and determining a first-order strategy; obtaining a reconstructed energy storage topology according to the aggregation coordination requirement, and transmitting the reconstructed energy storage topology back to the energy storage system; performing a multi-energy storage monomer aggregation decision based on the reconstructed energy storage topology, and determining a second-order strategy; and fusing the first-order strategy and the second-order strategy as an energy scheduling strategy, and executing energy storage scheduling management on the target area. The technical problems that in the prior art, a plurality of energy storage monomers are difficult to coordinate, and the energy supply and demand prediction accuracy is insufficient, so that the scheduling fineness, the real-time performance and the energy utilization efficiency of a power grid energy storage system are poor are solved. The technical effects of realizing fine scheduling of the intelligent power grid energy storage system and improving the scheduling reliability and the energy utilization efficiency are achieved.

Description

Technical Field

[0001] This application relates to the technical field of power dispatching optimization, and particularly to an energy dispatching optimization method for a energy storage system based on demand response. Background Art

[0002] Traditional power systems mainly rely on fossil fuel power generation, and their output is relatively stable and controllable. However, the intermittency and uncertainty of renewable energy sources such as wind energy and solar energy make the power supply and demand balance of the power system more complex. Energy storage systems can store and release electrical energy, providing additional power support during peak power demand or storing excess electrical energy during oversupply, thereby achieving the power supply and demand balance of the power system. However, for the current energy dispatching optimization of energy storage systems, centralized or distributed control strategies are usually adopted. Although the centralized control strategy can achieve global optimization, it faces problems of computational complexity and communication delay in large-scale systems; although the distributed control strategy can reduce computational complexity, it is often difficult to achieve global optimality when coordinating multiple energy storage units, thus affecting the overall energy utilization efficiency of the grid energy storage system.

[0003] In the related technologies at the present stage, there are technical problems such as difficulty in coordinating multiple energy storage units and insufficient accuracy in energy supply and demand prediction, resulting in poor dispatching fineness, real-time performance, and energy utilization efficiency of the grid energy storage system. Summary of the Invention

[0004] By providing an energy dispatching optimization method for an energy storage system based on demand response, this application solves the technical problems in the prior art, such as difficulty in coordinating multiple energy storage units and insufficient accuracy in energy supply and demand prediction, which lead to poor dispatching fineness, real-time performance, and energy utilization efficiency of the grid energy storage system, and achieves the technical effects of realizing refined dispatching of the smart grid energy storage system and improving dispatching reliability and energy utilization efficiency.

[0005] This application provides an energy dispatching optimization method for an energy storage system based on demand response, including: determining energy supply and demand data through source-load prediction of a target area and introducing energy storage elements, where the energy storage elements are the power regulation degree, adaptive balance degree, and capacity contribution degree of energy storage units; for the energy storage system, configuring a lightweight allocation module for the proximal data center of each energy storage unit, executing an allocation decision based on the energy supply and demand data, and determining a first-order strategy, where the first-order strategy includes an energy storage dispatching strategy and an aggregated coordination demand; according to the aggregated coordination demand, combining the energy storage elements to obtain a reconstructed energy storage topology and transmitting it back to the energy storage system; according to the lightweight coordination module built in the energy storage system, making a multi-energy storage unit aggregation decision based on the reconstructed energy storage topology to determine a second-order strategy; fusing the first-order strategy and the second-order strategy as an energy dispatching strategy to perform energy storage dispatching management on the target area.

[0006] In a possible implementation, the energy scheduling optimization method for the demand response-based energy storage system further performs the following processes: Interact with the energy scheduling records of the target area, conduct data-driven training, and construct an energy scheduling module; according to the source-load relationship, perform lightweight transfer learning on the energy scheduling module to determine the lightweight allocation module, where the mapping between the energy storage unit and the load side is used as the source-load relationship; perform lightweight transfer learning on the energy scheduling module through the aggregation analysis among the energy storage units to determine the lightweight coordination module.

[0007] In a possible implementation, the energy scheduling optimization method for the demand response-based energy storage system further performs the following processes: Deploy the lightweight allocation module in the proximal data center of each energy storage unit; receive the energy supply and demand data to the proximal data center of each energy storage unit, use the target energy storage unit as the source end, and use the allocation of the mapped load side as the decision-making target to determine the first-order strategy, where the load priority is used as the allocation condition.

[0008] In a possible implementation, the energy scheduling optimization method for the demand response-based energy storage system further performs the following processes: Interact with the energy storage topology of the target area; according to the aggregation coordination requirements, perform source-load topology node screening and topology reconstruction on the energy storage topology to determine the reconstructed energy storage topology.

[0009] In a possible implementation, the energy scheduling optimization method for the demand response-based energy storage system further performs the following processes: According to the aggregation coordination requirements, screen the scheduling source end and the scheduling load end based on the energy storage topology, where the supply-demand imbalance is used as the screening criterion; according to the energy storage elements, determine the element characteristic values of each scheduling source end according to the aggregation coordination requirements to determine the source end element coefficients; determine the load end consumption demand according to the aggregation coordination requirements; use the scheduling source end, the scheduling load end, the source end element coefficients, and the load end consumption demand to determine the reconstructed energy storage topology, where the reconstructed energy storage topology is a temporary decision-making topology.

