A hydrogen energy storage system configuration method based on load partition collaborative control of power grid

Abstract

The application discloses a hydrogen energy storage system configuration method based on load partition cooperative control of a power grid, and comprises the following steps: extracting the active-reactive comprehensive sensitivity matrix of each node of the power grid network; drawing a hierarchical clustering diagram of a clustering process according to the ward distance between various types, and determining the number of classification groups, i.e., node partition, according to the classification height comparison and merging of the flat period; taking the minimum network loss, the minimum voltage deviation and the minimum power fluctuation of the distribution network as the target to solve the node partition and the hydrogen energy storage system site selection; taking the initial investment cost, the maintenance cost and the operation cost of the distribution network containing distributed power as the objective function to solve the capacity configuration and the operation state; and using the meta-heuristic algorithm to iteratively solve the operation strategy of the three configuration schemes, i.e., wind and light curtailment hydrogen production, single hydrogen energy storage and hydrogen energy storage as the main electric energy storage supplemented by auxiliary hydrogen energy storage. The application optimally plans and arranges the hydrogen energy storage station containing distributed power, so as to realize the site selection and capacity determination strategy considering the flexibility and economic indicators.

Description

Technical Field

[0001] This invention relates to the field of hydrogen energy storage configuration technology for wind and solar power distribution network systems, and in particular to a method for configuring a hydrogen energy storage system for a power grid under load zoning coordinated control. Background Technology

[0002] In new power systems, energy characteristics have changed significantly, shifting from storable and transportable fossil fuels to meteorologically dependent and stochastic wind and solar energy resources. This results in a high degree of uncertainty surrounding the integration of high-proportion wind and solar power into the grid. As the penetration rate of photovoltaic and wind power in these new power systems continues to increase, the stochasticity and limited transmission capacity become increasingly acute, posing challenges to grid absorption. To address the balancing capacity requirements of these new power systems, efforts are being made to gradually decouple power generation through energy storage, continuously exploring the potential of energy storage as the deep regulation capacity of existing power systems is exhausted.

[0003] Currently, considerable research has been conducted on wind-solar hybrid power generation and optimized operation of hydrogen energy storage. The "Capacity Configuration of Integrated Electric-Hydrogen Energy Stations Including Photovoltaics and Hydrogen Storage" proposed by Liu Shangqi, Hu Jian, Zhang Xiaojie, et al.; the "Robust Optimization Configuration of Centralized Shared Energy Storage for Multi-Scenario Regulation Needs" proposed by Du Xili, Li Xiaozhu, Chen Laijun, et al.; and the "Multi-Indicator Comprehensive Evaluation Research of Distributed Energy Systems" proposed by Dong Fugui, Zhang Ye, Shang Meimei, etc., can assess the techno-economic feasibility of different energy storage capacity configuration schemes through comprehensive analysis of economic indicators, providing a scientific basis for system planning and operation. The research papers "Multi-objective Optimization Design of Wind-Solar Hybrid Power Supply System" proposed by Yang Qi, Zhang Jianhua, Liu Zifa, etc., "Hydrogen Energy Storage Integrated Hybrid Renewable Energy Systems: A Review Analysis for Future Research Directions" proposed by Arsad AZ, Hannan MA, Al-Shetwi AQ, etc., and "A Critical Review on the Current Technologies for the Generation, Storage, and Transportation of Hydrogen" proposed by Faye O, Szpunar J, Eduok U, summarize the current status of hydrogen production technology and coordination strategies for new energy complementary systems, and demonstrate the environmental and economic feasibility of hydrogen energy storage. The research paper "Planning Optimization Research of Distributed Photovoltaic and Hydrogen Hybrid Energy Storage Oriented to Multi-Energy Complementarity [J]. Power Grid Technology" proposed by Wang Yongli, Xiang Hao, Guo Lu, etc., provides an energy storage operation scheme for a photovoltaic seasonal hydrogen hybrid energy storage model with economic and environmental objectives. The "Optimal Configuration of Hydrogen Storage in Low-Carbon Industrial Park Integrated Energy Systems Considering Electrothermal-Gas Coupling Characteristics" proposed by Xiong Yufeng, Chen Laijun, Zheng Tianwen, and others, and the "Coordinated Scheduling of Multi-Building Smart Parks Considering Electricity-Hydrogen Complementarity under Dual Carbon Objectives" proposed by Fan Hong, Yu Weinan, Liu Lu, and others, use hydrogen storage as an energy conversion hub and coordinate its configuration in integrated energy park scenarios to reduce system energy supply costs and carbon emissions. The "An Integrated Electricity-Hydrogen Market Design for Renewable-Rich Energy System Considering Mobile Hydrogen Storage" proposed by Zhu J, Meng D, Dong X, and others, and the "Optimal Design of Electricity-Hydrogen Coordinated Energy Storage System for Chemical Systems to Absorb Renewable Energy" proposed by Wang Jing, Kang Lixia, Liu Yongzhong, establish a capacity configuration optimization model for electricity-hydrogen coupling systems with the goal of minimizing annual comprehensive costs and annual carbon emission costs.Most research on hydrogen energy storage focuses on economics and models for the configuration and operation of hydrogen energy storage, treating hydrogen energy storage as part of a comprehensive energy system for synergistic optimization. However, research on hydrogen energy storage configuration has not considered the impact of the existing power grid structure on the layout of hydrogen energy storage.

