Energy storage power grid intelligent management system based on big data driving

Abstract

The invention discloses an energy storage power grid intelligent management system based on big data driving, and belongs to the technical field of energy storage power grids, and the system comprises an obtaining module which is used for obtaining historical information and carbon dioxide emission factors; the prediction module is used for performing prediction based on historical information to obtain electrical load power, wind and light output power, electricity price information and a charge state of a storage battery pack in an energy storage power grid in a next scheduling period; the optimization module is used for solving a Pareto optimal solution set and determining a compromise solution by using a preset target energy optimization scheduling algorithm based on the carbon dioxide emission factor and the prediction data of the next scheduling period and taking the lowest operation and maintenance cost of the energy storage power grid of the next scheduling period and the minimum carbon dioxide emission as optimization objectives; and the control module is used for determining the charging and discharging strategy of the storage battery pack in the next scheduling period based on the compromise solution and controlling the charging and discharging of the storage battery pack. When the system is used for intelligent management of the energy storage power grid, both economic cost and environmental cost can be taken into account.

Description

Technical Field

[0001] This invention belongs to the field of energy storage grid technology, specifically relating to a big data-driven intelligent management system for energy storage grids. Background Technology

[0002] With the rapid development of renewable energy, wind turbines and photovoltaic (PV) power generation equipment are gradually increasing their share in the power supply. Due to the significant intermittency and randomness of wind turbines and PV, the stable operation of the power system (or external grid) poses a significant challenge. Energy storage grids, capable of storing excess energy during periods of overcapacity and releasing it during peak load periods, have become crucial for stabilizing the power system. Therefore, managing energy storage grids has become an important task. Existing technologies typically combine battery management within the energy storage grid with economic costs, aiming to achieve intelligent management of the grid by reducing battery management costs. However, this approach often only considers direct economic costs, such as battery charging and discharging losses, neglecting the environmental costs incurred when purchasing electricity from the external grid. Therefore, balancing economic and environmental costs is an urgent problem to be solved when implementing intelligent management of energy storage grids.

[0003] It should be noted that the information disclosed in the background section above is only used to enhance the understanding of the background of this application, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0004] The present invention aims to at least partially solve one of the technical problems in the related art.

[0005] This disclosure provides a big data-driven intelligent management system for energy storage grids, which can balance economic and environmental costs when intelligently managing energy storage grids.

[0006] In some embodiments, the big data-driven intelligent management system for energy storage grids includes: an acquisition module for acquiring historical information and carbon dioxide emission factors; the historical information includes historical wind and solar power output information, historical electricity load information, historical electricity price information, and the historical state of charge (SOC) of battery banks in the energy storage grid; a prediction module for predicting based on the historical information to obtain the electricity load power, wind and solar power output power, electricity price information, and SOC of battery banks in the energy storage grid for the next scheduling cycle; an optimization module for, based on the carbon dioxide emission factors and the electricity load power, wind and solar power output power, electricity price information, and SOC of battery banks in the next scheduling cycle, using the minimum energy storage grid operation and maintenance cost and the minimum carbon dioxide emissions for the next scheduling cycle as optimization objectives, solving for the Pareto optimal solution set using a preset target energy optimization scheduling algorithm, and determining the compromise solution in the Pareto optimal solution set; and a control module for determining the charging and discharging strategy of the battery banks in the next scheduling cycle based on the compromise solution, and controlling the charging and discharging of the battery banks in the next scheduling cycle based on the charging and discharging strategy.

[0007] In some embodiments, the historical wind and solar power output information includes historical status information, historical environmental information, and historical power output of the wind and solar power generation equipment; the prediction module includes a wind and solar sub-module; the wind and solar sub-module includes: a wind and solar prediction unit, used to predict the initial wind and solar power output and the corresponding typical power scenario of the wind and solar power generation equipment in the next scheduling cycle based on the historical status information, historical environmental information, and historical power output; a wind and solar error unit, used to obtain wind and solar error information of the typical power scenario corresponding to the next scheduling cycle; wherein, the wind and solar error information represents the error information between the predicted initial wind and solar power output and the actual wind and solar power output; and a wind and solar adjustment unit, used to adjust the initial wind and solar power output based on the wind and solar error information of the next scheduling cycle to obtain the wind and solar power output of the next scheduling cycle.

[0008] In some embodiments, the wind and solar error unit includes: a wind and solar determination subunit, configured to acquire multiple error data for each typical power scenario, and determine wind and solar error information corresponding to each typical power scenario based on the multiple error data; a wind and solar scenario subunit, configured to predict environmental information for the next scheduling cycle, and determine the corresponding typical power scenario based on the environmental information for the next scheduling cycle; and a wind and solar error subunit, configured to determine wind and solar error information corresponding to the next scheduling cycle based on the typical power scenario corresponding to the next scheduling cycle.

