Intelligent soft switching and electro-hydrogen hybrid energy storage configuration method based on space-time decoupling

Abstract

The invention relates to an intelligent soft switching and electro-hydrogen hybrid energy storage configuration method based on space-time decoupling. The method comprises the following steps: step 1, constructing an MEH-SOP capacity optimal configuration model based on two-stage space-time decoupling; according to the constructed MEH-SOP capacity optimal configuration model, the optimal configuration capacity of various flexible resources can be obtained through model calculation, including capacity configuration results of an intelligent soft switch SOP, battery energy storage BESS in a multi-time scale electricity-hydrogen hybrid energy storage system, hydrogen energy storage HS, seasonal hydrogen energy storage SHS and energy conversion equipment thereof. And all device annual energy / power flow conditions. According to the invention, a multi-level energy storage system covering intra-day to inter-week is constructed, and a multi-port SOP direct current side is accessed in a modularized manner, so that the structural flexibility and the operation flexibility of the DN are remarkably enhanced.

Description

Technical Field

[0001] This invention belongs to the field of power grid control technology, and in particular to a method for configuring intelligent soft switching and electric-hydrogen hybrid energy storage based on spatiotemporal decoupling. Background Technology

[0002] Against the backdrop of a global push for low-carbon transformation, the power system is rapidly shifting from traditional fossil fuels to an energy structure dominated by "zero-carbon" renewable energy. As the primary carrier for distributed generation (DG) access, the distribution network's penetration rate continues to climb globally. It is projected that by the end of 2025, several counties in China will have DG penetration rates exceeding 100%, a trend driving the transformation of the distribution network from a traditional centralized power supply model to a regionally autonomous model. However, the high volatility of DG is further amplified in areas with extremely high penetration rates, leading to complex operational risks across multiple temporal and spatial scales for the distribution network, including power and energy imbalances and severe voltage exceedances.

[0003] To address the aforementioned issues, it is necessary to enhance the spatiotemporal regulation capabilities of the distribution network. In the temporal dimension, short- and long-term power / energy regulation can be achieved through hybrid electric-hydrogen energy storage systems. In the spatial dimension, smart soft switches (SOPs) can be used to support resource sharing between feeders through power flow control. From an optimization scheduling perspective, the two systems have natural spatiotemporal complementarity in regulation capabilities, and coordinated configuration can significantly enhance the overall regulation capability of the distribution network. However, hybrid electric-hydrogen energy storage systems and SOPs differ significantly in their operating mechanisms and control strategies, and a unified and efficient configuration method has not yet been proposed. Furthermore, their spatiotemporal coupling characteristics significantly increase the complexity of the configuration model, especially since hydrogen energy storage needs to consider its 8760-hour annual operation, greatly increasing the difficulty of solving the problem. Therefore, in areas with extremely high distributed generation (DG) penetration, it is urgent to construct a collaborative architecture and configuration model for SOPs and hybrid electric-hydrogen energy storage systems oriented towards multi-spatiotemporal scale regulation, and to propose efficient solution strategies to support the safe, clean, and efficient operation of the distribution network.

[0004] Existing research on hybrid energy storage configurations using hydrogen and electricity primarily focuses on hydrogen energy storage modeling, while electric energy storage modeling is relatively well-established due to technological maturity. Typical methods often employ a two-tier energy storage system structure, treating hydrogen energy storage as a seasonal or monthly energy regulation resource, coordinating it with electric energy storage in a long-term, fixed charge / discharge state. This approach works well in scenarios with low to medium distributed gas (DG) penetration rates. However, in areas with extremely high DG penetration, the distribution network not only faces monthly or seasonal power imbalances but may also experience imbalances across multiple time scales, such as daily and weekly imbalances. In such cases, existing two-tier energy storage systems lack multi-level configurations in their structure and clear coordination mechanisms in their control, making it difficult to cope with complex energy fluctuations. For example, when hydrogen storage is in a discharging phase during a certain season or month, while DG output is surplus in a certain week and electric energy storage is constrained by daily balance limitations and cannot absorb the surplus power, existing two-tier hydrogen and electricity storage systems struggle to respond flexibly and promptly. Therefore, in areas with extremely high DG penetration, it is urgent to construct a refined hybrid energy storage system with multiple spatiotemporal scales and to clarify the collaborative scheduling mechanism of energy storage devices at different time scales.

[0005] In terms of spatial flexibility, Standard Operating Procedures (SOPs) can achieve flexible interconnection and inter-regional coordination optimization of the distribution network by actively controlling the power flow between feeders. In recent years, the DC side of the SOP, as the interface for DC resource integration, has spawned various flexible interconnection system structures, which can integrate energy storage on the DC side to improve short-term power flow regulation capabilities. Further development could extend the SOP to a multi-port structure to enhance its spatial flexibility. However, currently, the DC side of the SOP mainly integrates energy storage, lacking effective integration with long-term regulation capacity resources, making it difficult to meet the multi-temporal and spatial coordination needs of the distribution network under extremely high distributed generation (DG) penetration.

