A power storage system for optimizing power storage in coordination with voltage management and peak-valley arbitrage of a transformer area

Abstract

The present application relates to the technical field of energy storage optimization, in particular to a kind of energy storage system of substation voltage management and peak-valley arbitrage synergic optimization.It obtains the operation time sequence data of main network and the operation time sequence data of multiple distributed energy nodes, and establishes benchmark library correspondingly;Based on benchmark library, calibrate main network management demand, and correspondingly evaluate the management contribution degree of multiple distributed energy nodes;Obtain the operation specification information of multiple distributed energy nodes, and construct global synergic optimization strategy in combination with management contribution degree;Using multi-objective optimization algorithm, the maximum solution of synergic evaluation function is carried out in combination with global synergic optimization strategy, and the maximum solution result is global optimal charging and discharging plan, and is issued to multiple distributed energy nodes for execution.The present application can improve the comprehensive utilization efficiency of energy storage system under the premise of ensuring the safe and stable operation of power grid.

Description

Technical Field

[0001] This invention relates to the field of energy storage optimization technology, and in particular to an energy storage system that coordinates voltage management and peak-valley arbitrage optimization in transformer substations. Background Technology

[0002] Currently, the large-scale integration of distributed energy sources has highlighted the problem of voltage fluctuations in the power grid. The peak-valley electricity price difference provides arbitrage opportunities for energy storage systems. However, current technologies mostly focus on single-objective optimization. Single voltage management technologies have not explored the arbitrage potential of energy storage. Single arbitrage technologies are prone to exacerbating voltage fluctuations. Existing collaborative optimization technologies are difficult to balance grid security, economic benefits, and the enthusiasm of node participation. Summary of the Invention

[0003] This invention addresses the problem in existing technologies of balancing grid security, economic benefits, and node participation incentives by providing an energy storage system that coordinates and optimizes voltage management in distribution areas and peak-valley arbitrage.

[0004] The technical solution of the present invention to solve the above-mentioned technical problems is as follows:

[0005] This invention provides an energy storage system that synergistically optimizes transformer area voltage management and peak-valley arbitrage, comprising:

[0006] The benchmark establishment module is used to acquire runtime timing data of the main network and runtime timing data of multiple distributed energy nodes, and establish a benchmark library accordingly. The benchmark library includes at least a first typical time-series data set, a second typical time-series data set, and a greedy arbitrage scheme set.

[0007] The demand and contribution calibration module is used to calibrate the main network governance demand based on the benchmark library, and to evaluate the governance contribution of multiple distributed energy nodes accordingly.

[0008] The collaborative optimization configuration module is used to obtain the operating specification information of multiple distributed energy nodes and, in combination with the governance contribution, construct a global collaborative optimization strategy, wherein the global collaborative optimization strategy includes at least a collaborative evaluation function and collaborative constraints.

[0009] The solution and execution module is used to maximize the collaborative evaluation function by employing a multi-objective optimization algorithm combined with the global collaborative optimization strategy, outputting the maximized solution result as the globally optimal charging and discharging plan, and distributing it to multiple distributed energy nodes for execution.

[0010] Optionally, the runtime sequence data of the main network and multiple distributed energy nodes are acquired, and a benchmark library is established accordingly. The benchmark library includes at least a first set of typical time-series data, a second set of typical time-series data, and a greedy arbitrage scheme set, including:

[0011] Historical and real-time data are collected to obtain runtime sequence data of the main network and multiple distributed energy nodes. Statistical analysis is performed based on the runtime sequence data of the main network to establish the first typical time-series data of the main network. Statistical analysis is then performed on the runtime sequence data of multiple distributed energy nodes to obtain the second typical time-series data set, wherein the second typical time-series data corresponds one-to-one with each distributed energy node. Based on the first typical time-series data and the second typical time-series data set, each distributed energy node is driven to perform greedy optimization decisions with the goal of maximizing its own economic operating benefits, generating a greedy arbitrage scheme and its corresponding benchmark return for each distributed energy node, and outputting the greedy arbitrage scheme set.

[0012] Optionally, based on the benchmark library, the main grid governance requirements are calibrated, including: obtaining and integrating the node constraint information of multiple distributed energy nodes, configuring them as a virtual single microgrid; fusing the second typical time-series data set to form aggregated time-series data of the virtual single microgrid; and, with the goal of minimizing main grid voltage fluctuations, performing scheduling solutions on the virtual single microgrid in conjunction with the aggregated time-series data to obtain an aggregated power reference curve, which is then calibrated as the main grid governance requirements.

[0013] The evaluation of the governance contribution of multiple distributed energy nodes includes: performing simulation analysis based on the topology and electrical parameters of the main network and the multiple distributed energy nodes, combined with the aggregated power reference curve; calculating the voltage-power sensitivity response curve of each distributed energy node to the main network voltage, and calculating the corresponding response linearity; performing linear regression analysis on the voltage-power sensitivity response curve to obtain the apparent voltage-power sensitivity, and performing weighted correction using the normalized response linearity as a correction coefficient to obtain the governance contribution of the multiple distributed energy nodes.

[0014] Optionally, the steps for constructing the collaborative evaluation function include: extracting and regularizing governance evaluation rules based on the voltage governance task book of the main network to obtain voltage governance evaluation items; using the governance contribution as a coefficient, weighting and summing the estimated arbitrage income of multiple distributed energy nodes to define a weighted economic operation income item, wherein the estimated arbitrage income is calculated based on a preset arbitrage mathematical model; and defining the dimensionless normalized sum of the voltage governance evaluation item and the weighted economic operation income item as the collaborative evaluation function.

