Virtual capacity expansion control method for optical storage overcharge system based on energy storage peak load shifting

Abstract

The invention discloses an optical storage overcharge system virtual capacity expansion control method based on energy storage peak load shifting, belongs to the technical field of optical storage overcharge system control, and aims to solve the problems that the virtual capacity expansion effect is poor, the service life of an energy storage unit is influenced by excessive charging and discharging, the system operation stability is poor, the reliability is reduced, and power grid reverse transmission and energy storage unit damage occur. The method comprises the steps of constructing a real-time operation state monitoring system of the optical storage overcharge system, establishing a power grid power supply capacity constraint model, predicting total charging demand power change of a future overcharge terminal, formulating a dynamic charging and discharging scheduling strategy of an energy storage unit, and establishing a charging and discharging protection mechanism of the energy storage unit. According to the method, virtual capacity expansion is realized through energy storage peak clipping and valley filling, the limitation of the power supply capacity of the power grid is effectively broken through on the premise that the investment of the power grid is not increased, and a large amount of charging requirements in the charging peak period are met.

Description

Technical Field

[0001] This invention relates to the field of control technology for photovoltaic-storage-supercharging systems, specifically a virtual capacity expansion control method for photovoltaic-storage-supercharging systems based on energy storage peak shaving and valley filling. Background Technology

[0002] A photovoltaic-storage-supercharging system is a system that combines photovoltaic power generation, energy storage technology, and supercharging technology. It can effectively utilize clean energy and reduce the dependence of supercharging technology on the power grid. This development trend has become an important direction for the development of supercharging stations. However, existing photovoltaic-storage-supercharging systems face the problem of insufficient power grid capacity in actual operation. Especially during peak charging periods, the concentrated charging of a large number of new energy vehicles will cause the charging demand to exceed the power grid's power supply capacity limit. Therefore, this problem will result in the inability to meet users' fast charging needs.

[0003] To address the issue of insufficient grid power supply capacity, some photovoltaic-storage supercharging systems have adopted energy storage peak shaving and valley filling technologies. However, existing control methods still have several shortcomings in practical applications, specifically: First, the lack of accurate charging demand forecasting and dynamic assessment of grid power supply capacity leads to a lack of targeted energy storage charging and discharging scheduling, resulting in poor virtual capacity expansion effects. Second, the charging and discharging control strategies of energy storage units are relatively simple, failing to fully consider factors such as photovoltaic output fluctuations, energy storage SOC status, and energy storage unit lifespan. This can easily lead to overcharging and discharging of energy storage units, thus affecting their lifespan. Third, the control strategies lack dynamic correction mechanisms. When there is a significant deviation between actual operating data and predicted data, the control strategies cannot be adjusted in a timely manner, resulting in poor system stability and reduced reliability. Fourth, a comprehensive protection mechanism has not been established, which may lead to risks such as grid backfeeding and energy storage unit damage in extreme situations.

[0004] To address the above issues, a virtual capacity expansion control method for a photovoltaic-storage-supercharging system based on energy storage peak shaving and valley filling is proposed. Summary of the Invention

