Main circuit topology based on integrated over-charging device for light storage and charging and control method

Abstract

The invention relates to the field of new energy charging, in particular to a main circuit topology based on an integrated over-charging device for light storage and charging and a control method. The system is composed of the following parts: an ACDC module, a charging DCDC module, a photovoltaic DCDC module, a DCAC module and a battery, wherein the battery is used for storing redundant electric quantity and discharging the electric quantity to a charging pile for use; the invention provides a main circuit topology based on direct-current coupling, a very simple network architecture and an AI intelligent algorithm are fused to greatly expand the application scene of optical storage and charging in the overcharging field, meanwhile, a dynamic intelligent algorithm is combined, the dependence of an overcharging station on a power grid is reduced, and through an economic operation strategy, the system is more reliable and reliable. Efficient collaboration and accurate distribution of energy of all links of light, storage, charging and utilization are achieved, and the operation capacity and economical efficiency of a charging station are improved.

Description

Technical Field

[0001] This invention relates to the field of new energy charging, specifically to a main circuit topology and control method for an integrated supercharging device based on photovoltaic, energy storage, and charging. Background Technology

[0002] Against the backdrop of the electrification of transportation, society and users are increasingly demanding faster battery charging capabilities. Fast charging, especially ultra-fast charging technology that allows charging to be as fast as refueling, has attracted widespread attention and become a key technology for promoting the popularization of new energy vehicles, as well as a focus of research and development by major economies around the world. Currently, ultra-fast charging faces challenges such as high grid load, unstable grids in underdeveloped areas leading to charging difficulties, and the single business model and long payback period of ultra-fast charging stations.

[0003] Currently, charging and energy storage are mostly connected in parallel on the AC side, with energy storage and charging being independent units. This solution has high technological maturity, but it has disadvantages such as low level of intelligence, high cost, and large footprint. Summary of the Invention

[0004] To address the problems of existing technologies, this invention provides a main circuit topology and control method for an integrated photovoltaic-storage-charging-utilization supercharging device. This invention solves the problems existing in current ultrafast charging stations by achieving deep DC coupling between light, energy storage, charging, and power utilization.

[0005] To achieve the above objectives, the present invention adopts the following technical solution:

[0006] A main circuit topology for an integrated supercharging device based on photovoltaic, energy storage, and charging consists of the following parts:

[0007] ACDC module: Responsible for providing energy to the DC bus and supporting grid connection. When used in areas without electricity, the ACDC module can be removed.

[0008] Charging DC-DC module: responsible for providing power to the charger, can support parallel applications, and can support up to 60 units in parallel. The number of charging DC-DC modules can be adjusted according to project requirements.

[0009] Photovoltaic DC-DC module: Connected to photovoltaic panels, it provides photovoltaic power to the DC bus, and the photovoltaic power generation can be adjusted through control strategies;

[0010] DCAC module: The inverter of the DCAC module can convert DC to AC to power the auxiliary power supply of the charging pile, and can also provide power to other external power needs.

[0011] Battery: Stores excess power and discharges it to the charging station.

[0012] A control method based on the main circuit topology of an integrated photovoltaic, energy storage, charging, and utilization supercharging device includes the following steps:

[0013] Step 1: Photovoltaic-storage-charging-energy dispatching control method

[0014] First, the device conducts self-check. After a fault occurs in the self-check device, corresponding fault protection logic actions are executed. When the self-check is normal, following the principle of charging priority, it is judged whether charging is in progress. If not charging, considering the switching of the energy replenishment logic, the charging pile needs to be idle for 5 minutes. If charging, it is then judged whether the SOC of the battery is within the set range. If the battery capacity is outside the SOC range;

[0015] At this time, the required power of the charging pile needs to be limited to: EV_Demand = (P_dch_max + PV + P1 - P2);

[0016] The control method based on the main circuit topology of the integrated photovoltaic-storage-charging supercharging device is characterized by further including the following steps:

[0017] If the remaining capacity of the battery is within the range allowed by the SOC, it is then judged whether the required power of the charging pile meets the requirement of the formula EV_Demand < (P_dch_max + PV + P1 - P2). If it meets, it operates normally. If it does not meet, it means that the required power of the charging pile is greater than the maximum capacity provided by the current device itself, and the required power of the charging pile needs to be adjusted to: EV_Demand = (P_dch_max + PV + P1 - p2);

[0018] The control method based on the main circuit topology of the integrated photovoltaic-storage-charging supercharging device further includes the following steps:

[0019] Step 2: Energy storage replenishment control strategy SOC > 90%, at this time the battery is in a relatively full state, and the power of the ACDC module is set to 0; Condition 1: If the photovoltaic power generation is greater than the power consumption, P1_set = 0; Photovoltaic power PV_set = P2 + 2; Condition 2: PV > 5 && PV < P2, that is, the photovoltaic is generating electricity and the photovoltaic power generation is less than the power consumption. At this time, the photovoltaic power is not restricted; Condition 3: PV < 5, at this time the photovoltaic power generation is relatively small;

[0020] The control method based on the main circuit topology of the integrated photovoltaic-storage-charging supercharging device further includes the following steps:

[0021] 10% < SOC < 90%. Condition 1: When the ACDC module is in the operating state and the photovoltaic power generation is greater than the power consumption, under the current working condition, the photovoltaic power is not restricted, and the excess power is stored in the energy storage battery. Combining with the coordinated strategy of photovoltaic, energy storage, charging, and power consumption, it is predicted that there is no charging demand in a short period. At this time, the power of the ACDC module is limited to 0, and the excess photovoltaic power generation is used to charge the energy storage battery. If it is predicted that there is a charging demand in a short period, the power of the ACDC module is controlled to 100 kW to quickly charge the energy storage battery. Condition 2: PV > 5 && PV < P2, that is, the photovoltaic is generating electricity and the photovoltaic power generation is less than the power consumption. At this time, the photovoltaic power is not restricted. Combining with the prediction algorithm and electricity price policy, the power of the ACDC module is dynamically adjusted to charge the energy storage. The specific implementation strategy is as follows: when the electricity price is cheap, the ACDC module is set to 100 kW for rapid charging; when the electricity price is relatively expensive and there is no predicted charging demand, the power of the ACDC module is 0. After the SOC reaches the lower limit, the ACDC module is started to charge the energy storage, and the power is set to a small power for energy replenishment at this time; when the electricity price is relatively expensive and there is a predicted charging demand, the power of the ACDC module is set to 100 kW to quickly charge the energy storage. Condition 3: When PV < 5, the photovoltaic power generation is weak, and the strategy is the same as Condition 2.