[0015] Optionally, the collaborative constraints include at least objective hardware constraints and subjective incentive constraints, wherein: the objective hardware constraints are defined based on the operational specification information and include at least the energy storage charging and discharging power constraints and the state of charge (SOC) constraints of each distributed energy node; the subjective incentive constraints are bidirectional satisfaction constraints, including: a lower satisfaction constraint, which limits the estimated arbitrage profit of each distributed energy node under collaborative optimization to not be less than the product of the corresponding benchmark profit and a preset first subjective constraint coefficient; and an upper satisfaction constraint, which limits the estimated arbitrage profit of each distributed energy node under collaborative optimization to not be greater than the product of the corresponding benchmark profit and a second subjective constraint coefficient; wherein the second subjective constraint coefficient is greater than the first subjective constraint coefficient by more than 1.

[0016] The process of determining the value of the second subjective constraint coefficient includes:

[0017] Based on the greedy arbitrage scheme set, the corresponding arbitrage control parameters are extracted and output as an ideal parameter upper limit set; based on the preset incremental mapping function, the governance contribution is mapped to the subjective constraint coefficient on each parameter dimension, wherein the subjective constraint coefficient is less than or equal to 1; according to the subjective constraint coefficient, the ideal parameter upper limit set is iterated and adjusted, wherein the subjective constraint coefficient on each parameter dimension is configured differently.

[0018] The method employs a multi-objective optimization algorithm, combined with the global collaborative optimization strategy, to maximize the collaborative evaluation function. It also includes: defining a voltage risk threshold based on the main network's voltage management task list; real-time monitoring of the main network's operating voltage indicators; and ignoring the subjective initiative constraint in the maximization process if the operating voltage indicators exceed the voltage risk threshold.

[0019] The method of using a multi-objective optimization algorithm and combining the global collaborative optimization strategy to maximize the collaborative evaluation function also includes: under the subjective initiative constraint, if a feasible solution that satisfies the voltage governance task is not obtained after a preset number of iterations, the subjective initiative constraint is ignored in the maximization process.

[0020] The subjective motivation constraint is ignored in the maximization solution, including: firstly ignoring the lower limit constraint of satisfaction; if the maximization solution is still not obtained after ignoring the lower limit constraint of satisfaction, then the upper limit constraint of satisfaction is ignored.

[0021] By implementing this invention, it is possible to acquire runtime sequence data of the main network and multiple distributed energy nodes, and establish a corresponding benchmark library. The benchmark library includes at least a first set of typical time-series data, a second set of typical time-series data, and a set of greedy arbitrage schemes. This provides basic data support for subsequent main network demand calibration, node contribution evaluation, and optimization strategy construction, clarifies the arbitrage profit benchmark for each node's independent operation, and ensures that subsequent collaborative optimization has a reliable reference basis.

[0022] By implementing this invention, it is possible to calibrate the main grid governance requirements based on the benchmark library, evaluate the governance contribution of multiple distributed energy nodes accordingly, accurately identify the core requirements for main grid voltage governance, objectively quantify the contribution value of each node to main grid governance, and provide a fair and reasonable basis for the distribution of benefits and configuration of constraints in subsequent collaborative optimization.

[0023] By implementing this invention, it is possible to obtain the operational specification information of multiple distributed energy nodes and, in conjunction with their governance contributions, construct a global collaborative optimization strategy. The global collaborative optimization strategy includes at least a collaborative evaluation function and collaborative constraints, establishes an optimization objective that balances the main grid voltage governance effect with the economic benefits of nodes, and balances hardware capabilities and node participation enthusiasm through dual constraints to ensure the feasibility and fairness of the optimization strategy.

[0024] By implementing this invention, a multi-objective optimization algorithm can be used to maximize the collaborative evaluation function in conjunction with the global collaborative optimization strategy. The maximized solution is output as the globally optimal charging and discharging plan, which is then distributed to multiple distributed energy nodes for execution. This allows for the rapid identification of the optimal solution that balances governance effectiveness and economic benefits, ensuring the accurate implementation of the charging and discharging plan. Furthermore, a dynamic constraint adjustment mechanism balances fairness in conventional scenarios with governance priority in special scenarios.

[0025] In summary, by implementing this invention, the overall utilization efficiency of energy storage systems can be improved while ensuring the safe and stable operation of the power grid. Attached Figure Description

[0026] Figure 1 A schematic diagram of the structure of an energy storage system for the coordinated optimization of transformer area voltage management and peak-valley arbitrage provided by the present invention;

[0027] Figure 2 This is a flowchart illustrating the process of establishing a benchmark library in an energy storage system that provides synergistic optimization of transformer area voltage management and peak-valley arbitrage for this invention.

[0028] In the attached diagram, the components represented by each number are as follows:

[0029] The module includes: 11 (benchmark establishment), 12 (demand and contribution calibration), 13 (collaborative optimization configuration), and 14 (solution and execution). Detailed Implementation

[0030] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0031] In the description of this invention, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more features. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0032] In the description of this invention, the term "for example" is used to mean "used as an example, illustration, or description." Any embodiment described as "for example" in this invention is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is provided to enable any person skilled in the art to make and use the invention. Details are set forth in the following description for purposes of explanation. It should be understood that those skilled in the art will recognize that the invention can be made without using these specific details. In other instances, well-known structures and processes will not be described in detail to avoid obscuring the description of the invention with unnecessary detail. Therefore, the invention is not intended to be limited to the embodiments shown, but is consistent with the broadest scope of the principles and features disclosed herein.

[0033] Example: Figure 1 As shown, this embodiment of the invention provides an energy storage system that synergistically optimizes transformer area voltage management and peak-valley arbitrage, comprising:

[0034] The benchmark establishment module 11 is used to acquire the runtime timing data of the main network and the runtime timing data of multiple distributed energy nodes, and establish a benchmark library accordingly. The benchmark library includes at least a first typical time-series data set, a second typical time-series data set, and a greedy arbitrage scheme set.