[0005] The purpose of this invention is to provide a virtual capacity expansion control method for a photovoltaic-storage-supercharging system based on energy storage peak shaving and valley filling. By using this invention, the problems mentioned above, such as poor virtual capacity expansion effect, overcharging and discharging of energy storage units affecting their service life, poor system operation stability and reduced reliability, and grid backfeeding and damage to energy storage units, are solved.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a virtual capacity expansion control method for a photovoltaic-storage-supercharging system based on energy storage peak shaving and valley filling, comprising the following steps: S1: Construct a real-time operation status monitoring system for the photovoltaic-storage-supercharging system, and synchronously collect data on the actual output power of photovoltaic modules, the remaining power (SOC) of energy storage units, the charging and discharging power thresholds of energy storage units, the real-time charging power demand of supercharging terminals, the real-time power supply power of grid access points, and power supply capacity limitation parameters. S2: Based on the real-time operation data collected by S1, establish a power grid supply capacity constraint model and clarify the maximum power supply limit that the power grid is allowed to output to the photovoltaic-storage supercharging system at different times. This limit needs to be dynamically determined in combination with the rated capacity of the power grid-side transformer, the line current carrying capacity and the regional power grid load dispatching requirements, and the real-time update cycle shall not exceed 5 minutes. S3: Through statistical analysis of historical charging data, identify the type of battery in the vehicle being charged and input the user's charging time requirements, and predict the total charging power demand curve of the supercharging terminal within the next 15 minutes. S4: Based on the real-time output power of photovoltaic modules, the predicted charging demand power, and the upper limit of grid power supply capacity, determine whether there is a peak period where the charging demand power exceeds the upper limit of grid power supply capacity. When the predicted charging demand power is higher than the upper limit of grid power supply capacity for three consecutive collection cycles, the energy storage peak shaving and valley filling virtual capacity expansion control mode is triggered. S5: In the virtual capacity expansion control mode, based on the remaining power (SOC) of the energy storage unit, the charging and discharging power threshold and the current charging and discharging state, a dynamic charging and discharging scheduling strategy for the energy storage unit is formulated. When the charging demand power is at its peak, the energy storage unit is controlled to supplement the power supply to the supercharging terminal in a manner that does not exceed its maximum discharge power. The sum of the supplementary power supply and the grid power supply meets the real-time charging demand of the supercharging terminal. S6: When the output power of the photovoltaic module is greater than the real-time charging demand of the supercharging terminal, and the remaining power (SOC) of the energy storage unit is lower than the threshold of its rated capacity, the energy storage unit is controlled to absorb the excess output of the photovoltaic for charging, while avoiding the photovoltaic output from being fed back to the grid. S7: Establish a charging and discharging protection mechanism for the energy storage unit. When the remaining energy charge (SOC) of the energy storage unit is lower than the set minimum threshold, the energy storage unit is prohibited from discharging; when the remaining energy charge (SOC) of the energy storage unit is higher than the set maximum threshold, the energy storage unit is prohibited from charging; when the charging and discharging power of the energy storage unit exceeds its rated power, the power limiting protection is automatically triggered. S8: By monitoring the charging status of the supercharging terminal in real time, when one of the supercharging terminals completes charging or the user terminates charging, the charging and discharging power of the energy storage unit and the distribution ratio of the grid power supply can be adjusted immediately to ensure that the total power supply of the system matches the remaining charging demand power and avoid power waste. S9: Based on the 24-hour grid power supply capacity change pattern, photovoltaic power output change curve and historical charging demand data, establish a virtual capacity expansion control strategy optimization model, and generate energy storage charging and discharging scheduling plans for different time periods of the next day before the daily specified time nodes. The plans must include energy storage charging and discharging power thresholds, grid power supply power allocation ratios and virtual capacity expansion upper limits under different scenarios. S10: During system operation, the deviation between actual operating data and predicted data is continuously compared. When the deviation exceeds the set threshold within multiple consecutive acquisition cycles, the dynamic correction mechanism of the control strategy is automatically triggered to adjust the prediction model parameters and energy storage scheduling strategy to ensure the accuracy and stability of virtual capacity expansion control.

[0007] Furthermore, the real-time operational status monitoring system in S1 includes a hardware acquisition module and a data preprocessing module, as detailed below: The hardware acquisition module uses a high-precision power sensor, a SOC detection module, and a voltage and current monitoring device. The data preprocessing module filters, denoises, and removes outliers from the acquired raw data.

[0008] Furthermore, the process of constructing the power grid supply capacity constraint model in S2 is as follows: The rated capacity, short-circuit impedance, wiring method and other parameters of the transformer on the grid side are obtained. Combined with the line parameters of the regional distribution network, the load distribution and the dispatch instructions of the upper-level power grid, the maximum allowable power supply at the grid access point is determined by the power flow calculation method, while taking into account the load factor limit of the power grid at different time periods of peak, valley and flat periods.

[0009] Furthermore, the charging demand power prediction in S3 adopts a combined prediction model, which integrates time series analysis methods, machine learning algorithms, and user behavior analysis models, as detailed below: The time series analysis method predicts trends based on charging power data from the same period in the past cycle; the machine learning algorithm uses a random forest model with battery type, temperature, date type, and time period as input features; and the user behavior analysis model corrects for users' historical charging duration and charging power preference data. The weights of the three prediction results are 0.3, 0.5, and 0.2, respectively. After weighted summation, the final charging demand power change curve can be obtained.