[0009] In some embodiments, the optimization module includes a selection submodule; the selection submodule includes: a time period unit, used to determine the electricity price time period corresponding to the next scheduling cycle; and a selection unit, used to select a preset energy optimization scheduling algorithm as the target energy scheduling optimization algorithm based on the state of charge and the electricity price time period of the next scheduling cycle.

[0010] In some embodiments, the preset energy optimization scheduling algorithm includes a first optimization algorithm, a second optimization algorithm, and a third optimization algorithm; the state of charge includes charging required, discharging required, and rechargeable / dischargeable; and the electricity price period includes peak and off-peak hours. The selection unit includes: a first algorithm subunit, configured to select the first optimization algorithm as the target energy optimization scheduling algorithm if the state of charge in the next scheduling cycle is charging required or discharging required; and a second algorithm subunit, configured to determine the electricity price period corresponding to the next scheduling cycle if the state of charge in the next scheduling cycle is rechargeable / dischargeable; if the electricity price period is peak hour, select the second optimization algorithm as the target energy optimization scheduling algorithm; and if the electricity price period is off-peak hour, select the third optimization algorithm as the target energy optimization scheduling algorithm.

[0011] In some embodiments, the system further includes an algorithm construction module; the algorithm construction module is configured to: construct a first optimization algorithm, a second optimization algorithm, and a third optimization algorithm, wherein the first optimization algorithm, the second optimization algorithm, and the third optimization algorithm each include an objective function and constraints as follows: Objective function:

[0012]

[0013]

[0014] In the objective function of the first optimization algorithm

[0015] In the objective functions of the second and third optimization algorithms, +

[0016]

[0017]

[0018] In the second optimization algorithm,

[0019]

[0020] In the third optimization algorithm

[0021]

[0022] Constraints:

[0023]

[0024]

[0025]

[0026] in, To reduce the operation and maintenance costs of energy storage grids, The cost of charging and discharging battery packs. This is the maintenance cost coefficient for the battery pack. and These are the charging and discharging state variables of the battery pack during the scheduling cycle, with values ​​of 0 or 1, indicating whether the corresponding state has occurred. The charging power of the battery pack, This refers to the discharge power of the battery pack. The duration of a scheduling cycle, Depreciation costs for charging battery packs, For charging penalty function The depreciation cost of discharging the battery pack. Let be the discharge penalty function. This refers to the grid purchase price of electricity in the electricity price information. This refers to the electricity purchased by the microgrid from the external power grid. for For Sale quantity, This refers to carbon dioxide emissions. As a carbon dioxide emission factor, Let be the power of the electrical load at time t. Let t be the wind and solar power output at time t. Let be the tie-line power at time t. A positive value indicates that the microgrid purchases electricity from the external grid. A negative value indicates that the microgrid sells electricity to the external grid. Let be the charging and discharging power of the battery pack at time t. A positive value indicates that the battery pack is charging. A negative value indicates that the battery pack is discharging. In a charged state, At minimum state of charge, This is the maximum state of charge.

[0027] In some embodiments, the battery pack includes multiple batteries; the optimization module further includes a scheduling submodule; the scheduling submodule includes: a detection unit, used to detect the operating status of each battery in the battery pack, and determine whether there are abnormal batteries in the battery pack based on each operating status; and an update unit, used to, if there are abnormal batteries in the battery pack, use the remaining batteries other than the abnormal batteries as a new battery pack, and based on the new battery pack, use a preset target energy optimization scheduling algorithm to determine the charging and discharging strategy of the new battery pack in the next scheduling cycle.

[0028] The beneficial effects of this invention are as follows: Electricity load power, wind and solar power output, electricity price information, and the state of charge (SOC) of battery banks in the energy storage grid are key factors affecting the operation and maintenance costs of the energy storage grid. The carbon dioxide emission factor is a key factor affecting carbon dioxide emissions. The operation and maintenance costs of the energy storage grid affect its economic costs, while carbon dioxide emissions affect its environmental costs. Therefore, a module acquires historical information and the carbon dioxide emission factor to allow a prediction module to forecast electricity load power, wind and solar power output, electricity price information, and the SOC of battery banks for the next scheduling cycle. Then, an optimization module uses the minimum operation and maintenance cost and the minimum carbon dioxide emissions for the energy storage grid in the next scheduling cycle as its optimization objectives. The carbon dioxide emission factor, along with the predicted user load power, wind and solar power output, electricity price information, and battery SOC of the next scheduling cycle, are input into the target energy optimization scheduling algorithm for calculation. This algorithm solves for the Pareto optimal solution set and determines a compromise solution that balances the economic and environmental costs of the energy storage grid. Finally, the control module determines and controls the charging and discharging strategy of the battery pack for the next scheduling cycle based on a compromise solution, thereby balancing the economic and environmental costs of the energy storage grid when managing the energy storage grid intelligently.

