A power distribution method for energy storage power station using digital twinning technology

Abstract

This invention belongs to the field of energy management technology for energy storage power stations, specifically relating to a power allocation method for energy storage power stations using digital twin technology. Addressing the shortcomings of existing power allocation methods for energy storage power stations that do not consider both battery lifespan and converter efficiency, this invention adopts the following technical solution: a power allocation method for energy storage power stations using digital twin technology, comprising: constructing a digital twin model; acquiring real-time operating data of the energy storage batteries to obtain the current state of each level of the energy storage power station; constructing a dynamic programming model for power allocation that comprehensively considers the SOC balance of each battery stack and the overall efficiency of the energy storage power station; determining the relationship between the grid's power demand and the maximum power that the energy storage power station can provide, and solving for the optimal allocation strategy in the dynamic programming model; and distributing the power allocation strategy to the battery management system from the cloud and executing it. The beneficial effects of this invention are: extending the lifespan of the energy storage power station batteries while simultaneously ensuring the operational efficiency of the energy storage power station.

Description

Technical Field

[0001] This invention belongs to the field of energy management technology for energy storage power stations, and specifically relates to a power allocation method for energy storage power stations using digital twin technology. Background Technology

[0002] Energy management of energy storage systems has a significant impact on the safe, stable, and efficient operation of energy storage power stations. Among these strategies, power allocation strategies for energy storage batteries have consistently been a research hotspot in the field of energy management for energy storage systems. Currently, most power allocation strategies among energy storage units are based on fixed parameters, which have poor adaptability to different operating conditions of energy storage power stations. Alternatively, the allocated power may not allow the energy storage units to operate within their optimal SOC range, affecting battery lifespan, or it may not consider converter efficiency, failing to ensure the converter operates within its high-efficiency range, resulting in energy loss and waste.

[0003] Chinese invention patent application CN110957780A discloses a power allocation method for energy storage batteries based on AGC (Automatic Generation Control), including the following steps: S1. Dividing batteries into Class I and Class II batteries according to the remaining SOC (State of Charge) reference value; S2. Determining the charge / discharge state of the batteries and performing charging and discharging processes; S3. In charging mode, prioritizing charging of Class I batteries; S4. In discharging mode, prioritizing discharging of Class II batteries. The aforementioned AGC-based power allocation method for energy storage batteries can reasonably allocate charge / discharge power among the batteries, achieving the goal of balancing the SOC and extending battery life. However, it still does not consider inverter efficiency and cannot guarantee that the inverter operates in its high-efficiency range.

[0004] Furthermore, in practical applications, obtaining accurate initial state parameters and limit constraints of the battery is the primary prerequisite for power allocation. Summary of the Invention

[0005] This invention addresses the shortcomings of existing power allocation methods for energy storage power stations, such as difficulty in accurately obtaining required battery parameters and failure to consider both battery lifespan and converter efficiency. It provides a power allocation method for energy storage power stations using digital twin technology. This method combines digital twin technology with power allocation, utilizing a cloud-based digital twin model of the energy storage battery system to provide precise initial state parameters and limit constraints for the power allocation algorithm, ensuring its accurate operation. Furthermore, by introducing a dynamic programming algorithm, the method comprehensively considers the SOC balance of each battery stack and the overall power loss of the energy storage power station to find the optimal power allocation strategy. This effectively improves the working life of the energy storage batteries and reduces the power loss of the energy storage system.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a power allocation method for an energy storage power station using digital twin technology, wherein the power allocation method for an energy storage power station using digital twin technology includes:

[0007] Step S1: Construct a digital twin model of the energy storage battery of the energy storage power station in the cloud;

[0008] Step S2: Obtain the operating data of the energy storage battery in real time, and obtain the state of charge, health status and power status of each level of the energy storage power station under the current state based on the digital twin model;

[0009] Step S3: Construct a dynamic programming model for power allocation that comprehensively considers the SOC balance of each battery stack and the overall efficiency of the energy storage power station;

[0010] Step S4: Determine the relationship between the power demand of the power grid and the maximum power that the energy storage power station can provide, solve the optimal allocation strategy of the dynamic programming model, and obtain the optimal power allocation strategy.