[0010] Furthermore, the dynamic charging and discharging scheduling strategy for energy storage units in S5 adopts hierarchical control logic, as follows: The upper layer is the capacity allocation layer, which determines the total supplementary power supply that the energy storage unit needs to provide based on the difference between the total charging power demand and the grid power supply capacity. The lower layer is the power allocation layer. When the system contains multiple energy storage units, the charging and discharging power is allocated according to the remaining energy capacity (SOC), charging and discharging efficiency, and state of health (SOH) of each energy storage unit. Energy storage units with higher remaining energy capacity (SOC), higher charging and discharging efficiency, and a state of health (SOH) greater than 80% are given priority to undertake charging and discharging tasks. The power allocation ratio is positively correlated with the comprehensive performance score of each energy storage unit. Overall performance score = 0.4 × normalized value of remaining energy storage unit (SOC) + 0.3 × charge and discharge efficiency + 0.3 × normalized value of state of health (SOH).

[0011] Furthermore, the control logic in S6 to prevent photovoltaic power output from feeding back into the grid is as follows: The difference between the photovoltaic output power and the charging demand power of the supercharging terminal is calculated in real time. When the difference is greater than 0, the energy storage unit is given priority to absorb the power difference. If the energy storage unit has reached full charge, the output power of the photovoltaic module is reduced by adjusting the MPPT (maximum power point tracking) controller of the photovoltaic module to ensure that the photovoltaic output power does not exceed the charging demand power of the supercharging terminal, so that the power flowing into the grid is always greater than or equal to 0.

[0012] Furthermore, the energy storage unit charging and discharging protection mechanism in the S7 also includes a temperature protection function. When the temperature of a single cell in the energy storage unit exceeds the set threshold range, the rated power of the charging and discharging power can be automatically adjusted, and a heat dissipation or heating device can be activated according to the degree of heat generation. Once the temperature of the energy storage unit returns to the normal threshold range, the charging and discharging function can be restored.

[0013] Furthermore, the virtual capacity expansion control strategy optimization model in S9 takes the lowest system operating cost and the longest energy storage unit life as the dual objective function. The constraints include grid power supply capacity constraints, energy storage unit charging and discharging power constraints, energy storage SOC constraints, photovoltaic output constraints, and charging demand satisfaction constraints. The NSGA-II (Non-dominated sorting genetic algorithm II) is used to solve this multi-objective optimization problem, which can obtain the Pareto optimal solution set. Then, the optimal energy storage charging and discharging scheduling plan is selected according to the actual operation requirements.

[0014] Furthermore, the dynamic correction mechanism for the control strategy in S10 includes two parts: correction of prediction model parameters and adjustment of energy storage scheduling strategy, as detailed below: The prediction model parameter correction adopts a rolling optimization method, which substitutes the latest actual running data into the prediction model to retrain the model parameters and update the prediction weights; The energy storage dispatch strategy adjustment is based on the type of deviation and a correction plan is formulated accordingly. If it is a deviation in charging demand prediction, the response speed of energy storage charging and discharging power is adjusted; if it is a deviation in photovoltaic output prediction, the absorption strategy of excess photovoltaic output is optimized; if it is a deviation in grid power supply capacity, the upper limit of virtual capacity expansion is recalculated.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. This invention achieves virtual capacity expansion through energy storage peak shaving and valley filling, effectively breaking through the power grid supply capacity limitation without increasing grid investment, and meeting the large charging demand during peak charging periods.

[0016] 2. This invention optimizes the strategy for absorbing excess photovoltaic power, maximizing the conversion of photovoltaic power into charging energy, reducing curtailment, and decreasing dependence on the traditional power grid.

[0017] 3. This invention establishes a comprehensive energy storage charging and discharging protection mechanism and an optimized scheduling strategy to avoid overcharging and discharging, extreme temperature charging and discharging, and overload charging and discharging of energy storage units. This can effectively extend the cycle life of energy storage units and reduce system operation and maintenance costs.

[0018] 4. This invention, through accurate prediction models, real-time power matching mechanisms, and dynamic control strategy correction mechanisms, can ensure that the system maintains stable operation in complex and ever-changing operating environments, avoiding problems such as charging interruptions and grid backfeeding.