[0004] Power quality issues in distribution networks under new energy systems are becoming increasingly prominent, and grid regulation is placing greater emphasis on building dynamic regulation architectures. The concept of "zonal aggregation and hierarchical control" can incorporate the diversification of resource characteristics and spatiotemporal factors into centralized energy storage configuration schemes. The "Hierarchical Capacity Configuration of Wind-Solar-Storage Virtual Power Plants Considering Reliability and Flexibility" proposed by Bai Xueyan, Fan Yanfang, Liu Yujia, and others reveals that with the increasing penetration rate of distributed power sources in the grid, active distribution networks can achieve stable control of node voltages, thereby suppressing the intermittent output of wind and solar resources and optimizing active power losses. The "Hierarchical and Zonal Voltage Regulation Strategy for Distribution Networks Based on HEM Sensitivity" proposed by Chen Wenjin, Gan Wen, Zhang Jun, and others, and the "Two-Stage Voltage Control Zoning Method Based on Comprehensive Sensitivity" proposed by Li Yingliang, Wang Kang, Gao Zhaodi, and others, reveal how the comprehensive sensitivity of load zoning reflects the impact of active and reactive power output on zoning, providing theoretical support for microgrid partitioning. The "Active Distributed Reactive Power Optimization Method for Distribution Networks Based on Equal Network Loss Increment Rate" proposed by Wang Yong, Chen Ming, Zhang Ming, etc.; the "Distributed Power Optimization Configuration Considering Reconfiguration and Microgrid Partitioning" proposed by Bian Yiheng, Gui Hengli, Bie Chaohong, etc.; and the "Voltage Reactive Power Self-Discipline-Collaborative Control for New Energy Access to Distribution Networks" proposed by Song Minggang, Tao Jun, Zhang Huaying, etc., address load and new energy output fluctuations by implementing partitioned self-discipline control of dynamic reactive power resources in each sub-region to achieve "partitioned autonomy," effectively regulating new energy fluctuations. The "Data-Driven Load Partitioning for Energy Storage Capacity Configuration" proposed by Gao Ronggang, Yang Yang, Yuan Tiejiang, etc., starts from a data-driven approach, utilizing the anti-peak-shaving characteristics and large fluctuations of new energy as load partitioning indicators, verifying the accuracy and control precision of the energy storage configuration scheme under load partitioning. Hydrogen energy storage, as an adjustable load integrated into the grid, makes centralized optimization modes difficult to meet the reliable operation requirements of the grid. Based on partitioned and hierarchical optimization theory, hydrogen energy storage configuration can be matched to the grid architecture.

[0005] In summary, during the transformation of the new power system structure, it is particularly crucial to endow new energy sources with stability through energy storage and to seek configuration modes suitable for large-scale consumption. Summary of the Invention

[0006] To address the aforementioned technical problems, this invention provides a method for configuring a hydrogen energy storage system for a power grid under load zoning coordinated control.

[0007] To achieve the above objectives, the present invention is implemented according to the following technical solution:

[0008] A method for configuring a hydrogen energy storage system for a power grid under load-zone coordinated control includes the following steps:

[0009] S1. Extract the combined active and reactive power sensitivity matrix of each node in the power grid network;

[0010] S2. Each node to be partitioned in the power grid network is divided into its own category, that is, there are n initial groups, and each group has only one individual.