[0022] The control method based on the main circuit topology of the integrated photovoltaic, energy storage, charging, and power consumption supercharging device further includes the following steps:

[0023] C. When SOC < 10%, the battery is in the empty state, and energy is preferentially replenished to the battery. Condition 1: When the photovoltaic power generation is greater than the power consumption, the excess energy is used to replenish the battery; Condition 2: When the photovoltaic power generation is less than the power consumption, the power consumption DCAC is shut down to protect the battery at this time; Condition 3: When the photovoltaic does not generate electricity, the power consumption DCAC is shut down at this time.

[0024] The control method based on the main circuit topology of the integrated photovoltaic, energy storage, charging, and power consumption supercharging device further includes the following steps: Step 3: The coordinated control strategy of photovoltaic, energy storage, charging, and power consumption first performs multi-source data collection, cleans, preprocesses, and extracts feature quantities from the collected data. Through deep learning algorithms, the multi-source data is analyzed to predict the photovoltaic power generation, user charging demand, peak and valley periods of the power grid, and power consumption information in advance. When the light is sufficient, the photovoltaic power is preferentially used to charge the electric vehicle, and the excess electric energy is stored in the energy storage device; during the peak period of the power grid, the energy storage device releases electric energy to meet the charging demand, reducing the dependence on the power grid and lowering the power consumption cost. The health status of the energy storage device is monitored in real time, and the charge and discharge strategy of the energy storage device is dynamically adjusted. At the same time, the operation results are fed back to the deep learning regulation model in real time to optimize the energy scheduling model.

[0025] Compared with existing technologies, the beneficial effects of the invention are: the invention proposes a DC-coupled main circuit topology, a simplified network architecture, and integrates AI intelligent algorithms to greatly expand the application scenarios of photovoltaic, energy storage, charging and utilization in the supercharging field. At the same time, combined with dynamic intelligent algorithms, it reduces the dependence of supercharging stations on the power grid. Through economical operation strategies, it achieves efficient coordination and precise allocation of energy in the photovoltaic, energy storage, charging and utilization links, thereby improving the operation capability and economy of charging stations.

[0026] This invention employs a DC coupling method of light, energy storage, charging, and utilization. Through the regulating effect of photovoltaics and energy storage, it can greatly reduce the grid load pressure and mitigate the impact of high-power charging on the grid.

[0027] This invention supports grid-connected and off-grid operation through intelligent control strategies, and is suitable for charging applications in areas with weak grids and no electricity.

[0028] This invention uses deep learning algorithms to analyze multi-source data such as weather forecasts, grid load, user charging data, and energy storage monitoring status to dynamically adjust charging and discharging strategies, thereby achieving efficient coordination and precise allocation of energy in all aspects of photovoltaic, energy storage, charging, and energy use, and improving the operational capabilities and economic efficiency of charging stations. Attached Figure Description

[0029] Other features, objects, and advantages of the invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings.

[0030] Figure 1 This is a schematic diagram of the main circuit topology of the integrated photovoltaic storage and charging device of the present invention.

[0031] Figure 2 This is a schematic diagram of the network architecture of the integrated photovoltaic storage, charging, and utilization device of the present invention.

[0032] Figure 3 This is a diagram of the collaborative control strategy for photovoltaic storage, charging, and utilization in this invention.

[0033] Figure 4 This is the battery charging logic diagram of the present invention.

[0034] Figure 5 This is a diagram of the photoelectric storage-charging collaborative control method of the present invention.

[0035] Figure 6 This is a schematic diagram of the main circuit topology of the integrated photovoltaic storage and charging device of the present invention.

[0036] Figure 7 This is a schematic diagram of the main circuit topology of the integrated photovoltaic storage and charging device of the present invention.

[0037] Figure 8 This is a diagram of the collaborative control strategy for photovoltaic storage, charging, and utilization in this invention.

[0038] Figure 9This is the battery charging logic diagram of the present invention.

[0039] Figure 10 This is a diagram of the photoelectric storage-charging collaborative control method of the present invention. Detailed Implementation

[0040] The present invention will be further described in detail below through embodiments. These embodiments are only used to illustrate the present invention and do not limit the scope of the present invention.

[0041] A main circuit topology based on an integrated supercharging device for photovoltaic storage and charging.

[0042] like Figure 1 , Figure 2 As shown, it consists of the following parts:

[0043] ACDC module: Responsible for providing energy to the DC bus and supporting grid connection. When used in areas without electricity, the ACDC module can be removed.

[0044] Charging DC-DC module: Responsible for providing power to the charger, it can support parallel applications, with a maximum of 60 units in parallel. The number of charging DC-DC modules can be adjusted according to project requirements.

[0045] Photovoltaic DC-DC module: Connected to photovoltaic panels, it provides photovoltaic power to the DC bus, and the photovoltaic power generation can be adjusted through control strategies;

[0046] DCAC module: The inverter of the DCAC module can convert DC to AC to power the auxiliary power supply of the charging pile, and can also provide power to other external users.

[0047] Batteries: Store excess power and discharge it to the charging station. Batteries store excess energy on the DC bus and discharge it to the charging station or other loads when needed (such as when photovoltaic power generation is insufficient, when the grid is out of service, or when there is high charging demand).

[0048] In this scheme, all energy sources and loads are connected to a common DC bus, and energy is converted and distributed through different converters. This topology facilitates energy management and control.

[0049] A control method based on the main circuit topology of an integrated supercharging device for photovoltaic storage and charging.

[0050] like Figure 3 , Figure 4 , Figure 5 As shown:

[0051] The photovoltaic-storage-charging-utilization coordinated control method consists of three parts: photovoltaic-storage-charging-utilization energy dispatch control method, energy storage replenishment control strategy, and photovoltaic-storage-charging-utilization coordinated control strategy.

[0052] The energy dispatch and control method for photovoltaic, energy storage, charging, and utilization is as follows: First, the equipment performs a self-test. If a fault is found in the self-test equipment, the corresponding fault protection logic is executed. When the self-test is normal, the charging priority principle is followed to determine whether the battery is charging. If the battery is not charging, the charging pile needs to be idle for 5 minutes to consider the switching of the energy replenishment logic. This time can be adjusted according to the actual situation. If the battery is charging, the system charge (SOC) of the battery is determined to be within the set range. If the battery capacity is outside the SOC range, the system charge is determined to be charged.