[0019] 5. This invention can adjust parameters and optimize strategies according to different scales of photovoltaic-storage-supercharging systems, different grid conditions, and different application scenarios, and has wide applicability. Attached Figure Description

[0020] Figure 1 This is a schematic diagram of the method flow of the present invention. Detailed Implementation

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] Example 1: A virtual capacity expansion control method for a photovoltaic-storage-supercharging system based on energy storage peak shaving and valley filling, the specific implementation steps are as follows: Real-time operation status monitoring: High-precision power sensors are used to collect the output power of photovoltaic modules, the power supplied by the grid connection point, and the charging power demand of the supercharging terminal, with a collection frequency of 1 time / second; the SOC value of each energy storage battery pack is collected through the SOC detection module, with a collection frequency of 1 time / second; and the temperature of the individual battery cells of the energy storage unit is collected through temperature sensors, with a collection frequency of 1 time / 30 seconds. The data preprocessing module filters and denoises the collected raw data, which can remove outliers that deviate from the mean by ±20%.

[0023] Power grid supply capacity constraint model construction: For example, the rated capacity of the power grid transformer connected to the system is 630kVA. Based on the regional distribution network parameters and dispatch requirements, the upper limit of power grid supply capacity during peak hours (08:00-11:00, 18:00-21:00) is determined to be 504kW (630kVA×80%), during normal hours (06:00-08:00, 11:00-18:00, 21:00-23:00) it is 567kW (630kVA×90%), and during off-peak hours (23:00-06:00 the next day) it is 630kW (630kVA×100%). The model updates the upper limit of power grid supply capacity every 5 minutes.

[0024] Charging demand power prediction: A combined prediction model is used, which takes into account the charging power data of the same period in the past 7 days, the current battery type of the vehicle being charged, the temperature, the date type (weekday / holiday), and the user's charging duration requirement. The model is then weighted by time series analysis, random forest algorithm, and user behavior analysis model, with weights set to 0.3, 0.5, and 0.2, respectively, to predict the total charging demand power change curve in the next 15 minutes.

[0025] Virtual capacity expansion control mode triggering: When the predicted charging demand power is higher than 95% of the grid power supply capacity limit for three consecutive collection cycles (3 seconds), the virtual capacity expansion control mode is triggered. For example, at 08:30 on a weekday morning peak, the predicted total charging demand power for the next 15 minutes is 600kW, while the grid power supply capacity limit is 504kW. If 600kW is higher than 504kW×95%=478.8kW for three consecutive seconds, the virtual capacity expansion control is triggered.

[0026] Dynamic charging and discharging scheduling of energy storage: The upper capacity allocation layer calculates the difference between the total charging demand power and the upper limit of the grid power supply capacity as 600kW-504kW=96kW, and determines that the energy storage unit needs to provide 96kW of supplementary power supply. The lower power allocation layer allocates power based on the SOC, charge / discharge efficiency, and SOH of each energy storage battery pack. Assuming the comprehensive performance scores of the 10 energy storage battery packs are 0.9, 0.88, 0.85, 0.82, 0.8, 0.78, 0.75, 0.72, 0.7, and 0.68 respectively, and the total score is 8.08, the allocated discharge power for each battery pack is as follows: 96kW × (0.9 / 8.08) ≈ 10.77kW; 96kW×(0.88 / 8.08)≈10.53kW; 96kW×(0.85 / 8.08)≈10.11kW; 96kW × (0.82 / 8.08) ≈ 9.70kW; 96kW × (0.8 / 8.08) ≈ 9.48kW; 96kW × (0.78 / 8.08) ≈ 9.25kW; 96kW × (0.75 / 8.08) ≈ 8.89kW; 96kW × (0.72 / 8.08) ≈ 8.54kW; 96kW × (0.7 / 8.08) ≈ 8.29kW; 96kW × (0.68 / 8.08) ≈ 8.04kW; The total discharge power is 96kW, which is sufficient to provide supplementary power through energy storage.

[0027] Absorbing excess photovoltaic power: For example, at 12:00 noon, the output power of the photovoltaic module is 450kW, the real-time charging demand of the supercharging terminal is 300kW, and the current SOC of the energy storage unit is 60%, which is lower than 90%. The calculated difference is 450kW-300kW=150kW. At this time, the energy storage unit is controlled to absorb 150kW of excess photovoltaic power for charging. At this time, the grid power supply is 0, avoiding the photovoltaic power output from being fed back to the grid.

