Energy storage energy management method suitable for household light storage system

Abstract

The application discloses an energy storage energy management method suitable for a household light storage system. In the method, the household light storage system user adopts an economic scheduling mode, and in a scheduling period, based on all-day predicted data of load and photovoltaic power generation and real-time transmitted photovoltaic and load power actual data, a storage charging and discharging power plan for a remaining period of the day is formulated with the minimum household light storage load cooperative operation cost as the target; the optimization result of the current scheduling period is used to provide a storage power value as a reference, and an inverter sets the purchase and sale power based on the reference power; meanwhile, at several fixed time points of the scheduling day, corresponding inputs are updated and adjusted based on prediction, and the latest storage power plan is updated. The strategy is generally suitable for household renewable energy systems with storage, and compared with the existing technology, has the advantages of high efficiency, practicality, economy, environmental protection and full use of photovoltaic uncertainty resources.

Description

Technical Field

[0001] This invention belongs to the field of user-side multi-element energy scheduling, simulation and analysis, and in particular relates to an energy storage management method suitable for residential photovoltaic-storage systems. Background Technology

[0002] With the continuous advancement of distributed renewable energy generation technology and the support of relevant national policies, the penetration rate of distributed renewable energy generation on the user side is constantly increasing. Distributed renewable energy generation, mainly based on photovoltaics, is characterized by randomness, volatility, and intermittency, which brings significant uncertainty to the comprehensive energy dispatch of households. At the same time, the access of highly volatile loads in households also increases the uncertainty of the overall plan and places higher demands on power supply reliability. Photovoltaic and energy storage systems need to maintain economical and effective operation for a long period of time.

[0003] Distributed residential photovoltaic (PV) systems often experience a surge in surplus electricity fed into the grid at midday, exceeding the grid's absorption capacity and leading to curtailment. Furthermore, some countries' grid connection regulations explicitly stipulate that the grid-connected power of PV systems cannot exceed 50% or 70% of the system capacity, contributing to underutilization of PV resources and insufficient exploitation of energy storage potential. Therefore, load and PV forecasting are necessary. Based on the predicted electricity consumption and generation, the timing of battery charging can be determined. This allows for the charging and storage of PV power generated at midday that would otherwise exceed the electricity sales limit, while intelligently scheduling additional electricity sales periods in the morning to maximize PV power utilization and reduce resource waste.

[0004] The electricity pricing policies implemented by power grid companies have also placed higher demands on energy storage management. Many countries and regions implement a two-part electricity pricing system, which includes a basic charge and a per-unit charge. The basic charge is based on transformer capacity or the maximum contracted demand. The per-unit charge is determined by electricity consumption and the electricity price, and may be implemented through time-of-use pricing, time-of-use pricing, peak-valley pricing, or package pricing based on long-term electricity consumption.

[0005] Many researchers have conducted in-depth studies on using distributed photovoltaic (PV) power generation or energy storage for load balancing and peak shaving to improve economic efficiency and feasibility. For example, patent document CN202011322938.7 proposes an AC / DC hybrid architecture considering multiple load types and an intelligent energy coordination method, which can effectively promote the large-scale application of household PV-storage-charging terminals. Patent document CN201911275379.6 discloses a modular household PV-storage system and control method with automatic energy dispatch, which analyzes the current energy storage SOC state and the values ​​of PV and load to make decisions on the operating mode of the PV-storage system. Therefore, in current research, there is no energy management strategy for household PV-storage systems that makes decisions based on the prediction of PV and load to consider the overall economic efficiency of the household, in order to meet the actual needs of PV grid-connected curtailment conditions and energy storage economic optimization scheduling. These strategies and methods cannot fully utilize the flexibility of PV-storage systems and cannot take into account both the economic efficiency of the overall household electricity consumption and the utilization rate of PV resources. Summary of the Invention

[0006] The purpose of this invention is to address the shortcomings of existing technologies by providing an energy storage management method suitable for residential photovoltaic energy storage systems.

[0007] The objective of this invention is achieved through the following technical solution: an energy storage management method applicable to residential photovoltaic-storage systems, the method comprising a rolling optimization phase based on day-ahead forecasts and a rolling optimization phase based on multiple forecast updates for the energy storage management plan;

[0008] The rolling optimization phase based on day-ahead forecasts includes:

[0009] The energy management of residential photovoltaic and energy storage systems is based on a daily scheduling cycle. Before starting the energy management plan for the current scheduling day, the photovoltaic power forecast and load power forecast for the next 24 hours are obtained the night before. If there are fluctuating electricity prices in the local area, the time-of-use electricity price for the scheduling day also needs to be obtained. Before the scheduling period begins, the real-time photovoltaic power and real-time load power of the photovoltaic and energy storage system are obtained.