[0035] The demand and contribution calibration module 12 is used to calibrate the main network governance demand based on the benchmark library, and to evaluate the governance contribution of multiple distributed energy nodes accordingly.

[0036] The collaborative optimization configuration module 13 is used to obtain the operating specification information of multiple distributed energy nodes and, in combination with the governance contribution, construct a global collaborative optimization strategy, wherein the global collaborative optimization strategy includes at least a collaborative evaluation function and collaborative constraints.

[0037] The solution and execution module 14 is used to maximize the collaborative evaluation function by using a multi-objective optimization algorithm combined with the global collaborative optimization strategy, output the maximized solution result as the globally optimal charging and discharging plan, and distribute it to multiple distributed energy nodes for execution.

[0038] like Figure 2 As shown, in the benchmark establishment module 11 of this application embodiment, runtime timing data of the main network and runtime timing data of multiple distributed energy nodes are obtained, and a benchmark library is established accordingly. The benchmark library includes at least a first set of typical time-series data, a second set of typical time-series data, and a greedy arbitrage scheme set, including:

[0039] Collect historical and real-time data to obtain runtime sequence data of the main network and multiple distributed energy nodes;

[0040] Statistical analysis is performed based on the runtime sequence data of the mainnet to establish the first typical time-series data of the mainnet;

[0041] The runtime sequence data of multiple distributed energy nodes are traversed and statistically analyzed to obtain the second typical time-series data set, wherein the second typical time-series data corresponds one-to-one with the distributed energy nodes;

[0042] Based on the first set of typical time-series data and the second set of typical time-series data, each of the distributed energy nodes is driven to make greedy optimization decisions with the goal of maximizing its own economic operating benefits, generating a greedy arbitrage scheme and its corresponding benchmark benefit for each of the distributed energy nodes, and outputting the greedy arbitrage scheme set.

[0043] In the benchmark establishment module 11 of this application embodiment, the purpose of the above steps is to provide data benchmarks and benefit references for subsequent collaborative optimization, and to support the identification of main network governance requirements, evaluation of node contribution, and construction of global optimization strategies.

[0044] To achieve the above objectives, it is first necessary to collect historical and real-time data to obtain the runtime sequence data of the main grid and multiple distributed energy nodes. The runtime sequence data of the main grid includes time-varying sequences of voltage, frequency, load power, grid connection price, transmission losses, etc.

[0045] The runtime sequence data of distributed energy nodes includes: the power generation, power consumption, energy storage charging and discharging status, local load forecast, and the interaction power between the node and the main grid.

[0046] Specifically, the above two types of data can be collected in real time through sensors, smart meters, SCADA systems, etc., and combined with database storage as historical data to form a runtime sequence dataset with unified time granularity.

[0047] Next, statistical analysis needs to be performed based on the runtime sequence data of the mainnet to establish the first typical time-series data of the mainnet.

[0048] This involves segmenting and clustering the runtime sequence data of the main network, such as dividing it by weekdays / weekends, seasons, and peak / valley periods, and extracting typical time-series data of the same time period through time-series pattern recognition methods such as DTW dynamic time warping.

[0049] For example, the main grid load data is clustered into "typical summer peak curves" and "typical winter valley curves". Each curve contains the average power load, power fluctuation range, and electricity price change characteristics of that period, forming the first typical time series data.

[0050] Then, statistical analysis is performed on the runtime sequence data of multiple distributed energy nodes to obtain the second typical time-series data set, wherein the second typical time-series data corresponds one-to-one with the distributed energy nodes. That is, each distributed energy node is processed individually, its historical / real-time runtime sequence data is traversed, and statistical methods similar to those used for the main grid are adopted. In addition, the individual characteristics of distributed energy nodes need to be highlighted, such as the "typical power output curve on sunny days" for photovoltaic nodes, the "typical charge and discharge response curve" for energy storage nodes, and so on.

[0051] For example, if node A is a photovoltaic + energy storage node, its second typical time series data may include: power generation curves under different light intensities and energy storage charging and discharging strategy curves under different electricity price incentives. The typical data of each distributed energy node is stored independently, forming a set that corresponds one-to-one with the node, namely the second typical time series data set.

[0052] Then, based on the first typical time series data and the second typical time series data set, each of the distributed energy nodes is driven to make greedy optimization decisions with the goal of maximizing its own economic operating benefits, generating a greedy arbitrage scheme and its corresponding benchmark benefit for each of the distributed energy nodes, and outputting the greedy arbitrage scheme set.

[0053] That is, using the first typical time series data and the second typical time series data as input, a single-objective optimization model is constructed for each distributed energy node, where the objective function is to maximize the node revenue, and the node revenue = electricity sales revenue - electricity purchase cost - operating loss; the constraints are the node's own power generation / energy storage capacity limit and the upper limit of the power interaction with the main grid.

[0054] The single-objective optimization model is solved using heuristic algorithms such as particle swarm optimization and dynamic programming to obtain the optimal operation scheme for each distributed energy node under different typical scenarios, such as priority electricity sales during a certain period or energy storage charging during a certain period. The corresponding maximum revenue is calculated and used as the benchmark revenue. After summing the schemes and benchmark revenues of all distributed energy nodes, a greedy arbitrage scheme set is formed, which serves as the individual optimal reference for subsequent collaborative optimization.

[0055] In the requirement and contribution determination module 12 of this application embodiment, based on the benchmark library, the mainnet governance requirements are determined, including:

[0056] Obtain and integrate the node constraint information of multiple distributed energy nodes, and configure them as a virtual single microgrid;

[0057] The second typical time-series data set is integrated to form the aggregated time-series data of the virtual single micronet;

[0058] With the goal of minimizing main grid voltage fluctuations, the virtual single microgrid is scheduled and solved using the aggregated time-series data to obtain an aggregated power reference curve, which is then calibrated as the main grid governance requirement.

