Multi-region collaborative mobile alternative energy storage optimal configuration and operation method and device

Abstract

The invention relates to a multi-region collaborative mobile alternative energy storage optimal configuration and operation method and device. The method comprises the following steps: firstly, analyzing regional load characteristics, classifying load curves by adopting an ordered clustering method, and generating multi-scene typical daily load data; determining the deployment position of the energy storage unit by using an optimization algorithm in the equipment deployment stage; in the operation stage, based on the load curve, the electricity price and the energy storage state, the two-stage optimization configuration model is adopted, and the charging and discharging power of the energy storage unit in multiple periods is optimized. A basic configuration scheme of the energy storage equipment is generated in the equipment deployment stage, real-time scheduling optimization is carried out in the running stage, and quick response to load fluctuation and cost minimization are achieved. Through the architecture of centralized optimization and distributed execution, each region can make a decision independently based on local load and electricity price information, and meanwhile, the scheduling information is shared to improve the global optimization capability, so that the flexibility and economy of the system in multiple scenes are ensured.

Description

Technical Field

[0001] This invention relates to the field of power system energy storage optimization and dispatch technology, specifically to a method and apparatus for the optimized configuration and operation of multi-regional collaborative mobile alternative energy storage. Background Technology

[0002] With rapid economic development and improved living standards, the load demand of regional power distribution networks is constantly increasing, especially during seasonal peak periods such as summer and winter, when the distribution network often faces power supply bottlenecks and safety pressures. Traditional solutions typically involve upgrading distribution network equipment to increase power supply capacity, but this is not only costly and time-consuming to construct, but also significantly impacts the operation of the existing power grid.

[0003] In recent years, the development of energy storage technology has provided new solutions to this problem. Energy storage devices can charge during low-load periods and discharge during high-load periods, thereby achieving peak shaving and valley filling and improving grid operating efficiency. However, traditional energy storage systems configured in a single area have significant economic shortcomings, failing to fully utilize the flexibility and cross-regional dispatch potential of energy storage devices. If the configuration and dispatch of energy storage devices are not reasonable, especially if they cannot fully play their role during peak load periods, it will lead to low utilization rates of energy storage devices, thus affecting the return on investment. In addition, there are significant differences in load characteristics between different regions. For example, some regions may face high load demand in summer, while other regions may face high load demand in winter. This load difference between regions provides an opportunity for cross-regional energy storage dispatch. By migrating energy storage devices between different regions, charging can be performed in low-demand areas and discharging in high-demand areas, thereby achieving peak-valley balance and improving the utilization rate and economy of energy storage devices. Therefore, a more flexible and efficient energy storage configuration and operation method is needed to adapt to the complex needs of modern distribution networks.

[0004] In the prior art, research on the configuration and optimization of mobile energy storage has made some progress. For example, Chinese patent document CN116995706 A discloses a multi-objective optimization configuration method for mobile energy storage in distribution networks. This method constructs a multi-objective optimization model with the objectives of maximizing load peak reduction, maximizing power supply reliability improvement, maximizing energy storage device utilization, and minimizing the number of configurations. An enumeration method is then used to solve the model, thereby determining the number of mobile energy storage devices, their power level, and discharge time. Although this method can achieve capacity configuration optimization for mobile energy storage within a single distribution area, it mainly focuses on capacity planning within a fixed area, lacks a cross-regional collaborative scheduling mechanism, and fails to fully consider the differences in load characteristics and peak-shaving features between different areas. Furthermore, it is difficult to adapt to real-time operation optimization under complex conditions in multiple scenarios, and it does not involve intraday scheduling and economic analysis during the operation phase, resulting in limitations on the spatiotemporal flexibility and utilization rate of energy storage devices.

[0005] Chinese patent document CN118117623 B discloses a joint optimization method combining stationary and mobile energy storage systems. This scheme establishes a dual-objective optimization model under normal and fault scenarios by analyzing historical operating data of the distribution network. The model aims to minimize economic costs and resilience risks, and introduces a Nash equilibrium game model to solve for energy storage configuration and operation strategies, achieving synergistic optimization of stationary and mobile energy storage. However, this technical solution mainly focuses on the global trade-off between economic efficiency and resilience, failing to establish a model for the migration and dynamic scheduling of mobile energy storage across multiple regions, and also lacking a typical daily scenario analysis method based on ordered clustering, resulting in insufficient characterization of seasonal load differences across multiple regions. Furthermore, this scheme primarily considers emergency power supply optimization under extreme or fault scenarios, with limited attention paid to time-of-use pricing response and operating cost optimization during daily operation.

[0006] In summary, existing technologies for the configuration and operation optimization of mobile energy storage still have the following shortcomings: First, they lack comprehensive utilization of load complementarity characteristics across multiple regions, making it difficult to achieve cross-regional energy storage migration and sharing; second, they lack a spatiotemporal collaborative optimization mechanism based on typical daily scenarios, making it difficult to support dynamic scheduling under multiple scenarios; and third, they fail to simultaneously consider the collaborative optimization of the energy storage system during the configuration and operation phases, resulting in room for improvement in the utilization rate of energy storage devices and the economic efficiency of the system. Therefore, there is an urgent need for a two-stage optimization configuration and operation method that can achieve collaborative mutual assistance across multiple regions based on typical load scenarios, in order to fully leverage the flexibility and economic potential of mobile energy storage systems and improve overall power supply reliability and comprehensive benefits. Summary of the Invention

[0007] To address the shortcomings of existing single-region energy storage systems in terms of economic efficiency and the failure to fully utilize the flexibility and cross-regional dispatch potential of energy storage devices, this invention provides a method and apparatus for the optimized configuration and operation of mobile alternative energy storage in a multi-regional collaborative manner. By introducing mobile alternative energy storage devices, this method and apparatus achieve efficient configuration and dynamic operation optimization of energy storage systems, alleviate load fluctuations in the distribution network, and delay grid upgrades.

[0008] To achieve the above objectives, the technical solution of the present invention is as follows:

[0009] A method for optimizing the configuration and operation of mobile alternative energy storage with multi-regional collaboration, comprising the following steps:

[0010] S1. Use ordered clustering to classify the annual load curve of the region, determine typical load scenarios and scenario probabilities, and form typical days for multiple scenarios;

[0011] S2. Establish a two-stage optimization configuration model: The first stage is based on the probability load curves of typical days in multiple scenarios to obtain the spatiotemporal characteristics of regional load and determine the deployment location of energy storage devices with the goal of minimizing configuration costs; The second stage combines the probability load curves of typical days in multiple scenarios, the time-based electricity price response mechanism, and the real-time operating status of energy storage units to optimize the power purchase / load shedding at the current deployment location and the charging and discharging power of energy storage devices with the goal of minimizing operating costs.

[0012] S3. Utilize a two-stage optimization configuration model to deploy the location of energy storage devices and optimize their operating status.

[0013] As a further improvement to the above solution, step S1 specifically includes the following steps:

[0014] S11. Obtain annual load curves for multiple regions: Collect annual time-period load data for multiple regions and form a complete annual load curve dataset according to the time series.

[0015] S12. Use ordered clustering to classify the load curves of the two regions;

[0016] S13. For each load curve obtained by ordered clustering, take the average time dimension of all date load curves in the same category, calculate the typical daily load value under each scenario, form a typical daily load curve for multiple scenarios, and calculate the scenario probability of each scenario, that is, the proportion of days included in this scenario to the total number of days in the year.

[0017] As a further improvement to the above solution, step 12 specifically includes the following steps:

[0018] S121. Determine the degree of intra-class variability based on the Fraser distance and the sum of squared deviations.

