Traditional power grid regional charging dispatch system based on AI and green electricity synergy

Abstract

This invention discloses a regionalized charging dispatching system for traditional power grids based on AI and green electricity collaboration, comprising a multi-source data perception layer, an artificial intelligence decision-making layer, an intelligent dispatching execution layer, an energy accounting supervision layer, and an energy storage system. The aim is to provide a regionalized charging dispatching system capable of collaboratively perceiving green electricity generation data, energy storage system operation data, traditional power grid supply data, regional charging load data, green electricity procurement data, and electricity purchase settlement data, and thereby achieving coordinated dispatching among green electricity, energy storage systems, and the traditional power grid. This improves the power supply coordination, green electricity utilization level, and energy supervision capabilities within the regional charging network.

Description

Technical Field

[0001] This invention relates to the field of charging dispatching technology, and in particular to a regionalized charging dispatching system for traditional power grids based on AI and green electricity collaboration. Background Technology

[0002] Green electricity, such as photovoltaic and wind power, has been gradually applied to the power supply scenarios of electric vehicle charging networks to reduce dependence on traditional fossil fuels and promote the low-carbon transformation of the energy structure. However, the power generation process of green electricity is greatly affected by changes in sunlight intensity, wind speed, and meteorological conditions, exhibiting strong volatility and intermittency. This can easily lead to rapid changes in power generation, insufficient power supply continuity, and short-term output instability. Current technologies lack the ability to coordinate and predict the green electricity generation status, regional charging load changes, and energy storage operation status, making it difficult to promptly formulate power supply arrangements that adapt to regional charging demand. This results in low utilization of green electricity in charging scenarios and difficulty in stably matching continuously changing charging load demands.

[0003] The existing power dispatching systems for regional charging stations are mostly built around the traditional power grid, with insufficient consideration for the coordinated allocation of green electricity, energy storage systems, and the traditional power grid, and lack a dynamic balancing mechanism for the regional charging network. When green electricity output fluctuates or regional charging load changes rapidly, existing technologies often struggle to adjust the power supply ratio of various power sources in real time based on the real-time green electricity generation, the state of charge of energy storage systems, and the power supply capacity of the traditional power grid. This can easily lead to situations where there is a surplus or shortage of power supply in certain periods, or an excessive amount of electricity purchased from the traditional power grid. This not only affects the stable operation of charging stations but also hinders the priority consumption of green electricity and the effective utilization of energy storage systems.

[0004] Furthermore, with the increasing number of electric vehicles in the region and the random changes in charging behavior, the load on the charging network exhibits dynamic fluctuations. Currently, an integrated intelligent linkage system has not yet been formed between the charging network, green electricity, energy storage systems, and the traditional power grid. In particular, there is a lack of coordinated monitoring of green electricity procurement data, electricity purchase settlement data, energy supply data, and energy consumption data, making it difficult to achieve coordinated optimization of the allocation and accounting processes. Simultaneously, under conditions such as network anomalies or cloud-based dispatch failures, the existing system's continuous operation guarantee capability is relatively insufficient, making it difficult to simultaneously ensure the power supply stability, energy utilization rationality, and operational management accuracy of the regional charging network. Summary of the Invention

[0005] The purpose of this application is to propose a regionalized charging dispatch system for traditional power grids based on AI and green electricity collaboration. It aims to provide a regionalized charging dispatch system that can collaboratively perceive green power generation data, energy storage system operation data, traditional power grid supply data, regional charging load data, green power procurement data, and electricity purchase settlement data, and thereby realize coordinated dispatch among green electricity, energy storage systems, and traditional power grids. This will improve the power supply coordination, green electricity utilization level, and energy supervision capabilities in the regional charging network.