[0011] Step S5: The power allocation strategy is sent from the cloud to the battery management system and executed to complete the final allocation of power for each battery stack in the physical system of the energy storage power station.

[0012] Step S6: The running results are collected and processed by the battery management system and uploaded to the cloud. The cloud compares the actual consistency of the battery with the results of the digital twin model algorithm, and conducts fault diagnosis and aging analysis. At the same time, it updates the state of charge, health status and power status of the energy storage battery.

[0013] This invention presents a power allocation method for energy storage power stations using digital twin technology. It combines digital twin technology with power allocation in energy storage power stations, utilizing a digital twin model of the energy storage battery system built in the cloud to provide precise initial state parameters and limit constraints for the power allocation algorithm, ensuring its accurate and error-free operation. By introducing a dynamic programming algorithm in power allocation, it comprehensively considers the SOC balance of each battery stack and the overall power loss of the energy storage power station to find the optimal power allocation strategy, effectively improving the working life of the energy storage batteries and reducing the power loss of the energy storage system. Lithium-ion battery energy storage power stations can be divided into four levels: individual battery cells, modules, battery clusters, and battery compartments. Battery stacks typically refer to battery compartments, but can also be battery clusters.

[0014] As an improvement, step S1 includes:

[0015] Step S11: Collect the state characterization data of the operating energy storage battery, upload it to the cloud, and perform data cleaning and feature extraction;

[0016] Step S12: Using the previous offline battery test data, combined with battery characteristics and mechanisms, construct a general digital twin model of the battery cell;

[0017] Step S13: Using the historical operating data of the energy storage battery stored in the cloud, the general digital twin model of the battery cell is corrected based on the correction algorithm to obtain the corrected digital twin model of the battery cell.

[0018] Step S14: Combining the electrical structure of the energy storage battery system, perform multi-layer expansion on the single-unit digital twin model to obtain the digital twin model of the energy storage battery system.

[0019] As an improvement, in step S11, the data used to characterize the battery state specifically include: voltage, current, battery surface temperature, and ambient temperature.

[0020] As an improvement, in step S12, the general digital twin model of the battery cell is determined based on the previous offline battery test data and the battery's material system, specific model and application scenario to determine the battery characteristics, model structure and working environment.

[0021] As an improvement, in step S13, the general digital twin model is modified, specifically through machine learning, deep learning, and optimization algorithms.

[0022] As an improvement, in step S14, the establishment of the digital twin model of the energy storage battery system is based on the series-parallel group structure of the energy storage battery system, and the digital twin model of each individual cell is extended in multiple levels. Specifically, multi-level extension can be achieved through a feature-based individual cell strategy, which can better cope with scenarios with insufficient computing power or high real-time requirements. If the computing power can meet the requirements, other strategies can also be used for extension.

[0023] As an improvement, in step S2, the state of charge estimation, health state estimation, and power state estimation are specifically implemented through a state estimator, an error compensator, and machine learning.

[0024] As an improvement, step S3 includes:

[0025] Step S31: Construct the state variable x and decision variable u of the dynamic programming algorithm, using the remaining power to be allocated as the state variable and the power allocated to each battery stack as the decision variable.

[0026] Step S32: Construct the state transition equation, the mathematical expression of which is as follows:

[0027] x(k+1)=f(x(k),u(k))

[0028] In the formula, f is the state transition function, k indicates that we are currently in the kth stage of dynamic programming, that is, x(k) represents the remaining power to be allocated when allocating power to the kth battery stack, and u(k) represents the power allocated to the kth battery stack;

[0029] Step S33: Construct the cost function. When the battery stack discharges externally, the mathematical expression is as follows:

[0030]

[0031] The mathematical expression for charging the battery stack is as follows:

[0032]

[0033] In the formula, α is a weighting coefficient used to allocate the weights of SOC equalization and power loss in the cost function; soc(k) is the state of charge of the k-th battery stack; Δt represents the power allocation duration of the energy storage station, i.e., the charging and discharging duration of the battery stack; U is the charging and discharging voltage of the battery stack; Q is the rated capacity of each battery stack; and soc(k) represents the state of charge of the battery stack. ideal η represents the ideal SOC value. pcs The efficiency of the power storage converter PCS in the energy storage power station is related to the ratio of the power carried by the PCS to its rated power. The power carried by the PCS is the power allocated to each battery stack.