[0019] S1211, Calculate the Freiser distance

[0020] The annual load curve is segmented according to time series, and a sub-interval of load data after segmentation is determined as {x}. i , x i+1 ,…, x j}, j>i.

[0021] Where, x i Let x be the load curve for day i. i x i+1 Until x j These are load curve data points arranged in a time series.

[0022] The index set of the load data sub-intervals is as follows:

[0023] P = {i, i+1,…, j};

[0024] Where i, i+1,…, j are the time indices of the load data within the sub-interval.

[0025] The mean Friesian distance of this set of load sub-intervals is determined using the following formula:

[0026] ;

[0027] in, The mean Friesian distance of the load sub-interval; The number of combinations represents the number of pairs of any two curves selected from the j-i+1 curves; i and j are the time indices of the load data within this sub-interval. For from x i To x j The sum of the Fréchet distances between any two load curves; l is the index number corresponding to all curves.

[0028] S1212, Calculate the degree of intra-class differences

[0029] Using the sum of squared deviations as the intraclass diameter, the intraclass variability of the load subintervals is calculated using the following formula:

[0030] ;

[0031] in, Within-class sum of squared deviations of load from day i to day j; The number of combinations represents the number of pairs of any two curves selected from the (j-i+1) curves; t is time, and X... t Let Frège be the Frescher distance between the two load curves in the t-th group of the load sub-interval; is the mean Frescher distance of the load sub-interval.

[0032] S122. Based on the degree of intra-class differences, construct a segmentation objective function and generate the optimal segmentation scheme.

[0033] S1221. The piecewise objective function is shown in the following formula:

[0034] ;

[0035] in, To sum the intraclass diameters of all load sub-intervals throughout the year, This represents the partitioning scheme for the annual load data under a given number of categories r; n is the number of load sample data, r is the number of categories, and i v Represent the dividing point between classes, denoted as .

[0036] S1222. Based on the piecewise objective function, the optimal piecewise scheme is generated using the following formula:

[0037] ;

[0038] Where p(n,r) represents the optimal segmentation scheme, n is the number of load sample data, r is the number of categories; and i is the time index of the load data in this sub-interval. To sum the intraclass diameters of all load sub-intervals throughout the year, Let n be the partitioning scheme for the annual load data under a given number of categories r, where n is the number of load sample data.

[0039] S1223. Based on the aforementioned optimal segmentation scheme, the optimal segmentation is performed using the Fisher algorithm, as shown in the following formula:

[0040] Using Fisher's algorithm, the two recursive formulas for optimal splitting are as follows:

[0041] ;

[0042] in, This represents the minimum sum of squared differences within each category when the annual load is divided into two categories; p(n,2) represents the optimal segmentation scheme; n is the number of load sample data. Let be the sum of squared intra-class differences from sample 1 to sample (j-1). Let be the sum of squared intra-class differences from the j-th sample to the n-th sample; This represents the minimum sum of squared differences within each of the r categories when the annual load is divided into r categories. Let be the sum of squared intra-class differences from the (j-1)th sample to the (r-1)th sample.

[0043] S123. Determine the optimal number of clusters, optimize the solution by calculating the rate of change of slope under adjacent categories, and the point of abrupt change in the rate of change of slope is the optimal number of clusters r. id .

[0044] S1231. Calculate the rate of change of the slope of the objective function under adjacent classification numbers using the following formula:

[0045] ;

[0046] in, The slope of the deviation between the number of categories r and the number of categories r+1; This represents the minimum sum of squared differences within each class when the annual load is divided into r+1 categories; n is the number of load sample data; r is the number of categories; This represents the minimum sum of squared differences within each of the r categories when the annual load is divided into r categories.

[0047] S1232. Define the slope change rate of the adjacent cluster number as dif:

[0048] ;

[0049] S1233. Determine the optimal number of clusters r using the following formula. id :

[0050] ;

[0051] in, For the number of clusters is The rate of change of the slope at that time; The number of clusters; The rate of change of the slope of the objective function when the number of clusters is R. This represents the rate of change of the slope under the previous cluster number. This represents the rate of change of the slope under the next cluster number; This represents the optimal number of clusters.

[0052] As a further improvement to the above scheme, the typical daily load value is calculated using the following formula:

[0053] ;

[0054] Among them, L et (n,t) represents the load value at time t on a typical load day in the nth clustering scenario, in kW; n L represents the number of days included in the nth load scenario. e (y,t) represents the load value at time t on day y; T represents the total time of the day.

[0055] As a further improvement to the above technical solution, the first stage obtains the spatiotemporal characteristics of regional load based on the probability load curves of typical days in multiple scenarios, and determines the deployment location of energy storage equipment with the goal of minimizing configuration costs, including:

[0056] S21. Based on typical daily load curves of multiple scenarios, obtain the spatiotemporal characteristics of regional load.

[0057] S22. Build a cross-regional migration and dynamic scheduling model for energy storage, solve the model, and determine the deployment location of mobile alternative energy storage units.

[0058] Step S22 specifically includes the following steps:

[0059] S221. With the goal of minimizing the configuration cost of mobile alternative energy storage units, the objective function for the deployment phase is constructed as shown in the following equation:

[0060] ;

[0061] Where F1 is the objective function for the pre-deployment stage; C E C P These are the unit capacity cost and unit power cost of mobile alternative energy storage, respectively; E a P a These are the rated capacity and power of mobile alternative energy storage, respectively.

[0062] S222. The mobile alternative energy storage device has the ability to move across regions, and the migration state constraints of the energy storage device unit are established; the migration state constraints include regional docking mutual exclusion constraints and migration state consistency constraints.

[0063] The area docking mutual exclusion constraint is shown in the following formula:

[0064] ;

[0065] Where ωA t and ωB t represent whether the mobile energy storage device is docked in region A and region B at time t, respectively, and T is the total time within the scheduling cycle.

[0066] The consistency constraint for the migration state is shown in the following formula:

[0067] ;

[0068] Wherein, βA t and βB t represent whether the mobile alternative energy storage device leaves region A and region B at time t, respectively.

[0069] S223. Solve the cross-regional migration and dynamic scheduling model of energy storage, determine the deployment location of mobile alternative energy storage units, and output the cross-regional migration and scheduling strategy of energy storage.

[0070] As a further improvement to the above solution, the second stage combines the probability load curves of typical days in multiple scenarios, the time-of-use electricity price response mechanism, and the real-time operating status of the energy storage unit. With the goal of minimizing operating costs, it optimizes the power purchase / load shedding at the current deployment location and the charging and discharging power of the energy storage equipment, including:

[0071] S31. With the goal of minimizing the operating cost across multiple time periods, and taking into account both the cost of electricity purchase and the cost of load shedding penalties, construct an objective function for the operation phase.

[0072] S32. Establish operating constraints, which include transformer area capacity constraints, charging and discharging power constraints, power balance constraints, and state of charge constraints.

[0073] S33. Based on the objective function and operational constraints of the operational phase, and using the mixed-integer linear programming method, model with YALMIP and solve with GUROBI to output the power curves of power purchase / load shedding at the current deployment location in multiple scenarios, as well as the charging and discharging power curves of the energy storage device.

[0074] As a further improvement to the above scheme, the objective function of the running phase is shown in the following equation:

[0075] ;

[0076] Where M represents the total number of months; N represents the total number of scenarios; p represents the probability of region i in month m - scene n; t Let be the electricity price at time t; T be 24 hours; and I be the set of regions A and B. Pbuy i,m,n,t represents the electricity purchased by the energy storage deployment location at time t in the nth cluster scenario of month m in region i; Pshed i,m,n,t represents the load shedding amount of the energy storage deployment location at time t in the nth cluster scenario of month m in region i; C shed Load shedding penalty coefficient.