[0053] At this time, the power demand of the charging pile needs to be limited to: EV_Demand=(P_dch_max+PV+P1-P2);

[0054] P1_set should be set to the rated power of the ACDC module, which should not exceed the battery's allowed charging power.

[0055] If the remaining battery capacity is within the SOC allowable range, then determine whether the charging pile's required power meets the requirement of EV_Demand<(P_dch_max+PV+P1-P2). If it does, it will operate normally. If it does not, it means that the charging pile's required power is greater than the maximum capacity provided by the device itself, and the charging pile's required power needs to be adjusted to: EV_Demand=(P_dch_max+PV+P1-p2).

[0056] third:

[0057] The design principles of the energy storage and replenishment control strategy are: first, to utilize charging idle time; and second, to maximize the use of photovoltaic power generation.

[0058] It should be noted that

[0059] 1. The scope of SOC can be agreed upon according to the project; this invention does not limit the specific value of SOC.

[0060] 2. The power of the AC-CDC module, photovoltaic DC-CDC module, DC-AC module, etc. used in this article is not limited. For the convenience of describing the strategy logic, the AC-CDC module is 100kW as an example, the DC-CDC module is 40kW as an example, the quantity is 3, and the DC-AC module power is 100kW.

[0061] A. When SOC > 90%, the battery is in a relatively full charge state, and the power of the ACDC module is set to 0. Condition 1: If the photovoltaic power generation is greater than the power consumption (PV > 5 is the deadband set for control, and this deadband can be adjusted), that is, PV > 5 && PV > P2. Under the current working condition, the battery is about to be fully charged, and there is excessive photovoltaic power generation. It is necessary to control the power of the ACDC module and the photovoltaic power. The power setting of the ACDC module is: P1_set = 0; the photovoltaic power PV_set = P2 + 2; here, +2 is the control deadband, and the following is similar. Condition 2: PV > 5 && PV < P2, that is, the photovoltaic is generating electricity, and the photovoltaic power generation is less than the power consumption. At this time, there is no restriction on the photovoltaic power. Condition 3: PV < 5, at this time, the photovoltaic power generation is relatively small, and the processing logic is the same as Condition 2 (this step).

[0062] B. When 10% < SOC < 90%, the energy storage can be charged and discharged. Condition 1: When the ACDC module is in the operating state and the photovoltaic power generation is greater than the power consumption. Under the current working condition, there is no restriction on the photovoltaic power, and the excess power is stored in the energy storage battery. Combining the coordinated strategy of photovoltaic, energy storage, charging, and power consumption, if there is no charging demand predicted in a short time, the power of the ACDC module is limited to 0 at this time, and the excess photovoltaic power generation is used to charge the energy storage battery; if there is a charging demand predicted in a short time, the power of the ACDC module is controlled to 100 kW at this time to quickly charge the energy storage battery. Condition ②: PV > 5 && PV < P2, that is, the photovoltaic is generating electricity, and the photovoltaic power generation is less than the power consumption. At this time, there is no restriction on the photovoltaic power. Combining the prediction algorithm and the electricity price policy, the power of the ACDC module is dynamically adjusted to charge the energy storage. The specific implementation strategy is: when the electricity price is cheap, the ACDC module is set to 100 kW for quick charging; when the electricity price is relatively expensive + no charging demand is predicted, the power of the ACDC module is 0 at this time. After SOC reaches the lower limit, the ACDC module is started to charge the energy storage, and the power is set to small - power energy replenishment at this time; when the electricity price is relatively expensive + a charging demand is predicted, the power of the ACDC module is set to 100 kW at this time to quickly charge the energy storage. Condition ③: When PV < 5, the photovoltaic power generation is weak at this time, and the strategy is the same as Condition 2 (this step). It should be noted that if the ACDC module is in a non - operating state, it is necessary to automatically start the operation of the ACDC module through strategy distribution to meet the use under this condition;

[0063] C. When SOC < 10%, the battery is in a discharged state. Priority is given to replenishing the battery, using the same strategy as in B. It should be noted that the strategy must meet the power demand. If there is no predicted charging demand, the minimum condition is P1 = P2 - PV. When the AC / DC module is in a fault shutdown state, since the battery is in a discharged state, the strategy prioritizes battery protection. Condition 1: When the photovoltaic power generation is greater than the power consumption, the excess energy is used to replenish the battery. Condition 2: When the photovoltaic power generation is less than the power consumption, the battery protection will shut down the power-consuming DC / AC module. Condition 3: When the photovoltaic does not generate power, the power-consuming DC / AC module will be shut down.

[0064] Fourth: Photovoltaic-storage-charging-utilization coordinated control strategy

[0065] First, multi-source data is collected, including: meteorological forecast data, real-time grid load data, user charging reservation data, user electricity consumption data, and energy storage device health status data. The collected data undergoes data cleaning, preprocessing, and feature extraction. Deep learning algorithms are then used to analyze the multi-source data to predict photovoltaic power generation, user charging demand, grid peak and off-peak periods, and electricity consumption information in advance. When there is sufficient sunshine, photovoltaic power is prioritized for charging electric vehicles, and excess energy is stored in energy storage devices. During peak grid periods, energy storage devices release electricity to meet charging demand, reducing dependence on the grid and lowering electricity costs. The health status of energy storage devices is monitored in real time, and the charging and discharging strategies of energy storage devices are dynamically adjusted. At the same time, the operational results are fed back to the deep learning control model in real time to optimize the energy scheduling model and achieve efficient coordination and precise allocation of energy in all aspects of photovoltaic, storage, charging, and consumption.

[0066] The innovative points and technical features of this invention's needle:

[0067] This invention simplifies the network architecture through a DC-coupled topology and uses intelligent control strategies to reduce the pressure on the power grid and mitigate the impact of high-power charging on the power grid. It can support grid-connected and off-grid operation, is suitable for charging applications in weak grids and areas without electricity, and expands the application scenarios of photovoltaic energy storage and charging.

[0068] The dynamic energy scheduling method based on multi-source data fusion integrates AI algorithms with energy scheduling and utilization to achieve efficient coordination and precise allocation of energy in all aspects of photovoltaic, storage, charging and utilization.