[0010] In a possible implementation, the energy scheduling optimization method for the demand response-based energy storage system further performs the following processes: Deploy the lightweight coordination module in the data center of the energy storage system; through interactive communication, the data center determines the reconstructed energy storage topology and transmits it to the lightweight coordination module to perform the aggregation scheduling coordination of multiple energy storage units under the dual-mode parallel operation to determine the second-order strategy.

[0011] In a possible implementation, the energy scheduling optimization method for the demand response-based energy storage system further performs the following processes: Use energy scheduling and load transfer as the dual mode, and determine the second-order strategy through combined optimization based on the dual mode.

[0012] In a possible implementation, the energy scheduling optimization method for the energy storage system based on demand response further performs the following processes: determining an initialization strategy with single-mode scheduling; using the coordination target and coordination amplitude based on the source side as the adjustment method for the energy scheduling mode, and using the transfer target and transfer amplitude based on the load side as the adjustment method for the load transfer mode, adjusting and iterating the initialization strategy multiple times to determine multiple sets of optimization strategies; traversing the initialization strategy and the multiple sets of optimization strategies, and selecting the optimal strategy as the second-order strategy.

[0013] In a possible implementation, the energy scheduling optimization method for the energy storage system based on demand response further performs the following processes: interacting with the energy scheduling records of the target area to mine the periodic consumption curve; predicting the energy grid connection based on the environmental data and combining with the base station power conversion efficiency to determine the energy supply data; combining the periodic consumption curve to predict the load consumption volume based on time series to determine the energy demand data.

[0014] In a possible implementation, the energy scheduling optimization method for the energy storage system based on demand response further performs the following processes: determining the scheduling response information by performing energy storage scheduling tracking; evaluating the supply-demand balance of the scheduling response information to determine the supply-demand balance coefficient; adjusting the grid data of the energy storage system according to the supply-demand balance coefficient.

[0015] It is intended to use the energy scheduling optimization method for the energy storage system based on demand response proposed in this application to determine the energy supply and demand data by performing source-load prediction in the target area and introducing energy storage elements; execute the allocation decision based on the energy supply and demand data to determine the first-order strategy; obtain the reconstructed energy storage topology according to the aggregated coordination requirements and send it back to the energy storage system; perform the aggregation decision of multiple energy storage monomers based on the reconstructed energy storage topology to determine the second-order strategy; fuse the first-order strategy and the second-order strategy as the energy scheduling strategy to perform energy storage scheduling management on the target area. This solves the technical problems in the prior art, such as the difficulty in coordinating multiple energy storage monomers and the insufficient accuracy of energy supply and demand prediction, resulting in poor scheduling fineness, real-time performance, and energy utilization efficiency of the power grid energy storage system, and achieves the technical effects of realizing the refined scheduling of the smart grid energy storage system and improving the scheduling reliability and energy utilization efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings of the embodiments of the present invention will be briefly introduced below. Flowcharts are used in this application to illustrate the operations performed by the system according to the embodiments of the application. It should be understood that the operations in the front or below do not necessarily need to be executed precisely in sequence. On the contrary, according to the need, they can be executed in reverse order or simultaneously. At the same time, other operations can also be added to these processes, or one or several operations can be removed from these processes.

[0017] Figure 1 This is a schematic flowchart of the energy scheduling optimization method for a demand - response - based energy storage system provided by an embodiment of the present application; Figure 2 This is a schematic flowchart of determining the reconstructed energy storage topology in the energy scheduling optimization method for a demand - response - based energy storage system provided by an embodiment of the present application. Detailed implementation manners

[0018] The above description is only an overview of the technical solution of the present application. In order to be able to understand the technical means of the present application more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features and advantages of the present application more obvious and understandable, the following specifically gives the detailed implementation manners of the present application.

[0019] In order to make the purpose, technical solution and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limitations of the present application. All other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of the present application.

[0020] In the following description, "some embodiments" are involved, which describe a subset of all possible embodiments. However, it can be understood that "some embodiments" can be the same subset or different subsets of all possible embodiments, and can be combined with each other without conflict. The terms "first" and "second" involved are only used to distinguish similar objects and do not represent a specific order for the objects. The terms "including" and "having" and any variations thereof are intended to cover non - exclusive inclusion. For example, a process, method, product or server including a series of steps does not have to be limited to those steps clearly listed, but may include other steps not clearly listed or inherent to these processes, methods, products or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field of the present application. The terms used herein are only for the purpose of describing the embodiments of the present application.

[0021] An embodiment of the present application provides an energy scheduling optimization method for a demand - response - based energy storage system, as Figure 1 shown. The method includes: Step S100, by performing source - load prediction in a target area, determining energy supply - demand data, and introducing energy storage elements, where the energy storage elements are the power regulation degree, self - adaptive balance degree, and capacity contribution degree of energy storage monomers.

[0022] Preferably, predicting the source and load of the target area refers to predicting the power supply (source) and power demand (load) of the target area. Among them, the source includes traditional power generation (such as coal power and natural gas power generation) and renewable energy power generation (such as wind power and photovoltaic power generation), and the load includes the power demands of various users such as industrial, commercial, and residential users. Specifically, based on factors such as historical data, weather conditions, and economic indicators, algorithms such as machine learning and time series analysis are used for prediction. For example, the photovoltaic power generation can be predicted according to weather forecasts (such as sunshine intensity), and the power demand can be predicted according to historical load curves and factors such as holidays; then, analyze the supply-demand relationship of the power system to determine the energy supply-demand data, specifically including the power supply curve, which reflects the power supply capacity at different time periods; the power demand curve reflects the power demand at different time periods; the supply-demand gap reflects the difference between power supply and demand, that is, the situation of supply falling short of demand or supply exceeding demand; to clarify when the energy storage system needs to discharge to make up for power shortages and when it needs to charge to store excess electric energy.