[0006] Furthermore, the configuration of hydrogen energy storage typically relies on annual operational simulations. Currently, existing research mainly employs two time-series modeling strategies: scenario analysis, which simulates annual load curves by selecting and piecing together several typical days; and year-round time-series modeling, which directly utilizes the annual 8760-hour load curve for power balance simulation. Scenario analysis offers higher solution efficiency, but its reliance on typical day selection makes it difficult to accurately capture the dynamic changes of hydrogen storage at fine time scales such as daily and weekly, potentially leading to configuration errors and limiting its applicability in multi-tiered energy storage systems. In contrast, year-round time-series modeling accurately reflects the annual energy evolution of the energy storage system, making it more suitable for multi-tiered hybrid energy storage configuration problems across multiple time scales, and offering higher solution accuracy. However, this method suffers from high model dimensionality and computational complexity, especially in multi-tiered hybrid energy storage scenarios, where the solution difficulty increases significantly, making it difficult to obtain an effective solution within an acceptable timeframe. Summary of the Invention

[0007] The purpose of this invention is to overcome the shortcomings of the prior art and propose a smart soft-switching and electric-hydrogen hybrid energy storage configuration method based on spatiotemporal decoupling. By constructing a multi-level energy storage system covering the day to the week and modularly connecting it to the multi-port SOP DC side, the structural flexibility and operational flexibility of the DN are significantly enhanced.

[0008] The technical problem solved by this invention is achieved through the following technical solution: A method for configuring intelligent soft switching and electro-hydrogen hybrid energy storage based on spatiotemporal decoupling includes the following steps: Step 1: Construct the MEH-SOP model; Step 2: Construct the MEH-SOP capacity optimal configuration model based on two-stage spatiotemporal decoupling. The optimal configuration capacity of various flexible resources can be obtained through model calculation, including the capacity configuration results of smart soft switching (SOP), battery energy storage (BESS), hydrogen energy storage (HS), seasonal hydrogen energy storage (SHS) and their energy conversion equipment (such as electrolyzers and fuel cells) in the multi-timescale electric-hydrogen hybrid energy storage system, as well as the energy / power flow of all the above equipment for 8760 hours throughout the year. It also includes the investment cost of the above equipment, the operation and maintenance cost of the above equipment, the operation and maintenance revenue of the above equipment, the network loss cost of the distribution network, and the cost of curtailment of solar / wind power.

[0009] Furthermore, the specific implementation method of step 1 is as follows: Step 1.1: Analyze the structural characteristics and spatiotemporal regulation mechanism of MEH-SOP; Step 1.2: Based on the structural characteristics and spatiotemporal adjustment mechanism of MEH-SOP, establish a multi-port SOP model in MEH-SOP; Step 1.3: Based on the structural characteristics and spatiotemporal adjustment mechanism of MEH-SOP, establish the MEH system model in MEH-SOP.

[0010] Moreover, the specific implementation method of step 1 is as follows: the MEH in the MEH-SOP system consists of an energy conversion device and a multi-level energy storage unit: energy conversion device and multi-level energy storage unit.

[0011] Furthermore, the multi-port SOP model in step 1.2 MEH-SOP is as follows: Multi-port SOP power balance constraints: in, P t,i,SOP For SOP No. i The port is t Power at any moment; Ω k For SOP port set; P t,MEH for tThe exchange power between the MEH and the DC side of the SOP at any given time.

[0012] Multi-port capacity constraints: in, Q t,i,sop for t Time Port i reactive power, S i,SOP For port i capacity.

[0013] Furthermore, the MEH system model in the MEH-SOP in step 1.3 includes power balance constraints, BEES operation constraints, HS operation constraints, and SHS operation constraints.

[0014] Furthermore, step 2 includes the following steps: Step 2.1: Focusing on the time dimension, a MEH capacity optimization model is constructed based on 8760 hours of annual load data. The STL trend feature extraction algorithm is introduced to pre-determine the seasonal hydrogen storage (SHS) charge / discharge state, thereby reducing the model dimensionality and improving solution efficiency. The output of this step includes the optimal capacity of various electric-hydrogen storage devices and the power interaction process with the DC side of the SOP.

[0015] Step 2.2: Construct an optimal configuration model for the multi-port capacity of SOP that considers the power flow optimization constraints of the distribution network, and construct typical daily scenarios and the interactive power boundary between MEH and SOP to achieve spatial dimension optimization configuration of flexible interconnection resources.

[0016] Furthermore, step 2.1 includes the following steps: Step 2.1.1: Construct the MEH optimal configuration model; Step 2.1.2: Solve the model using a fast solution strategy based on STL.

[0017] The advantages and positive effects of this invention are: 1. This invention proposes a multi-timescale electro-hydrogen hybrid energy storage-intelligent soft-switching (MEH-SOP) system architecture with multi-temporal and spatial scale adjustment capabilities and modular integration. By constructing a multi-level energy storage system covering the day to the week and modularly connecting it to the DC side of a multi-port SOP, the structural flexibility and operational flexibility of the DN are significantly enhanced.

[0018] 2. This invention proposes a multi-temporal and spatial coordination control mechanism for MEH-SOP with an embedded DN autonomous operation strategy. It clarifies the scheduling priorities and collaborative paths of various MEH and SOP devices at different time scales, constructs a multi-temporal and spatial scale operation mechanism with orderly response and clear coupling, and improves the efficiency of resource coordination and utilization.