[0077] Step S32 specifically includes the following steps:

[0078] S321. Establish the power supply constraints for mobile alternative energy storage using the following formula:

[0079] ;

[0080] Where Pload i,t is the actual load of region i at time t; m is the month index; and t is the time. The maximum power supply capacity of the transformer substation in region i, where I is the set of regions A and B.

[0081] S322. Establish the charging and discharging power constraints for mobile alternative energy storage using the following formula:

[0082] ;

[0083] in, The charging power of a mobile alternative energy storage device at time t; This refers to the maximum charging power of mobile alternative energy storage devices. This represents the maximum discharge power of the mobile alternative energy storage device. For 0-1 variables, in =1 indicates that the mobile energy storage is charged at time t; For 0-1 variables, in When = 1, it means that the mobile energy storage is charged at time t. This constraint means that the charging and discharging of the mobile energy storage cannot be carried out simultaneously.

[0084] S323. Establish the power balance constraints for the current deployment location of mobile alternative energy storage using the following formula:

[0085] ;

[0086] in, The power consumption at the current deployment location; Discharge power for mobile alternative energy storage; The load at the current deployment location; This is the load shedding amount; Charging power for mobile alternative energy storage.

[0087] S324. Establish the state-of-charge constraints for mobile alternative energy storage using the following formula:

[0088] ;

[0089] Among them, SOC min SOC max These are the minimum and maximum values ​​of the state of charge for mobile alternative energy storage, respectively. For mobile alternative energy storage, the state of charge at time t satisfies =E t / E m E t E represents the remaining charge of the mobile energy storage at time t. m This refers to the rated capacity of the energy storage device.

[0090] S325. Establish the continuity constraint of the state of charge of mobile energy storage using the following formula:

[0091] ;

[0092] in, The state of charge of mobile alternative energy storage at time t; , These represent the charging and discharging power of mobile alternative energy storage at time t, respectively; η ch η dch These represent the charging and discharging efficiencies of mobile energy storage, respectively.

[0093] In a second aspect of the invention, a multi-regional collaborative mobile alternative energy storage optimization configuration and operation device is disclosed, the device comprising: a clustering module, a model building module and a segmented optimization module.

[0094] The clustering module is used to classify the annual load curve of a region using an ordered clustering method, determine typical load scenarios and scenario probabilities, and form typical days for multiple scenarios.

[0095] The model building module is used to establish a two-stage optimization configuration model. In the first stage, based on the probability load curves of typical days in multiple scenarios, the spatiotemporal characteristics of regional load are obtained, and the deployment location of energy storage devices is determined with the goal of minimizing configuration costs. In the second stage, combined with the probability load curves of typical days in multiple scenarios, the time-based electricity price response mechanism, and the real-time operating status of energy storage units, the power purchase / load shedding at the current deployment location and the charging and discharging power of energy storage devices are optimized with the goal of minimizing operating costs.

[0096] The segmented optimization module is used to deploy the location of energy storage devices and optimize the operating status of energy storage devices using a two-stage optimization configuration model.

[0097] In a third aspect of the invention, an electronic device is disclosed, comprising: at least one processor; and a memory storing instructions that, when executed by the at least one processor, cause the at least one processor to perform the multi-regional collaborative mobile alternative energy storage optimization configuration and operation method as described above.

[0098] In a fourth aspect of the invention, a machine-readable storage medium is disclosed, which stores executable instructions that, when executed, cause the machine to perform the multi-regional collaborative mobile alternative energy storage optimization configuration and operation method described above.

[0099] Compared with the prior art, the advantages of the present invention are:

[0100] (1) This invention proposes a method and device for optimizing the configuration and operation of mobile alternative energy storage in multiple regions. By constructing a flexible mutual assistance and emergency supply mechanism among multiple regions, dynamic scheduling of energy storage equipment in different regions is realized. This method not only improves the utilization rate of energy storage equipment, but also effectively alleviates the load fluctuation of the distribution network through cross-regional peak-valley mutual assistance, reduces the overall cost, improves the reliability of power supply, and thus delays the need for grid upgrades.

[0101] (2) This invention addresses the bottlenecks and power supply security pressures in distribution areas caused by seasonal peak loads in different regions. By using mobile alternative energy storage as a flexible energy storage resource, it ensures the reliability of power supply to critical loads and improves overall economic efficiency. Compared with traditional single-region independent configuration, this invention introduces multi-region collaborative configuration, utilizing the monthly portability of energy storage between multiple regions to achieve capacity sharing and peak-valley mutual assistance. During off-peak hours, mobile alternative energy storage devices can be charged in low-demand areas and released in high-demand areas to alleviate load fluctuations in the distribution network. This method improves the utilization rate and economic efficiency of energy storage devices by constructing flexible mutual assistance and emergency supply in multiple regions, forming a scheduling scheme that takes into account the power supply security of multiple scenarios, delaying or replacing the upgrading and transformation of transmission and transformation facilities, and is suitable for urban distribution networks with seasonal peak loads and inter-regional peak shifting characteristics. Attached Figure Description

[0102] Figure 1 This is a flowchart of the method for optimizing the configuration and operation of mobile alternative energy storage with multi-regional collaboration in this invention.

[0103] Figure 2 Five typical daily load curves after effective clustering of region A;

[0104] Figure 3 These are four typical daily load curves after effective clustering of region B;

[0105] Figure 4 A diagram illustrating the cross-regional migration and dynamic scheduling of mobile alternative energy storage devices;

[0106] Figure 5 The daily power curve for a typical day in region A;

[0107] Figure 6 This is the daily power curve for a typical day in region B. Detailed Implementation

[0108] To provide a better understanding of the structural features and effects achieved by the present invention, a detailed description is provided below, accompanied by preferred embodiments and accompanying drawings:

[0109] Example 1

[0110] This invention proposes a multi-regional collaborative method for the optimized configuration and operation of mobile alternative energy storage. It constructs an energy storage optimization framework based on a multi-regional collaborative mechanism, introducing a "multi-scenario + two-stage optimization" mechanism, designing different optimization strategies for the deployment and operation scheduling of mobile alternative energy storage devices. In the deployment stage, ordered clustering is used to classify regional load curves, generating typical daily load data for multiple scenarios. An optimization model based on cost minimization is then constructed to determine the deployment location and capacity of mobile alternative energy storage units. In the operation stage, load and electricity price data are collected in real time, and a two-stage optimization configuration model is used to optimize the charging and discharging power of energy storage units across multiple time periods. The two-stage optimization is achieved using an optimization algorithm framework, improving the adaptability and economy of the scheduling strategy.

[0111] like Figure 1 The method for optimizing the configuration and operation of multi-regional collaborative mobile alternative energy storage, as shown, includes the following steps:

[0112] S1. Use ordered clustering to classify the annual load curve of the region, determine typical load scenarios and scenario probabilities, and form typical days for multiple scenarios.

[0113] S2. Establish a two-stage optimization configuration model: The first stage is based on the probability load curves of typical days in multiple scenarios to obtain the spatiotemporal characteristics of regional load, and determines the deployment location of energy storage devices with the goal of minimizing configuration costs; The second stage combines the probability load curves of typical days in multiple scenarios, the time-based electricity price response mechanism, and the real-time operating status of energy storage units to optimize the power purchase / load shedding at the current deployment location and the charging and discharging power of energy storage devices with the goal of minimizing operating costs.

[0114] S3. Using a two-stage optimization configuration model, the location of energy storage devices is deployed, and the operating status of the energy storage devices is optimized. The operating status of the energy storage devices includes the charging and discharging power of the energy storage devices.