[0069] The principle details of this scheme are as follows: The control method based on the main circuit topology of the integrated photovoltaic, energy storage, charging and utilization supercharging device (photovoltaic, energy storage, charging and utilization collaborative control method) consists of three parts: photovoltaic, energy storage, charging and utilization energy scheduling control method, energy storage and replenishment control strategy, and photovoltaic, energy storage, charging and utilization collaborative control strategy.

[0070] Photovoltaic-storage-charging-utilization energy dispatch and control method:

[0071] 1. Initialization and self-test: First, the equipment (photovoltaic inverter, energy storage converter, battery, charging pile, etc.) performs a self-test. If a fault is found in the self-test equipment, the corresponding fault protection logic action is executed, such as stopping charging or disconnecting from the grid, to ensure safety. When all equipment self-tests are normal, the system enters the normal energy dispatch mode.

[0072] 2. Charging priority principle and charging pile status judgment:

[0073] Once the self-check is normal, the charging priority will be given to electric vehicles if energy is available; the system will also determine if any vehicles are currently charging.

[0074] Scenario 1: No vehicles are charging.

[0075] If the system is not charging, it will enter a waiting or ready state. Considering the switching of the charging logic, a charging pile idle time is set (e.g., the charging pile must be idle for 5 minutes if it is not charging for 5 minutes, and this time can be adjusted according to the actual situation).

[0076] Scenario 2: A vehicle is currently charging.

[0077] The system will continue to supply power to the current vehicle and proceed to the next step of the judgment.

[0078] 3. Battery status assessment and charging power adjustment (core logic)

[0079] When a vehicle is charging, the system checks the battery's remaining state of charge (SOC) to see if it is within a preset reasonable range (e.g., neither too full nor too empty).

[0080] Sub-case A: Battery SOC is outside the preset range (battery protection required):

[0081] The battery may need to be charged (SOC too low) or stopped charging (SOC too high) to protect its lifespan and performance.

[0082] The formula for calculating the power demand (EV_demand) of a charging station is as follows:

[0083] EV_Demand=(P_dch_max+PV+P1-P2);

[0084] P_dch_max: The maximum allowable discharge power of the battery at present (the value is positive if the battery needs to discharge; the value may be 0 or negative if the battery needs to charge. The formula implies the allowable discharge power of the battery in its current state).

[0085] PV: The actual power generated by the current photovoltaic array.

[0086] P1: Current actual active power output of the AC-CDC module (photovoltaic inverter to DC side).

[0087] P2: Current actual output active power (current power consumption) of the DCAC module (energy storage converter to AC side).

[0088] Other variable definitions: EV: the sum of current charging power;

[0089] P1_set: Current power setting of the ACDC module; Note: When the ACDC module is needed to charge the battery, P1_set will be set to the rated power (100kW in the example), but will not exceed the maximum charging power allowed by the battery, P_ch_max.

[0090] PV_set: The sum of the power settings for the photovoltaic modules;

[0091] EV_set_max: Current maximum operating power during charging;

[0092] P_ch_max: Maximum allowed charging power of the battery;

[0093] Sub-case B: Battery SOC is within the preset range (battery condition is good and can be charged):

[0094] The system allows the battery to charge and discharge as needed to match the charging process;

[0095] The system will check whether the charging pile's power demand (EV_demand) meets the following condition: EV_Demand < (P_dch_max + PV + P1 - P2).

[0096] If the following conditions are met: (EV_demand <= total available power):

[0097] This means that the current photovoltaic power generation, battery discharge capacity, and power provided by the AC / DC module, minus the power consumed by the DC / AC module, are sufficient to meet the needs of charging piles.

[0098] The system will operate normally and charge the electric vehicle according to the EV_demand power.

[0099] If the requirement of (EV_demand > total available power) is not met:

[0100] The power that the charging pile wants to absorb exceeds the maximum power that can be provided by all energy sources of the current system (battery discharge, photovoltaic, ACDC module) minus the current consumption (DCAC module).

[0101] To avoid potential grid instability or equipment overload, the system needs to adjust the charging pile's power demand, limiting it to the system's current maximum capacity. The adjusted power is: EV_Demand = (P_dch_max + PV + P1 - P2)

[0102] This ensures that the charging power is limited to the maximum energy supply level currently available to the system.

[0103] The control method based on the main circuit topology of the integrated photovoltaic-storage-charging supercharger is a real-time control logic based on real-time status and preset rules. This ensures successful device self-testing and fault protection. Charging priority: Charging demand is met as much as possible when energy is available. Battery protection: When the battery SOC is unsuitable, charging power is limited to protect the battery. When energy is insufficient, charging power is dynamically adjusted to ensure stable system operation and not exceed the power limits of each component. It continuously calculates and compares the relationship between EV_demand and P_dch_max + PV + P1 - P2 to determine whether to meet charging demand or limit charging power, and accordingly sets the power output of the ACDC module.

[0104] Energy storage and replenishment control strategy: The design principles of the energy storage and replenishment control strategy are: First, charging idle time (when there are no vehicles charging at the charging pile, it is a good time to carry out energy storage and replenishment, which can avoid energy competition); Second, maximize the use of photovoltaic power generation (prioritize the use of excess energy generated by photovoltaic power generation to charge the battery, thereby improving the utilization rate of renewable energy).

[0105] It should be noted that

[0106] 1. The scope of SOC can be agreed upon according to the project; this invention does not limit the specific value of SOC.

[0107] 2. The power of the AC-CDC module, photovoltaic DC-CDC module, DC-AC module, etc. used in this article is not limited. For the convenience of describing the strategy logic, the AC-CDC module is 100kW as an example, the DC-CDC module is 40kW as an example, the quantity is 3, and the DC-AC module power is 100kW.

[0108] SOC range control strategy:

[0109] A. When SOC > 90% (battery relatively fully charged), the control strategy aims to: avoid overcharging the battery to prevent damage from continuous charging; and manage excess energy: when photovoltaic power generation exceeds current electricity demand, this excess energy needs to be handled strategically, rather than being fully charged into the already nearly fully charged battery.

[0110] Control logic:

[0111] 1. Basic Settings:

[0112] Set the ACDC module power to 0; that is, P1_set = 0, to stop drawing power from the grid to charge the battery.

[0113] 2. Condition Judgment and Processing:

[0114] Condition 1: PV>5 && PV>P2

[0115] Meaning: Photovoltaic power generation (PV) is greater than a threshold (5, used to control hysteresis / dead zone, this value can be adjusted) and photovoltaic power generation is greater than P2.