[0059] The purpose of the above steps in the requirement and contribution rating module 12 of this application embodiment is to consolidate the distributed nodes, simplify the main network governance objects, and clarify the core requirements of the main network for the distributed cluster, so as to achieve a balance between the overall interests of the main network and the characteristics of the node cluster.

[0060] First, it is necessary to acquire and integrate the node constraint information of multiple distributed energy nodes and configure them as a virtual single microgrid. The node constraint information includes the physical and operational constraints of each distributed energy node. Physical constraints include maximum power generation / consumption, upper limit of energy storage capacity, and charging / discharging efficiency; operational constraints include minimum start / stop time and ramp rate limits.

[0061] The integration logic transforms the constraint information of multiple distributed energy nodes into the overall constraints of the virtual microgrid through aggregation rules. For example: the maximum power generation of the virtual microgrid = the sum of the maximum power generation of each distributed energy node, taking into account the complementarity between nodes, such as the peak-valley offsetting of photovoltaic and wind power output; the total energy storage capacity of the virtual microgrid = the sum of the capacities of each energy storage node, while retaining key individual constraints, such as the special start-stop restrictions of a certain node, which need to be transformed into the scheduling exclusion zone of the virtual microgrid.

[0062] Next, the second typical time-series data set needs to be integrated to form the aggregated time-series data of the virtual single micronet.

[0063] Based on the second typical time-series data set, aggregated time-series curves are generated through time-series overlay and complementarity analysis. For example, the "sunny day power output curve" of photovoltaic nodes, the "high wind power output curve" of wind power nodes, and the "peak electricity consumption curve" of load nodes are overlaid along the time dimension, while considering the correlation between nodes, such as the possibility that wind power may be higher when photovoltaic power output is low on cloudy or rainy days. This ultimately forms the overall power time-series characteristics of the virtual microgrid under different scenarios, such as the "typical aggregated power output curve on weekdays" and the "aggregated load curve in extreme weather," which serve as the aggregated time-series data for a single virtual microgrid. This aggregated time-series data reflects the overall operating pattern of the distributed cluster, providing realistic input conditions for subsequent scheduling solutions.

[0064] Then, with the goal of minimizing the main grid voltage fluctuation, the virtual single microgrid needs to be scheduled and solved using the aggregated time-series data to obtain the aggregated power reference curve, which is then calibrated as the main grid governance requirement.

[0065] The objective function aims to minimize the voltage fluctuation of the main grid, which can be quantified as the sum of squares of voltage deviations from the rated value, fluctuation frequency, and other indicators. The constraints are the aggregated time-series data of the virtual microgrid, the overall constraints of the virtual microgrid, and the security constraints of the main grid, such as the line transmission capacity.

[0066] Specifically, optimization algorithms such as Model Predictive Control (MPC) and genetic algorithms can be used to solve the above objective function, obtaining the power values ​​that the virtual microgrid should inject / absorb into the main grid at different times, forming a continuous aggregated power reference curve. For example, during periods of high main grid voltage, the reference curve requires the virtual microgrid to increase its purchased power to absorb excess power from the main grid; during periods of low voltage, it requires increasing its sold power to supplement the main grid's power gap. This aggregated power reference curve is a quantitative expression of the main grid governance requirements, clarifying how the distributed cluster needs to coordinate and adjust power to ensure the stability of the main grid voltage.

[0067] In the demand and contribution assessment module 12 of this application embodiment, the governance contribution of multiple distributed energy nodes is evaluated, including:

[0068] Based on the topology and electrical parameters of the main network and the multiple distributed energy nodes, simulation analysis is performed in conjunction with the aggregated power reference curve;

[0069] Calculate the voltage-power sensitivity response curve of each distributed energy node to the main grid voltage, and calculate the corresponding response linearity.

[0070] Linear regression analysis is performed on the voltage-power sensitivity response curve to obtain the apparent voltage-power sensitivity, and the normalized response linearity is used as a correction coefficient for weighted correction to obtain the governance contribution of multiple distributed energy nodes.

[0071] In the requirement and contribution rating module 12 of this application embodiment, the purpose of the above steps is to quantify the actual contribution of distributed energy nodes to the main network governance, solve the problem of contribution ambiguity, and associate the characteristics of distributed energy nodes with the benefits of the main network to improve the targeting of governance.

[0072] To achieve the above objectives, it is first necessary to conduct simulation analysis based on the topology and electrical parameters of the main network and the multiple distributed energy nodes, combined with the aggregated power reference curve.

[0073] Specifically, the input data for the simulation analysis includes the topology, electrical parameters, and aggregated power reference curve of the main grid and distributed energy nodes. The topology includes node access locations and line connections; electrical parameters include line impedance, node equivalent impedance, and transformer turns ratio; and the aggregated power reference curve is the target power curve for main grid voltage regulation. A simulation model including the main grid and each distributed energy node is then built using power system simulation tools such as PSCAD and DIgSILENT to simulate the dynamic impact of power changes at each node on the main grid voltage under the constraint of the aggregated power reference curve. For example, the simulation examines the magnitude and timing characteristics of voltage changes at key nodes in the main grid when a distributed energy node increases its power sales by 100kW, providing a data foundation for subsequent sensitivity calculations.

[0074] Next, it is necessary to calculate the voltage-power sensitivity response curve of each distributed energy node to the main grid voltage, and then calculate the corresponding response linearity.

[0075] The voltage-power sensitivity response curve is defined as the relationship between the power change (ΔP) of a distributed energy node and the voltage change (ΔU) of the main grid. By repeatedly adjusting the node power in the simulation and recording the corresponding main grid voltage changes, the ΔU-ΔP curve for each node is fitted. For example, the voltage-power sensitivity response curve of a distributed energy node might show an approximately linear relationship where a 1kW increase in power corresponds to a 0.02kV increase in main grid voltage, while the curve of a photovoltaic node might be more non-linear due to its remote location.