[0115] This invention employs energy storage units with autonomous mobility, enabling flexible deployment across different regions. Compared to traditional fixed energy storage systems, this mobile alternative energy storage device is no longer limited to a single geographical location. It can dynamically migrate and respond to energy based on multi-dimensional spatiotemporal information such as load changes and electricity price signals, thereby improving the spatial scheduling flexibility and depth of participation of energy storage resources. This invention proposes a two-stage optimization solution strategy, corresponding to two time scales: initial equipment configuration and intraday operation scheduling. The first stage is the equipment deployment stage, mainly addressing the location selection problem of the energy storage device. The second stage is the operation stage, which optimizes the charging and discharging power of the energy storage unit in multiple time periods based on load curves, electricity prices, and energy storage status, with the goal of minimizing costs.

[0116] In this embodiment, the energy storage device is a mobile alternative energy storage unit. A mobile alternative energy storage unit refers to a unit that is directly connected to the public power grid, primarily intended to delay or replace upgrades to transmission and transformation facilities. The functional positioning of a mobile alternative energy storage unit is equivalent to that of power grid infrastructure. It is mainly used in temporary, seasonal load scenarios or resource-constrained scenarios, providing functions such as capacity support, voltage regulation, and emergency backup to solve short-term power supply bottlenecks or improve grid reliability. A typical characteristic of mobile alternative energy storage units is their combination of temporary and reusable features, optimizing costs while delaying grid investment.

[0117] As a further improvement to the above scheme, for the annual load data of the region, this invention adopts an ordered clustering method. First, the load curves are classified into typical load scenarios with common characteristics and the probability of occurrence of each scenario is calculated. Then, the load curves of each scenario are averaged over time to generate typical daily load curves for the corresponding scenarios. Finally, typical days for multiple scenarios are obtained, providing a load data basis for subsequent energy storage optimization.

[0118] Step S1 specifically includes the following steps:

[0119] S11. Obtain annual load curves for multiple regions.

[0120] Collect year-round time-period load data from multiple regions (such as two regions), and form a complete year-round load curve dataset according to the time series, covering different seasons, time periods and electricity consumption scenarios, to provide basic data for subsequent cluster analysis.

[0121] S12. Use ordered clustering to classify the load curves of the two regions.

[0122] The ordered clustering algorithm includes three steps: calculating the degree of difference within each cluster, calculating the objective function, and optimizing the optimal number of clusters.

[0123] Step 12 specifically includes the following steps:

[0124] S121. Determine the degree of intra-class variability based on the Fraser distance and the sum of squared deviations.

[0125] S1211, Calculate the Freiser distance

[0126] The annual load curve is segmented according to time series, and a sub-interval of load data after segmentation is determined as {x}. i , x i+1 ,…, x j}, j>i.

[0127] Where, x i Let x be the load curve for day i. i x i+1 Until x jIt is a load curve data point arranged in time series (e.g., load curve of day i, day i+1...day j); j>i indicates that this sub-interval contains at least two consecutive load data units, which are used to analyze the degree of intra-class differences through ordered clustering algorithm, thereby realizing the classification of the annual load curve.

[0128] The index set of the load data sub-intervals is as follows:

[0129] P = {i, i+1,…, j};

[0130] Where i, i+1,…, j are the time indices of the load data within the sub-interval, such as the load curves for day i, day i+1,…, day j; the index set is used to identify the load sub-intervals from the i-th data to the j-th data in subsequent analysis, so as to calculate the degree of intra-class difference within the interval, and is a key symbol for defining the analysis unit in ordered clustering algorithms.

[0131] The mean Friesian distance of this set of load sub-intervals is determined using the following formula:

[0132] ;

[0133] in, The mean Friesian distance of the load sub-interval; The number of combinations represents the number of pairs of any two curves selected from the j-i+1 curves; i and j are the time indices of the load data within this sub-interval. For from x i To x j The sum of the Friesian distances between any two load curves; l is the index number of all curves. The smaller the Friesian distance, the higher the similarity between the two curves.

[0134] S1212, Calculate the degree of intra-class differences

[0135] In ordered clustering, the sum of squared deviations is used as the intra-cluster diameter, and the intra-cluster variability of the load sub-intervals is calculated using the following formula:

[0136] ;

[0137] in, Within-class sum of squared deviations of load from day i to day j; The number of combinations represents the number of pairs of any two curves selected from the (j-i+1) curves; t is time, and X... t Let Frège be the Frescher distance between the two load curves in the t-th group of the load sub-interval; This represents the mean Frescher distance across load sub-intervals. The Frescher distance defines the similarity of data within each load class segment.

[0138] S122. Based on the degree of intra-class differences, construct a segmentation objective function and generate the optimal segmentation scheme.

[0139] The sum of the intra-class diameters of each load sub-interval throughout the year is defined as the piecewise objective function. The smaller the piecewise objective function, the smaller the intra-class differences of the segmented data, and the more reasonable the classification. That is, the division method with the smallest piecewise objective function is the optimal segmentation scheme.

[0140] S1221. The piecewise objective function is shown in the following formula:

[0141] ;

[0142] in, To sum the intraclass diameters of all load sub-intervals throughout the year, This represents the partitioning scheme for the annual load data under a given number of categories r; n is the number of load sample data, r is the number of categories, and i v Represent the dividing point between classes, denoted as The smaller the objective function of the segmentation, the smaller the sum of squared deviations of the r groups of load data after segmentation, indicating that the classification is more reasonable.

[0143] S1222. Based on the piecewise objective function, the optimal piecewise scheme is generated using the following formula:

[0144] ;

[0145] Where p(n, r) represents the optimal segmentation scheme, n is the number of load sample data, r is the number of categories; and i is the time index of the load data in this sub-interval. To sum the intraclass diameters of all load sub-intervals throughout the year, Let n be the partitioning scheme for the annual load data under a given number of categories r, where n is the number of load sample data and r is the number of categories.

[0146] S1223. Based on the aforementioned optimal segmentation scheme, the optimal segmentation is performed using the Fisher algorithm, as shown in the following formula:

[0147] Using Fisher's algorithm, the two recursive formulas for optimal splitting are as follows:

[0148] ;

[0149] in, This represents the minimum sum of squared differences within each category when the annual load is divided into two categories; p(n,2) represents the optimal segmentation scheme; n is the number of load sample data. Let be the sum of squared intra-class differences from sample 1 to sample (j-1). Let be the sum of squared intra-class differences from the j-th sample to the n-th sample; This represents the minimum sum of squared differences within each of the r categories when the annual load is divided into r categories. Let be the sum of squared intra-class differences from the (j-1)th sample to the (r-1)th sample.

[0150] To reduce the computational cost of solving the piecewise objective function, Fisher's optimal segmentation algorithm is used, achieving efficient optimal segmentation through the above recursive formula. The recursive formula significantly reduces the computational cost while maintaining accuracy by narrowing the search range for the optimal segment in stages.

[0151] S123. Determine the optimal number of clusters. Optimize by solving for the rate of change of slope between adjacent clusters; the point of abrupt change in the rate of change of slope, r, is the optimal number of clusters, r. id .

[0152] S1231. Calculate the rate of change of the slope of the objective function under adjacent classification numbers using the following formula:

[0153] ;

[0154] in, The slope of the deviation between the number of categories r and the number of categories r+1; This represents the minimum sum of squared differences within each class when the annual load is divided into r+1 categories; n is the number of load sample data; r is the number of categories; This represents the minimum sum of squared differences within each of the r categories when the annual load is divided into r categories.

[0155] S1232. Define the slope change rate of the adjacent cluster number as dif:

[0156] ;

[0157] S1233. Determine the optimal number of clusters r using the following formula. id :

[0158] ;

[0159] in, For the number of clusters is The rate of change of the slope at that time; The number of clusters; The rate of change of the slope of the objective function when the number of clusters is R. This represents the rate of change of the slope under the previous cluster number. This represents the rate of change of the slope under the next cluster number; This represents the optimal number of clusters.