[0010] The scheduling day is divided into N time periods. Based on the power forecast information of photovoltaic and load for each time period in the next 24 hours and the real-time photovoltaic and load power data, the electricity cost, surplus power grid connection revenue, energy storage scheduling cost and curtailment loss are comprehensively considered. With the goal of minimizing the operating cost of the remaining time periods of the day, the plan takes into account power balance constraints, photovoltaic grid connection power limit constraints, photovoltaic and energy storage system equipment technical characteristic constraints and photovoltaic and energy storage system shared inverter constraints. The remaining time periods of the day are modeled as a mixed integer linear programming problem. The energy storage energy management plan for each remaining time period of the day is formulated, including energy storage charging and discharging power, energy storage battery state of charge (SOC), power purchased from and sold to the grid, and curtailment power.

[0011] Based on the energy storage energy management plan for the remaining time period of the day, the plan for the current time period is selected, and the energy storage charging and discharging power of the current time period in the plan result is transmitted to the inverter. The inverter follows the plan result, combines the real-time photovoltaic power and real-time load power in the current dispatch period, executes the charging and discharging of energy storage, obtains the actual power to purchase and sell electricity to the grid and executes it.

[0012] The rolling optimization phase based on multiple prediction updates includes:

[0013] Multiple forecast updates are performed at several fixed times within the scheduling day. The forecast objects are adjusted to the photovoltaic power and load power of each time period within the next 24 hours from each fixed time. The n fixed times are divided into n+1 forecast update periods. Each scheduling period on the day determines its current forecast update period and extracts the forecast update data for the corresponding 24 hours. The extraction range is from the current time period to the last time period of the day. Based on the extracted photovoltaic and load power data, combined with the real-time photovoltaic and load power data, an updated energy storage energy management plan for the remaining time periods of the day is formulated.

[0014] Furthermore, the input information required in the rolling optimization phase based on day-ahead predictions includes:

[0015] 1) Household photovoltaic power forecast and load power forecast with the same time scale as the scheduling period, one value for each period;

[0016] 2) Time-of-use electricity pricing for dispatch days with hourly granularity;

[0017] 3) Real-time photovoltaic power and load power at the start of each scheduling period;

[0018] 4) Real-time SOC at the start of each scheduling period.

[0019] Furthermore, the method for modeling the remaining time slots of a residential photovoltaic-storage system as a mixed-integer linear programming problem is as follows:

[0020] S1. Define all variables as follows, categorized into continuous variables and 0-1 variables:

[0021] Defined continuous variables Including: users in Power purchased from the grid during different time periods Photovoltaic storage system in Remaining power supplied to the grid during a given period Photovoltaic storage system in Periodic power curtailment Energy storage devices in Charging power during the period Energy storage devices in Discharge power during the period Energy storage devices in SOC during the period Among them, continuous variables , , , , , All are non-negative;

[0022] Defined 0-1 variables Including: energy storage devices Charging status during the period Energy storage devices in Discharge state during the period Users Electricity purchase status from the grid during a certain period Photovoltaic storage system in Remaining electricity status during a given time period ;

[0023] S2. The objective function for the mixed-integer linear programming problem, defined with the goal of minimizing the total operating cost for the remaining time period of the day, is as follows:

[0024]

[0025] in This represents the set of remaining time periods of the day at the current moment. Indicates that the user is The price of purchasing electricity from the grid during a given period. Indicates that the photovoltaic energy storage system is in The price of surplus electricity sold to the grid during certain time periods. Indicates the duration of a single scheduling period;

[0026] S3. Set the constraints for the mixed-integer linear programming problem as follows:

[0027] 1) Power balance constraints:

[0028]

[0029] in express Forecasted photovoltaic power for the period express Forecasted load power for the specified time period, and when When the value is set to 1, the predicted value is replaced by the actual photovoltaic and load power.

[0030] 2) Maximum charge / discharge power constraints for energy storage:

[0031]

[0032] in , These are the maximum charging power and maximum discharging power of energy storage, respectively.