[0076] Response linearity measures the degree of fit between the voltage-power sensitivity response curve and an ideal straight line, typically expressed as the goodness of fit R² or nonlinearity error. A higher linearity indicates a more stable and predictable impact of power regulation on voltage at distributed energy nodes, with R² being closer to 1.

[0077] Furthermore, linear regression analysis needs to be performed on the voltage-power sensitivity response curve to obtain the apparent voltage-power sensitivity, and weighted correction is performed using the normalized response linearity as a correction coefficient to obtain the governance contribution of multiple distributed energy nodes.

[0078] The apparent voltage-power sensitivity is obtained by performing a linear regression on the voltage-power sensitivity response curve, and the slope of the fitted line is the apparent sensitivity coefficient K. It represents the average intensity of the impact of distributed energy node power regulation on the main grid voltage. The larger the K value, the more significant the impact of unit power change on the main grid voltage.

[0079] Then, the response linearity of each distributed energy node, such as the R² value, is normalized, i.e., converted into a normalized linearity coefficient between 0 and 1, which serves as a confidence weight. Nodes with higher response linearity have more valuable sensitivity coefficients. The governance contribution can be defined as: Governance Contribution = Apparent Sensitivity Coefficient K × Normalized Linearity Coefficient. For example, node A has K = 0.03 kV / kW and normalized linearity = 0.9; node B has K = 0.04 kV / kW but a response linearity of 0.6. Their governance contributions are 0.027 and 0.024 respectively, indicating that node A contributes more.

[0080] Calculate the governance contribution metric of each distributed energy node to reflect its comprehensive role in main grid voltage governance.

[0081] The collaborative optimization configuration module 13 is used to obtain the operating specification information of multiple distributed energy nodes and, in combination with the governance contribution, construct a global collaborative optimization strategy, wherein the global collaborative optimization strategy includes at least a collaborative evaluation function and collaborative constraints.

[0082] In the collaborative optimization configuration module 13 of this application embodiment, the construction step of the collaborative evaluation function includes:

[0083] Based on the voltage governance task book of the main network, the governance evaluation rules are extracted and regularized to obtain voltage governance evaluation items.

[0084] Using the governance contribution as a coefficient, the estimated arbitrage returns of multiple distributed energy nodes are weighted and summed to define a weighted economic operation return item, wherein the estimated arbitrage returns are calculated based on a preset arbitrage mathematical model.

[0085] The dimensionless normalized sum of the voltage governance evaluation term and the weighted economic operation benefit term is defined as the collaborative evaluation function.

[0086] In the collaborative optimization configuration module 13 of this application embodiment, the purpose of the above steps is to construct a quantitative evaluation standard that balances the main network governance needs and the economic interests of distributed energy nodes, so as to provide a decision-making basis for global collaborative optimization.

[0087] To achieve the above objectives, it is first necessary to extract and regularize the governance evaluation rules based on the voltage governance task book of the main network to obtain voltage governance evaluation items.

[0088] This involves extracting key evaluation rules from the voltage management task book of the main grid, such as voltage fluctuation amplitude needing to be controlled within ±5% and voltage recovery time not exceeding 10 seconds, and converting these rules into quantifiable indicators, such as the compliance rate of fluctuation amplitude and recovery time.

[0089] These indicators are regularized, that is, the units and value ranges are unified. For example, all indicators are converted into values ​​between 0 and 1, where 1 represents full compliance and 0 represents serious exceedance, forming the final voltage governance evaluation item, which is used to measure the degree to which the collaborative strategy meets the requirements of the main grid voltage stability.

[0090] Next, the estimated arbitrage returns of multiple distributed energy nodes need to be weighted and summed using the governance contribution as a coefficient, and a weighted economic operation return term is defined, wherein the estimated arbitrage returns are calculated based on a preset arbitrage mathematical model.

[0091] This involves pre-setting an arbitrage mathematical model for each distributed energy node. Based on the operating characteristics of the distributed energy node and market electricity prices, the model estimates its economic benefits in collaborative operation, such as the net profit after deducting the cost of purchasing electricity from the revenue from electricity sales, and calculates the estimated arbitrage profit for each node.

[0092] Then, using the governance contribution of each node as a weight—that is, the higher the contribution, the greater the weight—the estimated arbitrage revenue of all distributed energy nodes is weighted and summed to obtain a weighted economic operating revenue item. This step links the revenue calculation to the node's contribution to the main network, reflecting the principle of more contribution, more revenue.

[0093] Then, it is necessary to define the dimensionless normalized sum of the voltage governance evaluation term and the weighted economic operation benefit term as the collaborative evaluation function.

[0094] This involves performing dimensionless normalization on the voltage management evaluation item and the weighted economic operation benefit item to ensure that the value ranges of the two are consistent, such as converting them both into scores of 0-100, thus eliminating the impact of different indicators due to differences in dimensions.

[0095] The normalized voltage management evaluation term and the weighted economic operating benefit term are then added together to obtain the collaborative evaluation function. The higher the value of this function, the better the comprehensive effect of the collaborative strategy in both meeting the main grid voltage management requirements and improving the overall economic benefits of the nodes.

[0096] In the collaborative optimization configuration module 13 of this application embodiment, the collaborative constraints include at least objective hardware constraints and subjective initiative constraints, wherein:

[0097] The objective hardware constraints are defined based on the operational specification information and include at least the energy storage charging and discharging power constraints and the state of charge (SOC) constraints of each distributed energy node.