[0029] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0030] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 This is a diagram of the combined system architecture of a microgrid and an external power grid; Figure 2 This is a schematic diagram of a big data-driven intelligent management system for energy storage grids. Detailed Implementation

[0031] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0032] For ease of understanding, combined with Figure 1 As shown, Figure 1 A system architecture diagram for a microgrid and an external power grid is provided. In this embodiment, the microgrid includes an energy storage grid, wind and solar power generation equipment, and electrical loads. The energy storage grid mainly utilizes battery banks for energy storage or release; therefore, the lowest operation and maintenance cost of the energy storage grid is related to the loss cost of charging and discharging the battery banks. When wind and solar power generation cannot meet the electrical load, both the energy storage grid (i.e., discharging) and the external power grid (i.e., selling electricity to the microgrid) can provide power to the electrical load. The external power grid (i.e., selling electricity to the microgrid) and the remaining wind and solar power generation (when wind and solar power generation exceeds the electrical load) can both provide power to the energy storage grid. It is understood that wind and solar power generation are renewable resources and produce almost no carbon dioxide; the external power grid mainly consists of thermal power generation, which produces a significant amount of carbon dioxide. Therefore, carbon dioxide emissions are related to the amount of electricity the microgrid purchases from the external power grid.

[0033] Combination Figure 2 As shown in the embodiments of this disclosure, a big data-driven intelligent management system for energy storage grids includes: an acquisition module, a prediction module, an optimization module, and a control module. The acquisition module acquires historical information and carbon dioxide emission factors; the historical information includes historical wind and solar power output, historical electricity load, historical electricity price, and the historical state of charge (SOC) of the battery banks in the energy storage grid. The prediction module predicts the electricity load, wind and solar power output, electricity price, and SOC of the battery banks in the energy storage grid for the next scheduling cycle based on the historical information. The optimization module, based on the carbon dioxide emission factors and the electricity load, wind and solar power output, electricity price, and SOC of the battery banks in the next scheduling cycle, uses the minimum energy storage grid operation and maintenance cost and the minimum carbon dioxide emissions for the next scheduling cycle as optimization objectives, and solves for the Pareto optimal solution set using a preset target energy optimization scheduling algorithm, and determines the compromise solution in the Pareto optimal solution set. The control module is used to determine the charging and discharging strategy of the battery pack in the next scheduling cycle based on the compromise solution, and to control the charging and discharging of the battery pack in the next scheduling cycle based on the charging and discharging strategy.

[0034] The energy storage grid intelligent management system based on big data driven by this disclosure provides that the key factors affecting the operation and maintenance cost of the energy storage grid are the power load, wind and solar power output, electricity price information, and the state of charge (SOC) of the battery packs in the energy storage grid. The carbon dioxide emission factor is the key factor affecting carbon dioxide emissions. The operation and maintenance cost of the energy storage grid affects its economic cost, while carbon dioxide emissions affect its environmental cost. Thus, a module acquires historical information and the carbon dioxide emission factor, allowing a prediction module to use this historical information to predict the power load, wind and solar power output, electricity price information, and SOC of the battery packs for the next scheduling cycle. Then, an optimization module uses the minimum operation and maintenance cost and the minimum carbon dioxide emissions for the energy storage grid in the next scheduling cycle as optimization objectives. The carbon dioxide emission factor and the predicted power load, wind and solar power output, electricity price information, and SOC of the battery packs for the next scheduling cycle are input into the target energy optimization scheduling algorithm for calculation. The Pareto optimal solution set is solved, and a compromise solution is determined, which balances the economic and environmental costs of the energy storage grid. Finally, the control module determines and controls the charging and discharging strategy of the battery pack for the next scheduling cycle based on a compromise solution, thereby balancing the economic and environmental costs of the energy storage grid when managing the energy storage grid intelligently.

[0035] It is understood that, since the carbon dioxide emission factor is relatively stable and its change frequency is much lower than that of the scheduling cycle, this embodiment does not predict the carbon dioxide emission factor for each scheduling cycle in order to reduce the amount of computation. The wind and solar power generation equipment in this embodiment includes wind turbine power generation equipment and photovoltaic power generation equipment.

[0036] Preferably, the historical wind and solar power output information includes historical status information, historical environmental information, and historical power output of the wind and solar power generation equipment. The prediction module includes a wind and solar sub-module; the wind and solar sub-module includes: a wind and solar prediction unit, a wind and solar error unit, and a wind and solar adjustment unit. The wind and solar prediction unit is used to predict the initial wind and solar power output and the corresponding typical power scenario for the next scheduling cycle based on the historical status information, historical environmental information, and historical power output. The wind and solar error unit is used to obtain the wind and solar error information for the typical power scenario corresponding to the next scheduling cycle; wherein the wind and solar error information represents the error information between the predicted initial wind and solar power output and the actual wind and solar power output. The wind and solar adjustment unit is used to adjust the initial wind and solar power output based on the wind and solar error information for the next scheduling cycle to obtain the wind and solar power output for the next scheduling cycle.