[0033] 3) Mutual exclusion constraints for energy storage charge and discharge states:

[0034]

[0035] 4) Energy storage SOC constraints:

[0036]

[0037] in Indicates the maximum energy storage capacity. This indicates the efficiency of the energy storage charging process. This indicates the efficiency of the energy storage and discharge process;

[0038] 5) Energy storage SOC upper and lower limits constraints:

[0039]

[0040] in This indicates the lower limit of the State of Charge (SOC) during the energy storage charging and discharging process. This indicates the upper limit of the State of Charge (SOC) during the energy storage charging and discharging process;

[0041] 6) Energy storage ramp-up rate constraints:

[0042]

[0043] in This represents the maximum value of the power increase during the energy storage charging process. This represents the maximum value of the power increase during the energy storage discharge process;

[0044] 7) Power purchase and sale constraints for photovoltaic-storage systems:

[0045]

[0046] in This indicates the maximum limit for purchasing electricity from the grid. This indicates the grid-connected capacity of surplus photovoltaic power, based on local policies and inverter capacity.

[0047] 8) Mutual exclusion constraints for power purchase and sale status of photovoltaic-storage systems:

[0048]

[0049] 9) Constraints on shared inverters for photovoltaic and energy storage systems:

[0050]

[0051] in This indicates the rated power limit of the inverter in a photovoltaic-storage system, which is typically 50%-70% of the rated power in some countries.

[0052] Furthermore, the selection process for the energy storage management plan during the current scheduling period includes:

[0053] Each scheduling period is based on the energy storage energy management plan for the remaining period of the day. The first value in the plan result is selected as the energy storage energy management plan for the current period. The energy storage charging and discharging power of the current period in the plan result is transmitted to the inverter. The inverter follows the plan result, combines the real-time photovoltaic power and real-time load power in the current scheduling period, and executes the charging and discharging of energy storage based on power balance constraints to obtain the actual power purchased and sold to the grid and execute it.

[0054] Meanwhile, a daily energy storage management plan is formulated for the remaining time period during each scheduling period. The first value in the energy storage charging and discharging power result in the plan is selected as the execution indicator and rolled forward over time.

[0055] The beneficial effects of this invention are as follows: In this method, users of the photovoltaic-storage system adopt an economical scheduling approach. Within the scheduling cycle, based on the all-day forecast data of load and photovoltaic power generation, and the real-time transmitted actual data of photovoltaic and load power, a storage charging and discharging plan for the remaining time period of the day is formulated with the goal of minimizing the cost of coordinated operation of household photovoltaic-storage-load. Based on the optimization results of the current scheduling period, and considering the typical operating mode of the inverter, a photovoltaic-storage system power generation mode for the current period is formulated. Simultaneously, at several fixed times on the scheduling day, the corresponding inputs are updated and adjusted based on forecasts to update the latest plan. This strategy is universally applicable to household renewable energy systems equipped with energy storage, and compared with existing technologies, it has advantages such as high efficiency and practicality, economic efficiency and environmental friendliness, and full utilization of uncertain photovoltaic resources. Attached Figure Description

[0056] Figure 1 This is a schematic diagram of the remaining time period optimization algorithm for a residential photovoltaic storage system;

[0057] Figure 2 This is a schematic diagram of a typical residential photovoltaic storage system in the embodiment;

[0058] Figure 3 This refers to the photovoltaic power generation and load power situation on that day;

[0059] Figure 4 This is the result of the rolling run optimization for the remaining time period;

[0060] Figure 5 The results are simulations of light wastage by residential users throughout the day.

[0061] Figure 6This is a comparison of the abandoned light power under the two operating modes. Detailed Implementation

[0062] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Many specific details are set forth in the following description to provide a thorough understanding of the present invention. However, the present invention can be practiced in many other ways different from those described herein, and those skilled in the art can make similar modifications without departing from the spirit of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below. Technical features in the various embodiments of the present invention can be combined accordingly without mutual conflict.

[0063] like Figure 1 As shown, in a preferred embodiment of the present invention, an energy storage management method suitable for residential photovoltaic energy storage systems is provided. The method includes a rolling optimization stage based on day-ahead forecasts and a rolling optimization stage based on multiple forecast updates for the energy storage management plan. The implementation process of each stage is described in detail below.