[0019] 3. This invention constructs a two-stage spatiotemporal decoupled configuration model for MEH-SOP, effectively alleviating the problems of high model dimensionality and computational complexity caused by strong spatiotemporal coupling. The first stage focuses on the time dimension, optimizing the power / capacity configuration of each MEH device based on the 8760-hour time-series operation simulation taking into account time-of-use pricing. The second stage focuses on the spatial dimension, extracting typical scenarios of high-power charging / discharging of MEH throughout the year (from the results of the first stage) and severe voltage overruns of DN throughout the year (based on power flow calculations), optimizing the configuration of SOP multi-port capacity.

[0020] 4. This invention proposes a fast solution strategy suitable for the first stage. Addressing the issues of large variable scale and complex solution in this stage, it utilizes the long-term power regulation characteristics of Seasonal Hydrogen Storage (SHS) and extracts the annual net load trend characteristics based on the Seasonal Trend Decomposition (STL) algorithm to predetermine the charge / discharge 0 / 1 states of the SHS in each week. This ensures the accuracy of the first-stage configuration while significantly reducing the solution time, and fully preserves the key characteristics of the annual maximum power interaction between MEH and SOP, ensuring the SOP configuration results in the second stage are unbiased. Attached Figure Description

[0021] Figure 1 This is a schematic diagram of the improved 31-node power distribution system in Taiwan according to an embodiment of the present invention; Figure 2 This is a schematic diagram illustrating the differences in charge and discharge states between Case 1 and Case 2 in this embodiment of the invention. Figure 3 This is a schematic diagram of the per-unit voltage values ​​at node 10 in Case 1 and Case 6 of the present invention. Figure 4 This is a schematic diagram of active power transmission in Case 1 SOP-8 of the present invention; Figure 5 This is a schematic diagram of reactive power transmission in Case 1 SOP-8 of the present invention. Detailed Implementation

[0022] The present invention will be further described in detail below with reference to the accompanying drawings.

[0023] Step 1: Construct the MEH-SOP model.

[0024] Step 1.1: Analyze the structural characteristics and spatiotemporal regulation mechanism of MEH-SOP.

[0025] Affected by production activities and weather fluctuations, in DN systems with extremely high DG penetration, DG supply and load demand often exhibit mismatches across multiple time scales, including intraday, interday, and interweekal periods, leading to severe wind / solar curtailment and significant voltage limit violations. To address this, this invention proposes an MEH-SOP system architecture. The MEH in the MEH-SOP system consists of an energy conversion device and multi-level energy storage units: a) Energy conversion devices: including alkaline electrolyzers (AEC), seasonal alkaline electrolyzers (SAEC), solid oxide fuel cells (SOFC), and seasonal solid oxide fuel cells (SSOFC).

[0026] b) Multi-level energy storage unit: Composed of bioelectrochemical energy storage (BESS), hydrogen energy storage (HS), and SHS. Additionally, the MEH is connected to the DC side of the multi-port SOP. The MEH-SOP is highly modular, allowing for flexible configuration of the capacity of each port and device according to actual needs, meeting deployment and adaptation requirements in various scenarios. In terms of operation, the MEH-SOP possesses coordinated adjustment capabilities in both time and space dimensions.

[0027] Table 1 Internal MEH Operating Characteristics

[0028] a. Time dimension: BESS is used to address hourly power fluctuations, HS regulates daytime power generation within the week, and SHS undertakes longer-term inter-weekly energy balance. When the inter-day energy imbalance exceeds the adjustment range of BESS and HS, SHS can shift surplus or deficit energy to the following week or several weeks, achieving seasonal cross-cycle adjustment. Table 1 lists the balance cycle and charge / discharge characteristics of the three types of energy storage. b. Spatial dimension: As a flexible interconnection device between feeders, the SOP further endows the MEH connected to its DC side with active power regulation capability in a wide area; in addition, the SOP itself has reactive power regulation capability, further realizing power flow optimization between feeders.

[0029] The aforementioned time and spatial dimensions correspond to the functional implementation of multi-level energy storage units and multi-port SOPs in the MEH-SOP system architecture, respectively: the time dimension achieves dynamic adjustment of energy imbalances at multiple time scales such as hourly, daily, and weekly through the hierarchical collaboration of BESS, HS, and SHS; the spatial dimension relies on the flexible interconnection capability of SOPs between feeders to enable MEHs connected to their DC side to have cross-regional active and reactive power regulation capabilities. Together, they constitute the core regulation mechanism of the MEH-SOP system based on its structural design.

[0030] Step 1.2: Based on the structural characteristics and spatiotemporal regulation mechanism of MEH-SOP, establish a multi-port SOP model in MEH-SOP. The power balance constraint of the multi-port SOP is as follows: (1) in, P t,i,SOP For SOP No. i The port is t Power at any moment; Ω k For SOP port set; P t,MEH for t The exchange power between the MEH and the DC side of the SOP at any given time.