[0023] Preferably, after determining the energy supply-demand data, energy storage elements are introduced to optimize the scheduling strategy of the energy storage system. Among them, the energy storage elements are quantitative descriptions of energy storage monomers, including power regulation degree, adaptive balance degree, and capacity contribution degree. Specifically, the power regulation degree refers to the power range that an energy storage monomer can regulate per unit time, that is, the flexibility and response speed of the charging and discharging power of the energy storage monomer, reflecting the regulation ability of the energy storage monomer to power fluctuations in the power system. An energy storage monomer with a high power regulation degree can quickly respond to the demand changes of the power system and is suitable for dealing with short-term power fluctuations or frequency regulation; the adaptive balance degree refers to the ability of an energy storage monomer to automatically adjust its charging and discharging strategy according to system requirements in a complex operating environment to achieve system energy balance, reflecting the adaptive ability of the energy storage monomer in a dynamically changing environment. An energy storage monomer with a high adaptive balance degree can optimize its own charging and discharging behavior according to real-time supply-demand data, avoid overcharging or over-discharging, and extend its service life; the capacity contribution degree refers to the effective energy capacity that an energy storage monomer can provide in the system, usually considering factors such as its available capacity, charging and discharging efficiency, and life attenuation, reflecting the energy storage ability of the energy storage monomer in the system. An energy storage monomer with a high capacity contribution degree can provide more energy support for the system and is suitable for long-term energy scheduling requirements.

[0024] Furthermore, step S100 further includes step S110, interacting with the energy scheduling records of the target area to mine the periodic consumption curve; step S120, according to the environmental data, combined with the base station power conversion efficiency, performing energy grid connection prediction to determine the energy supply data; step S130, combined with the periodic consumption curve, performing load consumption prediction based on time series to determine the energy demand data.

[0025] Preferably, data interaction is carried out with the energy scheduling system of the target area to obtain the historical energy scheduling records of this area, including detailed information such as the power generation power of each energy node, the charge and discharge status of energy storage devices, and the power consumption of the load side in different time periods. By using data analysis and mining, the periodic law of energy consumption is found. For example, the electricity consumption peak in the business area usually appears during the daytime working hours on weekdays, while the electricity consumption peak in the residential area may be in the evening. By mining the periodic consumption curve, the energy consumption pattern of the target area in different time periods can be clearly understood; environmental data has an important impact on energy production, especially for renewable energy sources (such as solar energy, wind energy). Environmental data includes information such as light intensity, temperature, wind speed, and humidity.

[0026] Preferably, for some distributed energy systems, energy production often needs to be converted and processed through a base station. The electrical conversion efficiency of the base station refers to the efficiency of the base station converting the original energy (such as the light energy collected by solar panels, the mechanical energy generated by wind turbines, etc.) into grid-connected electrical energy. Under different environmental conditions, the electrical conversion efficiency of different types of base stations and devices may be different; then, combining environmental data and the electrical conversion efficiency of the base station, a suitable prediction model (such as a machine learning model, a physical model, etc.) is used to predict the amount of energy that can be connected to the grid in the next period of time. For example, according to the light intensity and the electrical conversion efficiency of the solar base station, the power generation of the solar power station in the next day is predicted. In this way, the energy supply data of the target area, that is, the total amount of energy that can be provided for the load side in the future, is determined.

[0027] Preferably, considering the trends, seasonality, and periodicity of historical load data, the load consumption in different future time periods is predicted based on time series analysis, and then the load consumption in each future time period of the target area, that is, the energy demand data, is obtained. These data can reflect the energy demand of the load side at different time points and are crucial for the energy scheduling of the energy storage system. For example, during the peak electricity consumption period, the energy storage system needs to be prepared in advance to release the stored energy to meet the load demand; while during the low electricity consumption period, the energy storage device can be charged to improve the energy utilization efficiency. By mining the periodic consumption curve, predicting the energy grid connection, and predicting the load consumption, the energy supply data and energy demand data of the target area can be accurately determined, improving the refinement of the energy scheduling of the energy storage system.

[0028] Step S200, for the energy storage system, configure the lightweight allocation module for the proximal data center of each energy storage monomer, execute the allocation decision based on the energy supply and demand data, and determine the first-order strategy, where the first-order strategy includes the energy storage scheduling strategy and the aggregation coordination requirement.

[0029] Preferably, through distributed lightweight computing and local optimization, the efficient scheduling and coordination of energy storage units in the energy storage system are realized. Specifically, first, a lightweight allocation module of the proximal data center is configured for each energy storage unit. The proximal data center refers to a small computing node deployed near the energy storage unit, which is used to process local data and perform lightweight computing. The advantage of the proximal data center is to reduce communication latency, improve computing efficiency, and reduce the dependence on the central control system. The lightweight allocation module refers to a lightweight allocation model deployed in the proximal data center, which is used to perform local optimization and allocation decisions, aiming to reduce computing complexity and resource occupancy and make it suitable for a distributed environment. An independent lightweight allocation module is configured for each energy storage unit, and the functions of the module include data acquisition, local optimization calculation, and communication with the central control system.