[0064] I. The rolling optimization phase based on current day forecasts includes the following steps:

[0065] S1. The energy storage management of the residential photovoltaic-storage system is based on a daily scheduling cycle. Before starting the energy management plan for the current scheduling day, the photovoltaic power forecast and load power forecast for the next 24 hours are obtained the night before. If there are fluctuating electricity prices in the local area, the time-of-use electricity price for the scheduling day is also required. The real-time photovoltaic power and real-time load power of the photovoltaic-storage system are obtained before the scheduling period begins.

[0066] S2. Divide the scheduling day into N time periods. Based on the power forecast information of photovoltaic and load for each time period in the next 24 hours and the real-time photovoltaic and load power data, comprehensively consider the electricity cost, surplus power grid connection revenue, energy storage scheduling cost and curtailment loss. With the goal of minimizing the operating cost of the remaining time periods of the day, and taking into account power balance constraints, photovoltaic grid connection power limit constraints, photovoltaic-storage system equipment technical characteristic constraints and photovoltaic-storage system shared inverter constraints, the remaining time period plan of the day is modeled as a mixed integer linear programming problem. Formulate the energy storage energy management plan for each remaining time period of the day, including energy storage charging and discharging power, energy storage battery state of charge (SOC), power purchased from and sold to the grid, and curtailment power.

[0067] S3. Based on the energy storage energy management plan for the remaining time period of the day, select the plan for the current time period, and transmit the energy storage charging and discharging power of the current time period in the plan results to the inverter. The inverter follows the plan results, combines the real-time photovoltaic power and real-time load power in the current scheduling period, executes the charging and discharging of energy storage, obtains the actual power purchased and sold to the grid, and executes it.

[0068] Specifically, the input information required in this stage can be summarized as follows:

[0069] 1) Household photovoltaic power forecast and load power forecast with the same time scale as the scheduling period, one value for each period;

[0070] 2) Time-of-use electricity pricing for dispatch days with hourly granularity;

[0071] 3) Real-time photovoltaic power and load power at the start of each scheduling period;

[0072] 4) Real-time state of charge (SOC) of the energy storage battery at the beginning of each scheduling period.

[0073] Specifically, in S2, the method for modeling the remaining time slots of the residential photovoltaic storage system as a mixed-integer linear programming problem is as follows:

[0074] S21. Define all variables as follows, categorized into continuous variables and 0-1 variables:

[0075] Defined continuous variables Including: users in Power purchased from the grid during different time periods Photovoltaic storage system in Remaining power supplied to the grid during a given period Photovoltaic storage system in Periodic power curtailment Energy storage devices in Charging power during the period Energy storage devices in Discharge power during the period Energy storage devices in SOC during the period Among them, continuous variables , , , , , All are non-negative;

[0076] Defined 0-1 variables Including: energy storage devices Charging status during the period Energy storage devices in Discharge state during the period Users Electricity purchase status from the grid during a certain period Photovoltaic storage system in Remaining electricity status during a given time period ;

[0077] S22. With the goal of minimizing the total operating cost for the user's remaining time period on the day, the objective function for the mixed-integer linear programming problem is defined as follows:

[0078]

[0079] in This represents the set of remaining time periods of the day at the current moment. Indicates that the user is The price of purchasing electricity from the grid during a given period. Indicates that the photovoltaic energy storage system is in The price of surplus electricity sold to the grid during certain time periods. Indicates the duration of a single scheduling period;

[0080] S23. Set the constraints for the mixed-integer linear programming problem as follows:

[0081] 1) Power balance constraints:

[0082]

[0083] in express Forecasted photovoltaic power for the period express Forecasted load power for the specified time period, and when When the value is set to 1, the predicted value is replaced by the actual photovoltaic and load power.

[0084] 2) Maximum charge / discharge power constraints for energy storage:

[0085]

[0086] in , These are the maximum charging power and maximum discharging power of energy storage, respectively.

[0087] 3) Mutual exclusion constraints for energy storage charge and discharge states:

[0088]

[0089] 4) Energy storage SOC constraints:

[0090]

[0091] in Indicates the maximum energy storage capacity. This indicates the efficiency of the energy storage charging process. This indicates the efficiency of the energy storage and discharge process;

[0092] 5) Energy storage SOC upper and lower limits constraints:

[0093]

[0094] in This indicates the lower limit of the State of Charge (SOC) during the energy storage charging and discharging process. This indicates the upper limit of the State of Charge (SOC) during the energy storage charging and discharging process;

[0095] 6) Energy storage ramp-up rate constraints:

[0096]

[0097] in This represents the maximum value of the power increase during the energy storage charging process. This represents the maximum value of the power increase during the energy storage discharge process;

[0098] 7) Power purchase and sale constraints for photovoltaic-storage systems:

[0099]

[0100] in This indicates the maximum limit for purchasing electricity from the grid. This indicates the grid-connected capacity of surplus photovoltaic power, based on local policies and inverter capacity.