[0031] Multi-port capacity constraints are: (2) in, Q t,i,sop for t Time Port i reactive power, S i,SOP For port i capacity.

[0032] Step 1.3: Based on the structural characteristics and spatiotemporal regulation mechanism of MEH-SOP, establish the MEH system model in MEH-SOP, with the power balance constraint as follows: (3) in, P t,BESS+ and P t,BESS- For BEES t Constant charging and discharging power; P t,SOFC and P t,SSOFC For SOFC and SSOFC in t Discharge power at any given moment; P t,AEC and P t,SAEC For AEC and SAEC in t Hydrogen production capacity at any given time; The collection is for 8760 hours throughout the year.

[0033] BEES runtime constraints are (4) (5) (6) (7) (8) (9) in, I t,BEES The 0 / 1 variable represents the charge / discharge state of BEES; γ BEES The BEES power-to-capacity ratio; E 0,BESS , E 1,BESS , E t,BEES For BEES at the initial time, time 1 and t State of charge at any given moment; E BEES Rated capacity for BEES; η BEES To improve the charge and discharge efficiency of BEES; P 1,BESS , P t,BESS The power of BESS at time 1 and time t; Δ t For time intervals; D It is a collection for all days throughout the year.

[0034] (4)-(5) Constrain the charging and discharging power of BEES; (6)-(7) Describe the load state of BEES; (8) Constrain the capacity of BEES from exceeding the rated capacity; (9) Provide a balance constraint for the state of charge of BEES, indicating that BEES returns to its initial state of charge in the last hour of each day.

[0035] The operating constraints of HS are: (10) (11) (12) (13) (14) (15) (16) in, H 0,HS and H 1,HS The load states at the initial time and time 1 of HS; η AEC and η SOFC For AEC and SOFC efficiency; P1,AEC and P 1,SOFC The power of AEC and SOFC at time 1; H t,HS for t HS load status at any given time; H HS This is the rated capacity of HS; H 0,HS and H 168,HS The HS capacity at the initial time and time 168; W Collections for each week of the year; I t,HS The variable 0 / 1 represents the hydrogen charging / discharging state of HS, indicating that HS only exists in one charging / discharging state on that day.

[0036] (10)-(11) Describe the HS load state; (12) is the HS state of charge balance constraint, indicating that the HS returns to the initial state of charge in the last hour of each week; (13) restricts the HS capacity from exceeding the rated capacity; (14)-(15) restricts the AEC and SOFC power from exceeding the rated value; (16) indicates that the HS has only one charge / discharge state each day.

[0037] SHS operating constraints are: (17) (18) (19) (20) (twenty one) (twenty two) (twenty three) in, H 0,SHS and H 1,SHS The load states at the initial time and time 1 of HS; η SAEC and η SSOFC For SAEC and SSOFC efficiency; P 1,SAEC and P 1,SSOFC The power of SAEC and SSOFC at time 1; H t,SHS for t SHS load status at any time; H SHS This is the rated capacity of the SHS.H 0,SHS and H 168,SHS The HS capacity at the initial time and time 168; I t,SHS The 0 / 1 variables describe the hydrogen charging and discharging state of the SHS, indicating that the SHS has only one charging and discharging state each week.

[0038] (17)-(18) Describe the SHS load state; (19) is the SHS state of charge balance constraint, indicating that the SHS returns to the initial state of charge in the last hour of the year; (20) restricts the SHS capacity from exceeding the rated capacity; (21)-(22) limit the SAEC and SSOFC power; (23) indicates that the SHS has only one charge / discharge state per week.

[0039] Step 2: Construct an optimal capacity configuration model for MEH-SOP based on two-stage spatiotemporal decoupling. The model calculates the optimal capacity configuration for various flexible resources, including the capacity configuration results for Smart Soft Switching (SOP), battery energy storage (BESS), hydrogen energy storage (HS), seasonal hydrogen energy storage (SHS) in multi-timescale electric-hydrogen hybrid energy storage systems, and their energy conversion equipment (such as electrolyzers and fuel cells). It also includes the energy / power flow of all the above equipment over 8760 hours throughout the year, as well as the investment costs, operation and maintenance costs, and revenue from the operation and maintenance of the above equipment, network losses in the distribution network, and costs incurred due to curtailment of solar / wind power.

[0040] This invention proposes a two-stage MEH-SOP capacity configuration method based on spatiotemporal decoupling: the first stage focuses on the time dimension, performing capacity configuration of the MEH system and introducing a fast solution strategy based on STL to improve computational efficiency; the second stage focuses on the spatial dimension, constructing a multi-port SOP capacity configuration model to support flexible resource sharing in typical scenarios. Simultaneously, the quadratic terms of the two-stage model are processed, and the overall framework of the two-stage configuration model is given.

[0041] Step 2.1: Focusing on the time dimension, a MEH capacity optimization model is constructed based on 8760 hours of annual load data. The STL trend feature extraction algorithm is introduced to pre-determine the seasonal hydrogen storage (SHS) charge / discharge state, thereby reducing the model dimensionality and improving solution efficiency. The output of this step includes the optimal capacity of various electric-hydrogen energy storage devices and the power interaction process with the DC side of the SOP.