[0098] The subjective motivation constraint is a two-way satisfaction constraint, including:

[0099] The lower limit constraint of satisfaction is used to limit the estimated arbitrage income of each of the distributed energy nodes under collaborative optimization to be no less than the product of the corresponding benchmark income and the preset first subjective constraint coefficient.

[0100] The upper limit constraint on the degree of satisfaction is used to limit the estimated arbitrage income of each of the distributed energy nodes under collaborative optimization to no more than the product of the corresponding benchmark income and the second subjective constraint coefficient.

[0101] Wherein, the second subjective constraint coefficient is greater than the first subjective constraint coefficient, which is greater than 1.

[0102] In the collaborative optimization configuration module 13 of this application embodiment, the purpose of the above steps is to ensure that the global collaborative optimization strategy is technically feasible while taking into account the enthusiasm of node participation and the overall interests of the main network by setting collaborative constraints.

[0103] To achieve the above objectives, it is first necessary to construct objective hardware constraints. These objective hardware constraints are defined based on the operational specification information and include at least the energy storage charging and discharging power constraints and the state of charge (SOC) constraints for each distributed energy node.

[0104] Specifically, it is necessary to clarify the physical operating limitations of each node based on the operating specifications of the distributed energy nodes, such as the technical parameters in the equipment manual.

[0105] Among these constraints, the energy storage charging and discharging power limits the maximum charging and discharging power of the energy storage node for each time period. For example, a certain energy storage device may charge at a maximum of 50kW and discharge at a maximum of 80kW per hour to prevent excessive power from overloading the device. The State of Charge (SOC) constraint limits the energy storage node's charge range, such as maintaining the SOC between 20% and 90% to avoid overcharging or over-discharging. These constraints are directly derived from the device's hardware capabilities and represent the technical baseline that the collaborative strategy must meet.

[0106] Next, it is necessary to construct subjective motivation constraints. These subjective motivation constraints are two-way satisfaction constraints.

[0107] On the one hand, it is necessary to satisfy the lower limit constraint, which is used to limit the estimated arbitrage income of each distributed energy node under collaborative optimization to be no less than the product of the corresponding benchmark income and the preset first subjective constraint coefficient.

[0108] Specifically, the minimum return for a node to participate in collaboration can be calculated by multiplying its baseline return (maximum return when running alone) within the greedy arbitrage scheme set by a first subjective constraint coefficient greater than 1, such as 1.1. For example, if a node's baseline return is 100 yuan and the first subjective constraint coefficient is 1.1, then the estimated return under collaboration must not be less than 110 yuan. This ensures that the return for a node participating in collaboration is better than running alone, thus increasing its participation incentive.

[0109] On the other hand, it is necessary to satisfy the upper limit constraint, which is used to limit the estimated arbitrage income of each distributed energy node under collaborative optimization to no higher than the product of the corresponding benchmark income and the second subjective constraint coefficient.

[0110] Specifically, the benchmark return can be used as a reference, multiplied by a larger "second subjective constraint coefficient," such as 1.5, which is greater than the first subjective constraint coefficient, to set an upper limit for the estimated arbitrage return of a node. For example, the return of the aforementioned node should not exceed 150 yuan. This is to prevent individual nodes from obtaining unreasonably high returns by excessively occupying mainnet resources or exploiting loopholes in the coordination rules, and to ensure a balance of interests between the mainnet and other nodes.

[0111] In the collaborative optimization configuration module 13 of this application embodiment, the process of determining the value of the second subjective constraint coefficient includes:

[0112] Based on the greedy arbitrage scheme set, the corresponding arbitrage control parameters are extracted and output as the upper limit set of ideal parameters;

[0113] Based on a preset incremental mapping function, the governance contribution is mapped to a subjective constraint coefficient on each parameter dimension, wherein the subjective constraint coefficient is less than or equal to 1.

[0114] Based on the subjective constraint coefficients, the set of upper limits of the ideal parameters is iterated and adjusted, wherein the subjective constraint coefficients are configured differently for each parameter dimension.

[0115] In this embodiment of the application, the purpose of the above steps is to achieve differentiated management by scientifically setting a second subjective constraint coefficient, which allows for a moderate relaxation of the upper limit of income for high-contribution nodes and strict constraints on low-contribution nodes. This prevents nodes from excessively arbitrageing and incentivizes high-contribution nodes to participate more actively in collaboration, ultimately balancing the overall interests of the main network with the incentives for individual nodes.

[0116] Specifically, the first step is to extract the corresponding arbitrage control parameters based on the greedy arbitrage scheme set, outputting a set of ideal parameter upper limits. That is, from the greedy arbitrage scheme set in the benchmark library, key arbitrage control parameters are extracted for each distributed energy node when individually pursuing maximum profit, such as the upper limit of electricity sales, energy storage discharge duration, and peak power interaction with the main grid. These parameters reflect the ideal maximum arbitrage capability of the node without collaborative constraints, and are summarized into a set of ideal parameter upper limits, i.e., the theoretical maximum value under each parameter dimension.

[0117] Next, based on a preset incremental mapping function, the governance contribution needs to be mapped to a subjective constraint coefficient on each parameter dimension, wherein the subjective constraint coefficient is less than or equal to 1.

[0118] This involves pre-setting an incremental mapping function, where the larger the input value, the larger the output value. The governance contribution obtained from the previous steps is input into this incremental mapping function to obtain the subjective constraint coefficients for each node across various parameter dimensions, with values ​​ranging from 0 to 1. For example, nodes with high governance contributions have coefficients closer to 1, while nodes with low contributions have smaller coefficients, reflecting the principle that higher contributions result in more lenient constraint coefficients.

[0119] Then, based on the subjective constraint coefficients, it is necessary to iterate and adjust the upper limit set of the ideal parameters, wherein the subjective constraint coefficients on each parameter dimension are configured differently.