[0101] 8) Mutual exclusion constraints for power purchase and sale status of photovoltaic-storage systems:

[0102]

[0103] 9) Constraints on shared inverters for photovoltaic and energy storage systems:

[0104]

[0105] in This indicates the rated power limit of the inverter in a photovoltaic-storage system, which is typically 50%-70% of the rated power in some countries.

[0106] Specifically, the selection process for the energy storage management plan during the current scheduling period includes:

[0107] Each scheduling period is based on the energy storage energy management plan for the remaining period of the day. The first value in the plan result is selected as the energy storage energy management plan for the current period. The energy storage charging and discharging power of the current period in the plan result is transmitted to the inverter. The inverter follows the plan result, combines the real-time photovoltaic power and real-time load power in the current scheduling period, and executes the charging and discharging of energy storage based on power balance constraints to obtain the actual power purchased and sold to the grid and execute it.

[0108] Meanwhile, a daily energy storage management plan is formulated for the remaining time period during each scheduling period. The first value in the energy storage charging and discharging power result in the plan is selected as the execution indicator and rolled forward over time.

[0109] II. The rolling optimization phase based on multiple prediction updates includes:

[0110] Multiple forecast updates are performed at several fixed times within the scheduling day. The forecast objects are adjusted to the photovoltaic power and load power of each time period within the next 24 hours from each fixed time. The n fixed times are divided into n+1 forecast update periods. Each scheduling period on the day determines its current forecast update period and extracts the forecast update data for the corresponding 24 hours. The extraction range is from the current time period to the last time period of the day. Based on the extracted photovoltaic and load power data, combined with the real-time photovoltaic and load power data, an updated energy storage energy management plan for the remaining time periods of the day is formulated.

[0111] For example, forecast updates are performed at fixed times during the day: morning, noon, and evening. Specifically, these are set at 06:00, 12:00, and 18:00. These three fixed times are then divided into four forecast update periods: the first period is from 00:00 to 06:00, the second from 06:00 to 12:00, the third from 12:00 to 18:00, and the fourth from 18:00 to 24:00. Forecasts are made at each fixed time, predicting the photovoltaic power and load power for each time period within the next 24 hours from that fixed time. For example, if 07:00 on a scheduling day is determined to be the second forecast update period, the forecast update data for the 24 hours corresponding to the second forecast update period is extracted, with the extraction range being 07:00 to 24:00. Based on the extracted data, combined with real-time photovoltaic and load power data, an updated energy storage energy management plan for the remaining time periods of the day is formulated.

[0112] Example:

[0113] Consider as Figure 2 The residential photovoltaic (PV) energy storage system shown has a peak daily load of approximately 2500W, occurring between 18:00 and 22:00. Based on the described implementation method, energy management is performed on a typical day, and the rolling optimization results for the entire day are simulated. The daily PV and load power are as follows: Figure 3 As shown.

[0114] On the day before the scheduling date, the energy management algorithm optimizes the power generation mode for the next scheduling cycle based on load and photovoltaic forecast results (time granularity of 15 minutes). The electricity price information for that day is shown in Table 1. The energy storage system chooses to charge during off-peak hours and discharge during peak load or peak electricity price periods, maximizing its role in arbitrage against price differences. The correlation coefficients of residential photovoltaic-storage systems are shown in Table 2.

[0115] Meanwhile, load power and photovoltaic power forecasts are updated at 8:00, 14:00 and 20:00 on the dispatch day, and the latest energy management strategy for the photovoltaic and energy storage system is formulated based on the latest forecast results.