[0042] This invention establishes a MEH configuration model based on a full-network 8760 time-series operation simulation, and utilizes time-of-use pricing to guide MEHs to discharge during peak electricity consumption periods. Simultaneously, a fast solution strategy based on STL is proposed.

[0043] Step 2.1.1: Construct the MEH optimal configuration model.

[0044] The objective function for the first stage considers the equivalent annual investment cost of various equipment in the MEH. C inv,1 Maintenance costs of various equipment C main,1 The cost of abandoning light C apv,1 and MEH electricity sales revenue based on time-of-use pricing C m,1 The objective function is as follows: First-stage objective function: (twenty four) objective function C 1 The calculation methods for each sub-function are explained in detail in (25)-(29).

[0045] The investment cost, taking into account the time value of the investment amount, is calculated as follows: (25) (26) (27) Wherein, Ψ represents the collection of various types of equipment; τ i For equipment i Capital recovery factor, r For interest rates, LT-SOP i For equipment i Service life; C i,inv,1 For equipment i The investment cost, P e,i For equipment i Rated power / capacity.

[0046] Maintenance costs (28) in, C i,main,1 For equipment i Maintenance costs. C i,main,1 is the maintenance cost of device i . Cost of curtailing wind / solar energy (29) Where abs() is the absolute value function, and min(0,X) means retaining only the part where X<0 (i.e., when the MEH absorbs power). P t,MEH When the net load of the entire network is less than the DN and the power is negative, the difference is considered as wind / solar power curtailment. P t,net This represents the net load power of the entire network before the DN was connected to the MEH. f DG The cost of DG forgoing light / wind; P t,MEH The absorbed power is negative, therefore (29) P t,net and P t,MEH Make a mistake.

[0047] MEH (Multi-access Edge Harbor) operating revenue is achieved by guiding MEH to release electricity during peak hours through time-of-use pricing signals, thereby ensuring energy balance across multiple time scales while improving the system's economic efficiency and power flow optimization.

[0048] (30) in, P t,MEH,dis express t The power at which the MEH discharges to the DC side of the SOP at any given moment; λ t,grid It is a time-of-use electricity price.

[0049] The MEH system primarily aims to improve the utilization of DG energy in DN under extremely high renewable energy penetration. Therefore, the MEH only charges when the net load of the entire network is negative. Simultaneously, one assumption of this paper is that DG energy that would otherwise be unusable can be stored through the MEH without any electricity purchase cost. Therefore, (30) there is no electricity purchase cost.

[0050] Constraints (31) The constraints of this invention are mainly for MEH charging and discharging, that is, when DN... P t,net When >0, P t,MEH Discharge to DN, and conversely charge it.

[0051] Step 2.1.2: Solve the model using a fast solution strategy based on STL.

[0052] Because the first-stage model is large in scale and contains a large number of 0 / 1 decision variables with complex spatiotemporal coupling relationships, the solution difficulty is significantly increased, making it difficult to obtain the global optimal solution within an acceptable time. As one of the key decision variables, the charge and discharge state of the SHS is used to reflect its cross-cycle regulation effect on energy surplus or shortage on an annual scale. Compared with the high-frequency fluctuations dealt with by BESS and HS, the long-term trend fluctuations dealt with by SHS are more identifiable and easier to accurately identify. Therefore, this paper predicts and locks the annual charge and discharge state of SHS by analyzing the energy surplus / shortage trend in the annual power series, thereby determining the values ​​in formulas (21)–(22) before optimization. I t,SHS Once determined, the influence of equation (23) on the model size and complexity can be eliminated simultaneously, significantly improving the solvability and computational efficiency of the model. The trend characteristics of the unbalanced power of the source load throughout the year can be revealed through the decomposition and analysis of its time series. The STL algorithm (Seasonal and Trend decomposition using Loess) is an effective time series feature extraction method. It uses locally weighted scatterplot smoothing (LOESS) to separate the trend, period, and residual in the unbalanced power series. The core idea of ​​LOESS is to dynamically capture the local characteristics of the series by nonparametric smoothing based on the data of the local region and by assigning a weight regression function to each scatter point. The weight decreases according to the distance between points, so that the data points that are closer have a greater impact on the regression, thereby enhancing the local sensitivity in the series analysis. The STL algorithm extracts the trend component step by step. T t Periodic components S t and residual components R t , will be the whole year P t,net The time series can be decomposed into the following form: initialization St,0 =0 indicates that there are no periodic components in the initial stage. The subsequent steps begin... m In each iteration, all components are updated gradually until convergence.

[0053] Trend component calculation Tt To represent long-term changes in a sequence, the LOESS parameter is used. Pt,net The data obtained by smoothing the data after subtracting the components from the previous iteration cycle is as follows: (32) in, mThis indicates the number of iterations. LOESS is a locally weighted regression smoothing method that can capture the long-term trend of a time series.

[0054] 3) Seasonal Component Calculation St,m from Pt,net Remove from Tt,m Then according to the cycle np Perform smoothing: (33) And further ensure St,m The periodicity is determined by averaging the values ​​within each period: (34) in np Take 24 hours k The number of repetitions in a periodic cycle.