[0030] Preferably, the allocation decision is made based on energy supply and demand data. That is, based on the energy supply and demand data, the lightweight allocation module locally optimizes the charging and discharging behavior of the energy storage unit. The specific decisions may include the charging and discharging time (when to charge or discharge), the charging and discharging power (the rate of charging or discharging), and the energy allocation (how much energy is allocated for charging or discharging) to maximize the utilization efficiency of the energy storage unit and minimize the local supply and demand gap. Furthermore, a local optimization strategy (i.e., determining the first-order strategy) is generated, which mainly includes the energy storage scheduling strategy and the aggregation coordination requirements. Among them, the energy storage scheduling strategy refers to the charging and discharging plan for a single energy storage unit, which may include the charging and discharging schedule, the charging and discharging power curve, and the energy allocation plan. For example, when the power supply is in excess (such as at the peak of photovoltaic power generation at noon), the energy storage unit performs a charging operation, and when the power demand is at a peak (such as in the evening), the energy storage unit performs a discharging operation. The aggregation coordination requirements refer to the coordination requirements generated by the energy storage unit during the local optimization process for interaction with the central control system or other energy storage units, which may include the status information of the energy storage unit (such as the remaining capacity and the charging and discharging ability), the local optimization results (such as the charging and discharging plan), and the suggestions or requirements for global optimization. If the capacity of a certain energy storage unit is close to the upper limit, a coordination requirement is sent to the central control system to request an adjustment of the charging and discharging plans of other energy storage units. By configuring the lightweight allocation module of the proximal data center and performing the allocation decision based on the energy supply and demand data to determine the first-order strategy, the efficient scheduling and local optimization of the energy storage unit can be realized, which not only reduces the computing complexity and communication latency but also improves the scheduling efficiency and flexibility of the energy storage system.

[0031] Further, step S200 further includes step S201, recording the energy scheduling of the interaction target area, performing data-driven training, and constructing an energy scheduling module; step S202, according to the source-load relationship, performing lightweight transfer learning on the energy scheduling module to determine the lightweight allocation module, where the mapping between the energy storage monomer and the load side is used as the source-load relationship; step S203, performing lightweight transfer learning on the energy scheduling module through the aggregation analysis between the energy storage monomers to determine the lightweight coordination module.

[0032] Preferably, an efficient energy scheduling module is constructed through historical data and machine learning techniques, and it is adapted to specific application scenarios through lightweight transfer learning. Specifically, recording the energy scheduling of the interaction target area means obtaining the historical energy scheduling data of the target area, including power supply, power demand, charge and discharge behaviors of energy storage monomers, system operating status, etc. Using machine learning algorithms (such as deep learning, reinforcement learning) to analyze and model train the energy scheduling records to obtain an energy scheduling module that can generate optimized scheduling strategies based on real-time data; then, according to the determined source-load relationship, perform lightweight transfer learning on the energy scheduling module, including adapting the pre-trained energy scheduling module to specific application scenarios, while reducing the complexity of the model and the computational resource requirements, to use existing knowledge (source domain) to solve new problems (target domain). Through transfer learning, the energy scheduling module is adapted to the lightweight allocation module in the proximal data center to perform local optimization and allocation decisions; then, the pre-trained energy scheduling module is adapted to the global coordination task, while reducing the complexity of the model and the computational resource requirements. Through transfer learning, the energy scheduling module is adapted to the lightweight coordination module built into the energy storage system. The lightweight coordination module is used to perform global optimization and coordination tasks, thereby determining the lightweight coordination module. Thus, efficient scheduling and optimization of the energy storage system are achieved, and the adaptability and computational efficiency of the energy scheduling module are improved.

[0033] Further, step S200 further includes step S210, deploying the lightweight allocation module in the proximal data centers of each energy storage monomer; step S220, receiving the energy supply and demand data to the proximal data centers of each energy storage monomer, taking the target energy storage monomer as the source end, and taking the allocation of the mapped load side as the decision target to determine the first-order strategy, where the load priority is used as the allocation condition.

[0034] Preferably, deploy the lightweight allocation module to the proximal data center of each energy storage unit, configure the operating environment and parameters of the module to ensure that it can process energy supply and demand data in real time. Herein, the lightweight allocation module refers to an energy scheduling module processed by lightweight transfer learning, which is used to execute local optimization and allocation decisions, obtain energy supply and demand data from the central control system or prediction model, and transmit it to the proximal data center of each energy storage unit. Finally, taking the target energy storage unit as the source end (taking the current energy storage unit as the core to perform local optimization and allocation decisions), and taking the allocation of the mapped load side as the decision-making goal (determining the charge and discharge plan of the energy storage unit according to the demand of the load side to achieve supply-demand balance), determine the first-order strategy, including the energy storage scheduling strategy (the charge and discharge plan of a single energy storage unit) and the aggregated coordination demand (suggestions for global optimization). Among them, the allocation condition is the load priority, that is, classify and rank the loads according to the importance and urgency of the load side, and then determine the charge and discharge plan of the energy storage unit to achieve efficient scheduling and energy allocation of the energy storage unit, reduce the computational complexity and communication delay, and thus improve the scheduling efficiency and flexibility of the energy storage system.

