A shared scheduling management system and method for distributed energy storage devices

Abstract

This invention specifically relates to a shared scheduling management system and method for distributed energy storage devices, belonging to the field of distributed energy storage scheduling technology. It includes: scheduling optimization based on a hierarchical collaborative scheduling framework. The framework includes: selecting energy storage units that meet ownership constraints and clustering them according to spatial location and capacity level to form virtual energy storage clusters; using the virtual energy storage clusters as scheduling units to solve a multi-objective optimization model to generate a global scheduling strategy. In this invention, the ownership constraints of each energy storage unit are encoded and stored on the blockchain in the form of smart contracts. Before scheduling instructions are issued, they undergo item-by-item compliance verification by the smart contracts. Ownership constraints are transformed from passively followed protocols into automatically executed mandatory rules, preventing unauthorized scheduling from the source and solving the problem of difficulty in real-time verification and enforcement of scheduling compliance in multi-owner scenarios.

Description

Technical Field

[0001] This invention relates to the field of distributed energy storage scheduling technology, and in particular to a shared scheduling management system and method for distributed energy storage devices. Background Technology

[0002] With the large-scale integration of distributed energy storage resources into power distribution networks, the shared scheduling of energy storage resources in multi-owner scenarios has gradually become an important issue in power system operation. Existing scheduling methods typically treat energy storage resources as arbitrarily allocated public resources, or treat ownership factors merely as a penalty item in the optimization objective. This approach lacks the technical guarantee of mandatory enforcement of ownership constraints, making it difficult to complete real-time compliance verification before the issuance of scheduling instructions.

[0003] In terms of scheduling architecture, the traditional centralized optimization scheduling mode requires the status information of all energy storage units to be aggregated to the central node for unified solution. Distributed energy storage units are spatially dispersed across different transformer areas and feeders, and temporally constrained by the personalized preferences of their owners. The centralized mode is prone to disconnect between scheduling instructions and actual operating status under communication delay conditions, making it difficult to balance global economy and local autonomy, and suffers from insufficient scalability in large-scale distributed scenarios.

[0004] In terms of revenue distribution, the revenue calculation and distribution process of existing shared energy storage systems relies on the unilateral pricing of centralized operating platforms. The scheduling behavior lacks on-chain traceability records, and the revenue distribution process is not transparent to each owner, which restricts the willingness of each owner to participate and hinders the sustainable operation of shared energy storage systems. Summary of the Invention

[0005] The purpose of this invention is to provide a shared scheduling management system and method for distributed energy storage devices in order to solve the above-mentioned problems.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: A shared scheduling and management method for distributed energy storage devices includes: The operating parameters and property rights attribute data of each distributed energy storage unit are collected, and the property rights constraint set is embedded as the core attribute to generate a multi-dimensional state vector of each energy storage unit. The property rights constraints of each owner are encoded into smart contracts and deployed on the consortium blockchain. Before the scheduling instructions are executed, they must be verified for compliance by the smart contracts. Scheduling optimization is performed based on a hierarchical collaborative scheduling framework, which includes: selecting energy storage units that meet ownership constraints and clustering them into virtual energy storage clusters according to spatial location and capacity level; using the virtual energy storage clusters as scheduling units to solve a multi-objective optimization model to generate a global scheduling strategy; and decomposing the global scheduling strategy to each virtual energy storage cluster so that the local scheduling module can autonomously complete the task allocation within the cluster. In the multi-objective optimization model, the property rights constraint set of each energy storage unit is treated as a hard boundary of inequality constraints, and the scheduling strategy is only searched within the feasible domain defined by the property rights boundary.

[0007] Preferably, the multidimensional state vector includes a timestamp, a unique device identifier, a state of charge, a health status, a set of property constraints, spatial coordinates, a distribution network access node number, an upper limit of available power, a lower limit of available power, and an electricity price sensitivity coefficient; wherein, the set of property constraints includes the allowable charging and discharging time interval, power limit, minimum state of charge retention threshold, and lifetime depreciation tolerance coefficient set by the owner; based on the access node information of each energy storage unit, the electrical distance and communication delay between any two energy storage units are calculated to generate a spatial distribution topology map of the energy storage units.

[0008] Preferably, the step of encoding the property rights constraints of each owner into a smart contract and deploying it on the consortium blockchain includes: creating a property rights digital certificate for each distributed energy storage unit on the consortium blockchain, writing the basic attribute information of the device, the owner's identity information, the ownership of the device, the charging and discharging permission range, and the validity period of the scheduling authorization into the property rights digital certificate, and realizing the on-chain storage of property rights information through the hash chain of the blockchain; converting the scheduling constraints set by each owner into executable rules written in the smart contract language and deploying them on the consortium blockchain; the compliance verification is the process by which the smart contract compares the charging and discharging period, power value, and expected change in state of charge in the scheduling instruction with the corresponding constraints in the property rights constraint set one by one, and if any condition is not met, the scheduling instruction is automatically intercepted and an alarm is triggered.

[0009] Preferably, a configurable bidding participation strategy is embedded in the smart contract. Each energy storage unit owner sets an opportunity cost threshold for participating in scheduling and expresses their willingness to participate in charging and discharging to the scheduling platform in the form of a dynamic bidding function. The scheduling platform's call request is only allowed to be executed by the smart contract when the bid is not lower than the opportunity cost threshold. Under the premise of meeting the grid scheduling needs, the scheduling platform selects and calls energy storage units in order of their bids from low to high. Historical scheduling participation records, instruction execution fulfillment rates, and violation records are recorded for each energy storage unit owner on the blockchain to generate a reputation score, which is used as a weighting factor in scheduling selection and revenue distribution.

[0010] Preferably, the objective function of the multi-objective optimization model includes five components: grid supply cost, Gini coefficient of each owner's revenue, renewable energy curtailment, carbon emission cost, and variance of energy storage unit lifetime loss; wherein, the grid supply cost is the cumulative sum of the product of the purchased electricity and the corresponding electricity price in each time period within the scheduling cycle; the Gini coefficient is calculated based on the cumulative revenue of each owner at the end of the scheduling cycle; the renewable energy curtailment is the cumulative sum of the curtailment in each time period within the scheduling cycle; the carbon emission cost is the cumulative sum of the product of the charging amount and the corresponding carbon emission factor in each time period within the scheduling cycle; and the variance of lifetime loss is obtained by converting the cumulative charge-discharge cycles of each energy storage unit within the scheduling cycle.