[0037] In this way, by predicting the initial wind and solar power output and typical power scenarios corresponding to the next scheduling cycle through the wind and solar prediction unit, and then adjusting the initial wind and solar power output based on the wind and solar error information corresponding to the typical power scenarios, the accuracy of the predicted wind and solar power output for the next scheduling cycle can be improved.

[0038] In some embodiments, the state information of wind and solar power generation equipment characterizes the equipment's own power generation characteristics, such as output power and equipment fault status. Environmental information mainly includes wind speed, wind direction, temperature, solar irradiance, and sunshine duration. Based on historical state information, historical environmental information, and historical output power, the relationship between wind and solar power output and state and environmental information can be determined. This allows for prediction based on the state and environmental information of the previous or current scheduling cycle, using this relationship to obtain the initial wind and solar power output for the next scheduling cycle. It is understood that predicting the initial wind and solar power output to obtain the output power of wind and solar power generation equipment is a relatively conventional existing technology, and therefore will not be elaborated upon here. Typical power scenario characterization can represent scenarios that reflect the typical output power of wind and solar power generation equipment. For example: The output power of wind turbines in wind and solar power generation equipment is greatly affected by wind speed. Therefore, in this embodiment, the preset typical power scenarios corresponding to the wind turbines include a stable high wind speed scenario, a stable low wind speed scenario, a fluctuating high wind speed scenario, and a fluctuating low wind speed scenario. The stable high wind speed scenario represents a scenario where the wind speed is stable and relatively high (e.g., the wind speed fluctuation is small, and the wind speed is equal to or greater than 11 m / s and less than the cut-out wind speed); the stable low wind speed scenario represents a scenario where the wind speed is stable and relatively low (e.g., greater than the cut-in wind speed and less than 11 m / s); the fluctuating high wind speed scenario represents a scenario where the wind speed fluctuates significantly and is relatively high; and the fluctuating low wind speed scenario represents a scenario where the wind speed fluctuates slightly and is relatively low.

[0039] The output power of photovoltaic (PV) power generation equipment is greatly affected by the intensity and duration of sunlight. Therefore, in this embodiment, the preset typical power scenarios corresponding to PV include strong and long sunlight scenarios, strong and short sunlight scenarios, weak and long sunlight scenarios, and weak and short sunlight scenarios. Specifically, a strong and long sunlight scenario represents a scenario where the sunlight intensity is greater than or equal to a preset intensity (e.g., 1000 W / m²) and the sunlight duration is greater than or equal to a preset duration (0.8 * the duration corresponding to the scheduling cycle); a strong and short sunlight scenario represents a scenario where the sunlight intensity is greater than or equal to a preset intensity and the sunlight duration is less than a preset duration; a weak and long sunlight scenario represents a scenario where the sunlight intensity is less than a preset intensity and the sunlight duration is greater than or equal to a preset duration; and a weak and short sunlight scenario represents a scenario where the sunlight intensity is less than a preset intensity and the sunlight duration is less than a preset duration.

[0040] Preferably, the wind-solar error unit includes: a wind-solar determination subunit, a wind-solar scene subunit, and a wind-solar error subunit. The wind-solar determination subunit is used to acquire multiple error data points for each typical power scene and determine the corresponding wind-solar error information based on these error data points. The wind-solar scene subunit is used to predict the environmental information for the next scheduling cycle and determine the corresponding typical power scene based on the environmental information for the next scheduling cycle. The wind-solar error subunit is used to determine the wind-solar error information for the next scheduling cycle based on the typical power scene corresponding to the next scheduling cycle. Here, error data equals the difference between the actual wind and solar power output and the predicted initial wind and solar power output.

[0041] The output power of wind and solar power generation equipment exhibits certain similarities under the same typical power scenarios, and therefore, the wind and solar error information under the same typical power scenarios also shows similarities. Thus, for each typical power scenario, by determining the corresponding wind and solar error information, and then using the typical power scenario corresponding to the next scheduling cycle to determine the wind and solar error information for the next scheduling cycle, it is possible to quickly determine the wind and solar error information for the next scheduling cycle. This allows for timely adjustments to the initial wind and solar power output, resulting in a more accurate wind and solar power output for the next scheduling cycle.

[0042] In some embodiments, the wind and solar error information includes an upper limit and a lower limit for wind and solar errors. Specifically, the wind and solar determination subunit is used to determine the upper and lower limits of wind and solar errors using confidence intervals. For example, for each typical power scenario: based on multiple acquired error data, the corresponding sample mean and sample standard deviation are calculated. Based on the sample mean and sample standard deviation, a confidence interval is constructed using a T-distribution. The upper limit of this confidence interval is the upper limit of the wind and solar error, and the lower limit is the lower limit of the wind and solar error.