[0028] Energy storage charging and discharging protection: For example, when the SOC of a certain energy storage battery pack drops to 4.8% (below 5%), the EMS immediately prohibits the battery pack from discharging; when the SOC rises to 95.2% (above 95%), the charging of the battery pack is prohibited; when the discharge power of a certain battery pack reaches 55kW (exceeding 110% of the rated power of 50kW), the power limiting protection is triggered, reducing its discharge power to 50kW; at the same time, when the battery pack temperature rises to 46℃ (above 45℃), the charging and discharging power is reduced to 25kW (50% of the rated power); when the temperature rises to 56℃, the charging and discharging operation is stopped, the heat dissipation device is activated, and the normal charging and discharging function is restored when the temperature drops to 38℃.

[0029] Real-time power matching: For example, at 2:00 PM, a supercharging terminal completes charging, and the charging power demand drops from the current 420kW to 300kW. The EMS immediately adjusts the power allocation, the grid power supply drops from 350kW to 220kW, and the energy storage discharge power is adjusted to 80kW to ensure that the total power supply is 300kW, which can match the charging power demand.

[0030] Dispatch plan generation: For example, before 2:00 AM every day, the EMS uses the NSGA-II algorithm to solve a multi-objective optimization problem based on the grid power supply capacity change pattern, photovoltaic output change curve and historical charging demand data over the past 30 days, and generates energy storage charging and discharging dispatch plans for different time periods of the next day; during the morning peak from 08:00 to 11:00 the next day, the energy storage discharge power threshold is 100kW, the grid power supply allocation ratio is 84% ​​(504kW / 600kW), and the virtual expansion capacity limit is 96kW; during the off-peak period from 23:00 to 06:00 the next day, the energy storage charging power threshold is 150kW, the grid power supply allocation ratio is 100%, and the virtual expansion capacity limit is 0kW.

[0031] Dynamic correction of control strategy: For example, at 18:00 in the evening, due to sudden rainfall, the actual value of photovoltaic output is 30% lower than the predicted value, resulting in a 20% deviation in the predicted power demand for charging. If the deviation exceeds ±15% for 5 consecutive collection cycles (5 seconds), the dynamic correction mechanism is triggered. At this time, the EMS will substitute the latest actual photovoltaic output data into the prediction model, retrain the model parameters and adjust the prediction weights. At the same time, it will optimize the energy storage charging and discharging scheduling strategy to improve the energy storage discharge power response speed, thereby increasing the energy storage supplementary power supply from 96kW to 120kW to ensure that the charging demand is met.

[0032] Example 2: The present invention addresses the usage requirements of different scenarios as follows: Scenarios of sudden surge in charging demand during holidays: For example, at 10:00 AM on a holiday, a large number of tourist vehicles charge simultaneously, and the actual charging power demand suddenly increases from 300kW to 650kW, exceeding the grid power supply capacity limit of 504kW. The prediction model of this invention can complete the prediction correction within 3 seconds by updating data in real time and trigger the virtual capacity expansion control mode. The energy storage unit can respond within 500ms and quickly provide 146kW of supplementary power to ensure that all supercharging terminals work normally and there will be no charging interruption.

[0033] Temporary adjustment scenario of grid power supply capacity: For example, at 15:00 on a holiday, the grid dispatch center issues a temporary instruction to reduce the upper limit of the grid power supply capacity of the supercharging system from 567kW (normal period) to 450kW. By monitoring the grid power supply capacity parameters in real time, the EMS can update the constraint model within 100ms and adjust the energy storage charging and discharging scheduling strategy, thereby increasing the energy storage supplementary power supply from 50kW to 200kW to ensure that the charging demand of 650kW is met, thus ensuring that the user's charging experience is not affected.

[0034] Energy storage unit partial failure scenario: For example, at 17:00 on a holiday evening, two energy storage battery packs stop working due to a fault. The maximum discharge power of the remaining eight energy storage battery packs is 400kW. By monitoring the energy storage unit health status (SOH) in real time, the EMS can adjust the power allocation strategy in a timely manner and recalculate the comprehensive performance score of the eight energy storage battery packs to improve the power allocation ratio of battery packs with good health status. At the same time, it optimizes the power allocation of the grid supply to ensure that a total charging power of 600kW can still be provided.