[0011] Preferably, the multi-objective optimization model is solved using an improved multi-objective particle swarm optimization algorithm. The solution process includes: initializing the particle swarm, setting the position of each particle within the dispatchable power range of each virtual energy storage cluster, performing compliance checks on the initial positions against the hard property rights constraint boundary, and eliminating initial particles that do not meet the constraint conditions; calculating the candidate positions of each particle according to the velocity update formula; performing hard property rights constraint boundary checks on the candidate positions of each particle, accepting the candidate position directly if it meets all property rights constraint conditions, and projecting the position of the particle along the dimension of the violation constraint to the nearest feasible domain boundary if the candidate position violates any property rights constraint condition; calculating the objective function vector corresponding to the current position of each particle, updating the individual optimal position and the Pareto archive set; repeating the iteration until a preset number of iterations is reached, and outputting a Pareto optimal scheduling strategy set; from the Pareto optimal scheduling strategy set, using a scheme selection method based on ideal point distance, calculating the weighted distance between the objective function vector corresponding to each non-dominated solution and the ideal point, and selecting the scheme with the smallest weighted distance as the global scheduling strategy, wherein the ideal point is composed of the minimum value of each objective component in the Pareto archive set.

[0012] Preferably, the local scheduling module autonomously completes the task allocation within the cluster by: receiving the net power target issued by the scheduling platform, and using a consensus algorithm to perform secondary allocation of the net power target. The input of the consensus algorithm is the real-time state of charge of each energy storage unit in the virtual energy storage cluster, the local preference parameters set by the owner, and the net power target. The output is the charging and discharging power share undertaken by each energy storage unit in that time period. Each energy storage unit in the virtual energy storage cluster autonomously negotiates its own charging and discharging power share through information exchange between neighboring nodes. When an energy storage unit in the virtual energy storage cluster exits the scheduling due to a sudden anomaly, the local scheduling module autonomously reallocates the power share undertaken by that energy storage unit to the remaining available units in the virtual energy storage cluster.

[0013] Preferably, the method further includes: Once the dispatch instructions have passed compliance verification, they are sent to the field controllers of the energy storage units for execution. Real-time operational data on charging and discharging power and state of charge changes are collected and compared point-by-point with the target values ​​of the dispatch instructions. When the deviation between the actual executed power and the target value of the dispatch instructions exceeds a preset tolerance threshold, a tiered alarm is triggered. Specifically, when the relative deviation is between the preset tolerance threshold and the preset violation threshold, the record is identified as an execution anomaly and included in the reputation score update process for that energy storage unit. When the relative deviation exceeds the preset violation threshold, the specific violation behavior of the energy storage unit is identified and identified as a violation record, and the subsequent dispatch priority of that energy storage unit is downgraded. After each verified dispatch instruction is executed, the complete content of the dispatch instruction is written to the consortium blockchain for evidence storage.

[0014] Preferably, the method further includes: Based on the actual execution data of scheduling instructions, the revenue from electricity cost savings, demand response compensation, and carbon emission reduction is calculated and aggregated to obtain the total revenue. A contribution score is calculated for each energy storage unit, which is obtained by weighting the normalized data based on actual discharge volume, response speed, response accuracy, equipment utilization rate, and reputation score. The revenue share of each owner is the total revenue multiplied by the ratio of that owner's contribution score to the sum of the contribution scores of all participating energy storage units. The calculation rules and allocation algorithm are encoded into a smart contract, which automatically reads the scheduling instruction execution records on the chain and completes the revenue transfer when the preset settlement period arrives.

[0015] A shared scheduling and management system for distributed energy storage devices includes: The data acquisition and fusion module is used to collect the operating parameters and property attribute data of each distributed energy storage unit, embed the property constraint set as the core attribute, and generate a multi-dimensional state vector of each energy storage unit. The property rights constraint on-chain module is used to encode the property rights constraints of each owner into smart contracts and deploy them on the consortium blockchain to perform compliance verification of scheduling instructions; The hierarchical collaborative scheduling module is used to screen energy storage units that meet the property rights constraints and cluster them into virtual energy storage clusters. The virtual energy storage clusters are used as scheduling units to solve multi-objective optimization models to generate a global scheduling strategy. The global scheduling strategy is decomposed into each virtual energy storage cluster and the local scheduling module autonomously completes the task allocation within the cluster. The property rights constraint set of each energy storage unit is treated as a hard boundary of inequality constraints. The execution monitoring and processing module is used to send dispatch instructions that have passed compliance verification to the energy storage unit for execution, monitor the actual execution status, and classify and process execution deviations.

[0016] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are: 1. This invention encodes and stores the ownership constraints of each energy storage unit in the form of blockchain smart contracts. Before the scheduling instructions are issued, they undergo item-by-item compliance verification by the smart contracts. The ownership constraints are transformed from passively followed protocols into automatically executed mandatory rules, which avoids the occurrence of unauthorized scheduling from the source and solves the problem of difficulty in real-time verification and enforcement of scheduling compliance in multi-owner scenarios.

[0017] 2. This invention adopts a hierarchical collaborative scheduling framework. At the global level, multi-objective optimization is performed using virtual energy storage clusters as units. At the local level, each energy storage unit within the virtual energy storage cluster autonomously completes task allocation through a consensus algorithm. A reasonable division of labor interface is formed between global scheduling and local autonomy. Under the constraints of communication latency and computational overhead, the continuity of scheduling can still be maintained, overcoming the lack of scalability of traditional centralized optimization mode in large-scale distributed scenarios. Attached Figure Description

[0018] Further details, features, and advantages of this application are disclosed in the following description of exemplary embodiments in conjunction with the accompanying drawings, in which: Figure 1 This is a flowchart of the method of the present invention. Detailed Implementation

[0019] Several embodiments of this application will now be described in more detail with reference to the accompanying drawings to enable those skilled in the art to implement this application. This application may be embodied in many different forms and for various purposes and should not be limited to the embodiments set forth herein. These embodiments are provided to make this application thorough and complete, and to fully convey the scope of this application to those skilled in the art. The embodiments described do not limit this application.

[0020] Unless otherwise defined, all terms used herein (including technical and scientific terms) shall have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains. It will be further understood that terms such as those defined in commonly used dictionaries shall be interpreted as having a meaning consistent with their meaning in the relevant field and / or the context of this specification, and shall not be interpreted in an idealized or overly formal sense unless expressly defined herein.

[0021] Example 1

[0022] Its specific implementation method is combined with the appendix Figure 1 Please provide a detailed explanation.

[0023] In this embodiment, it includes: With the large-scale integration of distributed energy storage resources into power distribution networks, the shared scheduling of energy storage resources in multi-owner scenarios faces the following technical challenges: Each energy storage unit belongs to a different individual, enterprise, or property management company, with varying ownership boundaries. Existing scheduling methods often treat energy storage resources as arbitrarily allocated public resources, or merely treat ownership factors as a penalty in the optimization objective, lacking a technical guarantee for the enforcement of ownership constraints. Simultaneously, distributed energy storage units are spatially dispersed across different transformer areas and feeders, and temporally constrained by the individual preferences of their owners. Traditional centralized optimization scheduling models struggle to balance overall economic efficiency with local autonomy, easily leading to disconnected scheduling instructions due to communication delays. Furthermore, the lack of transparency in revenue distribution and the untraceability of scheduling behavior reduce the willingness of each owner to participate, hindering the sustainable operation of shared energy storage systems.