[0055] Residual component calculation Rt,m It is to remove Tt,m and St,m The latter part: (35) Robust processing To improve the robustness of the STL algorithm to outliers, in Tt,m and St,m Weight adjustment is introduced during the extraction process. wt,w Calculate using the Bisquare function: (36) The Bisquare function is defined as follows: (37) Adjusted Tt,m、St,m The formula is: (38) (39) Convergence conditions Set convergence criteria and determine Tt,m Whether the change between two consecutive iterations is below a threshold ε.

[0056] (40) If the condition is met, stop the iteration; otherwise, return to step 2 to continue the calculation.

[0057] Trend Statistics Based on the decomposed Tt,m Calculate inter-week imbalance electricity: (41) SHS charging and discharging actions confirmed (42) Therefore, the whole year can be calculated based on (42). Pt,net Long-term trend features are mapped to weekly charge and discharge actions of SHS, reducing the search range of the model solution and lowering the solution time.

[0058] (Is this part the same as step 2.1?) Step 2.2: Construct an optimal configuration model for the multi-port capacity of SOP that considers the power flow optimization constraints of the distribution network, and construct typical daily scenarios and the interactive power boundary between MEH and SOP to achieve spatial dimension optimization configuration of flexible interconnection resources.

[0059] Step 2.2.1: This section constructs the SOP optimal capacity configuration model.

[0060] The optimal SOP (Start of Production) configuration model needs to consider the annual investment cost of SOPs. Cinv, 2. Maintenance costs Cmain, 2. Active power loss cost Closs The objective function is as follows: Second-stage objective function (43) objective function C 2 The calculation methods for each sub-function are explained in detail in (44)-(48).

[0061] Investment costs (44) (45) (46) in, Ψ SOP This is a collection of ports for each SOP; LT-SOP SOP service life; C i,inv,2 SOP port i Investment costs; S i,SOP For port i capacity; τ SOP This indicates the capital recovery rate of the Standard Operating Procedure (SOP). LT-SOP This refers to the service life of SOP equipment.

[0062] Maintenance costs (47) in, C i,main,2 SOP port i Maintenance costs.

[0063] Network loss cost (48) in, op For each typical scenario of M, the first op One scenario; For the set of routes; for t Time-of-use electricity pricing; At time t mn The square of the line current; op represents the th scene in M. It is a resistor.

[0064] Step 2.2.2, Second Stage SOP Multi-Port Capacity Optimal Configuration Constraints DN power flow and safe operation constraints (49) (50) in, π(:, j) Let j represent the set of nodes whose endpoint is node j. δ(j, :) This represents the set of nodes starting from node j. P ij,op,t and Q ij,op,t Let represent the active power and reactive power of branch ij at time t, respectively. Let represent the net load of the node. η sop This indicates whether the node is connected to the SOP variable. P jk,op,t and Q jk,op,t These represent the active power and reactive power transmitted from the current node to the child node, respectively.

[0065] (51) (52) in, , It is the square of the voltage. It is the square of the current. , The square of the active and reactive power of line ij (53) in, Let ij be the maximum value of the line current. , These are the upper and lower limits of the voltage at node i. Let be the square of the voltage at node i.

[0066] MEH and SOP DC-side power interaction constraints The MEH is the DC side integrated within the SOP. Therefore, when establishing the SOP operation model, the interaction power between each MEH device and the SOP DC side, obtained from the first stage solution, must be used as a constraint. This constraint is shown in (1). Steps 2.2.3, 2.2.1, and 2.2.2 still lack the input for a typical day; therefore, a typical day is further set here. This paper selects the typical day for the second-stage SOP multi-port capacity configuration model based on two considerations: High-power scenario on the DC side of SOP: Considering that MEH is integrated on the DC side of SOP, when it needs to rapidly charge and discharge a large amount of electrical energy to the DC side of SOP in a short period of time, SOP should have corresponding power generation / absorption capabilities. Therefore, from the annual operating data of MEH obtained in the first stage, several typical days in which MEH exchanges high-power energy with the DC side of SOP are selected as key inputs for configuring the rated capacity of SOP.

[0067] Severe overvoltage scenario in DN: The SOP has multi-terminal active / reactive power coordinated control capabilities, which can be used to alleviate the problem of voltage exceeding limits at the feeder end. To demonstrate its voltage regulation efficiency, it is also necessary to select a typical day with severe node voltage exceeding limits throughout the year before the MEH is configured. This type of scenario can be used to guide the SOP to take voltage support capabilities into account in the second-stage capacity configuration.

[0068] Based on the aforementioned method for configuring intelligent soft switching and hybrid electric-hydrogen energy storage based on spatiotemporal decoupling, this invention was verified using a regional power distribution system. This system comprises three 11.4kV medium-voltage feeders, with tie switches located at nodes 8, 21, and 27, as follows... Figure 1 As shown in the figure. The MEH-SOP modification location configured in this paper is the tie switch location. The source and load basic data comes from a certain distribution network (8760 hours), and the DG penetration rate is approximately 102%. The relevant model parameters and TOU pricing are shown in Tables 2 and 3.