[0043] In some embodiments, the wind and solar scenario subunit is specifically used to predict environmental factors for the next scheduling cycle based on historical environmental information. For example, for wind turbines, the predicted environmental information is wind speed. Then, based on the predicted initial wind and solar power output for the next scheduling cycle, the volatility of the wind and solar power output for that scheduling cycle is calculated. The wind speed and volatility are used together to determine the typical power scenario corresponding to the next scheduling cycle. For photovoltaics, the predicted environmental information is solar irradiance and solar duration. Solar irradiance and solar duration are used together to determine the typical power scenario corresponding to the next scheduling cycle. It is understood that predicting environmental information is a relatively mature existing technology and will not be elaborated upon here.

[0044] In some embodiments, the wind and solar power adjustment unit is specifically used to take the sum of the predicted initial wind and solar power output for the next scheduling cycle and the upper limit of the wind and solar error as the wind and solar power output for the next scheduling cycle, or to take the sum of the predicted initial wind and solar power output for the next scheduling cycle and the lower limit of the wind and solar error as the wind and solar power output for the next scheduling cycle. It is understood that the wind and solar power output for the next scheduling cycle in this embodiment is a predicted value.

[0045] In some embodiments, for each preset typical load scenario, multiple error data points are acquired. Based on the error data corresponding to each typical load scenario, the upper limit (upper limit of confidence interval) and lower limit (lower limit of confidence interval) of the load error corresponding to each typical load scenario are determined using a confidence curve and the T-distribution method. Then, based on historical electricity load information, the initial electricity load power and typical load scenario for the next scheduling cycle are predicted, and the upper limit and lower limit of the load error for the typical load scenario corresponding to the next scheduling cycle are determined. The sum of the initial electricity load power and the upper limit of the load error is used as the predicted electricity load power for the next scheduling cycle, or the sum of the initial electricity load power and the lower limit of the load error is used as the predicted electricity load power for the next scheduling cycle. Wherein, the error data corresponding to the typical load scenario = actual electricity load power - initial electricity load power.

[0046] In this way, the power load for the next scheduling cycle is first predicted based on historical load information, and then the upper and lower limits of the load error are determined in combination with typical load scenarios. This allows for the adjustment of the initial power load, thereby improving the accuracy of the final power load for the next scheduling cycle.

[0047] It is understandable that predicting the initial electricity load for the next scheduling cycle based on historical electricity load information is a relatively conventional existing technology, so it will not be elaborated upon here. Historical electricity load information includes historical electricity load power, historical ambient temperature, and historical electricity usage scenarios. Typical load scenarios can also be determined based on ambient temperature and electricity usage scenarios. For example, when the ambient temperature is high and the scenario is a commercial office setting, the typical load scenario is a high-load scenario; when the ambient temperature is high and the scenario is a residential setting, the typical load scenario is a normal-load scenario; when the ambient temperature is moderate and the scenario is a commercial office setting, the typical load scenario is a normal-load scenario; and when the ambient temperature is moderate and the scenario is a residential setting, the typical load scenario is a low-load scenario.

[0048] Preferably, the optimization module includes a selection submodule; the selection submodule includes a time period unit and a selection unit. The time period unit is used to determine the electricity price period corresponding to the next scheduling cycle. The selection unit is used to select a preset energy optimization scheduling algorithm as the target energy scheduling optimization algorithm based on the state of charge and electricity price period of the next scheduling cycle.

[0049] Electricity prices exhibit typical time-of-day characteristics, with peak-hour prices significantly higher than off-peak prices. Overcharging or over-discharging of batteries both impact their lifespan. Therefore, electricity prices and the state of charge (SOC) of batteries have a significant impact on the operation and maintenance costs and CO2 emissions of energy storage grids. Thus, selecting a corresponding energy optimization scheduling algorithm based on the electricity price period and SOC of the next scheduling cycle is crucial for better balancing the economic and environmental costs of energy storage grids.

[0050] In some embodiments, the electricity price period includes peak and off-peak hours, with peak hours from 16:00 to 24:00 and off-peak hours from 00:00 to 16:00. The start time of the next scheduling cycle is 15:50, and the duration of one scheduling cycle is half an hour. Therefore, the next scheduling cycle will experience both off-peak and peak periods. If the next scheduling cycle includes both peak and off-peak electricity price periods, the time period unit is used to determine the corresponding electricity price period based on the start time of the next scheduling cycle. In this embodiment, the electricity price period of the next scheduling cycle is off-peak. Alternatively, the time period unit is used to calculate the electricity price duration occupied by peak and off-peak hours within the next scheduling cycle, and to take the longer electricity price period as the corresponding electricity price period. In this embodiment, the electricity price duration occupied by off-peak hours is 10 minutes, and the electricity price duration occupied by peak hours is 20 minutes, so the corresponding electricity price period is peak.