[0034] Step S34: Construct the optimal objective function, the mathematical expression of which is as follows:

[0035] J(x(k),u(k))=min(L(x(k),u(k))+J(x(k+1),u(k+1)))

[0036] Step S35: Construct constraints for state variables and decision variables, specifically as follows:

[0037]

[0038] In the formula, P min (k) represents the minimum allowable charge / discharge power for the k-th battery stack, P max (k) represents the maximum allowable charge / discharge power of the k-th battery stack, with the peak charge / discharge power derived from the power state estimation results of the digital twin model; soc min This refers to the lower limit of the state of charge (SOC) of the battery stack; SOC max This represents the upper limit of the SOC of the battery stack.

[0039] As an improvement, SOC ideal It is 0.5.

[0040] As an improvement, if the grid demand is greater than the maximum power that the energy storage station can provide, then each battery stack absorbs or releases energy at its maximum power, and this allocation strategy is considered the optimal allocation strategy. If the result is that the grid demand is less than the maximum power that the energy storage station can provide, then the optimal allocation strategy is solved using a dynamic programming model. The specific steps are as follows:

[0041] Step S41: Starting from the last stage n, calculate the optimal objective function for stage n and save the optimal decision for that stage. Its mathematical expression is as follows:

[0042] J(x(n),u(n))=min(L(x(n),u(n)))

[0043] Step S42: Solve the optimal objective function of each stage step by step in reverse and save the optimal decision of that stage. Use the state transition equation of step S32 to connect the state variables of two adjacent stages. The specific form of the state transition equation is as follows.

[0044] x(k+1)=u(k)-x(k);

[0045] Step S43: Using the known condition of the initial stage state variables, calculate the optimal state sequence and the optimal decision sequence in a forward manner, wherein the initial stage state variables are derived from the state-of-charge estimation results of the digital twin model.

[0046] The beneficial effects of the power allocation method for energy storage power stations using digital twin technology in this invention are as follows: A digital twin model of a single battery cell is constructed on the cloud-based big data platform of the energy storage power station. Based on the electrical structure of the energy storage battery system, the single cell model is extended to obtain a digital twin model of the battery system. This model is used to estimate the state of charge and power state of the energy storage battery in real time under different operating conditions and environmental conditions. Using this as initial state parameters and limit constraints, a dynamic programming power allocation model is built, providing accurate initial state parameters and limit constraints for the power allocation algorithm, thus ensuring the accurate operation of the algorithm. By comprehensively considering the SOC balance of each battery stack and the overall power loss of the energy storage power station, the optimal power allocation scheme for the energy storage power station is obtained, achieving optimal energy management for each battery stack. This extends the service life of the energy storage power station batteries to a certain extent while also considering the operating efficiency of the energy storage power station. Attached Figure Description

[0047] Figure 1 This is a flowchart of the power allocation method for an energy storage power station according to an embodiment of the present invention, including its architecture.

[0048] Figure 2 This is a flowchart of the power allocation method for an energy storage power station according to an embodiment of the present invention.

[0049] Figure 3 This is a flowchart of the dynamic programming algorithm for the power allocation method of an energy storage power station according to an embodiment of the present invention.

[0050] Figure 4 This is a comparison chart of the SOC equalization results obtained from simulations of the energy storage power station power allocation method of this invention and the traditional power allocation method.