[0024] A distributed energy storage resource sharing scheduling method according to an embodiment of the present invention includes the following steps: Step 1: Collect multi-source heterogeneous data from each distributed energy storage unit and generate a multi-dimensional state vector for each energy storage unit.

[0025] Smart terminals are deployed in each distributed energy storage unit to collect real-time operating parameters of the energy storage devices via IoT communication protocols. These parameters include state of charge (SBC), health status, charging and discharging power, voltage, current, and temperature. Simultaneously, the ownership attribute data of each energy storage unit is obtained through the energy storage device registration and management system. This data includes owner identity information, usage permission level, charging and discharging time preferences, power upper and lower limits, minimum SBC retention threshold, and lifetime depreciation tolerance coefficient. Furthermore, the geographical coordinates of each energy storage unit and its access node information in the distribution network are collected, including the transformer substation number, feeder number, and transformer node. Time-of-use electricity price data, demand response event notifications, and carbon emission factor data are also integrated into the data.

[0026] The aforementioned operating parameters, property ownership data, spatial location information, and grid-side data are correlated and fused within a unified spatiotemporal coordinate system to generate a multidimensional state vector for each energy storage unit. The multidimensional state vector is represented as follows: ; in, This refers to the serial number of the energy storage unit; For timestamps; A unique identifier for the device; It is in a charged state; In good health; It is a set of property rights constraints, including constraints set by the owner such as charging and discharging periods, upper and lower power limits, and minimum state of charge; Spatial coordinates; Numbering of distribution network access nodes; and These are the upper and lower limits of available power, respectively. This is the electricity price sensitivity coefficient.

[0027] It should be noted that the set of property rights constraints This approach embeds multi-owner property rights constraints and spatial topology information as the first core attribute into a multi-dimensional state vector, differing from existing methods that only include physical operational parameters in their state representation. (Property rights constraint set) In its specific form, it is a set of ordered constraints, represented as... ,in For the permitted charging and discharging time range, For power limits, A threshold is reserved for the lowest state of charge. This is the life depreciation tolerance factor.

[0028] In this embodiment, to maintain the real-time status information of each energy storage unit, the multi-dimensional state vector is updated at a preset period, providing a complete and real-time decision-making information basis for subsequent scheduling optimization. Furthermore, based on the access node information of each energy storage unit, the electrical distance and communication delay between any two energy storage units are calculated, generating a spatial distribution topology map of the energy storage units, providing a spatial constraint basis for proximity matching in subsequent scheduling.

[0029] Step 2: Encode the property rights constraints of each owner into a blockchain smart contract and deploy it on the consortium blockchain.

[0030] For the multidimensional state vectors of each energy storage unit generated in step 1, step 2 can be further performed. Step 2 may specifically include: Step 201: Create and store digital ownership certificates for each energy storage unit on the consortium blockchain.

[0031] A digital certificate of ownership is created for each distributed energy storage unit on the consortium blockchain. This certificate contains the device's basic attributes, owner information, ownership details, charging / discharging permissions, and scheduling authorization validity period. The blockchain's hash chain ensures that the ownership information is stored on-chain and protected against tampering. The distributed ledger nature of the blockchain guarantees that the ownership information cannot be unilaterally altered.

[0032] Step 202: Transform the scheduling constraints of each owner into executable smart contracts and deploy them.

[0033] The scheduling constraints set by each owner in Step 1 are converted into executable rules written in smart contract language and deployed on the consortium blockchain. These constraints include allowed charging / discharging periods, upper and lower power limits, minimum state of charge threshold, and lifetime depreciation tolerance. When the scheduling platform issues charging / discharging commands to the energy storage unit, the commands must undergo compliance verification by the smart contract before execution. If parameters such as power value, time period, and state of charge change in the charging / discharging command conflict with the constraints in the smart contract, the command will be automatically intercepted and an alarm will be triggered.

[0034] It should be noted that the aforementioned compliance verification refers to the smart contract comparing the parameters in the charge / discharge instructions with the set of property rights constraints. The process involves comparing and judging the corresponding constraints one by one. When the charging / discharging time segment in the charging / discharging command is entered... Within the range, power value not exceeding The expected state of charge after the change in state of charge is not lower than If the condition is met, the verification passes; otherwise, the verification fails. The programmability and automatic execution of smart contracts transform property rights constraints from passively followed protocols into actively executed automated rules, solving the problem of real-time verification of scheduling compliance in multi-owner scenarios.

[0035] In this embodiment, to transform energy storage resources from passively dispatchable to autonomously participating in transactions, a configurable bidding participation strategy is embedded in the smart contract. Each energy storage unit owner sets an opportunity cost threshold for participating in dispatch, expressing their willingness to participate in charging and discharging within a specific time period to the dispatch platform in the form of a dynamic bidding function. The dispatch platform's call request is only allowed to be executed by the smart contract if the bid is not lower than the opportunity cost threshold. Under the premise of meeting the grid dispatch requirements, the dispatch platform selects and calls energy storage units in ascending order of their bids, thereby optimizing resource allocation efficiency.

[0036] In this embodiment of the application, in order to establish a lightweight trust foundation in a large-scale multi-owner scheduling system, the following steps are also included: establishing a digital identity for each energy storage unit owner on the blockchain, recording their historical scheduling participation records, instruction execution fulfillment rate, and violation records, and generating a reputation score. The reputation score serves as a weighting factor in subsequent scheduling optimization and revenue distribution, forming a positive incentive where better performance results in higher revenue.

[0037] Step 3: Based on the hierarchical collaborative scheduling framework, optimize the scheduling of distributed energy storage resources and generate scheduling strategies.

[0038] For data that has passed step 2, step 3 can be further performed. Based on the property rights constraint boundaries and spatial topology established in steps 1 and 2, a hierarchical collaborative scheduling framework is used to perform collaborative optimization of global supply and demand matching and local autonomous scheduling of distributed energy storage resources. The hierarchical collaborative scheduling framework includes a three-level decision-making process. Step 3 may specifically include: Step 301: Filter the energy storage resources that can participate in scheduling within the current time period and perform virtual aggregation and grouping.