[0011] S3. Calculate the electrical distance matrix between n groups;

[0012] S4. Use Ward's method to merge two groups with the smallest electrical distance, i.e., calculate the increment of the sum of squared variances (ESS) within the groups before and after merging.

[0013] S5. Recalculate the population increment I = ESS before and after. 合并后 -ESS 合并前 This allows the two classes with the smallest increment to be merged;

[0014] S6. Repeat steps S4-S5 until they are merged into one category;

[0015] S7. Draw a hierarchical clustering diagram of the clustering process based on the ward distance between each category, and determine the number of classification groups, i.e., node partitions, based on the flat period that appears when the classification heights are compared and merged.

[0016] S8. Solve the node partitioning and hydrogen energy storage system site selection with the objectives of minimizing distribution network loss, minimum voltage deviation, and minimum power fluctuation;

[0017] S9. Using the initial investment cost, maintenance cost, and operation cost of a distribution network with distributed power sources as objective functions, solve for the capacity configuration and operating status. Finally, use metaheuristic algorithms to iteratively solve for the operating strategies of three configuration schemes: wind and solar power curtailment for hydrogen production, single hydrogen energy storage, and hydrogen energy storage as the main energy storage and power storage as the auxiliary energy storage.

[0018] Further, step S1 specifically includes:

[0019] S11. Constructing the Jacobian matrix of the power grid network:

[0020]

[0021] In the formula: ΔP and ΔQ are the unbalanced amounts of active and reactive power injected at the node, respectively; Δδ and ΔU / U are the correction amounts for voltage phase angle and amplitude; H is... Subarray, N is Jacobi subarray, J is Jacobi subarray, L is Sub-array;

[0022] S12. Constructing a traditional sensitivity matrix using the Jacobian matrix:

[0023] S = -L -1 (2)

[0024] In the formula: S is the sensitivity matrix constructed from the Jacobian matrix;

[0025] S13, Through The Jacobian submatrix N is used to construct the comprehensive sensitivity matrix. To ensure the accuracy of load node partitioning, only the m×m matrix N′ containing the load node dimension in the N matrix is ​​considered:

[0026]

[0027] Consider the combined active and reactive power sensitivity matrix S′ as follows:

[0028]

[0029] Furthermore, step S3 specifically includes:

[0030] S31. Calculate the electrical distance between load nodes using the combined active and reactive power sensitivity matrix. For the voltage coupling α between the voltage phase angle changes at nodes i and j... ij Represented as:

[0031]

[0032] In the formula: The sensitivity model integrates the weak coupling effect of reactive power on voltage amplitude and active power on voltage phase angle deviation, and its elements correspond to the elements of the S matrix.

[0033] S32. To extract the electrical coupling relationship between nodes in the power network, the electrical distance between nodes i and j is further obtained as follows:

[0034] D ij =-lg(α) ij α ji (6).

[0035] Furthermore, in step S8, the hydrogen energy storage system comprises an electrolyzer, a hydrogen storage tank, a fuel cell, and an electrical energy storage system;

[0036] Hydrogen production capacity of electrolyzer:

[0037] V el (t)=η el P el (t)ρ (7)

[0038] In the formula: V el (t) represents the amount of hydrogen produced during time period t, η el For the efficiency of the electrolyzer, P el(t) represents the electrolyzer power during time period t, and ρ represents the amount of hydrogen produced by 1 kW·h of electrical energy.

[0039] The mathematical model for fuel cell operation can be simplified as follows:

[0040]

[0041] In the formula: P discharge (t) represents the electrical power released by the SOFC during the time period t, in kW; m represents the amount of hydrogen consumed by the SOFC during time period t. 3 η D The energy conversion efficiency of SOFC; The amount of hydrogen required to produce 1 kWh of electricity, m 3 ;

[0042] Hydrogen storage tank input working status:

[0043]

[0044] Output working status:

[0045]

[0046] In the formula: V tank Let m be the volume of hydrogen gas in the storage tank. 3 σ represents the leakage rate of the hydrogen storage tank. Let m be the volume of hydrogen required within the fuel cell Δt. 3 ;

[0047] Energy storage system charging operation status:

[0048]

[0049] Discharge operating status:

[0050]

[0051] In the formula: W E (t) represents the amount of energy stored in the electrical system during time period t; and The charging and discharging power of the energy storage during time period t; η charge With η di sch arge To improve the charging and discharging efficiency of electrical energy storage;