[0069] This embodiment sets up 6 cases for analysis, and the differences between the cases are shown in Table 4.

[0070] Table 2 Equipment Parameters

[0071] Table 3 Time-of-use electricity prices

[0072] Table 4. Setup of all examples in this paper

[0073] To verify the effectiveness of extracting SHS charge / discharge state features based on the STL method, this embodiment selects Case 1 and Case 2 for comparative analysis. The solution time is shown in Table 5. Case 1 was solved within 6963.80 seconds (approximately 1.93 hours), while Case 2 failed to converge within the set maximum solution time of 12 hours (43200 seconds). Intermediate results are shown in Table 5. Figure 2 The results show that the two are basically consistent. This result indicates that the main difficulty in solving the first stage lies in handling the 0 / 1 variables of the SHS's charge and discharge state throughout the year. By adopting the STL method, this paper can accurately determine these difficult-to-handle 0 / 1 variables in advance, thereby effectively reducing the difficulty of solving the model.

[0074] Table 5. SOP Configuration Results for Case 1 and Case 2

[0075] To further verify the practical effect of this method, Table 6 lists the configuration of the MEH system in Case 1 and Case 2. The total cost of Case 1 increased by only 2.56% compared to Case 2. The results show that the two schemes have little difference in configuration capacity and power, indicating that the STL method does not introduce significant deviations while ensuring optimization accuracy.

[0076] In summary, Case 1, which uses the STL method to determine the annual charge and discharge status of the SHS in advance, not only ensures the accuracy of the optimization results but also significantly reduces the solution time of the model, proving that this method has good practicality and computational advantages in the configuration of large-scale, multi-device coupled MEH systems.

[0077] Table 6 MEH Configuration Results in Case 1 and Case 2

[0078] Table 7 SOP Configuration Results in Case 1 and Case 2

[0079] Based on the selection criteria for typical days (i.e., high-power scenarios on the DC side of the SOP and severe voltage exceedance scenarios on the DN side), this embodiment selected 10 typical days for the second-stage SOP capacity configuration. The results are shown in Table 7. Case 1 and Case 2 have completely identical configurations at each port, indicating that the fast solution strategy proposed in the first stage does not affect the accuracy of the optimization results in the second stage. Further analysis revealed that the maximum power moment at the SOP port in both cases occurred at the moment of minimum net load throughout the year. Tp,max Furthermore, no wind or solar power curtailment occurred at this time. This means that both cases absorbed all the surplus DG energy in the DN at this moment, therefore, their configuration results are the same. This analysis indicates that the optimization process in the second stage is highly dependent on the critical moment (…). T p,max The system's operating status was monitored. During these critical adjustment moments, Case 1 and Case 2 exhibited highly consistent operating characteristics. This indirectly verifies the accuracy and engineering applicability of the input results generated by the first-stage fast solution strategy.

[0080] To demonstrate the necessity of the multi-timescale hybrid energy storage system set up inside the MEH-SOP in this embodiment, various energy storage configuration methods inside the MEH-SOP were studied. The results are shown in Table 8.

[0081] Table 8 Configuration Results for Case 1 and Other Cases

[0082] Case 3, with its BEES and SHS configurations, inevitably faces severe wind / solar curtailment when several consecutive days of strong wind and solar power occur, due to the lack of a weekday energy transfer device. While SHS offers a more significant price advantage over BEES, it is constrained by weekday energy balance limitations and cannot fully compensate for the capacity loss of SHS. This results in a substantial improvement in both BEES and SHS in Case 3 compared to Case 1, leading to a higher configuration cost for Case 3. Case 4, with its BEES and SHS configurations, is limited by a single weekly charge / discharge cycle. SHS accumulates energy during the day and week, improving the absorption rate and significantly expanding the configuration capacity. Furthermore, during the SHS discharge cycle, BEES becomes the main device for absorbing new energy, resulting in a significant increase in BEES capacity. Therefore, Case 4 has a higher configuration cost than Case 1. Case 5 only has BEES, and the BEES energy remains balanced during the day, with no energy transfer between day and week. The absorption methods are extremely limited, resulting in severe wind / solar curtailment and failing to meet the reasonable utilization of DG energy under extremely high DG penetration. Furthermore, as shown in Table 7, the SOP cost for Case 1 in Phase 2 is 973,100 yuan, while the total cost difference between Case 1 and Case 3-6 in Phase 1 MEH configuration is significantly greater than 973,100 yuan. In other words, even comparing the total cost of the two-phase MEH-SOP for Case 1 with the MEH cost of only Phase 1 for Case 3-6, Case 1 still demonstrates higher economic efficiency. Therefore, Case 1 exhibits the best overall economic efficiency.

[0083] To verify the optimization effect of MEH-SOP on DN, Case 1 was compared with Case 6, which did not undergo any equipment configuration. Table 9 shows the maximum voltage deviation values ​​for 10 typical days selected for the second-stage SOP configuration. Among them, Case 1 reduced the maximum voltage deviation by 2.79% and reduced the 10-day active power loss cost by 75%.