[0039] Based on the multi-dimensional state vectors of each energy storage unit and the property rights constraints of smart contracts, energy storage units that meet the following conditions in the current time period are selected: the charging and discharging period is within the allowable range set by the owner, the state of charge meets the scheduling requirements, the health status is good, and the unit has not been suspended within the validity period of the owner's authorization. The selected energy storage units are then grouped according to spatial location and capacity level using a clustering algorithm to form several virtual energy storage clusters. The input to the clustering algorithm is the spatial coordinates of each energy storage unit. Access Node Number The output includes the rated capacity and the cluster tag to which each energy storage unit belongs. Each virtual energy storage cluster has a unified external scheduling interface.

[0040] It should be noted that the spatial location basis in the above clustering grouping refers to grouping energy storage units with an electrical distance less than a preset electrical distance threshold into the same virtual energy storage cluster based on the spatial distribution topology map generated in step 1. Energy storage units within the same virtual energy storage cluster, due to their close electrical distance, exhibit spatial consistency in their impact on the voltage and power flow of the distribution network during joint charging and discharging. The grouping basis for capacity level is the rated capacity range of each energy storage unit; energy storage units with rated capacities within the same range are grouped into the same capacity level.

[0041] Step 302: Using the virtual energy storage cluster as the scheduling unit, solve the multi-objective optimization model to generate a global scheduling strategy.

[0042] The system receives global scheduling requirements, including grid load forecasts, electricity price signals, supply-demand gaps in each distribution area, and demand response instructions. Using virtual energy storage clusters as the basic scheduling unit, the system uses the charging and discharging power of each virtual energy storage cluster in each time period as decision variables to solve a multi-objective optimization model, outputting the charging and discharging power allocation scheme for each virtual energy storage cluster in each time period.

[0043] The objective function of the multi-objective optimization model is expressed as: ; in, The objective function vector; To reduce the cost of power supply to the grid, the total cost of purchasing electricity from the grid is reduced by optimizing the storage of each virtual energy storage cluster during off-peak hours and the discharge during peak hours. The Gini coefficient represents the revenue of each owner and is used to measure the fairness of revenue distribution, preventing the scheduling scheme from being overly biased towards a particular virtual energy storage cluster. To address the issue of abandoned renewable energy, adjacent energy storage units will be prioritized for charging during peak periods of local photovoltaic and wind power output, thereby improving the local consumption rate of new energy. To minimize carbon emissions costs, and considering the grid carbon emission factors for different time periods, charging should be prioritized during periods of low carbon emissions. To reduce the variance of energy storage unit lifespan loss, the distributed charging and discharging tasks are balanced and scheduled to reduce the accelerated lifespan decay of some units caused by frequent scheduling.

[0044] It should be noted that the objective function Each component is calculated using the cumulative or statistical value across all time periods within the scheduling cycle. This is the cumulative sum of the products of the purchased electricity volume and the corresponding electricity price for each time period within the scheduling cycle; The Gini coefficient is calculated based on the cumulative revenue of each owner at the end of the scheduling cycle; This is the cumulative sum of the amount of electricity wasted in each time period within the scheduling cycle; This is the cumulative sum of the products of the charging amount and the corresponding carbon emission factor for each time period within the scheduling cycle; This represents the variance of the lifetime loss of each energy storage unit within the scheduling cycle. The lifetime loss is calculated by converting the cumulative charge-discharge cycles of each unit within the scheduling cycle.

[0045] It should be noted that in the multi-objective optimization model, the ownership constraints of each virtual energy storage cluster are treated as hard boundaries that cannot be optimized. Specifically, the set of ownership constraints for each energy storage unit within each virtual energy storage cluster... The inequality constraints are mapped to a multi-objective optimization model, and the scheduling strategy searches only within the feasible region defined by the property rights boundary, without producing any out-of-bounds optimization results. This approach differs from existing techniques that treat property rights constraints as a penalty term in the objective function.

[0046] The multi-objective optimization model is solved using an improved multi-objective particle swarm optimization (PSO) algorithm. The input to the improved PSO algorithm is the dispatchable power range of each virtual energy storage cluster, the hard constraint boundary of ownership, and the global scheduling requirements; the output is a set of Pareto-optimal scheduling strategies. The solution process of the improved PSO algorithm includes the following steps: Step 3021: Initialize the particle swarm, randomly set the position of each particle within the dispatchable power range of each virtual energy storage cluster, and perform compliance checks on the initial positions against the hard constraint boundary of ownership, eliminating initial particles that do not meet the constraint conditions.

[0047] Step 3022: Calculate the candidate positions of each particle according to the speed update formula of the standard multi-objective particle swarm optimization algorithm.

[0048] Step 3023: Perform hard boundary checks on the candidate positions of each particle. If a candidate position satisfies all property rights constraints, the position is directly accepted; if a candidate position violates any property rights constraint, the particle's position is projected along the dimension of the violation onto the nearest feasible region boundary, and the projected position is used as the particle's updated position, thereby ensuring that each candidate solution is within the feasible region.

[0049] Step 3024: Calculate the objective function vector corresponding to the current position of each particle. Update the individual optimal position and Pareto archive set.

[0050] Step 3025: Repeat steps 3022 to 3024 until the preset number of iterations is reached, and output the Pareto optimal scheduling strategy set.

[0051] It should be noted that the Pareto optimal scheduling strategy set contains multiple non-dominated solutions. When the scheduling platform selects the final execution plan from the Pareto optimal scheduling strategy set, it adopts a plan selection method based on ideal point distance. This method calculates the weighted distance between the objective function vector corresponding to each non-dominated solution and the ideal point, and selects the plan with the smallest weighted distance as the global scheduling strategy distributed to each virtual energy storage cluster. Here, the ideal point refers to the reference point composed of the optimal values ​​of each single objective, i.e., the point on the objective function vector... In this context, each component is a vector formed by taking the minimum value of the corresponding objective in the Pareto archive set; the weighted distance is the Euclidean distance between the objective function vector of each non-dominated solution and the ideal point, where the weight of each objective component is preset by the scheduling platform according to the current power grid operation priority.

[0052] Step 303: Decompose the global scheduling strategy to each virtual energy storage cluster, and let the local scheduling module autonomously complete the task allocation within the cluster.

[0053] The scheduling platform decomposes the global scheduling strategy into net power targets for each virtual energy storage cluster and issues them to the local scheduling modules of each virtual energy storage cluster. The local scheduling modules perform fine-grained secondary allocation of the net power targets based on the real-time state of charge of each energy storage unit within the virtual energy storage cluster, the local preferences set by the owner, and real-time communication conditions.

[0054] It should be noted that the aforementioned net power target refers to the total power value that the entire virtual energy storage cluster needs to charge or discharge within a specific time period, as mandated by the dispatch platform. The secondary allocation process of the local dispatch module is completed using a consensus algorithm. The inputs to the consensus algorithm are the real-time state of charge of each energy storage unit within the virtual energy storage cluster, the local preference parameters set by the owner, and the net power target. The output is the charging and discharging power share undertaken by each energy storage unit within that time period. Through information exchange between neighboring nodes, each energy storage unit within the virtual energy storage cluster autonomously negotiates its assigned charging and discharging power share within the error range allowed by the global command, without waiting for individual assignments from the dispatch platform.