[0051] Figure 5 This is a comparison chart of the overall efficiency of the energy storage power station power allocation method of this invention and the traditional power allocation method. Detailed Implementation

[0052] The technical solutions of the embodiments of the present invention will be explained and described below with reference to the accompanying drawings. However, the following embodiments are only preferred embodiments of the present invention and not all embodiments. Other embodiments obtained by those skilled in the art based on the embodiments in the implementation methods without creative effort are all within the protection scope of the present invention.

[0053] Example 1

[0054] See Figures 1 to 5 The present invention provides a power allocation method for an energy storage power station using digital twin technology, comprising the following steps:

[0055] Step S1: Construct a digital twin model of the energy storage battery of the energy storage power station in the cloud;

[0056] Step S2: Obtain the operating data of the energy storage battery in real time, and obtain the state of charge, health status and power status of each level of the energy storage power station under the current state based on the digital twin model;

[0057] Step S3: Construct a dynamic programming model for power allocation that comprehensively considers the SOC balance of each battery stack and the overall efficiency of the energy storage power station;

[0058] Step S4: Determine the relationship between the power demand of the power grid and the maximum power that the energy storage power station can provide, solve the optimal allocation strategy of the dynamic programming model, and obtain the optimal power allocation strategy.

[0059] Step S5: The power allocation strategy is sent from the cloud to the battery management system and executed to complete the final allocation of power for each battery stack in the physical system of the energy storage power station.

[0060] Step S6: The running results are collected and processed by the battery management system and uploaded to the cloud. The cloud compares the actual consistency of the battery with the results of the digital twin model algorithm, and conducts fault diagnosis and aging analysis. At the same time, it updates the state of charge, health status and power status of the energy storage battery.

[0061] In this embodiment, step S1 includes:

[0062] Step S11: Collect the state characterization data of the operating energy storage battery, upload it to the cloud, and perform data cleaning and feature extraction;

[0063] Step S12: Using the previous offline battery test data, combined with battery characteristics and mechanisms, construct a general digital twin model of the battery cell;

[0064] Step S13: Using the historical operating data of the energy storage battery stored in the cloud, the general digital twin model of the battery cell is corrected based on the correction algorithm to obtain the corrected digital twin model of the battery cell.

[0065] Step S14: Combining the electrical structure of the energy storage battery system, perform multi-layer expansion on the single-unit digital twin model to obtain the digital twin model of the energy storage battery system.

[0066] In this embodiment, the data used to characterize the battery state in step S11 specifically includes: voltage, current, battery surface temperature, and ambient temperature.

[0067] In this embodiment, in step S12, the general digital twin model of the battery cell is established based on considerations from the perspectives of electricity, electrochemistry, and thermodynamics. The general digital twin model of the battery cell is based on the previous offline battery test data and the battery's material system, specific model and application scenario to determine the battery characteristics, model structure and working environment.

[0068] In this embodiment, in step S13, the general digital twin model is modified, specifically through machine learning, deep learning, and optimization algorithms.

[0069] In this embodiment, in step S14, the establishment of the digital twin model of the energy storage battery system is based on the series-parallel group structure of the energy storage battery system, and is achieved by multi-level extension of the digital twin model of each individual unit through a feature-based individual unit strategy. The feature-based individual unit strategy can achieve good real-time performance with relatively low computing power.

[0070] In this embodiment, in step S2, the state of charge estimation, health state estimation, and power state estimation are specifically implemented through a state estimator, an error compensator, and machine learning.

[0071] In this embodiment, step S3 includes:

[0072] Step S31: Construct the state variable x and decision variable u of the dynamic programming algorithm, using the remaining power to be allocated as the state variable and the power allocated to each battery stack as the decision variable.