[0160] S13. Calculate typical daily load curves for multiple scenarios.

[0161] For each load curve obtained by ordered clustering, the average time dimension of all date load curves in the same category is taken to calculate the typical daily load value under each scenario, forming a typical daily load curve for multiple scenarios, and the scenario probability of each scenario is calculated, that is, the proportion of days included in this scenario to the total number of days in the year.

[0162] The load curve classification obtained through ordered clustering can divide the load of multiple regions (such as region A and region B) into k categories in an orderly manner. Furthermore, the load trend and peak and valley positions are basically the same in each scenario. Therefore, the load curve data of each typical day (i.e., the load value of a typical day) can be obtained by averaging the load under each scenario.

[0163] The typical daily load value is calculated using the following formula:

[0164] ;

[0165] Among them, L et (n,t) represents the load value at time t on a typical load day in the nth clustering scenario, in kW; n L represents the number of days included in the nth load scenario. e (y,t) represents the load value at time t on day y; T represents the total time of the day.

[0166] Based on the typical daily load values ​​for various scenarios calculated by the above formula, typical daily load curves for multiple scenarios are formed, and the proportion of days included in each type of curve to the total number of days in the year is statistically analyzed to obtain the probability load curve for typical days.

[0167] As a further improvement to the above scheme, this invention takes a multi-regional collaborative mobile alternative energy storage system as the core of optimization, integrates energy storage devices and distribution network load resources in different regions, and adopts a two-stage, multi-scenario decision-making architecture, divided into a device deployment stage and an operation and scheduling stage, and uses optimization algorithms for intelligent decision-making and optimized configuration in each stage. This invention proposes a multi-regional collaborative mobile alternative energy storage optimization configuration and operation method by introducing alternative energy storage devices with migration power and scheduling capabilities, combined with the load characteristics of two regions. In this method, a two-stage modeling approach is adopted. The first stage realizes the monthly optimal location of alternative energy storage devices, and the second stage optimizes the charging and discharging power of energy storage units in multiple time periods, effectively improving the utilization rate and economy of energy storage devices.

[0168] Specifically, the first stage obtains the spatiotemporal characteristics of regional load based on the probabilistic load curves of typical days in multiple scenarios, and determines the deployment location of energy storage equipment with the goal of minimizing configuration costs, including:

[0169] S21. Based on typical daily load curves of multiple scenarios, obtain the spatiotemporal characteristics of regional load.

[0170] The spatiotemporal characteristics include the peak and valley periods and fluctuation range of load in each scenario in the time dimension (such as high load from 2-4 pm in summer and low load from 0-6 am), and the load differences in different regions in the same scenario in the spatial dimension (such as high load in region A in winter and high load in region B in summer).

[0171] S22. Build a cross-regional migration and dynamic scheduling model for energy storage, solve the model, and determine the deployment location of mobile alternative energy storage units.

[0172] Using the spatiotemporal characteristics of regional load extracted in step S21 as constraints, a mathematical model for cross-regional migration and dynamic scheduling of energy storage is constructed. In this model, firstly, migration constraints for energy storage units are defined, such as regional docking node limitations; then, scheduling decision variables are established, such as the dwell time of energy storage in each region and charging / discharging triggering conditions; finally, objective functions are defined, such as minimizing migration costs and maximizing energy storage utilization, along with constraints such as regional grid capacity and the state of charge range of energy storage.

[0173] Step S22 specifically includes the following steps:

[0174] S221. With the goal of minimizing the configuration cost of mobile alternative energy storage units, the objective function for the deployment phase is constructed as shown in the following equation:

[0175] ;

[0176] Where F1 is the objective function for the pre-deployment stage; C E C P These represent the unit capacity cost and unit power cost of mobile energy storage, respectively; E a P a These are the rated capacity and power of mobile alternative energy storage, respectively.

[0177] S222. Mobile alternative energy storage devices possess cross-regional mobility capabilities, and migration state constraints are established for energy storage device units. These migration state constraints include regional docking mutual exclusion constraints and migration state consistency constraints.

[0178] The area docking mutual exclusion constraint is shown in the following formula:

[0179] ;

[0180] Where ωAt and ωBt represent whether the mobile energy storage device is docked in area A and area B at time t, respectively, and T is the total time within the scheduling cycle. The value of this variable is 1 when the vehicle is docked in area A. The mobile alternative energy storage device can only connect to one area at any given time, or move between area A and area B.

[0181] The consistency constraint for the migration state is shown in the following formula:

[0182] ;

[0183] Here, βAt and βBt represent whether the mobile alternative energy storage device leaves region A and region B at time t, respectively. The state of the mobile alternative energy storage device moving from one region to another needs to remain consistent; if it leaves region A at time t, it must arrive at region B at time t+1.

[0184] S223. Solve the cross-regional migration and dynamic scheduling model of energy storage, determine the deployment location of mobile alternative energy storage units, and output the cross-regional migration and scheduling strategy of energy storage.

[0185] Based on the cross-regional migration and scheduling strategy for energy storage, the migration paths of energy storage units in different scenarios and at different times are determined, such as migrating from a low-load area to a high-load area B in summer and migrating to area A in winter; scheduling rules during the migration process are determined, such as the migration timing should avoid regional load peaks to ensure that the power supply stability is not affected during the migration.

[0186] Furthermore, the second phase, combining the probability load curves of typical days in multiple scenarios, the time-of-use electricity price response mechanism, and the real-time operating status of the energy storage units, aims to minimize operating costs by optimizing the power purchase / load shedding at the current deployment location and the charging and discharging power of the energy storage devices, including:

[0187] S31. With the goal of minimizing multi-period operating costs, and taking into account both electricity purchase costs and load shedding penalty costs, the objective function for the operation phase is constructed as follows:

[0188] ;

[0189] Where M represents the total number of months; N represents the total number of scenarios; p represents the probability of region i in month m - scene n; t Let be the electricity price at time t; T be 24 hours; and I be the set of regions A and B. ; The amount of electricity purchased at time t for the energy storage deployment location in the nth cluster scenario of region i in the mth month; Let t be the load shedding amount of the energy storage deployment location at time t in the nth clustering scenario of the mth month of region i; This is the load shedding penalty coefficient.

[0190] S32. Establish operating constraints, which include transformer area capacity constraints, charging and discharging power constraints, power balance constraints, and state of charge constraints.

[0191] Step S32 specifically includes the following steps:

[0192] S321. Establish the power supply constraints for mobile alternative energy storage using the following formula:

[0193] ;

[0194] in, m represents the actual load of region i at time t; m is the month index; t is the time. Let I be the maximum power supply capacity of the distribution transformer area in region i, and let I be the set of regions A and B. During operation, the maximum capacity constraint of the distribution transformer area is considered. The mobile alternative energy storage device will only start supplying power when the regional load exceeds the upper limit of the distribution transformer area capacity.

[0195] S322. Establish the charging and discharging power constraints for mobile alternative energy storage using the following formula:

[0196] ;

[0197] in, The charging power of a mobile alternative energy storage device at time t; This refers to the maximum charging power of mobile alternative energy storage devices. This represents the maximum discharge power of the mobile alternative energy storage device. For 0-1 variables, in =1 indicates that the mobile energy storage is charged at time t; For 0-1 variables, in When =1, it means that the mobile energy storage is charged at time t. This constraint means that the charging and discharging of the mobile energy storage cannot be carried out simultaneously.