[0006] To address the aforementioned technical problems, this application provides a regionalized charging dispatch system for traditional power grids based on AI and green energy collaboration, employing the following technical solution: It includes a multi-source data perception layer, an artificial intelligence decision-making layer, an intelligent scheduling and execution layer, an energy accounting and supervision layer, and an energy storage system; The multi-source data sensing layer is used to collect green power generation data, energy storage system operation data, traditional power grid power supply data, regional charging load data, green power procurement data, and electricity purchase settlement data. The artificial intelligence decision-making layer is connected to the multi-source data perception layer and is used to generate a power supply ratio scheme between green electricity, energy storage system and traditional power grid based on the green power generation data, the energy storage system operation data, the traditional power grid power supply data, the regional charging load data, the green electricity procurement data and the electricity purchase settlement data. The intelligent scheduling execution layer is connected to the artificial intelligence decision-making layer and is used to control the power distribution equipment of the green power grid connection link, energy storage system and regional charging station according to the power supply ratio scheme, so as to complete the regional charging load distribution and power supply switching. The energy accounting supervision layer is connected to the artificial intelligence decision-making layer and the intelligent dispatch execution layer respectively. It is used to calculate the green electricity consumption, energy storage system charging and discharging, and traditional grid electricity purchase based on dispatch data and execution data, and generate dispatch optimization suggestions based on the calculation results and feed them back to the artificial intelligence decision-making layer. The energy storage system is connected to the intelligent scheduling execution layer to realize green power buffer storage and charging and discharging regulation. The multi-source data sensing layer, the artificial intelligence decision-making layer, the intelligent scheduling execution layer and the energy accounting supervision layer interact with data and synchronize instructions through a unified communication protocol.

[0007] Compared with the prior art, the embodiments of this application have the following main advantages: The regionalized charging and dispatching system for traditional power grids based on AI and green electricity collaboration disclosed in this application firstly establishes a multi-source data perception layer, an artificial intelligence decision-making layer, an intelligent dispatch execution layer, and an energy accounting supervision layer, and incorporates energy storage systems into a unified dispatching system. It can generate power supply ratio schemes based on green power generation data, energy storage system operation data, traditional power grid power supply data, regional charging load data, green power procurement data, and electricity purchase settlement data, thereby realizing coordinated power supply between green electricity, energy storage systems, and traditional power grids. This is beneficial to improving the power supply coordination and dispatching flexibility of regional charging networks under green power output fluctuation scenarios. Secondly, by setting up time-series prediction models, reinforcement learning models, and multi-objective optimization models in the artificial intelligence decision-making layer, a comprehensive analysis of the short-term output of green electricity, changes in regional charging load, and the state of charge of energy storage systems is conducted. Combined with preset discharge thresholds and preset charging thresholds, the power supply ratio scheme is adjusted, which is conducive to timely updating the allocation strategy when green electricity output changes and regional charging load fluctuates, thereby improving the power supply stability and energy storage regulation rationality of the regional charging network. Third, by setting up an intelligent scheduling execution layer, the power supply ratio scheme is transformed into control instructions for the green power grid-connected links, energy storage systems, and regional charging station power distribution equipment. Combined with the scheduled charging demand and off-peak charging demand, the power supply ratio scheme is distributed in an orderly manner, which helps to improve the load allocation rationality of each charging station, reduce the energy replenishment pressure of the traditional power grid during high load periods, and improve the utilization effect of green power and energy storage systems in the regional charging network. Attached Figure Description

[0008] To more clearly illustrate the solutions in this application, the accompanying drawings used in the description of the embodiments of this application will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0009] Figure 1 This is a schematic diagram of a structural embodiment of a traditional power grid regional charging dispatch system based on AI and green electricity collaboration according to this application; Figure 2 This is a schematic diagram of the operation of an embodiment of a traditional power grid regional charging dispatch system based on AI and green electricity collaboration according to this application. Detailed Implementation

[0010] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0011] Given that existing technologies for green electricity such as photovoltaic and wind power are greatly affected by natural conditions and have fluctuating output, it is difficult to directly match the ever-changing electricity demand of regional charging networks. At the same time, the existing dispatching system lacks unified and coordinated control between green electricity, energy storage systems and traditional power grids, which can easily lead to problems such as insufficient power supply, oversupply, or excessive purchases of electricity by traditional power grids in scenarios with changes in green electricity output and fluctuations in charging load.