[0116] Operating condition analysis: This means that photovoltaic power generation not only meets current electricity demand but also has a surplus. At the same time, the batteries are already very full and cannot receive much more energy.

[0117] Controlling actions:

[0118] P1_set = 0: Remain unchanged, do not draw power from the grid.

[0119] PV_set = P2 + 2: Sets the output power of the photovoltaic module to the current power consumption P2 plus a positive number (here, 2, also used as a control hysteresis). By limiting the actual output power of the photovoltaic module to slightly exceed the current power demand, it just meets the electricity requirement and avoids generating excessive excess energy. The extra 2 (or an adjustable hysteresis value) ensures stable photovoltaic output and avoids frequent fluctuations when PV approaches P2. In this way, excess energy is kept to a minimum, satisfying the power demand while avoiding charging a fully charged battery or wasting a lot of energy.

[0120] Condition 2: PV > 5 && PV <P2

[0121] Meaning: The photovoltaic power generation is greater than the threshold 5 but less than the current power consumption P2.

[0122] Operating condition analysis: Photovoltaic power generation is insufficient to fully meet current electricity demand. This means that the power gap needs to be supplemented by other sources, usually battery discharge or the power grid (but at this time P1_set = 0, the power grid does not supply power).

[0123] Control action: No limit is placed on photovoltaic power (PV_set remains unchanged, which is usually understood as allowing photovoltaics to generate electricity at its maximum capacity, i.e., PV_set = PV).

[0124] Objective: To allow photovoltaic power to generate electricity to its full potential. Any shortfall will be supplemented by battery discharge (because even when a battery is full, a SOC > 90% usually means it still has some discharge capacity, unless there is a special protection strategy that completely prohibits discharge when the SOC > 95% or higher).

[0125] Condition 3: PV < 5

[0126] Meaning: Photovoltaic power generation is less than the set threshold of 5.

[0127] Operating condition analysis: Photovoltaic power generation is very small and its contribution to meeting electricity demand is limited.

[0128] Control action: The processing logic is the same as condition two, that is, there is no limit on photovoltaic power.

[0129] Objective: To allow photovoltaic power generation to continue even when it is minimal. Electricity demand will primarily be met through battery discharge. The threshold of 5 is established to prevent interference with the control system or frequent switching of control strategies during periods of extremely low or fluctuating photovoltaic output.

[0130] When the battery SOC is greater than 90%, the core of the control strategy is to stop the grid input (P1_set = 0) to prevent unnecessary charging.

[0131] Electricity and overcharging risks. Intelligent management of photovoltaic output: When photovoltaic power generation is sufficient (exceeding electricity demand), photovoltaic output is limited to prevent it from escalating.

[0132] Slightly higher than electricity consumption to avoid overcharging or waste due to excessive energy generation. When photovoltaic power generation is insufficient (below electricity demand), it should not be used.

[0133] The strategy involves limiting photovoltaic output to generate as much electricity as possible, with batteries supplementing any shortfall. When batteries are fully charged, this strategy prioritizes meeting electricity demand.

[0134] Simultaneously handling excess photovoltaic power generation demonstrates a balance between battery protection and energy management efficiency.

[0135] B. When 10% < SOC < 90%, the battery is in a rechargeable and dischargeable operating range. The goal of the control strategy is:

[0136] Maximize the use of photovoltaic power generation.

[0137] While meeting current electricity demand, optimize battery charging and discharging to prepare for future needs (especially charging needs).

[0138] By combining electricity price information from the power grid, the overall cost of electricity can be reduced.

[0139] Ensure stable system operation;

[0140] Control logic decomposition:

[0141] Prerequisites: The ACDC module is in operation (power grid connection is normal).

[0142] Condition judgment and processing:

[0143] Condition 1: The ACDC module is running, and PV > 5 && PV > P2

[0144] Meaning: The photovoltaic power generation is sufficient, not only meeting the current electricity demand P2, but also having a surplus, and the power generation exceeds the threshold 5.

[0145] Operating condition analysis: The system has a surplus of energy, which can be used to charge the battery.

[0146] Controlling actions:

[0147] There are no restrictions on photovoltaics: photovoltaics are allowed to generate electricity at maximum power (or its set value), and renewable energy sources are given priority.

[0148] Excess electricity is stored in the battery: The portion of photovoltaic power generated that exceeds the electricity demand (PV-P2) is charged into the battery.

[0149] Combining the photovoltaic-storage-charging-utilization coordinated strategy, 1) it is predicted that there will be no charging demand in the short term:

[0150] Limit the AC / DC module power to 0 to stop drawing power from the grid. Since the photovoltaic system can already provide enough energy (to meet electricity needs and charge the battery), and there is no immediate need for charging, there is no need to draw additional energy from the grid. Utilize the surplus photovoltaic power to charge the battery: rely entirely on the surplus photovoltaic power to charge the battery.

[0151] 2) Predicted short-term charging demand:

[0152] Controlling the ACDC module power to 100kW: Obtaining the maximum allowable power from the grid (assuming 100kW is the rated power of the ACDC module or the maximum value of the system design); quickly "recharging" the battery to ensure that the battery has enough charge to discharge and support charging when the predicted charging demand arrives.

[0153] Condition 2: The ACDC module is running, and PV > 5 && PV <P2

[0154] Meaning: The photovoltaic system is generating electricity, but the amount of electricity generated is insufficient to fully meet the current electricity demand P2, and the amount of electricity generated exceeds the threshold 5.

[0155] Operating condition analysis: Photovoltaic power can partially meet electricity demand, but there is still a shortfall. The system needs to decide whether to draw power from the grid to supplement this shortfall and consider whether to use low-cost electricity to replenish the batteries.

[0156] Controlling actions:

[0157] There are no restrictions on photovoltaic power output: photovoltaics are allowed to generate electricity to the fullest extent.

[0158] By combining prediction algorithms and electricity pricing policies, the power of the ACDC module is dynamically adjusted to supplement the energy storage:

[0159] Electricity is cheap:

[0160] Setting the ACDC module to 100kW: During periods of low electricity prices, even if the photovoltaic system cannot fully meet the electricity demand, it will prioritize obtaining the maximum possible power (100kW) from the grid. This not only meets the current electricity shortage but also allows the remaining energy to quickly recharge the battery in preparation for subsequent periods of high prices or charging needs.