[0073] Step S32: Construct the state transition equation, the mathematical expression of which is as follows:

[0074] x(k+1)=f(x(k),u(k))

[0075] In the formula, f is the state transition function, k indicates that we are currently in the kth stage of dynamic programming, that is, x(k) represents the remaining power to be allocated when allocating power to the kth battery stack, and u(k) represents the power allocated to the kth battery stack;

[0076] Step S33: Construct the cost function. When the battery stack discharges externally, the mathematical expression is as follows:

[0077]

[0078] The mathematical expression for charging the battery stack is as follows:

[0079]

[0080] In the formula, α is a weighting coefficient used to allocate the weights of SOC equalization and power loss in the cost function; soc(k) is the state of charge of the k-th battery stack; Δt represents the power allocation duration of the energy storage station, i.e., the charging and discharging duration of the battery stack; U is the charging and discharging voltage of the battery stack; Q is the rated capacity of each battery stack; and soc(k) represents the state of charge of the battery stack. ideal η represents the ideal SOC value. pcs The efficiency of the power storage converter (PCS) in an energy storage power station is related to the ratio of the power handled by the PCS to its rated power. The power handled by the PCS is the allocated power of each battery stack. Its mathematical expression is as follows:

[0081]

[0082] In the formula P N Let be the rated power of the energy storage converter, and a and b be coefficients. This formula is a piecewise function, and the segmentation and the values ​​of a and b in different segment intervals are shown in Table 1.

[0083] Table 1

[0084] Step S34: Construct the optimal objective function, the mathematical expression of which is as follows:

[0085] J(x(k),u(k))=min(L(x(k),u(k))+J(x(k+1),u(k+1)))

[0086] Step S35: Construct constraints for state variables and decision variables, specifically as follows:

[0087]

[0088] In the formula, P min (k) represents the minimum allowable charge / discharge power for the k-th battery stack, P max(k) represents the maximum allowable charge / discharge power of the k-th battery stack, with the peak charge / discharge power derived from the power state estimation results of the digital twin model; soc min This refers to the lower limit of the state of charge (SOC) of the battery stack; SOC max This represents the upper limit of the SOC of the battery stack.

[0089] In this embodiment, soc ideal It is 0.5.

[0090] In this embodiment, if the power demand of the power grid is greater than the maximum power that the energy storage station can provide, then each battery stack absorbs or releases electrical energy at its own maximum power, and this allocation strategy is considered the optimal allocation strategy; if the determination result is that the power demand of the power grid is less than the maximum power that the energy storage station can provide, then the optimal allocation strategy is solved using a dynamic programming model, and the specific steps are as follows:

[0091] Step S41: Starting from the last stage n, calculate the optimal objective function for stage n and save the optimal decision for that stage. Its mathematical expression is as follows:

[0092] J(x(n),u(n))=min(L(x(n),u(n)))

[0093] Step S42: Solve the optimal objective function of each stage step by step in reverse and save the optimal decision of that stage. Use the state transition equation of step S32 to connect the state variables of two adjacent stages. The specific form of the state transition equation is as follows.

[0094] x(k+1)=u(k)-x(k);

[0095] Step S43: Using the known condition of the initial stage state variables, calculate the optimal state sequence and the optimal decision sequence in a forward manner, wherein the initial stage state variables are derived from the state-of-charge estimation results of the digital twin model.

[0096] To verify the feasibility of the power allocation method proposed in this application and its superiority over traditional power allocation methods, simulation analysis of power allocation for four battery stacks was conducted. The SOC equalization results of the power allocation method in this embodiment are compared with those of traditional power allocation methods. Figure 4 As shown. A comparison of the overall efficiency of the energy storage converter in this embodiment and the traditional power distribution method is shown below. Figure 5 As shown.

[0097] In the simulation analysis, the parameter α of the cost function in the dynamic programming algorithm was set to 0.5, and the other basic parameters of the battery stack are shown in Table 2. The traditional power allocation method used for comparison refers to the traditional iterative formula method that only aims at SOC equalization and does not combine power allocation with digital twins or consider the operating efficiency of the energy storage converter.

[0098] Table 2

[0099] Depend on Figure 4 As shown in the SOC equalization comparison chart, with the increase of power allocation times, the power allocation method in this embodiment makes the SOC of each battery stack continuously tend to be consistent and as close as possible to the SOC. ideal This method achieves the goal of SOC equalization. Furthermore, its equalization speed and effectiveness are superior to traditional methods. Because the method described in this patent incorporates digital twin technology, it possesses a precise digital twin model of the energy storage battery system. During each power allocation, it can obtain extremely accurate key parameters such as the battery stack's state of charge and power state from this model, resulting in faster and more effective SOC equalization.