[0052] Solving for node partitioning and hydrogen energy storage system site selection specifically includes:

[0053] Construct the objective function and constraints:

[0054] (1) Minimum network loss

[0055]

[0056] In the formula: n is the total number of nodes; k is the set of nodes that have a direct coupling relationship with node i; U i U j G represents the voltage magnitudes at nodes i and j; ij θ represents the line conductance between nodes i and j; ij The voltage phase difference between nodes i and j;

[0057] (2) Minimum voltage deviation

[0058]

[0059] In the formula: Let be the expected voltage value at node i;

[0060] (3) Minimum active power fluctuation

[0061]

[0062] In the formula: P i,max P i,min P represents the maximum and minimum power achievable by node i at time t; i,0 Let be the natural power of node i at time t;

[0063] 1) Power Constraint

[0064] In distribution network optimization, the active and reactive power balance constraints at each node are as follows:

[0065]

[0066] In the formula: P i Q i These represent the combined active and reactive power of the load at node i and the distributed generation (DG), respectively; G ij B ij These represent the conductance and susceptance of the line, respectively; δ ij Let be the voltage phase difference between nodes i and j.

[0067] 2) Node voltage constraints

[0068] U min ≤U m (t)≤U max (18)

[0069] In the formula: U min U max These represent the upper and lower limits of the voltage, respectively.

[0070] 5. The method for configuring a hydrogen energy storage system for a power grid based on load zoning coordinated control according to claim 4, wherein step S9 specifically includes:

[0071] Construct the objective function and constraints:

[0072] (1) Initial investment cost

[0073]

[0074] In the formula: C capex-ALL For the initial investment cost, LF n This is the capital recovery factor, where LF n =d(1+d) n / (1+d) n -1, d is the discount rate; n is the equipment lifespan, C capex-Hydrogen C represents the daily investment cost of a hydrogen energy storage system. capex-BESS C represents the investment cost of an energy storage system. fc C el C tank These are the unit costs of the fuel cell, electrolyzer, and hydrogen storage tank, respectively; P fc P el The power configurations for the fuel cell are as follows; Q tank Configure the capacity of the hydrogen storage tank; C ba C in These are the battery cost and the battery converter cost, respectively; E BESS P represents the battery's energy storage capacity. BESS Rated power for battery energy storage;

[0075] (2) Maintenance costs

[0076]

[0077] In the formula: C MC_ALL C represents the total maintenance cost of a hydrogen energy storage station. MC_Hydrogen Maintenance costs for hydrogen energy storage systems; C MC_BESS Maintenance costs for energy storage systems; ω MC_fc ω MC_el ω MC_tank ω MC_ba ω MC_in These are the proportions of maintenance costs for fuel cells, electrolyzers, hydrogen storage tanks, and batteries and converters to the initial investment cost;

[0078] (3) Operating costs

[0079]

[0080] In the formula: C opex_ALLC represents the total operating cost of a hydrogen energy storage station. opex_BESS C is the operating cost of the energy storage system. opex_Hydrogen For the operating cost of hydrogen energy storage systems; C price (t) represents the grid electricity price at time t; W buy (t), W sell (t) represents the amount of electricity purchased and sold by the energy storage system to the grid at time t, respectively; η D η C These are the gas-to-electricity efficiency and the electricity-to-gas efficiency of the hydrogen energy storage system, respectively. This refers to the discharge power of the hydrogen energy storage system. Power for charging hydrogen energy storage systems;

[0081] (1) Constraints related to hydrogen energy storage system: Constraints related to hydrogen energy storage system include upper and lower limits of hydrogen energy storage power, operating status constraints of hydrogen energy storage system, and capacity balance constraints of hydrogen energy storage system.

[0082] Hydrogen energy storage power upper and lower limits constraints:

[0083] P Hydrogen·min ≤P Hydrogen·min (t)≤P Hydrogen·max (twenty two)

[0084] Operating constraints of hydrogen energy storage systems:

[0085]

[0086] Hydrogen storage capacity balance constraints:

[0087]

[0088] In the formula: P Hydrogen·min (t), P Hydrogen·max P Hydrogen·min The charging and discharging power and upper and lower limits for hydrogen energy storage;

[0089] (2) Constraints related to energy storage systems

[0090]

[0091] SOC min ≤SOC≤SOC max (26)

[0092] (3) Power balance constraint: The power balance constraint means that the actual grid power demand and the sum of the actual load and grid loss in the system are equal. Here, the constraint conditions are simplified and the influence of grid loss is ignored:

[0093]

[0094] In the formula: P load(t) represents the real-time power demand of the electrical load, δ 1,t The variables are 0-1, representing sufficient peak shaving and insufficient peak shaving, respectively, P 1,t This represents the peak-shaving difference between the wind, solar, and hydrogen storage energy stations at time t.