[0055] In this embodiment, to maintain scheduling continuity when global scheduling cannot respond in real time due to communication delays or computational overhead, the local scheduling module can dynamically adjust within the virtual energy storage cluster after receiving the net power target. When an energy storage unit in the virtual energy storage cluster exits scheduling due to a sudden anomaly, the local scheduling module autonomously redistributes the power share borne by that energy storage unit to the remaining available units in the virtual energy storage cluster, maintaining the stability of the net power output of the virtual energy storage cluster.

[0056] Step 4: Perform compliance verification on scheduling instructions, monitor execution status, and handle execution deviations.

[0057] For scheduling strategies that have passed step 3, step 4 can be further performed. Step 4 may specifically include: Step 401: Submit the scheduling instruction to the blockchain smart contract for compliance verification.

[0058] The scheduling instructions are encapsulated as transactions to be verified and submitted to a smart contract on the consortium blockchain. The scheduling instructions include the target energy storage unit identifier, charging / discharging power values, charging / discharging start and end times, and expected state of charge (SBC) changes. The smart contract compares each scheduling instruction against the ownership constraints of the energy storage unit as documented in step 2. The comparison items include time period permissibility, power limit, and minimum SBC threshold. When all comparison items pass, the scheduling instruction is signed and confirmed, and sent to the energy storage unit's field controller. If any comparison item fails, the scheduling instruction is automatically rejected, an alarm log is recorded, and the scheduling platform re-optimizes the scheduling plan for that time period.

[0059] Step 402: Monitor the actual execution status of the scheduling instructions, detect execution deviations, and perform graded processing.

[0060] After receiving verified dispatch instructions, the field controller of the energy storage unit controls the bidirectional energy storage converter to perform charging or discharging operations according to the dispatch instruction parameters. During execution, the field controller collects operational data such as charging and discharging power, voltage, current, and state of charge changes in real time, and transmits them back to the cloud platform for point-by-point comparison with the target values ​​of the dispatch instructions.

[0061] When the deviation between the actual executed power and the target value of the scheduling command exceeds a preset tolerance threshold, a tiered alarm is triggered. If the deviation is minor, the record is identified as an execution anomaly and included in the reputation score update process of the energy storage unit; if the deviation is severe, the specific violation of the energy storage unit is identified and identified as a violation record, reported to the blockchain reputation management contract, and the subsequent scheduling priority of the energy storage unit is downgraded.

[0062] It should be noted that the aforementioned slight deviation refers to the situation where the relative deviation between the actual executed power and the target value of the dispatch instruction is between the preset tolerance threshold and the preset violation threshold; severe deviation refers to the situation where the relative deviation exceeds the preset violation threshold, including behaviors such as unauthorized stopping of discharge or continuing to discharge beyond the duration of the dispatch instruction. Both the preset tolerance threshold and the preset violation threshold are percentages of relative power deviation preset by the dispatch platform according to the grid operation requirements, and the preset tolerance threshold is less than the preset violation threshold.

[0063] In this embodiment, to achieve end-to-end traceability of scheduling behavior, each verified scheduling instruction, after being executed, has its complete content written to the consortium blockchain for evidence storage. This complete content includes instruction parameters, timestamps, signature information, and execution status. When disputes arise regarding scheduling behavior, historical scheduling instruction records can be traced and audited through a blockchain explorer.

[0064] In this embodiment of the application, based on step 4, the following steps are also included: Step 5: Based on the actual execution data of the scheduling instructions, calculate and automatically distribute the revenue through smart contracts.

[0065] Step 501: Calculate the total revenue generated by the shared scheduling system in each time period.

[0066] Based on the actual execution data of dispatch instructions, the following three types of revenue are calculated: electricity cost savings, which is the difference in electricity costs incurred by the energy storage system during off-peak hours when charging and during peak hours; demand response compensation, which is the compensation fee obtained by the energy storage system for participating in grid demand response; and carbon emission reduction, which is the carbon trading revenue corresponding to the energy storage system reducing grid carbon emissions by optimizing charging and discharging times. The total revenue is obtained by summing these three types of revenue. .

[0067] Step 502: Calculate the revenue share of each owner based on the actual contribution of each energy storage unit.

[0068] Calculate the contribution score for each energy storage unit. The contribution score is obtained by weighting the following indicators: actual discharge amount. , represents the actual discharge energy of the energy storage unit during the dispatch period; response speed , is the reciprocal of the time required for the energy storage unit to reach the target power after the dispatch command arrives; response accuracy. , is the reciprocal of the relative deviation between the actual discharge power and the target value of the dispatch command; equipment utilization rate , which is the ratio of the frequency at which the energy storage unit is invoked to the available scheduling time; reputation score This refers to the on-chain reputation score accumulated in step 2.

[0069] Since the five indicators mentioned above have different dimensions, before weighted calculation, each indicator is normalized using a mean normalization method based on its range. This unifies the indicators to the same dimensionless numerical range. The corresponding normalized values ​​are denoted as follows: , , , , Then, according to the weighting coefficients agreed upon by each owner and the scheduling platform. to After performing linear weighting, we get: ; in, For the first Contribution score of each energy storage unit; This is the normalized value of the actual discharge amount. This is a normalized value for the response speed. In response to the accuracy normalization value, This is the normalized value of equipment utilization. This is the normalized value for the credit score. , , , , These are the weighting coefficients corresponding to actual discharge volume, response speed, response accuracy, equipment utilization rate, and reputation score, respectively. And all weighting coefficients are not less than zero.

[0070] The share of profits for each owner is calculated as follows: ; in, For the first The share of profits for each owner; Total revenue; For the first Contribution score of each energy storage unit; This represents the total number of energy storage units participating in the current dispatch. To sum the index variable, iterate through all energy storage units participating in the current scheduling.

[0071] Step 503: The smart contract automatically triggers settlement and completes the transfer of profits.

[0072] The aforementioned accounting rules and allocation algorithm are encoded into a smart contract, which is automatically triggered and executed when the preset settlement period arrives. The smart contract reads the scheduling instruction execution records on the blockchain and calculates the share of revenue for each owner. The proceeds are automatically transferred through the on-chain digital currency wallet.

[0073] In this embodiment, to improve the understandability of revenue distribution and reduce the information verification cost of dispute resolution, a personalized revenue dashboard is provided for each owner, displaying the source of revenue, distribution calculation process, and final settlement amount for each scheduling order. If an owner has any objection to the revenue distribution result, they can initiate a review request through the appeal channel. The system automatically retrieves historical scheduling records on the chain for cross-validation and generates an audit report.