[0035] Step S300, according to the aggregated coordination demand, combine the energy storage elements to obtain a reconstructed energy storage topology and send it back to the energy storage system.

[0036] Preferably, optimize the coordination and energy allocation among multiple energy storage units by dynamically adjusting the topology structure of the energy storage system to obtain a reconstructed energy storage topology, that is, according to the aggregated coordination demand and energy storage elements, dynamically adjust the connection relationship and energy allocation method among the energy storage units in the energy storage system to optimize the coordination and energy allocation among the energy storage units and improve the overall performance and efficiency of the system. Specifically, analyze the aggregated coordination demand, identify the bottlenecks and optimization opportunities in the energy storage system according to the state information and local optimization results of the energy storage units. For example, the capacity of some energy storage units may be close to the upper limit and the charging power needs to be reduced; the capacity of some energy storage units may be low and they need to be discharged preferentially. Combine the energy storage elements, and evaluate the role and function of each energy storage unit in the energy storage system according to its power regulation degree, adaptive balance degree and capacity contribution degree. For example, energy storage units with a high power regulation degree are suitable for short-term power regulation, and energy storage units with a high capacity contribution degree are suitable for long-term energy support. Then generate a reconstructed energy storage topology, including the connection relationship of the energy storage units, the energy allocation scheme and the scheduling strategy, that is, according to the analysis results, adjust the connection relationship and energy allocation method among the energy storage units. For example, group multiple energy storage units, and each group is responsible for different scheduling tasks (such as power regulation, energy reserve). Finally, transmit the reconstructed topology to the proximal data center and energy storage units through the communication network to realize the dynamic optimization and coordination of the energy storage system, and thus improve the overall performance and efficiency of the energy storage system.

[0037] Further, step S300 further includes step S310 of interacting with the energy storage topology of the target area; and step S320 of performing source-load topology node screening and topology reconstruction on the energy storage topology according to the aggregation coordination requirement to determine the reconstructed energy storage topology.

[0038] Preferably, by dynamically adjusting the topology structure of the energy storage system, the coordination and energy distribution among multiple energy storage monomers are optimized, thereby improving the overall performance and efficiency of the system. Specifically, connect to the energy storage topology of the target area, obtain the current energy storage topology information, and analyze the connection relationship and energy distribution method among the energy storage monomers; then, according to the aggregation coordination requirement and the distribution of the source (power supply) and load (power demand), screen out the key topology nodes of the energy storage topology, including the nodes that have a greater impact on the system performance, such as the energy storage monomers with a high power regulation degree or the high-priority load nodes. Finally, according to the screening results, dynamically adjust the connection relationship and energy distribution method among the energy storage monomers in the energy storage system to generate a new energy storage topology, including the connection relationship among the energy storage monomers, the energy distribution scheme, and the scheduling task, so as to improve the overall performance and efficiency of the power grid energy storage system.

[0039] Further, as Figure 2 shown, step S320 further includes step S321 of screening the dispatching source end and the dispatching load end based on the energy storage topology according to the aggregation coordination requirement, where the screening criterion is the imbalance between supply and demand; step S322 of determining the element characteristic values of each dispatching source end according to the energy storage elements and the aggregation coordination requirement to determine the source end element coefficients; step S323 of determining the load end consumption demand according to the aggregation coordination requirement; and step S324 of determining the reconstructed energy storage topology based on the dispatching source end, the dispatching load end, the source end element coefficients, and the load end consumption demand, where the reconstructed energy storage topology is a temporary decision topology.

[0040] Preferably, according to the existing energy storage topology (i.e., the connection relationship and layout among components such as energy storage monomers, power sources, and loads in the energy storage system), suitable dispatching source ends and dispatching load ends are found based on the aggregation coordination requirements. That is, taking the supply-demand imbalance as the screening criterion, the parts with mismatches between energy supply and demand are searched. For example, in some areas, there may be an oversupply of energy (supply > demand), while in other areas, there may be an undersupply of energy (demand > supply). These areas will be determined as the dispatching source ends (areas with oversupply, which can output energy) and dispatching load ends (areas with overdemand, which need to input energy) for subsequent reasonable allocation of energy. Then, according to the aggregation coordination requirements, the energy storage elements of each dispatching source end are analyzed, that is, the energy storage elements of each dispatching source end are quantitatively analyzed to obtain corresponding element characteristic values. For example, the specific value of the power regulation degree, the self-adaptive balance degree index, and the capacity contribution degree of each dispatching source end are calculated. Then, based on these characteristic values, a comprehensive source end element coefficient is determined to reflect the importance and dispatchable ability of each dispatching source end in the energy storage system and other characteristics.