[0100] It should also be noted that although the initial state parameters and limit constraints used in this embodiment are the same as those in the traditional method in this simulation analysis, the role of the digital twin model is to provide accurate parameters such as state of charge and power state. This allows for a more precise calculation of how much power to allocate to each battery stack to achieve SOC equalization and maximum efficiency in each allocation. In contrast, the traditional iterative formula method can be considered to have some error in the initial conditions of each allocation, resulting in a worse actual power allocation effect and a slower achievement of equalization. Therefore, considering parameter accuracy, the method in this embodiment has greater advantages over the traditional method.

[0101] Depend on Figure 5 As can be seen from the comparison chart of the total efficiency of the energy storage converter in the energy storage power station, compared with the traditional power allocation method of the energy storage power station that does not consider the efficiency problem, the method described in this patent takes the efficiency of the converter into account in the dynamic programming cost function, thereby achieving higher power allocation efficiency while completing SOC balancing faster.

[0102] The beneficial effects of the power allocation method for energy storage power stations using digital twin technology in Embodiment 1 of this invention are as follows: First, a digital twin model of a battery cell is constructed on the cloud-based big data platform of the energy storage power station. Based on the electrical structure of the energy storage battery system, the cell model is extended to obtain a digital twin model of the battery system. This model is used to estimate the state of charge and power state of the energy storage battery in real time under different operating conditions and environmental conditions. Using this as the initial state parameters and limit constraints, a dynamic programming power allocation model is built, providing accurate initial state parameters and limit constraints for the power allocation algorithm, thus providing a prerequisite guarantee for the accurate operation of the power allocation algorithm. In power allocation, by introducing a dynamic programming algorithm, the SOC balance of each battery stack and the overall power loss of the energy storage power station are comprehensively considered to find the optimal power allocation strategy, effectively improving the working life of the energy storage battery and reducing the power loss of the energy storage system. In other words, it achieves optimal energy management of each battery stack, extending the service life of the energy storage power station batteries to a certain extent, while also taking into account the operating efficiency of the energy storage power station.

[0103] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Those skilled in the art should understand that the present invention includes, but is not limited to, the content described in the above specific embodiments. Any modifications that do not depart from the functional and structural principles of the present invention will be included within the scope of the claims.