[0198] S323. Establish the power balance constraints for the current deployment location of mobile alternative energy storage using the following formula:

[0199] ;

[0200] in, The power consumption at the current deployment location; Discharge power for mobile alternative energy storage; The load at the current deployment location; This is the load shedding amount; Charging power for mobile alternative energy storage;

[0201] S324. To ensure the safety and performance stability of energy storage systems and prevent equipment damage or energy loss due to overcharging or over-discharging, the range of state of charge (SCC) values ​​should be limited. The SCC constraints for mobile alternative energy storage are established using the following formula:

[0202] ;

[0203] Among them, SOC min SOC max These are the minimum and maximum values ​​of the state of charge for mobile alternative energy storage, respectively. For mobile alternative energy storage, the state of charge at time t satisfies =E t / E m E t E represents the remaining charge of the mobile energy storage at time t. m Rated capacity of the energy storage device;

[0204] S325. Establish the continuity constraint of the state of charge of mobile energy storage using the following formula:

[0205] ;

[0206] in, The state of charge of mobile alternative energy storage at time t; , These represent the charging and discharging power of mobile alternative energy storage at time t, respectively; η ch η dch These represent the charging and discharging efficiencies of the mobile energy storage, respectively. This constraint indicates that the state of charge of the mobile energy storage will not undergo abrupt changes.

[0207] S33. Based on the objective function and operational constraints of the operational phase, and using the mixed-integer linear programming method, model with YALMIP and solve with GUROBI to output the power curves of power purchase / load shedding at the current deployment location under multiple scenarios, as well as the charging and discharging power curves of the energy storage device.

[0208] This embodiment models and studies the load characteristics of two typical regional distribution networks and the response characteristics of mobile alternative energy storage systems. A two-stage optimization model is constructed to achieve coordinated optimization of power supply and mobile alternative energy storage charging and discharging during different load periods. In the first stage, the location selection of the energy storage devices is optimized. In the second stage, based on the load conditions of each period, the charging and discharging strategies and spatiotemporal migration paths of the devices are dynamically solved. Based on the energy storage unit operating data and actual regional load data, a feedback mechanism is used to compare the pre-defined target identification deviation of the cross-spatiotemporal coordinated control, and the scheduling scheme of the energy storage units is continuously corrected.

[0209] To ensure the feasibility and practicality of the model, based on the annual load of regions A and B, this study analyzes the impact of the optimized configuration strategy of mobile alternative energy storage on flexible mutual assistance and emergency supply in multiple regions. This allows for a direct verification of the effectiveness of the strategy and provides data support and practical reference for actual engineering applications.

[0210] Figure 2 The results are obtained from the ordered cluster analysis of the annual load curve of Region A in stage S1, and the load characteristics are divided into five typical daily load types. Region A exhibits a clear "bimodal characteristic" in most typical days, that is, the load is higher in the morning and evening, while the load is relatively lower at noon, and the peak-valley difference is very significant. The clustering results clearly show that the load of Region A has obvious intraday time-varying differences and seasonal variation patterns, providing a typical load scenario basis for subsequent capacity configuration and charging and discharging strategy optimization of energy storage systems.

[0211] Figure 3 The results of the S1 stage ordered clustering analysis of the annual load curve of Region B are presented, ultimately classifying it into four typical daily load types. Compared with Region A, the load curve of Region B is generally flatter, but a significant single-peak characteristic still exists in some typical days. These results provide a basis for subsequent cross-regional energy storage coordinated scheduling. By identifying the differences in load peaks and valleys between regions, energy mutual assistance and peak-valley misalignment optimization of energy storage can be achieved across different regions.

[0212] In regions A and B, the two-stage configuration results are as follows: Figure 4 As shown; Figure 4 This demonstrates the process of mobile alternative energy storage migrating and dynamically scheduling between region A and region B after phase S3. It shows that the energy storage devices switch between charging and discharging based on regional load levels and electricity price signals at different times: charging in low-demand regions during low-load or low-price periods, and discharging in high-demand regions during high-load or high-price periods, thus achieving energy flow and load balance across multiple regions. The energy storage system's charging and discharging efficiency is 97%; the initial state of charge (SOC) is 25% of the rated capacity, and it is required to maintain it within a safe range of 10%-95% throughout the entire operating cycle. Within the scheduling cycle, typical peak-valley load distribution characteristics are considered at a 24-hour daily granularity. The model responds in real-time to changes in electricity price and load, charging during off-peak periods and meeting load demands during peak periods through migration and discharging, maximizing the utilization and economy of the energy storage devices, forming a scheduling scheme that considers power supply security in multiple scenarios.

[0213] Figure 5This presentation showcases the energy storage charging and discharging power scheduling results obtained after solving the optimized model S3 under a typical daily load scenario in Region A. Energy storage charges during off-peak hours at night and midday, and discharges during morning and evening peak hours, closely aligning with electricity price and load trends. Optimized energy storage charging and discharging achieves precise peak shaving and valley filling while minimizing costs based on typical load characteristics. When the load in Region A exceeds the transformer capacity limit, energy storage is promptly deployed to ensure power supply security; during off-peak hours, charging restores energy storage capacity, preparing for scheduling in the next time period.

[0214] Figure 6 This paper demonstrates the changes in energy storage charging and discharging power after solving the optimized model S3 under a typical daily load scenario in Region B. Compared with Region A, the energy storage operation strategy in Region B exhibits different temporal distribution characteristics. Its discharging period is mainly concentrated in the afternoon peak when the load rises, while the charging phase occurs more frequently during the night or morning off-peak hours. However, when the peak load exceeds the upper limit of the energy storage discharge capacity, the deployment area will trigger load shedding measures to reduce some non-critical loads. This model can adaptively schedule according to the load characteristics and electricity price signals of different regions, automatically adjusting the energy storage charging and discharging strategy to achieve overall operational economy and stability.

[0215] As can be seen from the above, the mobile alternative energy storage device optimization configuration strategy based on a two-stage optimization framework provided by this invention can effectively solve the bottleneck of power distribution areas and the pressure on power supply security caused by regional seasonal load peaks, and realize the coordinated control of energy storage devices in both time and space dimensions. On the one hand, it improves the configuration utilization rate and response flexibility of energy storage devices; on the other hand, it reduces overall costs and improves the return on investment of energy storage devices through cross-regional peak-valley mutual assistance and optimized scheduling. The solution of this invention not only reduces the overall operating cost, but also maximizes the economic benefits of mobile alternative energy storage.

[0216] In summary, this invention proposes an innovative multi-regional collaborative method for optimizing the configuration and operation of mobile alternative energy storage. It achieves flexible scheduling and dynamic optimization of energy storage devices across different regions through a two-stage integrated mixed-integer linear programming model. Specifically, the system first analyzes regional load characteristics, classifies load curves using ordered clustering, and generates typical daily load data for multiple scenarios. Considering the capacity and power cost of energy storage devices, an optimization algorithm is used to determine the deployment location of energy storage units during the equipment deployment phase. During the operation phase, based on load curves, electricity prices, and energy storage status, a two-stage optimization configuration model is used to optimize the charging and discharging power of energy storage units across multiple time periods. The two-stage optimization structure achieves regional collaboration through the interaction of load characteristics and electricity price information between regions: the equipment deployment phase generates a basic configuration scheme for energy storage devices, and the operation phase performs real-time scheduling optimization within the day, achieving rapid response to load fluctuations and cost minimization. Through a centralized optimization and distributed execution architecture, each region can make independent decisions based on local load and electricity price information, while sharing scheduling information to improve global optimization capabilities, ensuring the system's flexibility and economy across multiple scenarios. In the future, the system is expected to be further expanded to more regions and more types of energy storage devices, and through optimization of algorithm parameters and model structure, more intelligent and flexible energy storage scheduling can be achieved.