[0161] High electricity prices + no anticipated demand for charging:

[0162] The ACDC module power is set to 0: To avoid increasing electricity costs and because there is no immediate need for charging (meaning the battery is fully charged or does not require immediate preparation), power will not be drawn from the grid for the time being. The power shortage will be supplemented by battery discharge. Once the SOC reaches its lower limit, the ACDC module will be activated to charge the energy storage, at which point the power will be set to low-power replenishment.

[0163] High electricity prices + anticipated charging demand:

[0164] The ACDC module power is set to 100kW: Although electricity prices are high, charging demand is predicted to arrive soon, necessitating ensuring sufficient battery charge. In this case, prioritizing charging demand, maximum power is drawn from the grid to quickly replenish the battery, reducing reliance on battery discharge during future charging (or ensuring sufficient capacity to handle the charging load).

[0165] Condition 3: The ACDC module is running, and PV < 5

[0166] Meaning: Photovoltaic power generation is very small, below the set threshold of 5.

[0167] Operating condition analysis: The contribution of photovoltaics to meeting electricity demand is almost negligible.

[0168] Control actions: The strategy is the same as in condition two. That is, it mainly relies on the power grid and batteries to meet the power demand, and determines the power setting of the AC / DC module based on electricity prices and charging demand forecasts to replenish the battery.

[0169] Additional notes: ACDC module is not running.

[0170] If the ACDC module is not currently running (e.g., the grid is disconnected or not connected), but the system determines that the ACDC module needs to participate based on the above conditions (e.g., to use low-cost electricity to charge the battery, or to draw power from the grid to meet predicted charging demand), then the control strategy needs to include a step: automatically starting the ACDC module. This is to ensure that the system can execute subsequent power setting commands.

[0171] Summarize:

[0172] In the range of 10% < SOC < 90%, the control strategy becomes more complex and dynamic. It no longer simply meets the current demand but rather proactively manages the battery state, making optimal energy scheduling decisions by integrating photovoltaic output, grid electricity prices, and charging demand forecasts. The core lies in:

[0173] Prioritize the utilization of photovoltaic power.

[0174] When there is a surplus, decide whether to use the photovoltaic surplus or grid power to charge the battery based on the forecast.

[0175] When there is a shortfall, decide whether to draw power from the grid and how much to draw (to make up the shortfall and / or replenish the battery energy) based on the electricity price and the forecast.

[0176] Ensure that the battery has sufficient power when needed (especially when charging demand is predicted).

[0177] This set of strategies embodies the characteristics of intelligent and economical operation of the photovoltaic-storage-charging-using system.

[0178] C. In the case of extremely low battery power (SOC < 10%), the primary task is to protect the battery from over-discharging and damage while尽可能满足基本用电需求. SOC < 10% is regarded as the "fully discharged state";

[0179] Detailed explanation of the control logic:

[0180] Prioritize replenishing the battery energy, using the same strategy as in step B:

[0181] This indicates that even in the critical state of SOC < 10%, once conditions permit, the battery should be charged first.

[0182] The specific charging method (such as charging current, voltage, whether trickle charging is required, etc.) follows the previously defined strategy B. Strategy B is usually the charging method when the battery is in the normal operating range (10% - 90%), which may include a faster charging rate. Here, it shows that even when the battery power is extremely low, the charging method still follows strategy B, suggesting that it may be necessary to charge with a small current or perform trickle charging first to activate the battery, and then gradually increase the charging current.

[0183] The demand for electricity needs to be met. If there is no predicted charging demand, the minimum condition is to satisfy P1 = P2 - PV:

[0184] When there is no predicted upcoming charging demand, at this time, the system only draws power from the grid just enough to meet the current electricity demand minus the photovoltaic

[0185] It should be noted that there is an unclear expression "尽可能满足基本用电需求" in the original text. The above translation is for reference only.Differential power for power generation. Objective: To limit the energy obtained from the power grid, prioritize the use of photovoltaic power generation, and use the excess energy (if any) to charge the battery instead of meeting additional and unforeseen charging demands. This is an energy-saving and waiting strategy.

[0186] When the AC-DC module is in a fault shutdown state, since the battery is in a discharged state at this time, the strategy is to execute the strategy of giving priority to protecting the battery:

[0187] Core strategy: Under this double pressure (extremely low battery, critical equipment failure), protecting the battery becomes the absolute priority.

[0188] Condition 1: When the photovoltaic power generation is greater than the power consumption (PV > P2), the excess energy is used to charge the battery:

[0189] Situation: The photovoltaic power generation is sufficient, which can not only meet all power consumption demands but also have a surplus.

[0190] Action: All the excess part of the photovoltaic power generation (PV - P2) is used to charge the battery.

[0191] Objective: To use the only available energy source (photovoltaic) to restore the battery's power.

[0192] Condition 2: When the photovoltaic power generation is less than the power consumption (PV < P2), at this time, to protect the battery, the power consumption DC-AC module is shut down:

[0193] Situation: The photovoltaic power generation is not enough to meet all power consumption demands, and the battery power is extremely low and cannot provide supplementation.

[0194] Action: Shut down the DC-AC module of the AC load (possibly switching the AC load to the backup power supply or directly disconnecting it).

[0195] Objective: To prevent the battery from over-discharging. In the case of being unable to obtain power from the power grid (AC-DC module failure) and insufficient photovoltaic power generation, cutting off the AC load is the only option to protect the battery from discharging. This ensures that the remaining power of the battery will not be further consumed, reserving a basis for subsequent possible charging (such as when Condition 1 occurs).

[0196] Condition 3: When there is no photovoltaic power generation (PV ≈ 0), at this time, the power consumption DC-AC module is shut down.

[0197] Action: Similar to Condition 2, shut down the power consumption DC-AC module.

[0198] Objective: Similar to Condition 2, this is the ultimate means to protect the battery from discharging in the case of completely no energy input. The system completely relies on the battery for power supply at this time, but the battery power is extremely low and cannot support any load, so the load must be disconnected.

[0199] Under the dual pressures of extremely low battery power and critical equipment failure, battery safety is placed as the absolute top priority.

[0200] Solar power is the sole energy source: Due to the failure of the AC / DC module, the system relies entirely on solar power to maintain operation and charge the battery.

[0201] Electricity demand can only be met when photovoltaic power generation is sufficient.

[0202] Limited grid interaction: Due to the failure of the ACDC module, there does not appear to be an option to use the grid as a backup power source in the strategy (or even if there is an ACDC module, power draw from the grid may be limited when the SOC is less than 10% to protect the battery).