[0120] For each parameter dimension in the set of ideal parameter upper limits, such as electricity sales power and discharge duration, the corresponding subjective constraint coefficient is used to adjust the upper limit of the ideal parameter. For example, the subjective constraint coefficient is multiplied by the ideal upper limit value to obtain the final parameter upper limit. Since different nodes have different governance contributions, their subjective constraint coefficients for the same parameter dimension will differ; the subjective constraint coefficients for the same node in different parameter dimensions may also differ due to different mapping rules. For example, the upper limit adjustment coefficient for electricity sales power of a high-contribution node is 0.9, while that of a low-contribution node is 0.6, ultimately forming the second subjective constraint coefficient for each node.

[0121] In the solution and execution module 14 of this application embodiment, a multi-objective optimization algorithm is used, combined with the global collaborative optimization strategy, to maximize the collaborative evaluation function, and it further includes:

[0122] Based on the voltage management task book of the main network, define the voltage risk critical line;

[0123] The operating voltage indicators of the main network are monitored in real time. If the operating voltage indicators exceed the voltage risk threshold, the subjective initiative constraint is ignored in the maximization solution.

[0124] In this embodiment of the application, the purpose of the above steps is to establish a flexible balance mechanism between the core objective of ensuring the safe and stable operation of the main grid and the enthusiasm of distributed energy nodes to participate by dynamically adjusting and optimizing the constraints.

[0125] To achieve the above objectives, it is first necessary to define the voltage risk threshold based on the voltage management task book of the main grid. Specifically, according to the voltage management task book of the main grid, the boundaries between the safe operating range and the risk range should be clearly defined. For example, if the voltage fluctuation exceeds ±10% of the rated value, the risk threshold can be defined, thus dividing the safe operating range and the risk range of the voltage into two categories. This will serve as the standard for judging whether the main grid is facing an emergency.

[0126] Then, it is necessary to monitor the operating voltage indicators of the main network in real time. If the operating voltage indicators exceed the voltage risk threshold, the subjective initiative constraint is ignored in the maximization solution. Specifically, it is necessary to continuously track the actual operating voltage of the main network, such as the voltage values ​​and fluctuation amplitudes of key nodes. If the monitoring results exceed the risk threshold, that is, the main network enters a risk state, then in the multi-objective optimization solution, the subjective initiative constraint, that is, the upper and lower limits of the estimated arbitrage income of the nodes, is temporarily not considered, and the collaborative strategy is calculated with the goal of meeting the voltage governance requirements first.

[0127] In the solution and execution module 14 of this application embodiment, a multi-objective optimization algorithm is used, combined with the global collaborative optimization strategy, to maximize the collaborative evaluation function, and it further includes:

[0128] Under the aforementioned subjective initiative constraint, if a feasible solution satisfying the voltage management task is not obtained after a preset number of iterative solutions, the subjective initiative constraint is ignored in the maximization solution.

[0129] To achieve this step, you first need to preset an iteration threshold, that is, set a reasonable preset iteration number, such as 50 times, as a standard to judge whether the optimization has reached a deadlock.

[0130] Then, under the premise of strictly adhering to the subjective initiative constraint, if after a preset number of iterations, no feasible solution that can meet the voltage management task book is found, that is, it is impossible to simultaneously meet the main grid voltage requirements and the estimated arbitrage profit constraint of the node, then the subjective initiative constraint is ignored and the solution is re-solved to prioritize ensuring the feasibility of the main grid voltage management target.

[0131] In the solution and execution module 14 of this application embodiment, ignoring the subjective initiative constraint in the maximization solution includes:

[0132] Prioritize ignoring the lower limit constraint of satisfaction. If the maximum solution result is still not obtained after ignoring the lower limit constraint of satisfaction, then ignore the upper limit constraint of satisfaction.

[0133] The "priority ignoring of the lower limit of satisfaction" means that when it is necessary to ignore the subjective initiative constraint, the execution of the "lower limit of satisfaction" constraint is suspended first. This allows the estimated arbitrage income of the node to be temporarily lower than the product of the benchmark income and the first subjective constraint coefficient, and attempts to solve the problem while retaining only the upper limit constraint.

[0134] The further ignoring of the upper limit constraint of satisfaction means that if an effective solution cannot be obtained after ignoring the lower limit constraint of satisfaction, the execution of the upper limit constraint of satisfaction will continue to be suspended. That is, the estimated arbitrage income of the node is allowed to exceed the product of the benchmark income and the second subjective constraint coefficient. The solution is obtained under the condition of no income constraint at all, so as to ensure that the main grid voltage governance target is achieved first.

[0135] It should be noted that the descriptions of each embodiment in the above embodiments have different focuses. For parts that are not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0136] Those skilled in the art will understand that embodiments of the present invention can provide methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0137] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, as well as combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0138] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0139] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0140] Although preferred embodiments of the invention have been described, those skilled in the art, once they have learned the basic inventive concept, can make other changes and modifications to these embodiments.

[0141] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of this invention and its equivalents, this invention also intends to include these modifications and variations.