[0074] Example 2

[0075] During the peak summer dispatch period of 20XX, a city's power distribution network management agency (hereinafter referred to as "the platform provider") implemented shared dispatch for distributed energy storage units belonging to different owners within its jurisdiction. There are a total of 5 distributed energy storage units within the distribution network's jurisdiction, belonging to 3 owners: Owner A is a property management company in an industrial and commercial park, holding energy storage units 1 and 2; Owner B is a residential community owners' committee, holding energy storage unit 3; and Owner C is a photovoltaic power station operator, holding energy storage units 4 and 5. Each energy storage unit is connected to different transformer substations and is spatially distributed across feeders A and B. The dispatch period is from 08:00 to 20:00 daily, comprising 12 time slots, each lasting 1 hour.

[0076] Implementation of Step 1

[0077] The platform uses smart terminals deployed in each energy storage unit to collect operating parameters and ownership attribute data of each unit at 08:00, and correlates them with time-of-use electricity price information from the grid side to generate a multi-dimensional state vector for each energy storage unit. ,in Number the energy storage units. This refers to the time of data collection.

[0078] The original multidimensional state vector data of each energy storage unit are shown in Table 1 (at 08:00): Table 1: Original data of multidimensional state vectors for each energy storage unit:

[0079] Based on the access node information of each unit, the electrical distance between each unit is calculated. Energy storage unit 1 and energy storage unit 2 are both connected to feeder node 3 of feeder A, with an electrical distance of 0.12 pu; energy storage unit 4 and energy storage unit 5 are both connected to feeder node 2 of feeder B, with an electrical distance of 0.09 pu; the electrical distance between energy storage unit 1 and energy storage unit 3 is 0.38 pu; the electrical distance between any two units on feeder A and feeder B exceeds 0.60 pu. The above spatial distribution topology provides a basis for subsequent cluster partitioning.

[0080] Implementation of step 2

[0081] Step 201 Implementation: The consortium blockchain creates a digital certificate of ownership for each energy storage unit, and writes the device identifier, owner identity, scope of authority, and authorization validity period from Table 1 into the certificate and stores it on the blockchain. Taking energy storage unit 3 as an example, after the hash value of its digital certificate of ownership is written into the blockchain, any unilateral modification to the allowed charging and discharging period or minimum state of charge will fail the hash verification.

[0082] Step 202 implementation: The platform will set the property rights constraints of each unit in Table 1. Transform it into smart contract rules and deploy them, where Includes the allowed charging and discharging period Power limit and state of charge limit Three constraints. Taking the charging and discharging command at 10:00 as an example, the scheduling platform issues a discharging command to energy storage unit 3, with a target power of 55kW and an expected state of charge (SOC) reduction from 72% to 63%. The smart contract compares the following three criteria one by one: the time period 10:00 falls within... (10:00-18:00), passed; power 55kW not exceeded. (60kW), passed; expected state of charge not less than 63%. (30%), passed. All three items passed, and the instruction was signed and confirmed.

[0083] During the same period, if the dispatch platform mistakenly issues a discharge command for energy storage unit 2 at 08:30 (the allowed time period for energy storage unit 2 is from 09:00), the smart contract will detect that the time period is not met. The system automatically intercepts and generates alarm logs, allowing the scheduling platform to readjust its plans.

[0084] The compliance verification records for smart contracts are shown in Table 2 (example for the 10:00 time slot): Table 2: Smart Contract Compliance Verification Record Table

[0085] The opportunity cost thresholds for each owner have been configured in the smart contract: owner A's is set at 0.58 yuan / kWh, owner B's at 0.52 yuan / kWh, and owner C's at 0.61 yuan / kWh. The current time-of-use electricity price is 0.89 yuan / kWh. Since all three owners' bids are lower than the current price, the dispatch platform prioritizes resource allocation based on bids from lowest to highest (owner B, owner A, owner C), thus improving resource allocation efficiency.

[0086] Implementation of step 3

[0087] Step 301 Implementation: At 08:00, the platform selects energy storage units that can participate in scheduling based on the multi-dimensional state vectors of each unit and smart contract constraints. All five units meet the time period, charge state, and health state conditions in the current time period and pass the selection. Based on the spatial distribution topology, the preset electrical distance threshold is 0.50 pu. The electrical distances of energy storage units 1, 2, and 3 are all within feeder A. The electrical distance between energy storage units 1 and 2 is 0.12 pu, and the electrical distance between energy storage units 2 and 3 is 0.31 pu. All three are less than the threshold and are assigned to cluster A. The electrical distance between energy storage units 4 and 5 is 0.09 pu and are assigned to cluster B.

[0088] The results of the virtual energy storage cluster partitioning are shown in Table 3: Table 3: Results of Virtual Energy Storage Cluster Partitioning

[0089] Step 302 Implementation: The scheduling platform receives global scheduling requests: 14:00-16:00 is the peak demand response period, with a supply-demand gap of 180kW in area A and 240kW in area B; 08:00-10:00 is the off-peak electricity price period (0.32 yuan / kWh), and 14:00-18:00 is the peak electricity price period (0.89 yuan / kWh). Using cluster A and cluster B as decision variables, solve the multi-objective optimization model.

[0090] The improved multi-objective particle swarm optimization algorithm performs hard property rights checks on all particles during initialization, and will detect those that do not meet the requirements. Particle elimination. During the iteration process, if a candidate particle position causes the power of cluster A to exceed 260kW, then that dimension is projected onto the 260kW boundary. After 150 iterations, the Pareto archive contains 23 non-dominated solutions. The scheduling platform adopts a scheme selection method based on ideal point distance, with each objective weight set as follows: , , , , ,in For the first The weight coefficients of each objective are determined. The ideal point is formed by the minimum value of each objective component in the Pareto archive set. The weighted Euclidean distance between each non-dominated solution and the ideal point is calculated, and the scheme with the smallest weighted distance is selected as the global scheduling strategy.

[0091] The resulting global scheduling strategy (critical periods) is shown in Table 4: Table 4: Global Scheduling Strategy Table

[0092] Step 303 Implementation: The scheduling platform issues the net power target of 180kW for Cluster A during the period from 14:00 to 16:00 to the local scheduling module of Cluster A. The local scheduling module uses a consensus algorithm to autonomously negotiate and allocate power within the cluster based on the real-time state of charge (SOC) of energy storage unit 1 (68%), energy storage unit 2 (55%), and energy storage unit 3 (72%), as well as the local preference parameters of each owner. The negotiation result is: energy storage unit 1 undertakes 82kW, energy storage unit 2 undertakes 58kW, and energy storage unit 3 undertakes 40kW, totaling 180kW, consistent with the net power target. When energy storage unit 2's SOC drops to 25% at 15:30, it triggers... When constrained and out of scheduling, the local scheduling module autonomously redistributes the remaining portion of its 58kW share to energy storage unit 1 (increases by 32kW) and energy storage unit 3 (increases by 26kW), and the net power output of cluster A remains stable.