[0203] This strategy ensures that the system can survive in extreme situations (protecting the battery), taking into account different scenarios with and without predicted charging demand, and adopting a more conservative power consumption strategy (P1 = P2 - PV) when there is no predicted demand.

[0204] The photovoltaic-storage-charging-utilization coordinated control strategy first involves collecting multi-source data, including weather forecasts, real-time grid load data, user charging reservation data, user electricity consumption data, and energy storage device health status data. The collected data undergoes cleaning, preprocessing, and feature extraction. Deep learning algorithms are then used to analyze the multi-source data, predicting photovoltaic power generation, user charging demand, grid peak and off-peak periods, and electricity consumption information in advance. During periods of sufficient sunlight, photovoltaic power is prioritized for charging electric vehicles, with excess energy stored in energy storage devices. During peak grid periods, the devices release electricity to meet charging demands, reducing dependence on the grid and lowering electricity costs. The health status of the energy storage devices is monitored in real time, and the charging and discharging strategies are dynamically adjusted. Simultaneously, the operational results are fed back to the deep learning control model in real time to optimize the energy scheduling model, achieving efficient coordination and precise allocation of energy across the photovoltaic, storage, charging, and utilization stages.

[0205] The following are the relevant model algorithm formulas and processing flow:

[0206] 1. Photovoltaic power generation prediction model

[0207] Using an LSTM time series forecasting model, input meteorological forecast data (such as light intensity, temperature, humidity, etc.) and historical power generation data, the future photovoltaic power generation output is predicted.

[0208] LSTM model formula:

[0209] Input: i t =σ(W_i*[h_{t-1},x_t]+b_i)

[0210] Output: o_t=σ(W_o*[h_{t-1},x_t]+b_o)

[0211] Cell status update:

[0212] g t =tanh(W_g * [h_{t-1},x_t]+b_g)c_t=f_t⊙c_{t-1}+i_t⊙g_t

[0213] Hidden state update: h_t = o_t ⊙ tanh(c_t)

[0214] in:

[0215] x_t is the input data at the current moment (such as light intensity, temperature, etc.).

[0216] h_{t-1} is the hidden state of the previous time step.

[0217] c_{t-1} is the cell state at the previous time step.

[0218] W_i,W_f,W_o,W_g are weight matrices, and b_i,b_f,b_o,b_g are bias terms.

[0219] σ is the sigmoid activation function, tanh is the hyperbolic tangent function, and ⊙ represents element-wise multiplication.

[0220] 2. User charging demand prediction model

[0221] Using a time series forecasting model, combined with historical user charging behavior data and reservation data, we can predict future user charging needs.

[0222] Transformer model formula:

[0223] Self-attention mechanism:

[0224] Q = XW_Q, K = XW_K, V = XW_V

[0225]

[0226] Location coding:

[0227] PE(pos,2i)=sin(pos / 10000^{2i / d_model})

[0228] PE(pos,2i+1)=cos(pos / 10000^{2i / d_model})

[0229] in:

[0230] X is the input sequence.

[0231] W_Q, W_K, W_V are learnable weight matrices.

[0232] d_k is a scaling factor used to prevent the gradient from vanishing due to an excessively large inner product.

[0233] PE stands for Position Encoding, which is used to represent the position information of elements in a sequence.

[0234] 3. Power Grid Peak-Valley Prediction Model

[0235] A classification model (XGBoost or deep learning classification model) is used to analyze the power grid load data and predict the peak and valley periods of the power grid.

[0236] Classification model formula:

[0237] Objective function:

[0238] in:

[0239] L is the loss function.

[0240] Ω(f_k) is a regularization term used to prevent overfitting.

[0241] f_k is the output of the k-th tree.

[0242] 4. Energy Storage Equipment Health Status Assessment Model

[0243] The state of health (SOH) of energy storage devices is assessed using regression models (linear regression, random forest, or neural network), with input data including battery voltage, current, temperature, and number of charge / discharge cycles.

[0244] Linear regression model formula:

[0245] SOH=β_0+β_1*V+β_2*I+β_3*T+β_4*Cycles+ε

[0246] in:

[0247] V is the battery voltage.

[0248] I is the battery current.

[0249] T is the battery temperature.

[0250] Cycles refers to the number of charge-discharge cycles.

[0251] β_0,β_1,β_2,β_3,β_4 are regression coefficients.

[0252] ε is the error term.

[0253] 5. Dynamic charging and discharging strategy optimization model

[0254] Reinforcement learning (DQN) or optimization algorithms (linear programming, dynamic programming) are used to dynamically optimize the charging and discharging strategies of energy storage devices.

[0255] Reinforcement learning model formula:

[0256] Objective function: L^{CLIP}(θ)=E_t[min(r_t(θ)A_t,clip(r_t(θ),1-ε,1+ε)A_t)]

[0257] in:

[0258] r_t(θ)=π_θ(a_t|s_t) / π_{θ_old}(a_t|s_t) is the probability ratio.

[0259] A_t is the dominant function.

[0260] ε is the clip parameter, used to limit the update magnitude.

[0261] 6. Energy Dispatch Optimization Model

[0262] Linear programming or mixed integer programming (MIP) is used to optimize the energy dispatch of photovoltaic-storage-charging-utilization systems, with the goal of minimizing electricity costs or maximizing the utilization rate of renewable energy.

[0263] Linear programming model formula:

[0264] Objective function: min∑_{t=1}^T(C_grid*P_grid_t+C_pv*P_pv_t+C_ess*P_ess_t)

[0265] Constraints:

[0266] P_pv_t + P_ess_t + P_grid_t = P_load_t (energy balance)

[0267] SOC_min≤SOC_t≤SOC_max (Energy Storage SOC Constraint)

[0268] P_ess_min≤P_ess_t≤P_ess_max (Energy storage power constraint)

[0269] P_grid_min≤P_grid_t≤P_grid_max (Grid power constraint)

[0270] in:

[0271] C_grid, C_pv, and C_ess represent the unit costs of the power grid, photovoltaic system, and energy storage, respectively.

[0272] P_grid_t, P_pv_t, and P_ess_t represent the grid, photovoltaic, and energy storage power at time t.

[0273] P_load_t is the load power at time t.

[0274] SOC_t is the state of charge of the stored energy.