[0041] Preferably, according to the aggregation coordination requirements, the determined dispatching load ends are analyzed to clarify the amount of energy that each load end can absorb. For example, considering factors such as the equipment capacity, operating status, and current energy demand of the load end, it is determined how much more energy each load end can receive, that is, the load end absorption demand. Accurately determining the load end absorption demand helps to reasonably arrange the energy transmission from the dispatching source end to the load end and avoid the situation of excessive or insufficient energy transmission. Finally, by synthesizing the information of the previously determined dispatching source end (energy output end), dispatching load end (energy receiving end), source end element coefficient (reflecting the characteristics of the dispatching source end), and load end absorption demand (the amount of energy that the load end can receive), the original energy storage topology is reconstructed to obtain a reconstructed energy storage topology. Among them, the reconstructed energy storage topology is a temporary decision-making topology, which is determined temporarily for the optimal dispatching of the current energy storage system and may be adjusted again according to new situations and requirements in the future. Through this reconstruction, the energy flow of the energy storage system can be made more reasonable, the energy utilization efficiency and the system stability can be improved to meet the aggregation coordination requirements and realize the optimal operation of the energy storage system.

[0042] Step S400: Based on the lightweight coordination module built in the energy storage system, perform a multi-energy storage monomer aggregation decision based on the reconstructed energy storage topology to determine a second-order strategy.

[0043] Preferably, by performing global optimization and coordination (multi-energy storage monomer aggregation decision based on reconstructed energy storage topology), a second-order strategy is determined. Specifically, the lightweight coordination module built into the energy storage system receives the aggregation coordination requirements and reconstructed energy storage topology from the proximal data center, executes the global optimization algorithm, generates the aggregation decision of multiple energy storage monomers, and then transmits the optimization result (second-order strategy) to the energy storage system for guiding the operation of the energy storage monomers. Among them, the multi-energy storage monomer aggregation decision refers to the lightweight coordination module performing global optimization and coordination on multiple energy storage monomers based on the reconstructed energy storage topology, generating the optimal scheduling strategy to maximize the overall performance of the energy storage system and minimize the operation cost and energy loss of the system. Specifically, considering the power regulation degree, adaptive equilibrium degree, and capacity contribution degree of the energy storage monomers, as well as the energy supply and demand data of the system, the lightweight algorithm (such as distributed optimization algorithm, heuristic algorithm) is used to perform global optimization on multiple energy storage monomers, and according to the reconstructed energy storage topology, the energy distribution and scheduling tasks between the energy storage monomers are adjusted. For example, multiple energy storage monomers are grouped, and each group is responsible for different scheduling tasks (such as power regulation, energy storage), and then a second-order strategy is generated, including the global optimization result and coordination strategy, for guiding the operation of the energy storage monomers, which can achieve the global optimization and coordination of the energy storage system and improve the scheduling efficiency and flexibility of the energy storage system.

[0044] Further, step S400 further includes step S410 of deploying the lightweight coordination module in the data center of the energy storage system; step S420 of determining the reconstructed energy storage topology by the data center through interactive communication and transmitting it to the lightweight coordination module to perform the aggregation scheduling coordination of multiple energy storage monomers in the dual-mode parallel and determine the second-order strategy.

[0045] Step S420 further includes step S421 of taking energy scheduling and load transfer as the dual mode and determining the second-order strategy based on the combined optimization in the dual mode.

[0046] Preferably, the lightweight coordination module is a module obtained after performing specific lightweight transfer learning (based on the aggregation analysis between energy storage monomers) on the energy scheduling module, and is used for performing aggregation scheduling coordination on multiple energy storage monomers. The lightweight coordination module is deployed in the data center of the energy storage system to obtain and process various data in the energy storage system, so as to realize the unified management and coordinated scheduling of multiple energy storage monomers in the entire energy storage system. The data center obtains the relevant information of the reconstructed energy storage topology through interactive communication with other relevant components (such as devices or modules that may store the reconstructed energy storage topology information), and then transmits the information of the reconstructed energy storage topology to the lightweight coordination module, so that the lightweight coordination module can perform subsequent scheduling coordination operations based on this and perform the aggregation scheduling coordination of multiple energy storage monomers in the dual-mode parallel.

[0047] Preferably, the dual mode refers to including an energy scheduling mode and a load transfer mode. Specifically, for the energy scheduling mode: it mainly focuses on managing the generation, storage, and distribution of energy in the energy storage system. For example, according to the energy storage status, energy supply and demand of each energy storage unit, etc., it decides how to allocate energy among different energy storage units and how to deliver the energy in the energy storage unit to the load side that needs it; the load transfer mode focuses on transferring the load from one area or time period to another area or time period. For example, when the load in some areas is too high during a certain time period while there is surplus energy and capacity in other areas, through reasonable control strategies, part of the load is transferred to other areas to achieve load balance of the entire system and efficient utilization of energy.

[0048] Further, step S421 further includes step A, determining an initialization strategy with single-mode scheduling; step B, using the coordination target and coordination amplitude based on the source side as the adjustment method for the energy scheduling mode, and using the transfer target and transfer amplitude based on the load side as the adjustment method for the load transfer mode, adjusting and iterating the initialization strategy multiple times to determine multiple sets of optimization strategies; step C, traversing the initialization strategy and the multiple sets of optimization strategies, and selecting the optimal strategy as the second-order strategy.