[0093] Implementation of step 4

[0094] Implementation of step 401: Taking the scheduling instruction of energy storage unit 1 in cluster A at 14:00 as an example, the instruction content is: target unit energy storage unit 1, discharge power 82kW, time period 14:00-16:00, expected state of charge to decrease from 68% to 42%. Smart contract comparison: time segment input. Passed; 82kW did not exceed (120kW), passed; expected state of charge not less than 42%. (20%), passed. The instruction was signed and written to the consortium blockchain, with the complete content including instruction parameters, timestamp, and signature information.

[0095] Step 402 Implementation: The field controller of energy storage unit 1 discharges according to the target of 82kW, and the cloud platform collects the actual power data every 5 minutes. At 14:30, the actual discharge power is 79.5kW, with a relative deviation of 3.05%, which is within the preset tolerance threshold (5%), and the execution is normal. At 14:50, the actual discharge power drops to 68kW, with a relative deviation of 17.07%, exceeding the preset violation threshold (15%), triggering a severe deviation alarm. The system identifies this behavior as an abnormal power drop and reports it to the blockchain reputation management contract, and the subsequent scheduling priority of energy storage unit 1 is downgraded.

[0096] The operational status monitoring record of energy storage unit 1 is shown in Table 5 (14:00-16:00 period): Table 5: Execution Status Monitoring Record of Energy Storage Unit 1

[0097] Implementation of step 5

[0098] Step 501 Implementation: After the scheduling cycle ends, the smart contract reads the on-chain execution record and calculates three types of revenue: electricity cost savings, calculated as the difference in electricity costs between off-peak charging and peak discharging. Cluster A and Cluster B's combined actual charging and discharging volume is approximately 3860 kWh, with an electricity price difference of approximately 0.57 yuan / kWh, resulting in electricity cost savings of approximately 2200 yuan; demand response compensation revenue, calculated based on the grid compensation standard, is 860 yuan; and carbon emission reduction revenue, calculated based on the carbon trading price, is 310 yuan. Total Revenue Yuan.

[0099] Implementation of step 502: Taking 5 energy storage units as an example, based on 5 indicators—actual discharge amount, response speed, response accuracy, equipment utilization rate, and credit score—after range normalization, the results are weighted according to the following coefficients. , , , , Weighted calculation of contribution scores for each unit ,in For the first The weighting coefficient of each indicator.

[0100] The contribution calculation data for each energy storage unit is shown in Table 6: Table 6: Contribution Calculation Data for Each Energy Storage Unit:

[0101] Taking energy storage unit 4 as an example, its contribution score is: The sum of the contribution scores for each unit is: The share of profits for each owner is as follows: Calculation, where For the owner The share of profits, The total contribution score of the units they hold. The sum of the contribution scores of all units.

[0102] The distribution of profits among the owners is shown in Table 7: Table 7: Profit Distribution Results for Each Owner

[0103] It should be noted that the calculation process for the share of profits held by owner A is as follows: Yuan, owned by person B Yuan, owned by C Yuan, the sum of the three and Consistent.

[0104] Step 503 Implementation: When the preset settlement period (20:00) arrives, the smart contract automatically reads the on-chain scheduling execution record, triggers the revenue transfer according to the calculation results in Table 7, and transfers 1281 yuan, 578 yuan, and 1511 yuan to the on-chain digital currency wallets of owners A, B, and C, respectively. All three owners can view the revenue source and allocation calculation process for each scheduling order through the revenue dashboard. Owner A submits a review application for the severe deviation record of energy storage unit 1 at 14:50. The system automatically retrieves historical on-chain scheduling records for cross-validation, generates an audit report, confirming that the deviation has been included in the contribution deduction, and the review result is objective and verifiable.

[0105] Throughout the process, data starts from the multi-dimensional state vector in step 1, carrying property rights constraint information, and enters the smart contract notarization in step 2. In the multi-objective optimization in step 3, the search space is based on the hard boundary constraints of property rights. After compliance verification and execution monitoring in step 4, a trusted on-chain execution record is formed. Finally, in step 5, revenue calculation and automatic allocation are completed based on the actual execution data. The data flow between each step is based on the state parameters and property rights constraints of the same batch of energy storage units, maintaining consistency throughout the entire chain from collection, notarization, scheduling, execution to settlement.

[0106] The foregoing has only described certain exemplary embodiments of the present invention by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the foregoing drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

[0107] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0108] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A shared scheduling and management method for distributed energy storage devices, characterized in that, include: The operating parameters and property rights attribute data of each distributed energy storage unit are collected, and the property rights constraint set is embedded as the core attribute to generate a multi-dimensional state vector of each energy storage unit. The property rights constraints of each owner are encoded into smart contracts and deployed on the consortium blockchain. Before the scheduling instructions are executed, they must be verified for compliance by the smart contracts. Scheduling optimization is performed based on a hierarchical collaborative scheduling framework, which includes: selecting energy storage units that meet ownership constraints and clustering them into virtual energy storage clusters according to spatial location and capacity level; using the virtual energy storage clusters as scheduling units to solve a multi-objective optimization model to generate a global scheduling strategy; and decomposing the global scheduling strategy to each virtual energy storage cluster so that the local scheduling module can autonomously complete the task allocation within the cluster. In the multi-objective optimization model, the property rights constraint set of each energy storage unit is treated as a hard boundary of inequality constraints, and the scheduling strategy is only searched within the feasible domain defined by the property rights boundary.

2. The shared scheduling and management method for distributed energy storage devices according to claim 1, characterized in that, The multidimensional state vector includes timestamp, unique device identifier, state of charge, health status, property constraint set, spatial coordinates, distribution network access node number, upper limit of available power, lower limit of available power, and electricity price sensitivity coefficient; wherein, the property constraint set includes the allowable charging and discharging time interval, power limit, minimum state of charge retention threshold, and lifetime depreciation tolerance coefficient set by the owner; based on the access node information of each energy storage unit, the electrical distance and communication delay between any two energy storage units are calculated to generate a spatial distribution topology map of the energy storage units.

3. The shared scheduling and management method for distributed energy storage devices according to claim 1, characterized in that, Encoding the property rights constraints of each owner into smart contracts and deploying them on the consortium blockchain includes: creating a digital property rights certificate for each distributed energy storage unit on the consortium blockchain, writing the basic attribute information of the device, the owner's identity information, the ownership of the device, the scope of charging and discharging permissions, and the validity period of the scheduling authorization into the digital property rights certificate, and realizing the on-chain storage of property rights information through the hash chain of the blockchain; converting the scheduling constraints set by each owner into executable rules written in smart contract language and deploying them on the consortium blockchain; the compliance verification is the process by which the smart contract compares the charging and discharging period, power value, and expected change in state of charge in the scheduling instruction with the corresponding constraints in the property rights constraint set one by one, and if any condition is not met, the scheduling instruction is automatically intercepted and an alarm is triggered.