[0095] Compared with the prior art, the present invention has the following beneficial effects:

[0096] This invention treats the hydrogen production and storage system as an adjustable load, and determines the hydrogen energy storage site selection scheme based on the clustering hierarchical partitioning theory, thereby achieving precise energy storage configuration and autonomous adjustment in each region.

[0097] (2) This invention proposes a two-stage optimization configuration model for the distribution network of the wind-solar-hydrogen energy storage system that considers the optimal absorption of new energy. The upper layer of the model solves the partition model by optimizing the distribution network’s active and reactive power flow distribution. The lower layer model aims to optimize the overall cost of distribution network optimization configuration and scheduling with distributed power sources and the regional power self-management index. It completes the optimal planning and layout of hydrogen energy storage stations with distributed power sources to achieve a site selection and capacity determination strategy that takes into account both flexibility and economic indicators.

[0098] (3) Based on the metaheuristic improved algorithm, this invention verifies the effectiveness of the proposed model and the superiority of the method in this embodiment on the extended IEEE 33-node case, and performs iterative calculations on the two configuration methods. Under the requirement of rapid solution of hydrogen energy storage configuration, it is determined that the configuration scheme with hydrogen energy storage as the main component and electric energy storage as the auxiliary component has better economic efficiency. Attached Figure Description

[0099] Figure 1 This is a two-layer configuration strategy framework based on load zoning theory.

[0100] Figure 2 This is an extended diagram of the IEEE 33-node network.

[0101] Figure 3 Sensitivity analysis for distribution network nodes.

[0102] Figure 4 This is a clustering process.

[0103] Figure 5 This is the load balance curve for region A under autonomous mode.

[0104] Figure 6 Load balance curve for Region B under autonomous mode

[0105] Figure 7 This is the load balance curve for region C under autonomous mode.

[0106] Figure 8 This is the load balance curve for region A under a single hydrogen energy storage mode.

[0107] Figure 9 This is the load balance curve for Region B under a single hydrogen energy storage mode.

[0108] Figure 10 This is the regional C load balance curve under a single hydrogen energy storage mode.

[0109] Figure 11 This is the load balance curve for region A under the wind and solar curtailment mode.

[0110] Figure 12 This is the load balance curve for Region B under the wind and solar curtailment mode.

[0111] Figure 13 This is the load balance curve for region C under the wind and solar curtailment mode. Detailed Implementation

[0112] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to embodiments. The specific embodiments described herein are for illustrative purposes only and are not intended to limit the invention.

[0113] like Figure 1 As shown in the figure, this embodiment proposes a method for configuring a hydrogen energy storage system for a power grid based on load zoning coordinated control, including the following steps:

[0114] S1. Extract the combined active and reactive power sensitivity matrix of each node in the power grid network;

[0115] S11. Constructing the Jacobian matrix of the power grid network:

[0116]

[0117] In the formula: ΔP and ΔQ are the unbalanced amounts of active and reactive power injected at the node, respectively; Δδ and ΔU / U are the correction amounts for voltage phase angle and amplitude; H is... Subarray, N is Jacobi subarray, J is Jacobi subarray, L is Sub-array;

[0118] S12. Constructing a traditional sensitivity matrix using the Jacobian matrix:

[0119] S = -L -1 (2)

[0120] In the formula: S is the sensitivity matrix constructed from the Jacobian matrix;

[0121] S13, Through The Jacobian submatrix N is used to construct the comprehensive sensitivity matrix. To ensure the accuracy of load node partitioning, only the m×m matrix N′ containing the load node dimension in the N matrix is ​​considered:

[0122]

[0123] Consider the combined active and reactive power sensitivity matrix S′ as follows:

[0124]

[0125] S2. Each node to be partitioned in the power grid network is divided into its own category, that is, there are n initial groups, and each group has only one individual.