[0012] Combination Figure 1 and Figure 2 As shown, the purpose of this invention is to provide a regionalized charging and dispatching system based on artificial intelligence and green power in coordination with the traditional power grid. By constructing a multi-source data perception layer, an artificial intelligence decision-making layer, an intelligent dispatching and execution layer, and an energy accounting supervision layer, and combining an energy storage system, local edge computing nodes, and a cloud dual-backup redundancy architecture, the system achieves coordinated dispatching of green power, energy storage systems, and the traditional power grid in the regionalized charging network. This improves power supply stability, optimizes power supply ratio, enhances the utilization level of green power, and enables linked supervision of the energy supply process and the electricity purchase and settlement process.

[0013] The regionalized charging dispatch system in this embodiment includes a multi-source data perception layer, an artificial intelligence decision-making layer, an intelligent dispatch execution layer, an energy accounting supervision layer, an energy storage system, local edge computing nodes, a cloud-based dual-backup redundancy architecture, and a human-machine interface. The multi-source data perception layer collects green electricity generation data, energy storage system operation data, traditional power grid supply data, regional charging load data, green electricity procurement data, and electricity purchase settlement data. The artificial intelligence decision-making layer generates a power supply ratio scheme between green electricity, the energy storage system, and the traditional power grid based on the aforementioned data. The intelligent dispatch execution layer controls the power distribution equipment of the green electricity grid connection link, the energy storage system, and regional charging stations according to the power supply ratio scheme. The energy accounting supervision layer calculates the dispatch and execution data and generates dispatch optimization suggestions. Local edge computing nodes perform basic dispatch in case of network anomalies. The cloud-based dual-backup redundancy architecture backs up operational data and dispatch strategies. The human-machine interface displays the operating status and receives manual intervention commands. All the above layers and modules interact with each other and synchronize commands through a unified communication protocol.

[0014] In this embodiment, the park is equipped with distributed photovoltaic (PV) power generation equipment and wind power generation equipment, with a PV installed capacity of 10MW and a wind power installed capacity of 5MW; the energy storage system capacity is 2MWh; three charging stations are set up in the park, equipped with DC charging piles and AC charging piles, to provide charging services for electric vehicles in and around the park. The traditional power grid is connected to the park's distribution system as a supplementary energy source, supplementing the charging stations when green electricity is insufficient and the energy storage system can no longer discharge. The above capacity, quantity, and site layout are only examples; in other embodiments, adjustments can be made according to the park size, charging load level, and green electricity installed capacity.

[0015] The multi-source data sensing layer includes a photovoltaic (PV) power output monitoring terminal, a wind power acquisition module, a regional power grid load sensor, a charging pile electricity metering node, and an energy storage system data acquisition module. The PV power output monitoring terminal collects PV power generation and power generation prediction deviations; the wind power acquisition module collects wind power and related wind speed information; the regional power grid load sensor collects the remaining power supply capacity and power supply of the traditional power grid; the charging pile electricity metering node collects real-time electricity consumption, real-time charging load, scheduled charging data, and operating status of each charging pile; and the energy storage system data acquisition module collects the state of charge, charging and discharging power, battery temperature, and charging and discharging status of the energy storage system. The multi-source data sensing layer also connects to the green electricity procurement platform of purchasing companies and the power grid company's electricity purchase settlement system to obtain green electricity procurement data and electricity purchase settlement data, including the traditional power grid electricity purchase price and real-time energy account data.

[0016] In this embodiment, the multi-source data perception layer collects various types of data according to a preset collection cycle, which can be 1 second. The collected data is first sent to the local edge computing node, where the local edge computing node performs data cleaning, deduplication, and outlier removal to form standardized data packets. Subsequently, the standardized data packets are cached on the local edge computing node and simultaneously sent to the data management module in the cloud dual-backup redundant architecture, and then uploaded to the artificial intelligence decision-making layer for subsequent scheduling and computation.