[0051] In some embodiments, ,in, This represents the state of charge of the battery pack for the next scheduling cycle. This refers to the total capacity of the battery pack. This refers to the remaining capacity of the battery pack at the end of the current scheduling cycle. Since the state of charge (SOC) for the next scheduling cycle needs to be determined in advance, the remaining capacity at the end of the current scheduling cycle needs to be calculated or predicted. In this embodiment, the remaining capacity at the end of the current scheduling cycle can be calculated or predicted based on the charge / discharge amount corresponding to the determined charge / discharge strategy of the current scheduling cycle and the remaining capacity at the end of the previous scheduling cycle. For example, if the current scheduling cycle is the 5th scheduling cycle, the previous scheduling cycle is the 4th scheduling cycle, and the next scheduling cycle is the 6th scheduling cycle. The charge / discharge amount corresponding to the charge / discharge strategy of the 5th scheduling cycle is taken as the target capacity. Based on this target capacity and the remaining capacity at the end of the 4th scheduling cycle, the remaining capacity of the battery pack at the end of the 5th scheduling cycle is determined or predicted in advance before the arrival of the 6th scheduling cycle, so as to obtain the SOC of the battery pack at the end of the 6th scheduling cycle.

[0052] Preferably, the preset energy optimization scheduling algorithm includes a first optimization algorithm, a second optimization algorithm, and a third optimization algorithm. The state of charge includes charging, discharging, and rechargeable / dischargeable. The electricity price period includes peak and off-peak hours. The selection unit includes a first algorithm unit and a second algorithm unit. The first algorithm subunit is used to select the first optimization algorithm as the target energy optimization scheduling algorithm if the state of charge in the next scheduling cycle is charging or discharging. The second algorithm subunit is used to determine the corresponding electricity price period for the next scheduling cycle if the state of charge in the next scheduling cycle is rechargeable / dischargeable. If the electricity price period is peak hour, the second optimization algorithm is selected as the target energy optimization scheduling algorithm; if the electricity price period is off-peak, the third optimization algorithm is selected as the target energy optimization scheduling algorithm. Specifically, a state of charge of less than or equal to 20% is considered to require charging (over-discharging); a state of charge greater than 20% and less than 80% is considered rechargeable / dischargeable; and a state of charge greater than 80% is considered to require discharging (overcharging).

[0053] When the battery pack's state of charge (SOC) is either in a state requiring charging or discharging, it is either over-discharged or over-charged. The first optimization algorithm is used as the target energy optimization scheduling algorithm to control the charging or discharging of the battery pack while considering both the economic and environmental costs of the energy storage grid. When the battery pack's SOC is either rechargeable or dischargeable, it is in a state that can both be charged and discharged. Based on the electricity price period corresponding to the next scheduling cycle, the second or third optimization algorithm is used as the target energy optimization scheduling algorithm to control the charging or discharging of the battery pack while considering both the economic and environmental costs of the energy storage grid.

[0054] Preferably, the system further includes an algorithm construction module; the algorithm construction module is used to: construct a first optimization algorithm, a second optimization algorithm, and a third optimization algorithm, each of which includes an objective function and constraints as follows: Objective function:

[0055]

[0056]

[0057] In the objective function of the first optimization algorithm

[0058] In the objective functions of the second and third optimization algorithms, +

[0059]

[0060]

[0061] In the second optimization algorithm,

[0062]

[0063] In the third optimization algorithm

[0064]

[0065] Constraints:

[0066]

[0067]

[0068]

[0069] in, To reduce the operation and maintenance costs of energy storage grids, The cost of charging and discharging battery packs. This is the maintenance cost coefficient for the battery pack. and These are the charging and discharging state variables of the battery pack during the scheduling cycle, with values ​​of 0 or 1, indicating whether the corresponding state has occurred. The charging power of the battery pack, This refers to the discharge power of the battery pack. The duration of a scheduling cycle, Depreciation costs for charging battery packs, For charging penalty function The depreciation cost of discharging the battery pack. Let be the discharge penalty function. This refers to the grid purchase price of electricity in the electricity price information. This refers to the electricity purchased by the microgrid from the external power grid. for For Sale quantity, This refers to carbon dioxide emissions. As a carbon dioxide emission factor, Let be the power of the electrical load at time t. Let t be the wind and solar power output at time t. Let be the tie-line power at time t. A positive value indicates that the microgrid purchases electricity from the external grid. A negative value indicates that the microgrid sells electricity to the external grid. Let be the charging and discharging power of the battery pack at time t. A positive value indicates that the battery pack is charging. A negative value indicates that the battery pack is discharging. In a charged state, At minimum state of charge, This is the maximum state of charge.