[0049] Preferably, during the operation of the dual mode in parallel, there are multiple possible combinations of energy scheduling and load transfer. The lightweight coordination module searches and evaluates these combinations (combination optimization). Specifically, it determines an initialization strategy with single-mode scheduling (that is, separately considering the energy scheduling mode or the load transfer mode alone) as the starting point for subsequent optimization. Using the coordination target (such as determining the energy output target of each scheduling source side) and coordination amplitude (the adjustment range of energy output) based on the source side as the adjustment method for the energy scheduling mode, and using the transfer target (such as determining the target of each load side receiving energy) and transfer amplitude (the adjustment range of load transfer) based on the load side as the adjustment method for the load transfer mode, it adjusts and iterates the initialization strategy multiple times. In each iteration, it evaluates the operation effect of the energy storage system under different combination methods (such as indicators like energy utilization efficiency and system stability). Finally, it traverses all the initialization strategies and multiple sets of optimization strategies obtained from multiple iterations, selects the optimal strategy from these strategies, and determines it as the second-order strategy, which can achieve more efficient aggregated scheduling of multiple energy storage units in the energy storage system to achieve better operation effects and optimization goals.

[0050] Step S500, fusing the first-order strategy and the second-order strategy as the energy scheduling strategy to perform energy storage scheduling management on the target area.

[0051] Preferably, the first-order strategy and the second-order strategy are integrated to generate an energy scheduling strategy, so as to take into account the advantages of local optimization and global optimization, perform energy storage scheduling management on the target area, realize the efficient operation of the energy storage system and the supply-demand balance of the target area. Specifically, first, the energy storage scheduling strategy (charging and discharging plan of a single energy storage unit) in the first-order strategy is used as the basis to ensure real-time performance and response speed. Then, the global optimization result and coordination strategy in the second-order strategy are applied to the energy storage system to ensure the optimal overall performance of the system. According to real-time data and the state of the energy storage system, the energy scheduling strategy is dynamically adjusted to adapt to supply-demand changes, and then an energy scheduling strategy is generated, including the charging and discharging plan of a single energy storage unit, the coordination rules between multiple energy storage units, and the energy allocation scheme and scheduling tasks of the system. Then, energy storage scheduling management is carried out on the target area, that is, according to the comprehensive energy scheduling strategy, the energy storage system in the target area is scheduled and managed, including controlling the charging and discharging behavior of the energy storage unit according to the energy scheduling strategy. For example, during the peak power demand, the energy storage unit with a high power regulation degree is scheduled to discharge, and during the period of over-abundant power supply, the energy storage unit with a high capacity contribution degree is scheduled to charge. According to the coordination strategy, the energy allocation and scheduling tasks between the energy storage units are adjusted. For example, when the capacity of a certain energy storage unit is close to the upper limit, the charging and discharging plans of other energy storage units are adjusted to avoid overcharging or over-discharging. The operation state and energy supply-demand data of the energy storage system are monitored in real time, and according to the monitoring results, the energy scheduling strategy is dynamically adjusted to achieve the supply-demand balance of the target area and improve the operation management efficiency and flexibility of the energy storage system.

[0052] Further, step S500 further includes step S510 of determining scheduling response information by performing energy storage scheduling tracking; step S520 of performing supply-demand balance evaluation on the scheduling response information to determine a supply-demand balance coefficient; and step S530 of adjusting grid data of the energy storage system according to the supply-demand balance coefficient.

[0053] Preferably, after executing energy storage dispatch management on the target area, the operating status of the energy storage system and the actual distribution of energy are continuously monitored and tracked, including real-time monitoring of the charging and discharging status of the energy storage unit, the transmission of energy from the energy storage system to the load side, and the operating parameters of each part of the equipment, so as to obtain dispatch response information. For example, whether the actual discharge power of the energy storage unit in a certain period of time has reached the expected value, whether the energy has been accurately transmitted to the corresponding load end according to the established dispatch strategy, and whether there has been any equipment failure or abnormality during the dispatch process, etc., all belong to the category of dispatch response information; then, the dispatch response information is evaluated for supply and demand balance, including evaluating the comparative relationship between the total energy supply (such as the actual output of the energy storage system and the input of other energy sources) and the actual total consumption on the load side, and determining the supply and demand balance coefficient based on the evaluation results. If the energy supply just meets the load demand, the supply and demand balance coefficient may be 1; if the supply is greater than the demand, the coefficient will be greater than 1; if the demand is greater than the supply, the coefficient is less than 1, which intuitively reflects the balance state of energy supply and demand after the current energy storage dispatch.

[0054] Preferably, the current energy supply and demand situation is judged according to the determined supply and demand balance coefficient, so as to decide whether and how to adjust the grid data of the energy storage system. Specifically, if the coefficient shows that the energy supply is in excess (greater than 1), it may be necessary to reduce the output power of the energy storage system to the grid, or adjust the voltage and other parameters of the grid interface so that the excess energy can be properly handled, such as reverse charging and storing the energy storage device. If the coefficient indicates that the energy supply is insufficient (less than 1), it may be necessary to increase the output power of the energy storage system, or introduce more energy from other energy sources to meet the load demand; when adjusting the voltage and power parameters of the grid interface, although it may affect the stability of the energy supply to a certain extent (for example, a sudden change in voltage may affect some voltage-sensitive equipment), it plays an important role in dealing with the imbalance of energy supply and demand, especially in eliminating peak problems. By evaluating and feedback-adjusting the effect of energy storage scheduling, the operation of the energy storage system is continuously optimized to better achieve the reasonable allocation of energy and the balance of supply and demand, and improve the stability and efficiency of the entire energy system.