[0017] The AI ​​decision-making layer comprises a time-series forecasting model, a reinforcement learning model, and a multi-objective optimization model. The time-series forecasting model predicts green power output for a predetermined future period based on real-time meteorological data, historical green power generation data, and energy storage system operation data. The reinforcement learning model generates a power supply allocation scheme under constraints such as real-time green power output, energy storage system state of charge, energy storage system charging and discharging power thresholds, remaining power supply capacity of the traditional power grid, dynamic changes in regional charging load, and electricity purchase costs. The multi-objective optimization model corrects the power supply allocation scheme with objectives including prioritizing green power utilization, peak shaving and valley filling by energy storage systems, optimizing power supply costs, and achieving supply-demand balance.

[0018] In this embodiment, the time-series forecasting model can employ an LSTM time-series forecasting model. This model receives photovoltaic power generation, wind power generation, power generation forecast deviation, meteorological data, and energy storage system operation data as inputs, and outputs a short-term green power output forecast for the next 15 minutes. The reinforcement learning model receives real-time green power output, the aforementioned forecasts, the energy storage system's state of charge, real-time regional charging load, scheduled charging demand, traditional grid supply capacity, and electricity purchase cost data as state inputs, and outputs green power supply, energy storage system charging or discharging, and traditional grid replenishment energy as action outputs, thereby forming a power supply allocation scheme. A multi-objective optimization model performs constraint verification on the power supply allocation scheme to ensure that the generated power supply allocation scheme meets the energy storage system's operating boundaries, power supply balance conditions, and electricity purchase cost constraints.

[0019] In this embodiment, the AI ​​decision-making layer is configured with preset thresholds, including a preset discharge threshold and a preset charging threshold. The preset discharge threshold can be set to 20%, and the preset charging threshold can be set to 90%. When the state of charge (SBC) of the energy storage system is less than or equal to the preset discharge threshold, the AI ​​decision-making layer restricts the energy storage system from continuing to discharge; when the SBC of the energy storage system is greater than or equal to the preset charging threshold, the AI ​​decision-making layer restricts the energy storage system from continuing to charge. The AI ​​decision-making layer can also set thresholds for changes in green power output and regional charging load. When the green power output, regional charging load, or the SBC of the energy storage system exceeds the corresponding preset threshold, the AI ​​decision-making layer outputs an updated power supply ratio scheme.

[0020] The intelligent dispatch execution layer includes intelligent dispatch terminals. These terminals connect to the power distribution control system, grid connection switch, charging pile control module, energy storage system converter, and battery management system of the regional charging station. After receiving the power supply allocation scheme output by the artificial intelligence decision-making layer, the intelligent dispatch terminal converts the scheme into corresponding control commands and distributes them to each controlled device to execute power supply link switching and power regulation.

[0021] In one operational scenario, when the output of green electricity is greater than or equal to the regional charging load, and the state of charge of the energy storage system is lower than a preset charging threshold, the intelligent dispatch terminal controls the green electricity grid connection to be established, prioritizing green electricity to supply power to the charging stations. If there is still surplus power after the green electricity meets the charging load, the intelligent dispatch terminal controls the energy storage system to enter charging mode to absorb the remaining green electricity. Simultaneously, the intelligent dispatch terminal cuts off the power supply from the traditional power grid or reduces the power supply from the traditional power grid to a preset lower limit. At this time, the energy supplement from the traditional power grid in the power supply ratio scheme is zero or close to zero.

[0022] In another operating scenario, when the output of green electricity is less than the regional charging load, and the state of charge of the energy storage system is greater than or equal to a preset discharge threshold, the intelligent dispatch terminal maintains the grid-connected supply of green electricity and controls the energy storage system to discharge according to the load gap, thereby compensating for the insufficient output of green electricity. In this scenario, the intelligent dispatch terminal can dynamically adjust the discharge power of the energy storage system according to the size of the load gap, so that the sum of the green electricity supply and the energy storage system discharge meets the regional charging load demand.