[0070] Thus, by introducing corresponding charge / discharge penalty functions into the second and third optimization algorithms, when the state of charge (SOC) is rechargeable and the electricity price is at its peak in the next scheduling cycle, the discharge penalty function is larger and the charging penalty function is smaller when the SOC is low, thereby suppressing battery discharge and encouraging battery charging; conversely, when the SOC is high, the discharge penalty function is lower and the charging penalty function is larger, thereby encouraging battery discharge and suppressing battery charging. This ensures that when the SOC is rechargeable and the electricity price is off-peak in the next scheduling cycle, under the same SOC conditions, the discharge penalty function is larger and the charging penalty function is smaller than during peak periods, guiding batteries to charge more and thus reducing the operation and maintenance costs of the energy storage grid.

[0071] In some embodiments, when the state of charge (SBC) of the next scheduling cycle is in the state requiring charging, the optimization module determines a compromise solution in the Pareto optimal solution set based on a first optimization algorithm; the control module determines the charging strategy of the battery pack in the next scheduling cycle based on this compromise solution and controls the charging of the battery pack. When the SBC of the next scheduling cycle is in the state requiring discharging, the optimization module determines a compromise solution in the Pareto optimal solution set based on the first optimization algorithm; the control module determines the discharging strategy of the battery pack in the next scheduling cycle based on this compromise solution and controls the discharging of the battery pack. In this embodiment, the Pareto optimal solution set can be solved using the Particle Swarm Optimization (PSO) algorithm. After determining the compromise solution, the charging / discharging amount or charging / discharging power (i.e., charging / discharging strategy) of the battery pack in the next scheduling cycle can be determined based on the energy storage grid operation and maintenance cost and carbon dioxide emissions corresponding to the compromise solution.

[0072] Understandably, when solving for the Pareto optimal solution set, it is necessary to normalize the operation and maintenance costs and carbon dioxide emissions of the energy storage grid separately in order to unify the dimensions of the two objectives.

[0073] Preferably, the battery pack includes multiple batteries; the optimization also includes a scheduling submodule; the scheduling submodule includes a detection unit and an update unit. The detection unit is used to detect the operating status of each battery in the battery pack and determine whether there are any abnormal batteries in the battery pack based on the operating status. The update unit is used to, if there are abnormal batteries in the battery pack, designate the remaining batteries (excluding the abnormal batteries) as a new battery pack, and based on the new battery pack, use a preset target energy optimization scheduling algorithm to determine the charging and discharging strategy of the new battery pack in the next scheduling cycle.

[0074] In this way, by monitoring the operation of each battery through the detection unit, abnormal batteries can be detected in a timely manner. Since abnormal batteries are no longer suitable for participating in the charging and discharging process of the energy storage grid, the remaining batteries other than the abnormal batteries are used as a new battery group, and the charging and discharging strategy of this new battery group is determined so that the charging and discharging of the new battery group can be controlled in the next scheduling cycle.

[0075] In some embodiments, the remaining capacity of each battery can be estimated using a battery management system (BMS). If the remaining capacity of a battery shows significant abnormalities during normal use, such as a sudden drop or failure to reach the expected depth of charge / discharge, it indicates that the battery is operating abnormally. Alternatively, the number of battery cycles can be recorded; when the maximum number of cycles is reached, it indicates that the battery is operating abnormally. Batteries with abnormal operating conditions are designated as abnormal batteries.

[0076] In other embodiments, when controlling the charging and discharging of the battery pack, batteries with fewer cycles can be prioritized to participate in the charging and discharging process based on the cycle count of each battery. This ensures that the cycle counts of each battery in the battery pack are similar, making it easier for maintenance personnel to replace the batteries in one go and reducing replacement costs.

[0077] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0078] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A big data-driven intelligent management system for energy storage power grids, characterized in that, include: The acquisition module is used to acquire historical information and carbon dioxide emission factors; the historical information includes historical wind and solar power output information, historical electricity load information, historical electricity price information, and historical state of charge of battery packs in the energy storage grid; The prediction module is used to make predictions based on the historical information to obtain the power load, wind and solar power output, electricity price information, and state of charge of the battery packs in the energy storage grid for the next scheduling cycle. The optimization module is used to solve the Pareto optimal solution set based on the carbon dioxide emission factor and the power load, wind and solar power output, electricity price information and the state of charge of the battery pack in the energy storage grid in the next scheduling cycle, with the optimization objectives being the lowest operation and maintenance cost and the lowest carbon dioxide emissions of the energy storage grid in the next scheduling cycle. It uses a preset target energy optimization scheduling algorithm to solve the Pareto optimal solution set and determine the compromise solution in the Pareto optimal solution set. The control module is used to determine the charging and discharging strategy of the battery pack in the next scheduling cycle based on the compromise solution, and to control the charging and discharging of the battery pack in the next scheduling cycle based on the charging and discharging strategy.