[0275] The above model ensures efficient coordination and precise allocation of the photovoltaic, energy storage, charging, and utilization systems.

[0276] While specific embodiments of the present invention have been described above, those skilled in the art should understand that these are merely illustrative examples, and the scope of protection of the present invention is defined by the appended claims. Those skilled in the art can make various changes or modifications to these embodiments without departing from the principles and essence of the present invention, but all such changes and modifications fall within the scope of protection of the present invention.

Claims

1. A main circuit topology for an integrated supercharging device based on photovoltaic storage and charging, characterized in that, Consisting of the following parts Composition: AC / DC module: Responsible for providing energy to the DC bus and supporting grid connection. When applied in a power-off area, the AC / DC module can be removed; Charging DC / DC module: Responsible for providing power for charging, supports parallel application, and can support up to 60 units in parallel. The number of charging DC / DC modules can be adjusted according to project requirements; PV DC / DC module: Connects to photovoltaic panels, provides photovoltaic power generation to the DC bus, and can adjust the photovoltaic power generation power through control strategies; DC / AC module: The inverter of the DC / AC module can convert DC into AC, supply power to the auxiliary power supply of the charging pile, and can also provide power to other electricity demands externally; Battery: Stores excess power and discharges it for use by the charging pile.

2. A control method based on the main circuit topology of an integrated photovoltaic-storage-charging-use supercharging device, characterized in that, Including the following steps: Step 1: Energy scheduling control method for photovoltaic energy storage charging and utilization First, the device conducts self-check. After a fault is detected in the self-check device, the corresponding fault protection logic actions are executed. When the self-check is normal, following the principle of charging priority, it is judged whether it is in the charging state. If not charging, considering the switching of the energy replenishment logic, the charging pile needs to be idle for 5 minutes. If it is charging, at this time, it is judged whether the SOC of the battery is within the set range. If the battery capacity is outside the SOC range; At this time, the required power of the charging pile needs to be limited to: EV_Demand = (P_dch_max + PV + P1 - P2).

3. The control method based on the main circuit topology of the integrated photovoltaic-storage-charging supercharging device according to claim 2, characterized in that, It also includes the following steps: If the remaining capacity of the battery is within the range allowed by the SOC, at this time, it is judged whether the required power of the charging pile meets the requirement of the formula EV_Demand < (P_dch_max + PV + P1 - P2). If it meets, it operates normally. If it does not meet, it means that the required power of the charging pile is greater than the maximum capacity provided by the current device itself, and the required power of the charging pile needs to be adjusted to: EV_Demand = (P_dch_max + PV + P1 - p2).

4. The control method based on the main circuit topology of the integrated supercharging device for photovoltaic storage and charging as described in claim 3, characterized in that, It also includes the following steps: Step 2: Energy storage replenishment control strategy A. SOC > 90%, at this time, the battery is in a relatively full state, and the power of the AC / DC module is set to 0; Condition 1: If the photovoltaic power generation is greater than the power consumption, P1_set = 0; Photovoltaic power PV_set = P2 + 2; Condition 2: PV > 5 && PV < P2, that is, the photovoltaic is generating power and the photovoltaic power generation power is less than the power consumption. At this time, the photovoltaic power is not restricted; Condition 3: PV < 5, at this time, the photovoltaic power generation power is relatively small.

5. The control method based on the main circuit topology of the integrated photovoltaic-storage-charging-use supercharging device according to claim 4, characterized in that, It also includes the following steps: B. When 10% < SOC < 90%, Condition 1: When the ACDC module is in the operating state and the photovoltaic power generation is greater than the power consumption, under the current working conditions, the photovoltaic power is not restricted, and the excess power is stored in the energy storage battery. Combining with the coordinated strategy of photovoltaic, energy storage, charging, and power consumption, it is predicted that there is no charging demand in a short period. At this time, the power of the ACDC module is limited to 0, and the excess photovoltaic power generation is used to charge the energy storage battery. If it is predicted that there is a charging demand in a short period, the power of the ACDC module is controlled to be 100 kW at this time to quickly charge the energy storage battery. Condition 2: PV > 5 && PV < P2, that is, the photovoltaic is generating electricity and the photovoltaic power generation is less than the power consumption. At this time, the photovoltaic power is not restricted. Combining with the prediction algorithm and electricity price policy, the power of the ACDC module is dynamically adjusted to charge the energy storage. The specific implementation strategy is as follows: when the electricity price is cheap, the ACDC module is set to 100 kW for rapid charging; when the electricity price is relatively expensive and it is predicted that there is no charging demand, the power of the ACDC module is 0 at this time. After the SOC reaches the lower limit, the ACDC module is started to charge the energy storage, and the power is set to low-power charging at this time; when the electricity price is relatively expensive and it is predicted that there is a charging demand, the power of the ACDC module is set to 100 kW at this time to quickly charge the energy storage. Condition 3: When PV < 5, the photovoltaic power generation is weak at this time, and the strategy is the same as Condition 2.

6. The control method based on the main circuit topology of the integrated photovoltaic-storage-charging-use supercharging device according to claim 5, characterized in that, It also includes the following steps: C. When SOC < 10%, the battery is in the empty state at this time, and the battery is preferentially charged. Condition 1: When the photovoltaic power generation is greater than the power consumption, the excess energy is used to charge the battery. Condition 2: When the photovoltaic power generation is less than the power consumption, the power consumption DCAC is shut down to protect the battery at this time. Condition 3: When the photovoltaic is not generating electricity, the power consumption DCAC is shut down at this time.

7. The control method based on the main circuit topology of the integrated photovoltaic-storage-charging-use supercharging device according to claim 6, characterized in that, It also includes the following steps: Step 3: The coordinated control strategy of photovoltaic, energy storage, charging, and power consumption first performs multi-source data collection, cleans, preprocesses the collected data, and extracts feature quantities. The multi-source data is analyzed through a deep learning algorithm to predict the photovoltaic power generation, user charging demand, grid peak and valley periods, and power consumption information in advance. When the light is sufficient, the photovoltaic power is preferentially used to charge the electric vehicle, and the excess electric energy is stored in the energy storage device; during the grid peak period, the energy storage device releases electric energy to meet the charging demand, reduces the dependence on the grid, and reduces the power consumption cost. The health status of the energy storage device is monitored in real time, the charge and discharge strategy of the energy storage device is dynamically adjusted, and the operation result is fed back to the deep learning control model in real time to optimize the energy scheduling model.

Images