[0023] In another operating scenario, when the green power output is zero and the state of charge of the energy storage system is less than or equal to a preset discharge threshold, the intelligent dispatch terminal controls the energy storage system to stop discharging and initiates supplementary power from the traditional power grid. The supplementary power from the traditional power grid is adjusted according to the real-time charging load of the region to meet the power supply demand of the regional charging network. At this time, the green power supply in the power supply matching scheme is zero, the energy storage system discharge is zero, and the supplementary energy from the traditional power grid matches the regional charging load.

[0024] The intelligent scheduling execution layer also allocates scheduled charging demand and off-peak charging demand in an orderly manner. Specifically, the intelligent scheduling terminal adjusts the start-up and shutdown times and output power of each charging pile based on the scheduled charging time, scheduled charging power, current charging station load, and available green electricity supply capacity to reduce load peaks. For charging piles with fault conditions, the intelligent scheduling terminal can remove them from the set of available charging piles and transfer the corresponding load to other available charging piles.

[0025] The energy accounting regulatory layer includes an automated energy accounting model. This model receives power distribution plan data from the AI-driven decision-making layer and actual execution data from the intelligent dispatch execution layer, and performs calculations according to a preset calculation cycle, which can be 1 minute. The automated energy accounting model at least calculates the actual consumption of green electricity, the charging and discharging of energy storage systems, the actual electricity purchased from the traditional power grid, and the energy supply and consumption data of each charging station. It then matches the statistical results with the green electricity purchase quotas of electricity-purchasing companies and their traditional power grid purchase invoices.

[0026] In this embodiment, the automatic energy accounting model determines whether an energy debt or energy surplus status exists based on the accounting results. An energy debt or energy surplus status includes situations where green electricity procurement is not fully utilized, excessive electricity purchases from the traditional grid, oversupply of energy at charging stations, energy shortage at charging stations, and excessively high charging and discharging operation and maintenance costs for energy storage systems. After identifying the corresponding status, the automatic energy accounting model generates allocation optimization suggestions and feeds these suggestions back to the artificial intelligence decision-making layer. The artificial intelligence decision-making layer adjusts the green electricity supply, energy storage system charging and discharging arrangements, and traditional grid replenishment strategies in subsequent power supply allocation plans based on the allocation optimization suggestions.

[0027] A dual-backup redundancy architecture, consisting of local edge computing nodes and a cloud backup, is used to improve the system's continuous operation capability. Under normal conditions, data collected by the multi-source data sensing layer is synchronously sent to both the local edge computing nodes and the cloud backup redundancy architecture, where the cloud side performs complete power supply allocation calculations and policy management. When the network is interrupted or the cloud experiences an anomaly, the local edge computing nodes switch to independent operation and execute the basic allocation algorithm. The basic allocation algorithm prioritizes green electricity supply, ensures that energy storage systems charge and discharge within threshold ranges, and utilizes the traditional power grid as a supplementary power source to maintain continuous power supply to regional charging stations.

[0028] In one exemplary emergency operation mode, when a local edge computing node detects that the communication interruption with the cloud has lasted for more than a preset duration, it activates a local emergency mode. In the local emergency mode, the edge computing node calls the most recently cached collected data and preset basic allocation rules to generate an emergency power supply allocation plan, and then sends this plan to the intelligent scheduling execution layer for execution. After the network is restored, the local edge computing node sends the collected data, execution records, and accounting data from the emergency operation period back to the cloud's dual-backup redundant architecture to complete data completion and policy recovery.

[0029] The human-machine interface (HMI) displays green power generation capacity, energy storage system status of charge, regional charging load, power supply ratio scheme, energy accounting results, and alarm information. Maintenance personnel can use the HMI to query the operating status of each charging station, view the power supply link status, and input manual intervention commands in special scenarios such as grid maintenance, extreme weather, or equipment maintenance. Manual intervention commands may include limiting the output power of some charging stations, adjusting the operating mode of the energy storage system, or forcibly connecting to the traditional grid for supplemental energy. The intelligent dispatch execution layer receives the manual intervention commands and executes the corresponding controls.