2. The system according to claim 1, characterized in that, The historical wind and solar power output information includes historical status information, historical environmental information, and historical power output of the wind and solar power generation equipment; the prediction module includes a wind and solar sub-module; the wind and solar sub-module includes: The wind and solar forecasting unit is used to predict the initial wind and solar power output and the corresponding typical power scenario of the wind and solar power generation equipment in the next scheduling cycle based on the historical state information, historical environmental information and historical power output. The wind and solar error unit is used to acquire wind and solar error information for a typical power scenario corresponding to the next scheduling cycle; wherein, the wind and solar error information represents the error information between the predicted initial wind and solar power output and the actual wind and solar power output. The wind and solar power adjustment unit is used to adjust the initial wind and solar power output based on the wind and solar error information of the next scheduling cycle, so as to obtain the wind and solar power output of the next scheduling cycle.

3. The system according to claim 2, characterized in that, The wind and solar error unit includes: The wind and solar determination subunit is used to acquire multiple error data for each typical power scenario and determine the wind and solar error information corresponding to each typical power scenario based on the multiple error data. The wind and light scene subunit is used to predict the environmental information of the next scheduling cycle and determine the corresponding typical power scene based on the environmental information of the next scheduling cycle. The wind and solar error subunit is used to determine the wind and solar error information corresponding to the next scheduling cycle based on the typical power scenario corresponding to the next scheduling cycle.

4. The system according to claim 1, characterized in that, The optimization module includes a selection sub-module; The selection submodule includes: Time period unit, used to determine the electricity price time period corresponding to the next scheduling cycle; The selection unit is used to select a preset energy optimization scheduling algorithm as the target energy scheduling optimization algorithm based on the state of charge and electricity price period of the next scheduling cycle.

5. The system according to claim 4, characterized in that, The preset energy optimization scheduling algorithm includes a first optimization algorithm, a second optimization algorithm, and a third optimization algorithm; the state of charge includes charging required, discharging required, and rechargeable / dischargeable; and the electricity price period includes peak and off-peak hours; the selection unit includes: The first algorithm subunit is used to use the first optimization algorithm as the target energy optimization scheduling algorithm if the state of charge in the next scheduling cycle is that it needs to be charged or discharged. The second algorithm subunit is used to determine the electricity price period corresponding to the next scheduling period if the state of charge of the next scheduling period is rechargeable and dischargeable. If the electricity price period is a peak time, the second optimization algorithm is used as the target energy optimization scheduling algorithm. If the electricity price period is an off-peak time, the third optimization algorithm is used as the target energy optimization scheduling algorithm.

6. The system according to claim 5, characterized in that, The system further includes an algorithm construction module; the algorithm construction module is used for: Construct a first optimization algorithm, a second optimization algorithm, and a third optimization algorithm. Each of these algorithms includes an objective function and the following constraints: Objective function: In the objective function of the first optimization algorithm In the objective functions of the second and third optimization algorithms, + In the second optimization algorithm, In the third optimization algorithm Constraints: in, To reduce the operation and maintenance costs of energy storage grids, The cost of charging and discharging battery packs. This is the maintenance cost coefficient for the battery pack. and These are the charging and discharging state variables of the battery pack during the scheduling cycle, with values ​​of 0 or 1, indicating whether the corresponding state has occurred. The charging power of the battery pack, This refers to the discharge power of the battery pack. The duration of a scheduling cycle, Depreciation costs for charging battery packs, For charging penalty function The depreciation cost of discharging the battery pack. Let be the discharge penalty function. This refers to the grid purchase price of electricity in the electricity price information. This refers to the electricity purchased by the microgrid from the external power grid. for For Sale quantity, This refers to carbon dioxide emissions. As a carbon dioxide emission factor, Let be the power of the electrical load at time t. Let t be the wind and solar power output at time t. Let be the tie-line power at time t. A positive value indicates that the microgrid purchases electricity from the external grid. A negative value indicates that the microgrid sells electricity to the external grid. Let be the charging and discharging power of the battery pack at time t. A positive value indicates that the battery pack is charging. A negative value indicates that the battery pack is discharging. In a charged state, At minimum state of charge, This is the maximum state of charge.

7. The system according to claim 1, characterized in that, The battery pack includes multiple batteries; the optimization module further includes a scheduling submodule; the scheduling submodule includes: The detection unit is used to detect the operating status of each battery in the battery pack and determine whether there are any abnormal batteries in the battery pack based on the operating status of each battery. The update unit is used to, if there is an abnormal battery in the battery pack, take the remaining batteries other than the abnormal battery as a new battery pack, and, based on the new battery pack, use a preset target energy optimization scheduling algorithm to determine the charging and discharging strategy of the new battery pack in the next scheduling cycle.

Images