[0126] S3. Calculate the electrical distance matrix between n groups;

[0127] S31. Calculate the electrical distance between load nodes using the combined active and reactive power sensitivity matrix. For the voltage coupling α between the voltage phase angle changes at nodes i and j... ij Represented as:

[0128]

[0129] In the formula: The sensitivity model integrates the weak coupling effect of reactive power on voltage amplitude and active power on voltage phase angle deviation, and its elements correspond to the elements of the S matrix.

[0130] S32. To extract the electrical coupling relationship between nodes in the power network, the electrical distance between nodes i and j is further obtained as follows:

[0131] D ij =-lg(α) ij α ji (6).

[0132] S4. Use Ward's method to merge two groups with the smallest electrical distance, i.e., calculate the increment of the sum of squared variances (ESS) within the groups before and after merging.

[0133] S5. Recalculate the population increment I = ESS before and after. 合并后 -ESS 合并前 This allows the two classes with the smallest increment to be merged;

[0134] S6. Repeat steps S4-S5 until they are merged into one category;

[0135] S7. Draw a hierarchical clustering diagram of the clustering process based on the ward distance between each category, and determine the number of classification groups, i.e., node partitions, based on the flat period that appears when the classification heights are compared and merged.

[0136] S8. Solve the node partitioning and hydrogen energy storage system site selection with the objectives of minimizing distribution network loss, minimum voltage deviation, and minimum power fluctuation;

[0137] Hydrogen energy storage systems consist of an electrolyzer, a hydrogen storage tank, a fuel cell, and an electrical energy storage system.

[0138] Hydrogen production capacity of electrolyzer:

[0139] V el (t)=η el P el (t)ρ (7)

[0140] In the formula: V el (t) represents the amount of hydrogen produced during time period t, η el For the efficiency of the electrolyzer, P el (t) represents the electrolyzer power during time period t, and ρ represents the amount of hydrogen produced by 1 kW·h of electrical energy.

[0141] The mathematical model for fuel cell operation can be simplified as follows:

[0142]

[0143] In the formula: P discharge (t) represents the electrical power released by the SOFC during the time period t, in kW; m represents the amount of hydrogen consumed by the SOFC during time period t. 3 η D The energy conversion efficiency of SOFC; The amount of hydrogen required to produce 1 kWh of electricity, m 3 ;

[0144] Hydrogen storage tank input working status:

[0145]

[0146] Output working status:

[0147]

[0148] In the formula: V tank Let m be the volume of hydrogen gas in the storage tank. 3 σ represents the leakage rate of the hydrogen storage tank. Let m be the volume of hydrogen required within the fuel cell Δt. 3 ;

[0149] Energy storage system charging operation status:

[0150]

[0151] Discharge operating status:

[0152]

[0153] In the formula: W E (t) represents the amount of energy stored in the electrical system during time period t; and The charging and discharging power of the energy storage during time period t; η charge With η discharge To improve the charging and discharging efficiency of electrical energy storage;

[0154] Solving for node partitioning and hydrogen energy storage system site selection specifically includes:

[0155] Construct the objective function and constraints:

[0156] (1) Minimum network loss

[0157]

[0158] In the formula: n is the total number of nodes; k is the set of nodes that have a direct coupling relationship with node i; U i U j G represents the voltage magnitudes at nodes i and j; ij θ represents the line conductance between nodes i and j; ij The voltage phase difference between nodes i and j;

[0159] (2) Minimum voltage deviation

[0160]

[0161] In the formula: Let be the expected voltage value at node i;

[0162] (3) Minimum active power fluctuation

[0163]

[0164] In the formula: P i,max P i,min P represents the maximum and minimum power achievable by node i at time t; i,0 Let be the natural power of node i at time t;

[0165] 1) Power Constraint

[0166] In distribution network optimization, the active and reactive power balance constraints at each node are as follows:

[0167]

[0168] In the formula: P i Q i These represent the combined active and reactive power of the load at node i and the distributed generation (DG), respectively; G ijB ij These represent the conductance and susceptance of the line, respectively; δ ij Let be the voltage phase difference between nodes i and j.

[0169] 2) Node voltage constraints

[0170] U min ≤U m (t)≤U max (18)

[0171] In the formula: U min U max These represent the upper and lower limits of the voltage, respectively.

[0172] S9. Using the initial investment cost, maintenance cost, and operation cost of a distribution network with distributed power sources as objective functions, solve for the capacity configuration and operating status. Finally, use metaheuristic algorithms to iteratively solve for the operating strategies of three configuration schemes: wind and solar power curtailment for hydrogen production, single hydrogen energy storage, and hydrogen energy storage as the main energy storage and power storage as the auxiliary energy storage.