[0035] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0036] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A virtual capacity expansion control method for a light storage supercharging system based on energy storage peak shaving and filling, characterized in that, Comprising the following steps: S1: Constructing the real-time operation state monitoring system of the light storage overcharging system, and synchronously collecting the actual output power of the photovoltaic assembly, the residual electric quantity (SOC) of the energy storage unit, the charging and discharging power threshold of the energy storage unit, the real-time charging demand power of the overcharging terminal, the real-time power supply power of the power grid access point, and the power supply capacity limiting parameter; S2: Establishing the power grid power supply capacity constraint model based on the real-time operation data collected in S1, and clearly defining the maximum power supply power upper limit allowed by the power grid to output to the light storage overcharging system in different time periods, and the upper limit needs to be dynamically determined in combination with the rated capacity of the power grid side transformer, the line load flow, and the regional power grid load scheduling requirement, and the real-time update period is not more than 5 minutes; S3: Identifying the type of the current charging vehicle battery through statistical analysis of the historical charging data, inputting the user charging time length demand, and predicting the total charging demand power change curve of the overcharging terminal in the next 15 minutes; S4: According to the real-time output power of the photovoltaic assembly, the predicted charging demand power, and the power grid power supply capacity upper limit, it is judged whether there is a peak period in which the charging demand power exceeds the power grid power supply capacity upper limit, and when the predicted charging demand power is higher than the threshold value of the power grid power supply capacity upper limit for 3 consecutive collection periods, the virtual expansion control mode of energy storage peak shaving and valley filling is triggered; S5: In the virtual expansion control mode, based on the residual electric quantity (SOC) of the energy storage unit, the charging and discharging power threshold, and the current charging and discharging state, a dynamic charging and discharging scheduling strategy of the energy storage unit is formulated, when the charging demand power is at the peak, the energy storage unit is controlled to supply power to the overcharging terminal in a way that does not exceed its maximum discharging power, and the sum of the supplementary power supply power and the power grid power supply power meets the real-time charging demand of the overcharging terminal; S6: When the output power of the photovoltaic assembly is greater than the real-time charging demand power of the overcharging terminal, and the residual electric quantity (SOC) of the energy storage unit is lower than the threshold value of its rated capacity, the energy storage unit is controlled to absorb the excess output of the photovoltaic to charge, while avoiding the output of the photovoltaic to the power grid; S7: Establishing a charging and discharging protection mechanism for the energy storage unit, when the residual electric quantity (SOC) of the energy storage unit is lower than the set minimum threshold value, the energy storage unit is prohibited from discharging; when the residual electric quantity (SOC) of the energy storage unit is higher than the set maximum threshold value, the energy storage unit is prohibited from charging; when the charging and discharging power of the energy storage unit exceeds its rated power, the power limitation protection is automatically triggered; S8: By monitoring the charging state of the overcharging terminal in real time, when one of the overcharging terminals completes charging or the user terminates charging, the allocation ratio of the charging and discharging power of the energy storage unit and the power grid power supply power can be immediately adjusted to ensure that the total power supply power matches the remaining charging demand power, and power waste can be avoided; S9: Based on the power grid power supply capacity variation law throughout the day, the photovoltaic output variation curve, and the historical charging demand data, a virtual expansion control strategy optimization model is established, the generation of the energy storage charging and discharging scheduling plan for different time periods the next day is completed before the specified time node of each day, and the plan needs to include the energy storage charging and discharging power threshold, the power grid power supply power allocation ratio, and the virtual expansion capacity upper limit under different scenarios. S10: In the process of system operation, the deviation between actual operation data and predicted data is continuously compared. When the deviation exceeds the set threshold value in continuous multiple collection periods, the dynamic correction mechanism of control strategy is automatically triggered to adjust the prediction model parameters and energy storage scheduling strategy, so as to ensure the accuracy and stability of virtual expansion control. 2.The virtual capacity expansion control method of a light storage supercharging system based on energy storage peak load shifting according to claim 1, characterized in that: The real-time operation state monitoring system in S1 includes a hardware acquisition module and a data preprocessing module, which are as follows: The hardware acquisition module uses high-precision power sensors, SOC detection modules of energy storage units, and voltage and current monitoring devices. The data preprocessing module filters, denoises, and removes outliers from the collected raw data. 3.The virtual capacity expansion control method of a light storage supercharging system based on energy storage peak load shifting according to claim 2, characterized in that: The process of constructing the power grid power supply capacity constraint model in S2 is as follows: The rated capacity, short-circuit impedance, connection mode, and other parameters of the grid-side transformer are obtained. Combined with the line parameters, load distribution, and dispatching instructions of the regional distribution network and the upper-level power grid, the maximum allowable power supply power of the grid access point is determined using the power flow calculation method, while considering the load rate restrictions of the grid in peak-valley-flat different periods.