[0030] For example, before the morning charging peak, the AI ​​decision-making layer determines that the regional charging load will increase during a preset period based on scheduled charging data and green electricity forecasts. At this time, the energy storage system can be pre-charged during periods when green electricity is relatively abundant. During peak hours, if the output of green electricity is insufficient, the energy storage system will prioritize compensating for the load gap. Only when green electricity and the energy storage system are still insufficient to meet the load will the traditional power grid supplement the energy supply. Through the above scheduling method, the utilization rate of green electricity can be increased and the purchase of electricity from the traditional power grid can be reduced while ensuring continuous power supply to the charging network.

[0031] For example, when wind power output is high at night and regional charging load is low, the AI ​​decision-making layer can control the energy storage system to use the remaining green electricity for charging. During the morning peak hours of the following day, if photovoltaic output has not yet fully increased and regional charging load is rapidly increasing, the intelligent dispatch execution layer controls the energy storage system to discharge according to the power supply ratio scheme, thereby reducing the proportion of traditional grid power replenishment during high-price periods. The power supply ratio adjustment results and actual execution data in the above process are simultaneously recorded and calculated by the energy accounting regulatory layer for subsequent allocation optimization.

[0032] In this embodiment, green electricity can be generated from photovoltaic power generation and wind power generation, or one of these can be selected as the sole source of green electricity depending on the actual application scenario; the energy storage system can use solid-state batteries, semi-solid-state batteries, sodium-ion batteries, or flow batteries, in addition to lithium iron phosphate batteries. For different energy storage systems, the preset discharge threshold, preset charging threshold, and charging / discharging power threshold can be adjusted according to their rated capacity, allowable charge / discharge rate, and safe operating range.

[0033] The above are merely preferred embodiments of this application and are not intended to limit this application. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A traditional power grid regionalized charging deployment system based on AI and green electricity cooperation, characterized in that, It includes a multi-source data perception layer, an artificial intelligence decision-making layer, an intelligent scheduling and execution layer, an energy accounting and supervision layer, and an energy storage system; The multi-source data sensing layer is used to collect green power generation data, energy storage system operation data, traditional power grid power supply data, regional charging load data, green power procurement data, and electricity purchase settlement data. The artificial intelligence decision-making layer is connected to the multi-source data perception layer and is used to generate a power supply ratio scheme between green electricity, energy storage system and traditional power grid based on the green power generation data, the energy storage system operation data, the traditional power grid power supply data, the regional charging load data, the green electricity procurement data and the electricity purchase settlement data. The intelligent scheduling execution layer is connected to the artificial intelligence decision-making layer and is used to control the power distribution equipment of the green power grid connection link, energy storage system and regional charging station according to the power supply ratio scheme, so as to complete the regional charging load distribution and power supply switching. The energy accounting supervision layer is connected to the artificial intelligence decision-making layer and the intelligent dispatch execution layer respectively. It is used to calculate the green electricity consumption, energy storage system charging and discharging, and traditional grid electricity purchase based on dispatch data and execution data, and generate dispatch optimization suggestions based on the calculation results and feed them back to the artificial intelligence decision-making layer. The energy storage system is connected to the intelligent scheduling execution layer to realize green power buffer storage and charging and discharging regulation. The multi-source data sensing layer, the artificial intelligence decision-making layer, the intelligent scheduling execution layer and the energy accounting supervision layer interact with data and synchronize instructions through a unified communication protocol.

2. The system of claim 1, wherein, The multi-source data perception layer includes a photovoltaic power output monitoring terminal, a wind power acquisition module, a regional power grid load sensor, a charging pile electricity metering node, and an energy storage system data acquisition module. It is connected to the green electricity procurement platform of the electricity purchasing enterprise and the power grid company's electricity purchase settlement system. The multi-source data perception layer collects and synchronizes green power generation data, energy storage system operation data, traditional power grid power supply data, regional charging load data, charging pile operation data, green electricity procurement data, and electricity purchase settlement data in real time through the communication network, and constructs a dynamic database for the artificial intelligence decision-making layer to call.