Claims

1. An energy storage system for synergistic optimization of transformer area voltage management and peak-valley arbitrage, characterized in that, include: The benchmark establishment module is used to acquire runtime timing data of the main network and runtime timing data of multiple distributed energy nodes, and establish a benchmark library accordingly. The benchmark library includes at least a first typical time-series data set, a second typical time-series data set, and a greedy arbitrage scheme set. The demand and contribution calibration module is used to calibrate the main network governance demand based on the benchmark library, and to evaluate the governance contribution of multiple distributed energy nodes accordingly. The collaborative optimization configuration module is used to obtain the operating specification information of multiple distributed energy nodes and, in combination with the governance contribution, construct a global collaborative optimization strategy, wherein the global collaborative optimization strategy includes at least a collaborative evaluation function and collaborative constraints. The solution and execution module is used to maximize the cooperative evaluation function by using a multi-objective optimization algorithm combined with the global cooperative optimization strategy, output the maximized solution result as the globally optimal charging and discharging plan, and distribute it to multiple distributed energy nodes for execution; This includes acquiring runtime sequence data from the main network and multiple distributed energy nodes, and establishing a corresponding benchmark library, including: Collect historical and real-time data to obtain runtime sequence data of the main network and multiple distributed energy nodes; Statistical analysis is performed based on the runtime sequence data of the mainnet to establish the first typical time-series data of the mainnet; The runtime sequence data of multiple distributed energy nodes are traversed and statistically analyzed to obtain the second typical time-series data set, wherein the second typical time-series data corresponds one-to-one with the distributed energy nodes; Based on the first typical time series data and the second typical time series data set, each of the distributed energy nodes is driven to make greedy optimization decisions with the goal of maximizing its own economic operating benefits, generating a greedy arbitrage scheme and its corresponding benchmark benefit for each of the distributed energy nodes, and outputting the greedy arbitrage scheme set. The steps for constructing the collaborative evaluation function include: Based on the voltage governance task book of the main network, the governance evaluation rules are extracted and regularized to obtain voltage governance evaluation items. Using the governance contribution as a coefficient, the estimated arbitrage returns of multiple distributed energy nodes are weighted and summed to define a weighted economic operation return item, wherein the estimated arbitrage returns are calculated based on a preset arbitrage mathematical model. The dimensionless normalized summation of the voltage governance evaluation term and the weighted economic operation benefit term is defined as the collaborative evaluation function. The collaborative constraints include at least objective hardware constraints and subjective motivation constraints, wherein: The objective hardware constraints are defined based on the operational specification information and include at least the energy storage charging and discharging power constraints and the state of charge (SOC) constraints of each distributed energy node. The subjective motivation constraint is a two-way satisfaction constraint, including: The lower limit constraint of satisfaction is used to limit the estimated arbitrage income of each of the distributed energy nodes under collaborative optimization to be no less than the product of the corresponding benchmark income and the preset first subjective constraint coefficient. The upper limit constraint on the degree of satisfaction is used to limit the estimated arbitrage income of each of the distributed energy nodes under collaborative optimization to no more than the product of the corresponding benchmark income and the second subjective constraint coefficient. Wherein, the second subjective constraint coefficient is greater than the first subjective constraint coefficient, which is greater than 1.

2. The energy storage system for coordinated optimization of transformer area voltage management and peak-valley arbitrage as described in claim 1, characterized in that, Based on the aforementioned benchmark library, the mainnet governance requirements are defined, including: Obtain and integrate the node constraint information of multiple distributed energy nodes, and configure them as a virtual single microgrid; The second typical time-series data set is integrated to form the aggregated time-series data of the virtual single micronet; With the goal of minimizing main grid voltage fluctuations, the virtual single microgrid is scheduled and solved using the aggregated time-series data to obtain an aggregated power reference curve, which is then calibrated as the main grid governance requirement.

3. The energy storage system for coordinated optimization of transformer area voltage management and peak-valley arbitrage as described in claim 2, characterized in that, Evaluate the governance contribution of multiple distributed energy nodes, including: Based on the topology and electrical parameters of the main network and the multiple distributed energy nodes, simulation analysis is performed in conjunction with the aggregated power reference curve; Calculate the voltage-power sensitivity response curve of each distributed energy node to the main grid voltage, and calculate the corresponding response linearity. Linear regression analysis is performed on the voltage-power sensitivity response curve to obtain the apparent voltage-power sensitivity, and the normalized response linearity is used as a correction coefficient for weighted correction to obtain the governance contribution of multiple distributed energy nodes.

4. The energy storage system for coordinated optimization of transformer area voltage management and peak-valley arbitrage as described in claim 1, characterized in that, The process of determining the value of the second subjective constraint coefficient includes: Based on the greedy arbitrage scheme set, the corresponding arbitrage control parameters are extracted and output as the upper limit set of ideal parameters; Based on a preset incremental mapping function, the governance contribution is mapped to a subjective constraint coefficient on each parameter dimension, wherein the subjective constraint coefficient is less than or equal to 1. Based on the subjective constraint coefficients, the set of upper limits of the ideal parameters is iterated and adjusted, wherein the subjective constraint coefficients are configured differently for each parameter dimension.

5. The energy storage system for coordinated optimization of transformer area voltage management and peak-valley arbitrage as described in claim 1, characterized in that, The algorithm employs a multi-objective optimization approach, combined with the global collaborative optimization strategy, to maximize the collaborative evaluation function, and further includes: Based on the voltage management task book of the main network, define the voltage risk critical line; The operating voltage indicators of the main network are monitored in real time. If the operating voltage indicators exceed the voltage risk threshold, the subjective initiative constraint is ignored in the maximization solution.

6. The energy storage system for coordinated optimization of transformer area voltage management and peak-valley arbitrage according to claim 1, characterized in that, The algorithm employs a multi-objective optimization approach, combined with the global collaborative optimization strategy, to maximize the collaborative evaluation function, and further includes: Under the aforementioned subjective initiative constraint, if a feasible solution satisfying the voltage management task is not obtained after a preset number of iterative solutions, the subjective initiative constraint is ignored in the maximization solution.

7. The energy storage system for coordinated optimization of transformer area voltage management and peak-valley arbitrage according to claim 1, characterized in that, Ignoring the subjective motivation constraint in the maximization solution includes: Prioritize ignoring the lower limit constraint of satisfaction. If the maximum solution result is still not obtained after ignoring the lower limit constraint of satisfaction, then ignore the upper limit constraint of satisfaction.

Images