[0217] Example 2

[0218] A multi-regional collaborative mobile alternative energy storage optimization configuration and operation device includes a clustering module, a model building module, and a segmented optimization module. The clustering module uses ordered clustering to classify the annual load curves of a region, determine typical load scenarios and their probabilities, and form typical days for multiple scenarios. The model building module establishes a two-stage optimization configuration model. In the first stage, based on the probabilistic load curves of typical days for multiple scenarios, the spatiotemporal characteristics of the regional load are obtained, and the deployment location of the energy storage equipment is determined with the goal of minimizing configuration costs. In the second stage, combining the probabilistic load curves of typical days for multiple scenarios, the time-based electricity price response mechanism, and the real-time operating status of the energy storage units, the power purchase / load shedding at the current deployment location and the charging / discharging power of the energy storage equipment are optimized with the goal of minimizing operating costs. The segmented optimization module uses the two-stage optimization configuration model to deploy the energy storage equipment and optimize its operating status.

[0219] It should be noted that the above modules correspond to the steps described in Embodiment 1, and the examples and application scenarios implemented by the above modules and corresponding steps are the same, but are not limited to the content disclosed in Embodiment 1. It should also be noted that the above modules, as part of the system, can be executed in a computer system such as a set of computer-executable instructions.

[0220] Example 3

[0221] An electronic device includes: at least one processor; and a memory storing instructions that, when executed by the at least one processor, cause the at least one processor to perform the aforementioned multi-regional collaborative mobile alternative energy storage optimization configuration and operation method.

[0222] In this embodiment, the electronic device may include, but is not limited to: personal computer, server computer, workstation, desktop computer, laptop computer, notebook computer, mobile computing device, smartphone, tablet computer, cellular phone, personal digital assistant (PDA), handheld device, messaging device, wearable computing device, consumer electronic device, etc.

[0223] Example 4

[0224] A machine-readable storage medium storing executable instructions that, when executed, cause the machine to perform the aforementioned multi-regional collaborative mobile alternative energy storage optimization configuration and operation method.

[0225] Specifically, a system or apparatus equipped with a readable storage medium storing software program code that implements the functions of any of the embodiments described above, and enabling the computer or processor of the system or apparatus to read and execute instructions stored in the readable storage medium. In this case, the program code read from the readable medium itself can implement the functions of any of the embodiments described above; therefore, the machine-readable code and the readable storage medium storing the machine-readable code constitute a part of this specification. Embodiments of the readable storage medium include floppy disks, hard disks, magneto-optical disks, optical disks (such as CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RAM, DVD-RW, DVD-RW), magnetic tapes, non-volatile memory cards, and ROMs. Alternatively, the program code can be downloaded from a server computer or the cloud via a communication network. Those skilled in the art will understand that embodiments of the present invention can be provided as 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. The present 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, and 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 processor, 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, produce a machine for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 The 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 operate 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.

[0226] 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 1One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0227] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention. The scope of protection claimed by the appended claims and their equivalents is defined.

Claims

1. A method for optimizing the configuration and operation of mobile alternative energy storage with multi-regional collaboration, characterized in that, The method includes the following steps: S1. Use ordered clustering to classify the annual load curve of the region, determine typical load scenarios and scenario probabilities, and form typical days for multiple scenarios; S2. Establish a two-stage optimization configuration model: The first stage is based on the probability load curves of typical days in multiple scenarios to obtain the spatiotemporal characteristics of regional load and determine the deployment location of energy storage devices with the goal of minimizing configuration costs; The second stage combines the probability load curves of typical days in multiple scenarios, the time-based electricity price response mechanism, and the real-time operating status of energy storage units to optimize the power purchase / load shedding at the current deployment location and the charging and discharging power of energy storage devices with the goal of minimizing operating costs. S3. Utilize a two-stage optimization configuration model to deploy the location of energy storage devices and optimize their operating status.

2. The method for optimizing the configuration and operation of multi-regional collaborative mobile alternative energy storage according to claim 1, characterized in that, The method includes the following steps: Step S1 specifically includes the following steps: S11. Obtain annual load curves for multiple regions: Collect annual time-period load data for multiple regions and form a complete annual load curve dataset according to the time series. S12. Use ordered clustering to classify the load curves of the two regions; S13. For each load curve obtained by ordered clustering, take the average time dimension of all date load curves in the same category, calculate the typical daily load value under each scenario, form a typical daily load curve for multiple scenarios, and calculate the scenario probability of each scenario, that is, the proportion of days included in this scenario to the total number of days in the year.

3. The method for optimizing the configuration and operation of multi-regional collaborative mobile alternative energy storage according to claim 2, characterized in that, The method includes the following steps: Step 12 specifically includes the following steps: S121. Determine the degree of intra-class variability based on Fraser distance and sum of squared deviations; S1211, Calculate the Frescher distance The annual load curve is segmented according to time series, and a sub-interval of load data after segmentation is determined as {x}. i , x i+1 ,…, x j }, j>i; Where, x i Let x be the load curve for day i. i x i+1 Until x j These are load curve data points arranged in a time series. The index set of the load data sub-intervals is as follows: P = {i, i+1,…, j}; Where i, i+1,…, j are the time indices of the load data within the sub-interval; The mean Friesian distance of this set of load sub-intervals is determined using the following formula: ； in, The mean Friesian distance of the load sub-interval; The number of combinations represents the number of pairs of any two curves selected from the j-i+1 curves; i and j are the time indices of the load data within this sub-interval. For from x i To x j The sum of the Fréchet distances between any two load curves; l is the index number corresponding to all curves; S1212, Calculate the degree of intra-class differences Using the sum of squared deviations as the intraclass diameter, the intraclass variability of the load subintervals is calculated using the following formula: ； in, Within-class sum of squared deviations of load from day i to day j; The number of combinations represents the number of pairs of any two curves selected from the (j-i+1) curves; t is time, and X... t Let Frège be the Frescher distance between the two load curves in the t-th group of the load sub-interval; The mean Friesian distance of the load sub-interval; S122. Based on the degree of intra-class differences, construct a segmentation objective function and generate the optimal segmentation scheme; S1221. The piecewise objective function is shown in the following formula: ； in, To sum the intraclass diameters of all load sub-intervals throughout the year, This represents the partitioning scheme for the annual load data under a given number of categories r; n is the number of load sample data, r is the number of categories, and i v Represent the dividing point between classes, denoted as ; S1222. Based on the piecewise objective function, the optimal piecewise scheme is generated using the following formula: ； Where p(n,r) represents the optimal segmentation scheme, n is the number of load sample data, r is the number of categories; and i is the time index of the load data in this sub-interval. To sum the intraclass diameters of all load sub-intervals throughout the year, This represents the partitioning scheme for the annual load data under a given number of categories r, where n is the number of load sample data. S1223. Based on the aforementioned optimal segmentation scheme, the optimal segmentation is performed using the Fisher algorithm, as shown in the following formula: Using Fisher's algorithm, the two recursive formulas for optimal splitting are as follows: ； in, This represents the minimum sum of squared differences within each category when the annual load is divided into two categories; p(n,2) represents the optimal segmentation scheme; n is the number of load sample data. Let be the sum of squared intra-class differences from sample 1 to sample (j-1). Let be the sum of squared intra-class differences from the j-th sample to the n-th sample; This represents the minimum sum of squared differences within each of the r categories when the annual load is divided into r categories. Let be the sum of squared intra-class differences from the (j-1)th sample to the (r-1)th sample; S123. Determine the optimal number of clusters, optimize the solution by calculating the rate of change of slope under adjacent categories, and the point of abrupt change in the rate of change of slope is the optimal number of clusters r. id ; S1231. Calculate the rate of change of the slope of the objective function under adjacent classification numbers using the following formula: ； in, The slope of the deviation between the number of categories r and the number of categories r+1; This represents the minimum sum of squared differences within each class when the annual load is divided into r+1 categories; n is the number of load sample data; r is the number of categories; This represents the minimum sum of squared differences within each of the r categories when the annual load is divided into r categories. S1232. Define the slope change rate of the adjacent cluster number as dif: ； S1233. Determine the optimal number of clusters r using the following formula. id : ； in, For the number of clusters is The rate of change of the slope at that time; The number of clusters; The rate of change of the slope of the objective function when the number of clusters is R. This represents the rate of change of the slope under the previous cluster number. This represents the rate of change of the slope under the next cluster number; This represents the optimal number of clusters.