[0173] Construct the objective function and constraints:

[0174] (1) Initial investment cost

[0175]

[0176] In the formula: C capex-ALL For the initial investment cost, LF n This is the capital recovery factor, where LF n =d(1+d) n / (1+d) n -1, d is the discount rate; n is the equipment lifespan, C capex-Hydrogen C represents the daily investment cost of a hydrogen energy storage system. capex-BESS C represents the investment cost of an energy storage system. fc C el C tank These are the unit costs of the fuel cell, electrolyzer, and hydrogen storage tank, respectively; P fc P el The power configurations for the fuel cell are as follows; Q tank Configure the capacity of the hydrogen storage tank; C ba C in These are the battery cost and the battery converter cost, respectively; E BESS P represents the battery's energy storage capacity. BESS Rated power for battery energy storage;

[0177] (2) Maintenance costs

[0178]

[0179] In the formula: CMC_ALL C represents the total maintenance cost of a hydrogen energy storage station. MC_Hydrogen Maintenance costs for hydrogen energy storage systems; C MC_BESS Maintenance costs for energy storage systems; ω MC_fc ω MC_el ω MC_tank ω MC_ba ω MC_in These are the proportions of maintenance costs for fuel cells, electrolyzers, hydrogen storage tanks, and batteries and converters to the initial investment cost;

[0180] (3) Operating costs

[0181]

[0182] In the formula: C opex_ALL C represents the total operating cost of a hydrogen energy storage station. opex_BESS C is the operating cost of the energy storage system. opex_Hydrogen For the operating cost of hydrogen energy storage systems; C price (t) represents the grid electricity price at time t; W buy (t), W sell (t) represents the amount of electricity purchased and sold by the energy storage system to the grid at time t, respectively; η D η C These are the gas-to-electricity efficiency and the electricity-to-gas efficiency of the hydrogen energy storage system, respectively. This refers to the discharge power of the hydrogen energy storage system. Power for charging hydrogen energy storage systems;

[0183] (1) Constraints related to hydrogen energy storage system: Constraints related to hydrogen energy storage system include upper and lower limits of hydrogen energy storage power, operating status constraints of hydrogen energy storage system, and capacity balance constraints of hydrogen energy storage system.

[0184] Hydrogen energy storage power upper and lower limits constraints:

[0185] P Hydrogen·min ≤P Hydrogen·min (t)≤P Hydrogen·max (twenty two)

[0186] Operating constraints of hydrogen energy storage systems:

[0187]

[0188] Hydrogen storage capacity balance constraints:

[0189]

[0190] In the formula: P Hydrogen·min (t), P Hydrogen·max P Hydrogen·min The charging and discharging power and upper and lower limits for hydrogen energy storage;

[0191] (2) Constraints related to energy storage systems

[0192]

[0193] SOC min ≤SOC≤SOC max (26)

[0194] (3) Power balance constraint: The power balance constraint means that the actual grid power demand and the sum of the actual load and grid loss in the system are equal. Here, the constraint conditions are simplified and the influence of grid loss is ignored:

[0195]

[0196] In the formula: P load (t) represents the real-time power demand of the electrical load, δ 1,t The variables are 0-1, representing sufficient peak shaving and insufficient peak shaving, respectively, P 1,t This represents the peak-shaving difference between the wind, solar, and hydrogen storage energy stations at time t.

[0197] This embodiment selects IEEE-33 node (see...) Figure 2 The model was validated using a simulation example. A 0.75MW photovoltaic power generation system was connected at node 8, and two 0.9MW wind power generation systems were connected at nodes 21 and 27. The system rated voltage was 12.66kV, the three-phase power base value was SB = 10MVA, and the allowable node voltage range was 0.95–1.05 pu. A two-stage configuration program for hydrogen energy storage based on load partitioning clustering was developed using Matlab.