[0055] The above specific embodiments do not constitute a limitation on the protection scope of the present application. Those skilled in the art should understand that various modifications, combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principle of the present application shall be included within the protection scope of the present application. In some cases, the actions or steps recited in the present application can be executed in an order different from that in the embodiments and still achieve the desired results. Additionally, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. Energy dispatch optimization method for energy storage system based on demand response, characterized in that: The method comprises: By predicting the source and load of the target area, the energy supply and demand data is determined, and energy storage elements are introduced, wherein the energy storage elements are the power regulation degree, adaptive balance degree and capacity contribution degree of the energy storage monomer; For the energy storage system, a lightweight allocation module is configured for the proximal data center of each energy storage unit, and an allocation decision based on energy supply and demand data is executed to determine a first-order strategy, wherein the first-order strategy includes an energy storage scheduling strategy and aggregated coordination requirements; According to the aggregation coordination requirements and in combination with the energy storage elements, a reconstructed energy storage topology is obtained and transmitted back to the energy storage system; According to the built-in lightweight coordination module of the energy storage system, a multi-energy storage monomer aggregation decision is made based on the reconstructed energy storage topology to determine a second-order strategy; The first-order strategy and the second-order strategy are integrated as an energy scheduling strategy to perform energy storage scheduling management on the target area.

2. The energy storage system energy scheduling optimization method based on demand response according to claim 1 is characterized in that: Before the energy storage unit performs lightweight distribution module configuration of the proximal data center, the method further includes: Interact with the target area’s energy scheduling records, conduct data-driven training, and build an energy scheduling module; According to the source-load relationship, lightweight transfer learning is performed on the energy scheduling module to determine the lightweight allocation module, wherein the mapping between the energy storage monomer and the load side is the source-load relationship; Based on the aggregation analysis between energy storage monomers, lightweight transfer learning is performed on the energy scheduling module to determine the lightweight coordination module.

3. The energy storage system energy scheduling optimization method based on demand response according to claim 2 is characterized in that: The determining of the first-order strategy comprises: The lightweight distribution module is arranged in a proximal data center of each energy storage unit; The energy supply and demand data is received to the proximal data center of each energy storage unit, and the target energy storage unit is used as the source end, and the allocation of the mapped load side is used as the decision target to determine the first-order strategy, in which the load priority is used as the allocation condition.

4. The energy storage system energy scheduling optimization method based on demand response according to claim 1, characterized in that: According to the aggregation coordination requirements, a reconstructed energy storage topology is obtained, including: interacting with the energy storage topology of the target area; According to the aggregation coordination requirements, the energy storage topology is screened based on source-load topology nodes and topology reconstruction is performed to determine the reconstructed energy storage topology.

5. The energy storage system energy scheduling optimization method based on demand response according to claim 4 is characterized in that: The energy storage topology is screened and reconstructed based on source-load topology nodes, including: According to the aggregated coordination requirements, screening the dispatch source end and the dispatch load end based on the energy storage topology, wherein the imbalance between supply and demand is used as the screening criterion; According to the energy storage element and the aggregated coordination demand, the element characteristic value of each dispatch source is determined to determine the source element coefficient; Determine the load-end consumption demand based on the aggregated coordination demand; The reconstructed energy storage topology is determined based on the dispatching source end, the dispatching load end, the source end factor coefficient, and the load end consumption demand, wherein the reconstructed energy storage topology is a temporary decision topology.

6. The energy storage system energy scheduling optimization method based on demand response according to claim 2 is characterized in that: Identify second-order strategies, including: Deploy the lightweight coordination module in a data center of the energy storage system; Through interactive communication, the data center determines the reconstructed energy storage topology and transmits it to the lightweight coordination module, executes the aggregated scheduling coordination of multiple energy storage units under dual-mode parallelism, and determines the second-order strategy.

7. The energy storage system energy scheduling optimization method based on demand response according to claim 6 is characterized in that: With energy dispatch and load transfer as dual modes, the second-order strategy is determined by combining optimization under the dual modes.

8. The energy dispatch optimization method for energy storage system based on demand response according to claim 7, characterized in that: The second-order strategy is determined by combinatorial optimization based on the dual mode, including: Determine the initialization strategy in single mode scheduling; The coordination target and coordination range based on the source end are used as the adjustment method based on the energy dispatch mode, and the transfer target and transfer range based on the load end are used as the adjustment method based on the load transfer mode, the initialization strategy is adjusted and iterated multiple times to determine multiple groups of optimization strategies; The initialization strategy and the multiple groups of optimization strategies are traversed, and the optimal strategy is selected as the second-order strategy.

9. The energy storage system energy scheduling optimization method based on demand response according to claim 2, characterized in that: By forecasting the source and load of the target area, energy supply and demand data are determined, including: Interact with the target area’s energy scheduling records to mine periodic consumption curves; Based on environmental data and combined with the base station power conversion efficiency, energy access forecast is conducted to determine energy supply data; Combined with the periodic consumption curve, a load consumption forecast based on time series is performed to determine energy demand data.

10. The energy storage system energy dispatch optimization method based on demand response according to claim 1, characterized in that: After energy storage dispatch is performed in the target area, it includes: Determine dispatch response information by tracking energy storage dispatch; Performing supply-demand balance evaluation on the dispatch response information to determine a supply-demand balance coefficient; According to the supply-demand balance coefficient, the energy storage system is regulated for grid data.

Images