[0116] Table 1. Time-of-use electricity prices on a certain day in March

[0117]

[0118] Table 2 Main parameters of residential photovoltaic energy storage system

[0119]

[0120] The scheduling day is divided into multiple time periods with a 15-minute cycle. Based on the power / energy forecast information of photovoltaics and loads, as well as the real-time transmitted photovoltaic and load power data, and comprehensively considering electricity costs, surplus power grid connection revenue, energy storage scheduling costs, and curtailment losses, the goal is to minimize the operating cost of the remaining time periods of the day. Taking into account power balance constraints, photovoltaic grid-connected power limits, photovoltaic-energy storage system equipment technical characteristics constraints, and constraints on the shared inverter for the photovoltaic-energy storage system, the optimization results for the remaining time periods are obtained. The energy storage charging and discharging power and SOC status throughout the entire day are as follows: Figure 4 As shown.

[0121] Based on the energy storage management plan for the remaining time period of the day, the energy storage power plan for the current time period is selected. Considering the real-time load and photovoltaic output, the reference value of the charging and discharging power of the energy storage for the current time period is output. The power purchased from the grid, the power sold, and the curtailment of photovoltaic power for each dispatch period of the day are as follows: Figure 5 As shown.

[0122] To demonstrate the superiority of the energy management strategy for residential photovoltaic energy storage systems in this invention, the following two strategies are considered:

[0123] Case 1: Energy storage absorbs excess photovoltaic power from the morning until the SOC is fully charged, and begins to discharge during the peak load period in the evening;

[0124] Case 2: The energy management strategy proposed in this invention.

[0125] For the entire photovoltaic-storage system, energy storage can consider electricity price differences to make intelligent mode decisions, thus reducing operating costs. A comparison of the daily curtailment situation under the two modes is shown below. Figure 6As shown in Table 3, compared to Case 1, Case 2 can effectively reduce the curtailment rate and improve the utilization rate of photovoltaic resources.

[0126] Table 3 Comparison of photovoltaic grid integration rates under two different energy management strategies

[0127]

[0128] Based on one year's data and forecasts, operational simulations show that the electricity cost without using the photovoltaic and energy storage system is £443.55, the cost using the morning charging strategy (Case 1) is £203.47, and the cost using the strategy of this invention (Case 2) is £105.32. This demonstrates that more economic benefits can be obtained while increasing the photovoltaic absorption rate.

[0129] The above description is merely a preferred embodiment of the present invention. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make many possible variations and modifications to the technical solutions of the present invention using the methods and techniques disclosed above, or modify them into equivalent embodiments with equivalent changes, without departing from the scope of the technical solutions of the present invention. Therefore, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solutions of the present invention shall still fall within the protection scope of the technical solutions of the present invention.

Claims

1. A method for energy management of residential photovoltaic-storage systems, characterized in that, The method includes a rolling optimization phase based on day-ahead forecasts and a rolling optimization phase based on multiple forecast updates for energy storage energy management plans; The rolling optimization phase based on day-ahead forecasts includes: The energy management of residential photovoltaic and energy storage systems is based on a daily scheduling cycle. Before starting the energy management plan for the current scheduling day, the photovoltaic power forecast and load power forecast for the next 24 hours are obtained the night before. If there are fluctuating electricity prices in the local area, the time-of-use electricity price for the scheduling day also needs to be obtained. Before the scheduling period begins, the real-time photovoltaic power and real-time load power of the photovoltaic and energy storage system are obtained. The scheduling day is divided into N time periods. Based on the power forecast information of photovoltaic and load for each time period in the next 24 hours and the real-time photovoltaic and load power data, the electricity cost, surplus power grid connection revenue, energy storage scheduling cost and curtailment loss are comprehensively considered. With the goal of minimizing the operating cost of the remaining time periods of the day, the plan takes into account power balance constraints, photovoltaic grid connection power limit constraints, photovoltaic and energy storage system equipment technical characteristic constraints and photovoltaic and energy storage system shared inverter constraints. The remaining time periods of the day are modeled as a mixed integer linear programming problem. The energy storage energy management plan for each remaining time period of the day is formulated, including energy storage charging and discharging power, energy storage battery SOC, power purchased from and sold to the grid and curtailed power. Based on the energy storage energy management plan for the remaining time period of the day, the plan for the current time period is selected, and the energy storage charging and discharging power of the current time period in the plan result is transmitted to the inverter. The inverter follows the plan result, combines the real-time photovoltaic power and real-time load power in the current dispatch period, executes the charging and discharging of energy storage, obtains the actual power to purchase and sell electricity to the grid and executes it. The rolling optimization phase based on multiple prediction updates includes: Multiple forecast updates are performed at several fixed times within the scheduling day. The forecast objects are adjusted to the photovoltaic power and load power of each time period within the next 24 hours from each fixed time. The n fixed times are divided into n+1 forecast update periods. Each scheduling period on the day determines its current forecast update period and extracts the forecast update data for the corresponding 24 hours. The extraction range is from the current time period to the last time period of the day. Based on the extracted photovoltaic and load power data, combined with the real-time photovoltaic and load power data, an updated energy storage energy management plan for the remaining time periods of the day is formulated.