4. The virtual capacity expansion control method of the light storage and super charging system based on energy storage peak load shifting according to claim 3, characterized in that: The charging demand power prediction in S3 uses a combined prediction model that integrates time series analysis methods, machine learning algorithms, and user behavior analysis models, as follows: The time series analysis method is based on the charging power data of the same period in the past cycle for trend prediction. The machine learning algorithm uses a random forest model with battery type, temperature, date type, and time period as input features. The user behavior analysis model corrects the historical charging duration and charging power preference data. The weights of the three prediction results are 0.3, 0.5, and 0.2, respectively. After weighted summation, the final charging demand power variation curve can be obtained.

5. The virtual capacity expansion control method of the energy storage peak load shifting and valley load filling based optical storage super charging system according to claim 4, characterized in that: The dynamic charging and discharging scheduling strategy for energy storage units in S5 uses a hierarchical control logic, as follows: The upper layer is the capacity allocation layer, which determines the total supplementary power supply power provided by the energy storage units based on the difference between the total charging demand power and the power supply capacity of the grid. The lower layer is the power allocation layer. When the system contains multiple energy storage units, the charging and discharging power is allocated based on the remaining capacity (SOC), charging and discharging efficiency, and state of health (SOH) of each energy storage unit. Energy storage units with higher remaining capacity (SOC), higher charging and discharging efficiency, and state of health (SOH) greater than 80% are given priority for charging and discharging tasks. The power allocation ratio is positively correlated with the comprehensive performance score of each energy storage unit. Comprehensive performance score = 0.4 × normalized value of remaining capacity (SOC) of energy storage unit + 0.3 × charging and discharging efficiency + 0.3 × normalized value of state of health (SOH).

6. The virtual capacity expansion control method of the energy storage peak load shifting and valley load filling based optical storage and super charging system according to claim 5, characterized in that: The control logic for avoiding PV output from being sent back to the grid in S6 is as follows: The difference between the photovoltaic output power and the charging demand power of the supercharging terminal is calculated in real time. When the difference is greater than 0, the energy storage unit is preferentially controlled to absorb the difference power. If the energy storage unit has reached the full charging state, the MPPT (maximum power point tracking) controller of the photovoltaic module is adjusted to reduce the output power of the photovoltaic module, so as to ensure that the photovoltaic output power does not exceed the charging demand power of the supercharging terminal, so that the grid inflow power is always greater than or equal to 0.

7. The virtual capacity expansion control method of the energy storage peak load shifting and valley load filling based optical storage and super charging system according to claim 6, characterized in that: The energy storage unit charge-discharge protection mechanism in S7 also includes a temperature protection function. When the temperature of the single battery of the energy storage unit exceeds the set threshold range, the rated power of the charge-discharge power can be automatically adjusted, and the heating or cooling device can be started according to the heating degree. After the temperature of the energy storage unit returns to the normal threshold range, the charge-discharge function can be restored. 8.The virtual capacity expansion control method of a light storage supercharging system based on energy storage peak load shifting according to claim 7, characterized in that: The virtual expansion control strategy optimization model in S9 takes the minimum system operation cost and the longest energy storage unit life as the dual objective function, wherein the constraint conditions include the power grid power supply capacity constraint, the energy storage unit charge-discharge power constraint, the energy storage SOC constraint, the photovoltaic output constraint and the charging demand satisfaction constraint. The NSGA-II (Non-dominated Sorting Genetic Algorithm II) is used to solve the multi-objective optimization problem, and the Pareto optimal solution set can be obtained. Then, according to the actual operation demand, the optimal energy storage charge-discharge scheduling plan can be selected. 9.The virtual capacity expansion control method of a light storage supercharging system based on energy storage peak load shifting according to claim 8, characterized in that: The control strategy dynamic correction mechanism in S10 includes two parts: prediction model parameter correction and energy storage scheduling strategy adjustment. Specifically as follows: The prediction model parameter correction adopts a rolling optimization method, which substitutes the latest actual operation data into the prediction model to retrain the model parameters and update the prediction weights; The energy storage scheduling strategy adjustment formulates a correction scheme according to the deviation type. If it is a charging demand prediction deviation, the response speed of the energy storage charge-discharge power is adjusted. If it is a photovoltaic output prediction deviation, the absorption strategy of the excess photovoltaic output is optimized. If it is a power grid power supply capacity deviation, the upper limit of the virtual expansion capacity is recalculated.

Images