Claims

1. A power allocation method for an energy storage power station using digital twin technology, characterized in that: The power allocation method for energy storage power stations using digital twin technology includes: Step S1: Construct a digital twin model of the energy storage battery of the energy storage power station in the cloud; Step S2: Obtain the operating data of the energy storage battery in real time, and obtain the state of charge, health status and power status of each level of the energy storage power station under the current state based on the digital twin model; Step S3: Construct a dynamic programming model for power allocation that comprehensively considers the SOC balance of each battery stack and the overall efficiency of the energy storage power station; Step S4: Determine the relationship between the power demand of the power grid and the maximum power that the energy storage power station can provide, solve the optimal allocation strategy of the dynamic programming model, and obtain the optimal power allocation strategy. Step S5: The power allocation strategy is sent from the cloud to the battery management system and executed to complete the final allocation of power for each battery stack in the physical system of the energy storage power station. Step S6: The running results are collected and processed by the battery management system and uploaded to the cloud. The cloud compares the actual consistency of the battery with the results of the digital twin model algorithm, and conducts fault diagnosis and aging analysis. At the same time, it updates the state of charge, health status and power status of the energy storage battery. Step S1 includes: Step S11: Collect the state characterization data of the operating energy storage battery, upload it to the cloud, and perform data cleaning and feature extraction; Step S12: Using the offline battery test data from the previous stage, combined with the battery characteristics and mechanisms, construct a general digital twin model of the battery cell; Step S13: Using the historical operating data of the energy storage battery stored in the cloud, the general digital twin model of the battery cell is corrected based on the correction algorithm to obtain the corrected digital twin model of the battery cell. Step S14: Combine the electrical structure of the energy storage battery system and perform multi-layer expansion on the single-cell digital twin model to obtain the digital twin of the energy storage battery system. Step S3 includes: Step S31, constructing state variables of dynamic programming algorithm x and decision variables u， Taking the remaining power to be allocated as the state variable and the allocated power of each battery stack as the decision variable Step S32: Construct the state transition equation, the mathematical expression of which is as follows: In the formula, f This is the state transition function. k This indicates that we are currently in the first stage of dynamic programming. k Stage, i.e. x ( k ) indicates that in giving the first k The remaining power to be allocated when allocating power to a battery stack. u ( k ) indicates to give the first k The power allocated to each battery stack; Step S33: Construct the cost function. When the battery stack discharges externally, the mathematical expression is as follows: The mathematical expression for charging the battery stack is as follows: In the formula, α This is a tradeoff coefficient used to allocate the weights of SOC equalization and power loss in the cost function; soc ( k ) is the first k The state of charge of a battery stack; t This indicates the duration of power distribution at the energy storage power station, i.e., the duration of battery stack charging and discharging. U The charging and discharging voltage of the battery stack. Q The rated capacity for each battery stack, This represents the ideal SOC value. This indicates the efficiency of the power storage converter PCS in an energy storage power station. Its value is related to the ratio of the power handled by the PCS to its rated power. Step S34: Construct the optimal objective function, the mathematical expression of which is as follows: Step S35: Construct constraints for state variables and decision variables, specifically as follows: In the formula, For the first k The minimum allowable charge and discharge power for a single battery stack. For the first k The maximum allowable charge and discharge power of each battery stack, with the peak charge and discharge power derived from the power state estimation results of the digital twin model; This is the lower limit of the SOC of the battery stack; This represents the upper limit of the SOC of the battery stack; In step S4, if the grid demand is greater than the maximum power that the energy storage station can provide, each battery stack absorbs or releases energy at its maximum power, and this allocation strategy is considered the optimal allocation strategy. If the result is that the grid demand is less than the maximum power that the energy storage station can provide, then the optimal allocation strategy is solved using a dynamic programming model. The specific steps are as follows: Step S41, from the last stage n Start, calculate the first n The optimal objective function of the stage and save the record of the optimal decision of the stage, the mathematical expression is as follows: Step S42: Solve the optimal objective function of each stage step by step in reverse and save the optimal decision of that stage. Use the state transition equation of step S32 to connect the state variables of two adjacent stages. The specific form of the state transition equation is as follows. ； Step S43: Using the known condition of the initial stage state variables, calculate the optimal state sequence and the optimal decision sequence in a forward manner, wherein the initial stage state variables are derived from the state-of-charge estimation results of the digital twin model.

2. The energy storage power station power distribution method using digital twin technology according to claim 1, characterized in that: In step S11, the data used to characterize the battery state specifically include: voltage, current, battery surface temperature, and ambient temperature.

3. The energy storage power station power distribution method using digital twin technology according to claim 1, characterized in that: In step S12, the general digital twin model of the battery cell determines the battery characteristics, model structure and working environment based on the previous offline battery test data and the battery's material system, specific model and application scenario.

4. The energy storage power station power distribution method using digital twin technology according to claim 1, characterized in that: In step S13, the general digital twin model is modified, specifically through machine learning, deep learning, and optimization algorithms.

5. The energy storage power station power distribution method using digital twin technology according to claim 1, characterized in that: In step S14, the digital twin model of the energy storage battery system is established by extending the digital twin model of each individual cell through multiple levels, based on the series-parallel group structure of the energy storage battery system.

6. The power allocation method for an energy storage power station using digital twin technology according to claim 1, characterized in that: In step S2, the state of charge estimation, health state estimation, and power state estimation are specifically implemented through a state estimator, an error compensator, and machine learning.

7. The power allocation method for an energy storage power station using digital twin technology according to claim 1, characterized in that: It is 0.5.

Images