2. The energy storage management method for residential photovoltaic-storage systems according to claim 1, characterized in that, The input information required in the rolling optimization phase based on day-ahead forecasts includes: 1) Household photovoltaic power forecast and load power forecast with the same time scale as the scheduling period, one value for each period; 2) Time-of-use electricity pricing for dispatch days with hourly granularity; 3) Real-time photovoltaic power and load power at the start of each scheduling period; 4) Real-time SOC at the start of each scheduling period.

3. The energy storage management method for residential photovoltaic-storage systems according to claim 1, characterized in that, The method for modeling the remaining time slots of a residential photovoltaic-storage system as a mixed-integer linear programming problem is as follows: S1. Define all variables as follows, categorized into continuous variables and 0-1 variables: Defined continuous variables Including: users in Power purchased from the grid during different time periods Photovoltaic storage system in Remaining power supplied to the grid during a given period Photovoltaic storage system in Periodic power curtailment Energy storage devices in Charging power during the period Energy storage devices in Discharge power during the period Energy storage devices in SOC during the period Among them, continuous variables , , , , , All are non-negative; Defined 0-1 variables Including: energy storage devices Charging status during the period Energy storage devices in Discharge state during the period Users Electricity purchase status from the grid during a certain period Photovoltaic storage system in Remaining electricity status during a given time period ; S2. The objective function for the mixed-integer linear programming problem, defined with the goal of minimizing the total operating cost for the remaining time period of the day, is as follows: in This represents the set of remaining time periods of the day at the current moment. Indicates that the user is The price of purchasing electricity from the grid during a given period. Indicates that the photovoltaic energy storage system is in The price of surplus electricity sold to the grid during certain time periods. Indicates the duration of a single scheduling period; S3. Set the constraints for the mixed-integer linear programming problem as follows: 1) Power balance constraints: in express Forecasted photovoltaic power for the period express Forecasted load power for each time period; 2) Maximum charge / discharge power constraints for energy storage: in , These are the maximum charging power and maximum discharging power of energy storage, respectively. 3) Mutual exclusion constraints for energy storage charge and discharge states: 4) Energy storage SOC constraints: in Indicates the maximum energy storage capacity. This indicates the efficiency of the energy storage charging process. This indicates the efficiency of the energy storage and discharge process; 5) Energy storage SOC upper and lower limits constraints: in This indicates the lower limit of the State of Charge (SOC) during the energy storage charging and discharging process. This indicates the upper limit of the State of Charge (SOC) during the energy storage charging and discharging process; 6) Energy storage ramp-up rate constraints: in This represents the maximum value of the power increase during the energy storage charging process. This represents the maximum value of the power ramp-up during the energy storage discharge process; 7) Power purchase and sale constraints for photovoltaic-storage systems: in This indicates the maximum limit for purchasing electricity from the grid. This indicates the grid-connected capacity of surplus photovoltaic power, based on local policies and inverter capacity. 8) Mutual exclusion constraints for power purchase and sale status of photovoltaic-storage systems: 9) Constraints on shared inverters for photovoltaic and energy storage systems: in This indicates the rated power limit of the inverter in the photovoltaic-storage system.

4. The energy storage management method for residential photovoltaic-storage systems according to claim 1, characterized in that, The selection process for the energy storage management plan during the current scheduling period includes: Each scheduling period is based on the energy storage energy management plan for the remaining period of the day. The first value in the plan result is selected as the energy storage energy management plan for the current period. The energy storage charging and discharging power of the current period in the plan result is transmitted to the inverter. The inverter follows the plan result, combines the real-time photovoltaic power and real-time load power in the current scheduling period, and executes the charging and discharging of energy storage based on power balance constraints to obtain the actual power purchased and sold to the grid and execute it. Meanwhile, a daily energy storage management plan is formulated for the remaining time period during each scheduling period. The first value in the energy storage charging and discharging power result in the plan is selected as the execution indicator and rolled forward over time.

Images