4. The shared scheduling and management method for distributed energy storage devices according to claim 3, characterized in that, A configurable bidding participation strategy is embedded in the smart contract. Each energy storage unit owner sets an opportunity cost threshold for participating in scheduling and expresses their willingness to participate in charging and discharging to the scheduling platform in the form of a dynamic bidding function. The scheduling platform's call request is only allowed to be executed by the smart contract if the bid is not lower than the opportunity cost threshold. Under the premise of meeting the grid scheduling needs, the scheduling platform selects and calls energy storage units in order of their bids from low to high. Historical scheduling participation records, instruction execution fulfillment rates, and violation records are recorded for each energy storage unit owner on the blockchain to generate a reputation score. The reputation score is used as a weighting factor in scheduling selection and revenue distribution.

5. The shared scheduling and management method for distributed energy storage devices according to claim 1, characterized in that, The objective function of the multi-objective optimization model includes five components: grid supply cost, Gini coefficient of each owner's revenue, renewable energy curtailment, carbon emission cost, and variance of energy storage unit lifetime loss. Specifically, the grid supply cost is the cumulative sum of the products of electricity purchased and electricity price for each time period within the scheduling cycle; the Gini coefficient is calculated based on the cumulative revenue of each owner at the end of the scheduling cycle; the renewable energy curtailment is the cumulative sum of the curtailment for each time period within the scheduling cycle; the carbon emission cost is the cumulative sum of the products of charging amount and carbon emission factor for each time period within the scheduling cycle; and the variance of lifetime loss is calculated based on the cumulative charge-discharge cycles of each energy storage unit within the scheduling cycle.

6. The shared scheduling and management method for distributed energy storage devices according to claim 5, characterized in that, The multi-objective optimization model is solved using an improved multi-objective particle swarm optimization algorithm. The solution process includes: initializing the particle swarm, setting the position of each particle within the dispatchable power range of each virtual energy storage cluster, performing compliance checks on the initial positions against the hard property rights constraints, and eliminating initial particles that do not meet the constraints; calculating the candidate positions of each particle according to the velocity update formula; performing hard property rights constraint boundary checks on the candidate positions of each particle, accepting the candidate position directly if it meets all property rights constraints, and projecting the particle's position along the dimension of the violation to the nearest feasible region boundary if it violates any property rights constraint; calculating the objective function vector corresponding to the current position of each particle, and updating the individual optimal position and Pareto archive set; repeating the iteration until the preset number of iterations is reached, and outputting the Pareto optimal scheduling strategy set; from the Pareto optimal scheduling strategy set, using a scheme selection method based on ideal point distance, calculating the weighted distance between the objective function vector corresponding to each non-dominated solution and the ideal point, and selecting the scheme with the smallest weighted distance as the global scheduling strategy, where the ideal point is composed of the minimum value of each objective component in the Pareto archive set.

7. The shared scheduling and management method for distributed energy storage devices according to claim 1, characterized in that, The local scheduling module autonomously completes the task allocation within the cluster, including: receiving the net power target issued by the scheduling platform, and using a consensus algorithm to perform secondary allocation of the net power target. The input of the consensus algorithm is the real-time state of charge of each energy storage unit in the virtual energy storage cluster, the local preference parameters set by the owner, and the net power target. The output is the charging and discharging power share undertaken by each energy storage unit in that time period. Each energy storage unit in the virtual energy storage cluster autonomously negotiates its own charging and discharging power share through information exchange between neighboring nodes. When an energy storage unit in the virtual energy storage cluster exits the scheduling due to a sudden anomaly, the local scheduling module autonomously reallocates the power share undertaken by that energy storage unit to the remaining available units in the virtual energy storage cluster.

8. The shared scheduling and management method for distributed energy storage devices according to claim 1, characterized in that, Also includes: Once the dispatch instructions have passed compliance verification, they are sent to the field controllers of the energy storage units for execution. Real-time operational data on charging and discharging power and state of charge changes are collected and compared point-by-point with the target values ​​of the dispatch instructions. When the deviation between the actual executed power and the target value of the dispatch instructions exceeds a preset tolerance threshold, a tiered alarm is triggered. Specifically, when the relative deviation is between the preset tolerance threshold and the preset violation threshold, the record is identified as an execution anomaly and included in the reputation score update process for that energy storage unit. When the relative deviation exceeds the preset violation threshold, the specific violation behavior of the energy storage unit is identified and identified as a violation record, and the subsequent dispatch priority of that energy storage unit is downgraded. After each verified dispatch instruction is executed, the complete content of the dispatch instruction is written to the consortium blockchain for evidence storage.

9. The shared scheduling and management method for distributed energy storage devices according to claim 1, characterized in that, Also includes: Based on the actual execution data of scheduling instructions, the revenue from electricity cost savings, demand response compensation, and carbon emission reduction is calculated and aggregated to obtain the total revenue. A contribution score is calculated for each energy storage unit, which is obtained by weighting the normalized data based on actual discharge volume, response speed, response accuracy, equipment utilization rate, and reputation score. The revenue share of each owner is the total revenue multiplied by the ratio of that owner's contribution score to the sum of the contribution scores of all participating energy storage units. The calculation rules and allocation algorithm are encoded into a smart contract, which automatically reads the scheduling instruction execution records on the chain and completes the revenue transfer when the preset settlement period arrives.

10. A shared scheduling and management system for distributed energy storage devices, comprising a shared scheduling and management method for distributed energy storage devices according to any one of claims 1-9, characterized in that, include: The data acquisition and fusion module is used to collect the operating parameters and property attribute data of each distributed energy storage unit, embed the property constraint set as the core attribute, and generate a multi-dimensional state vector of each energy storage unit. The property rights constraint on-chain module is used to encode the property rights constraints of each owner into smart contracts and deploy them on the consortium blockchain to perform compliance verification of scheduling instructions; The hierarchical collaborative scheduling module is used to screen energy storage units that meet the property rights constraints and cluster them into virtual energy storage clusters. The virtual energy storage clusters are used as scheduling units to solve multi-objective optimization models to generate a global scheduling strategy. The global scheduling strategy is decomposed into each virtual energy storage cluster and the local scheduling module autonomously completes the task allocation within the cluster. The property rights constraint set of each energy storage unit is treated as a hard boundary of inequality constraints. The execution monitoring and processing module is used to send dispatch instructions that have passed compliance verification to the energy storage unit for execution, monitor the actual execution status, and classify and process execution deviations.

Images