4. The method for optimizing the configuration and operation of multi-regional collaborative mobile alternative energy storage according to claim 3, characterized in that, The typical daily load value is calculated using the following formula: ； Among them, L et (n,t) represents the load value at time t on a typical load day in the nth clustering scenario, in kW; n L represents the number of days included in the nth load scenario. e (y,t) represents the load value at time t on day y; T represents the total time of the day.

5. The method for optimizing the configuration and operation of multi-regional collaborative mobile alternative energy storage according to claim 4, characterized in that, The first stage, based on the probabilistic load curves of typical days in multiple scenarios, obtains the spatiotemporal characteristics of regional load and determines the deployment locations of energy storage devices with the goal of minimizing configuration costs, including: S21. Based on typical daily load curves of multiple scenarios, obtain the spatiotemporal characteristics of regional load; S22. Build a cross-regional migration and dynamic scheduling model for energy storage, solve the model, and determine the deployment location of mobile alternative energy storage units; Step S22 specifically includes the following steps: S221. With the goal of minimizing the configuration cost of mobile alternative energy storage units, the objective function for the deployment phase is constructed as shown in the following equation: ； Where F1 is the objective function for the pre-deployment stage; C E C P These are the unit capacity cost and unit power cost of mobile alternative energy storage, respectively; E a P a These are the rated capacity and power of mobile alternative energy storage, respectively. S222. The mobile alternative energy storage device has the ability to move across regions, and the migration state constraints of the energy storage device unit are established; the migration state constraints include regional docking mutual exclusion constraints and migration state consistency constraints. The area docking mutual exclusion constraint is shown in the following formula: ； Where ωA t and ωB t represent whether the mobile energy storage device is docked in region A and region B at time t, respectively, and T is the total time within the scheduling cycle; The consistency constraint for the migration state is shown in the following formula: ； Wherein, βA t and βB t represent whether the mobile alternative energy storage device leaves region A and region B at time t, respectively; S223. Solve the cross-regional migration and dynamic scheduling model of energy storage, determine the deployment location of mobile alternative energy storage units, and output the cross-regional migration and scheduling strategy of energy storage.

6. The method for optimizing the configuration and operation of multi-regional collaborative mobile alternative energy storage according to claim 5, characterized in that, The second phase combines the probability load curves of typical days in multiple scenarios, the time-of-use electricity price response mechanism, and the real-time operating status of the energy storage units. With the goal of minimizing operating costs, it optimizes the power purchase / load shedding at the current deployment location and the charging and discharging power of the energy storage devices, including: S31. With the goal of minimizing multi-period operating costs, and taking into account both electricity purchase costs and load shedding penalty costs, construct an objective function for the operation phase. S32. Establish operating constraints, which include transformer area capacity constraints, charging and discharging power constraints, power balance constraints, and state of charge constraints. S33. Based on the objective function and operational constraints of the operational phase, and using the mixed-integer linear programming method, model with YALMIP and solve with GUROBI to output the power curves of power purchase / load shedding at the current deployment location in multiple scenarios, as well as the charging and discharging power curves of the energy storage device.

7. The method for optimizing the configuration and operation of multi-regional collaborative mobile alternative energy storage according to claim 6, characterized in that, The objective function for the operational phase is shown in the following equation: ； Where M represents the total number of months; N represents the total number of scenarios; p represents the probability of region i in month m - scene n; t Let be the electricity price at time t; T be 24 hours; and I be the set of regions A and B. Pbuy i,m,n,t represents the electricity purchased by the energy storage deployment location at time t in the nth cluster scenario of month m in region i; Pshed i,m,n,t represents the load shedding amount of the energy storage deployment location at time t in the nth cluster scenario of month m in region i; C shed Shear load penalty factor; Step S32 specifically includes the following steps: S321. Establish the power supply constraints for mobile alternative energy storage using the following formula: ； Where Pload i,t is the actual load of region i at time t; m is the month index; and t is the time. Let I be the maximum power supply capacity of the transformer substation in region i, and let I be the set of regions A and B. S322. Establish the charging and discharging power constraints for mobile alternative energy storage using the following formula: ； in, The charging power of a mobile alternative energy storage device at time t; This refers to the maximum charging power of mobile alternative energy storage devices. This represents the maximum discharge power of the mobile alternative energy storage device. For 0-1 variables, in =1 indicates that the mobile energy storage is charged at time t; For 0-1 variables, in When =1, it means that the mobile energy storage is charged at time t. This constraint means that the charging and discharging of the mobile energy storage cannot be carried out simultaneously. S323. Establish the power balance constraints for the current deployment location of mobile alternative energy storage using the following formula: ； in, The power consumption at the current deployment location; Discharge power for mobile alternative energy storage; The load at the current deployment location; This is the load shedding amount; Charging power for mobile alternative energy storage; S324. Establish the state-of-charge constraints for mobile alternative energy storage using the following formula: ； Among them, SOC min SOC max These are the minimum and maximum values ​​of the state of charge for mobile alternative energy storage, respectively. For mobile alternative energy storage, the state of charge at time t satisfies =E t / E m E t E represents the remaining charge of the mobile energy storage at time t. m Rated capacity of the energy storage device; S325. Establish the continuity constraint of the state of charge of mobile energy storage using the following formula: ； in, The state of charge of mobile alternative energy storage at time t; , These represent the charging and discharging power of mobile alternative energy storage at time t, respectively; η ch η dch These represent the charging and discharging efficiencies of mobile energy storage, respectively.

8. A multi-regional collaborative mobile alternative energy storage optimization configuration and operation device, characterized in that, The device includes: a clustering module, a model building module, and a segmentation optimization module; The clustering module is used to classify the annual load curve of the region using an ordered clustering method, determine typical load scenarios and scenario probabilities, and form typical days for multiple scenarios; The model building module is used to establish a two-stage optimization configuration model. In the first stage, based on the probabilistic load curves of typical days in multiple scenarios, the spatiotemporal characteristics of regional load are obtained, and the deployment location of energy storage devices is determined with the goal of minimizing configuration costs. In the second stage, combined with the probabilistic load curves of typical days in multiple scenarios, the time-period electricity price response mechanism, and the real-time operating status of energy storage units, the power purchase / load shedding at the current deployment location and the charging and discharging power of energy storage devices are optimized with the goal of minimizing operating costs. The segmented optimization module is used to deploy the location of energy storage devices and optimize the operating status of energy storage devices using a two-stage optimization configuration model.

9. An electronic device, characterized in that, include: At least one processor; And a memory storing instructions that, when executed by the at least one processor, cause the at least one processor to perform the multi-regional collaborative mobile alternative energy storage optimization configuration and operation method as described in any one of claims 1 to 7.

10. A machine-readable storage medium, characterized in that, It stores executable instructions that, when executed, cause the machine to perform the multi-regional collaborative mobile alternative energy storage optimization configuration and operation method as described in any one of claims 1 to 7.

Images