[0084] Table 9. Optimization effects of Case 1 and Case 2 on DN

[0085] Figure 3 The per-unit voltage values ​​for node 10 under Case 1 and Case 6 configurations are shown. The voltage deviation in Case 1 is significantly smaller than that in Case 6. This is due to the MEH-SOP being connected at nodes 8, 21, and 27, enabling continuous optimization of the power flow of the three feeders through active / reactive power interaction.

[0086] 5.3.2 Analysis of Multi-Port SOP Operation in MEH-SOP Figure 4 and Figure 5 This demonstrates the power interaction between SOP-8 and feeder 1 (with outflow node 8 as the positive direction). At midday, SOP-8 absorbs active power from the DG, stores it through the MEH on the DC side, and releases it at appropriate times. Simultaneously, at midday, SOP-8 absorbs reactive power to lower the voltage, and at night and other times, it generates reactive power to raise the voltage. This analysis shows that MEH-SOP can effectively optimize the DN operating state and achieve flexible adjustment of the DN.

[0087] This invention significantly alleviates the problem of power and energy imbalance across multiple time scales, effectively improves power flow distribution, and achieves rapid solution while maintaining optimization accuracy. Furthermore, the proposed two-stage spatiotemporal decoupled configuration model and its rapid solution strategy are particularly suitable for optimization problems with large variable scales and complex spatiotemporal coupling, demonstrating good engineering adaptability and potential for widespread application.

[0088] It should be emphasized that the embodiments described in this invention are illustrative rather than limiting. Therefore, this invention includes, but is not limited to, the embodiments described in the specific implementation. Any other implementations derived by those skilled in the art based on the technical solutions of this invention are also within the scope of protection of this invention.

Claims

1. A method for configuring intelligent soft switching and electro-hydrogen hybrid energy storage based on spatiotemporal decoupling, characterized in that: Includes the following steps: Step 1: Construct the MEH-SOP capacity optimal configuration model based on two-stage spatiotemporal decoupling; Step 2: Construct the MEH-SOP capacity optimal configuration model. The optimal configuration capacity of various flexible resources can be obtained through model calculation, including the capacity configuration results of intelligent soft switching SOP, battery energy storage BESS, hydrogen energy storage HS, seasonal hydrogen energy storage SHS and their energy conversion equipment in multi-timescale electric-hydrogen hybrid energy storage system, as well as the annual energy / power flow of all equipment.

2. The method for configuring intelligent soft switching and electro-hydrogen hybrid energy storage based on spatiotemporal decoupling according to claim 1, characterized in that: The specific implementation method of step 1 is as follows: Step 1.1: Analyze the structural characteristics and spatiotemporal regulation mechanism of MEH-SOP; Step 1.2: Based on the structural characteristics and spatiotemporal adjustment mechanism of MEH-SOP, establish a multi-port SOP model in MEH-SOP; Step 1.3: Based on the structural characteristics and spatiotemporal adjustment mechanism of MEH-SOP, establish the MEH system model in MEH-SOP.

3. The method for configuring intelligent soft switching and electro-hydrogen hybrid energy storage based on spatiotemporal decoupling according to claim 1, characterized in that: The specific implementation method of step 1 is as follows: the MEH in the MEH-SOP system consists of an energy conversion device and a multi-level energy storage unit: energy conversion device and multi-level energy storage unit.

4. The method for configuring intelligent soft switching and electro-hydrogen hybrid energy storage based on spatiotemporal decoupling according to claim 1, characterized in that: The multi-port SOP model in step 1.2 MEH-SOP is as follows: Multi-port SOP power balance constraints: ； in, P t,i,SOP For SOP No. i The port is t Power at any moment; Ω k For SOP port set; P t,MEH for t The exchange power between the MEH and the DC side of the SOP at any given time; Multi-port capacity constraints: ； in, Q t,i,sop for t Time Port i reactive power, S i,SOP For port i capacity.

5. The method for configuring intelligent soft switching and electro-hydrogen hybrid energy storage based on spatiotemporal decoupling according to claim 1, characterized in that: In step 1.3, the MEH system model in MEH-SOP includes power balance constraints, BEES operation constraints, HS operation constraints, and SHS operation constraints.

6. The method for configuring intelligent soft switching and electro-hydrogen hybrid energy storage based on spatiotemporal decoupling according to claim 1, characterized in that: Step 2 includes the following steps: Step 2.1 Focusing on the time dimension, we construct a MEH capacity optimization model based on 8760 hours of load data throughout the year, and introduce the STL trend feature extraction algorithm to preset the seasonal hydrogen storage SHS charge and discharge state, thereby reducing the model dimension and improving the solution efficiency. Step 2.2: Construct an optimal configuration model for the multi-port capacity of SOP that considers the power flow optimization constraints of the distribution network, and construct typical daily scenarios and the interactive power boundary between MEH and SOP to achieve spatial dimension optimization configuration of flexible interconnection resources.

7. The method for configuring intelligent soft switching and electro-hydrogen hybrid energy storage based on spatiotemporal decoupling according to claim 7, characterized in that: Step 2.1 includes the following steps: Step 2.1.1: Construct the MEH optimal configuration model; Step 2.1.2: Solve the model using a fast solution strategy based on STL.

Images