3. The system of claim 1, wherein, The artificial intelligence decision-making layer includes a time-series prediction model, a reinforcement learning model, and a multi-objective optimization model. The time-series prediction model is used to predict the short-term output of green power based on real-time meteorological data, historical green power generation data, and energy storage system operation data. The reinforcement learning model is used to generate the power supply ratio scheme under the constraints of real-time green power output, energy storage system state of charge, energy storage system charging and discharging power thresholds, remaining power supply capacity of the traditional power grid, dynamic changes in regional charging load, and electricity purchase cost. The multi-objective optimization model is used to correct the power supply ratio scheme with the objectives of prioritizing the use of green power, peak shaving and valley filling of energy storage system, power supply cost optimization, and supply and demand balance. When the green power output, regional charging load, or energy storage system state of charge exceeds a preset threshold, the updated power supply ratio scheme is output. The preset thresholds include a preset discharge threshold and a preset charging threshold.

4. The system of claim 3, wherein, The intelligent scheduling execution layer includes an intelligent scheduling terminal, which is connected to the power distribution control system, grid connection switch, charging pile control module, energy storage system converter and battery management system of the regional charging station; It is configured to: when the green power output is greater than or equal to the regional charging load and the state of charge of the energy storage system is lower than the preset charging threshold, control the green power to be connected to the grid and control the charging of the energy storage system, while cutting off or reducing the power supply of the traditional grid. When the output of green electricity is less than the regional charging load and the state of charge of the energy storage system is greater than or equal to the preset discharge threshold, the green electricity is controlled to be connected to the grid and the energy storage system is controlled to discharge according to the load gap. When the output of green electricity is zero and the state of charge of the energy storage system is less than or equal to the preset discharge threshold, the energy storage system is controlled to stop discharging and the traditional power grid is started to replenish energy. The scheduled charging demand and off-peak charging demand are allocated in an orderly manner to smooth out charging load peaks.

5. The system of claim 1, wherein, The energy accounting supervision layer includes an automatic energy accounting model. This model is used to connect with the operational data of the artificial intelligence decision-making layer and the intelligent dispatch execution layer. It performs real-time statistics on the actual consumption of green electricity, the charging and discharging of energy storage systems, the actual electricity purchased by the traditional power grid, and the energy supply and consumption data of each charging station. It automatically matches the green electricity purchase quota of the electricity purchasing company with the electricity purchase bill of the traditional power grid, and generates allocation optimization suggestions based on the energy debt status or energy surplus status, so as to feed back to the artificial intelligence decision-making layer to adjust the power supply ratio scheme.

6. The system of claim 1, wherein, It also includes local edge computing nodes, a cloud dual-backup redundancy architecture, and a human-machine interface. The local edge computing nodes are used to independently run basic dispatching algorithms when the network is interrupted. The cloud dual-backup redundancy architecture is used to back up system operation data and dispatching strategies. The human-machine interface is used to display system operation data, power supply ratio schemes, and accounting data, and to receive manual intervention commands.

7. The system of claim 2, wherein, The green power generation data includes green power generation capacity and power generation forecast deviation; the energy storage system operation data includes state of charge, charging and discharging power, battery temperature, and charging and discharging status; the regional charging load data includes real-time charging load and charging demand reservation data; and the electricity purchase settlement data includes traditional grid electricity purchase price and real-time energy account data.

8. The system of claim 3, wherein, The power supply ratio scheme specifies the green power supply, energy storage system charging or discharging amount, and traditional grid supplementary energy for each charging station. The preset discharge threshold is no more than 20%, and the preset charging threshold is no less than 90%.

9. The system of claim 5, wherein, The energy debt status or energy surplus status includes the incomplete consumption of green electricity purchases, excessive purchases of electricity from traditional power grids, oversupply of energy at charging stations, shortage of energy supply at charging stations, and excessively high charging and discharging operation and maintenance costs of energy storage systems.

10. The system according to claim 1, characterized in that, The green electricity includes at least one of photovoltaic power generation and wind power generation, and the energy storage system includes solid-state batteries, semi-solid-state batteries, lithium iron phosphate batteries, sodium-ion batteries, or flow batteries.

Images