[0198] 5.2 Load Partitioning Results Based on Hierarchical Clustering

[0199] Based on the proposed definition of electrical distance, a hierarchical clustering algorithm is used to conduct a partitioned study of a 33-node distribution network. Sensitivity analysis of each node is as follows: Figure 3 As shown, the calculation process for the partitioning method is to first calculate the load nodes and active power source nodes, and then the reactive power source nodes. Figure 4To ensure strong coupling within each region and weak coupling between regions during the hierarchical clustering process of the system in this embodiment, the transition from 5 partitions to 4 partitions is relatively smooth, indicating a high merging cost. Further merging might disrupt the strong coupling within the regions, thus decomposing the system into 5 partitions. However, considering that node 1 provides network baseline parameters for the leading node, and that physical connections within partition 1 are poor during clustering, the distribution network is ultimately divided into three regions based on the physical connections and network structure: Region A includes nodes 13, 14, 15, 16, and 17; Region B includes nodes 5, 6, 7, 8, 26, 27, 28, 29, 30, 31, 32, and 33; and Region C includes nodes 9, 10, 11, and 12. The upper layer performs load zoning on the distribution network to achieve global collaborative optimization, which is beneficial for fully utilizing the precise control of energy storage zones.

[0200] Based on the load zoning results, the regional distribution network central nodes are determined according to the sensitivity of the load zoning results to serve as the site selection results for energy storage sites. The upper-level iteration results determine the site selection nodes for hydrogen energy storage sites as 12, 17, and 25. After the energy storage system is connected according to the load zoning, the energy storage operation mode can be further optimized under the premise of ensuring the continuous and stable operation of the system.

[0201] 6.3 Hydrogen energy storage station configuration results

[0202] Based on load zoning and clustering, the configuration strategies for wind-solar-hydrogen energy storage power stations are divided. Two energy storage configuration methods are considered: Method 1: Hydrogen storage is the primary energy storage, with electrical storage as a secondary energy storage; Method 2: No electrical storage is considered, and the entire power station uses hydrogen storage as the sole energy storage; Method 3: Hydrogen production is achieved using curtailed wind and solar power, i.e., no fuel cells are configured, and only hydrogen production and storage are considered using wind and solar energy. To meet the requirement of rapid solution for distribution network site selection and capacity determination, this embodiment uses a metaheuristic algorithm to solve for the two configuration strategies.

[0203] During the configuration process, energy storage is considered to reduce the operating cost of energy storage stations. The configuration scheme of Mode 1 obtained through simulation calculation is as follows: For regional network A, 0.15MW of energy storage is configured, and the capacity of fuel cells and alkaline electrolyzers are 0.12MW and 0.1MW, respectively; for regional network B, 0.17MW of energy storage is configured, and the capacity of fuel cells and alkaline electrolyzers are 0.1MW and 0.75MW, respectively; for regional network C, 0.12MW of energy storage is configured, and the capacity of fuel cells and alkaline electrolyzers are 0.9MW and 0.6MW, respectively.

[0204] The operating results of each energy storage mode after grid configuration of energy storage are as follows: Figure 5-7 As shown in the diagram, in the divided areas A, B, and C, it can be seen that the energy storage in each area meets the requirements of balancing the power grid load curve of that area, and the output of hydrogen energy storage and electrical energy storage is rationally arranged. Figure 5In Central Region A, the main renewable energy source is photovoltaic power. During midday, the hydrogen energy storage system utilizes surplus photovoltaic power to produce hydrogen. In Regions B and C, wind power output fluctuates significantly. The hydrogen energy storage station can adjust its energy storage plan in real time to smooth out wind power fluctuations. The composite energy storage system, with hydrogen storage as the main source and solar energy storage as a supplement, can effectively regulate wind and solar power resources within the region, thereby enabling on-site consumption of wind and solar resources and improving the grid compatibility of renewable energy sources.

[0205] Option two does not consider electrical energy storage; the entire facility uses hydrogen energy storage as its sole energy source. The configuration scheme is as follows: Figure 8-10 As shown, regional grid A is equipped with a 0.45MW electrolyzer and a 0.5MW fuel cell, regional grid B is equipped with a 0.35MW electrolyzer and a 0.45MW fuel cell, and regional grid C is equipped with a 0.3MW electrolyzer and a 0.4MW fuel cell.

[0206] Operation modes of hydrogen energy storage stations under various regional power grids, such as Figure 8-10 As shown, the electrolyzer needs to purchase electricity from the grid throughout the day, and then the fuel cell generates electricity at full capacity and connects to the grid to regulate the load. On the one hand, this not only increases the capacity of hydrogen energy storage configuration, leading to increased investment costs for the site, but also the increase in daily operating time and frequency will shorten the service life of hydrogen energy storage-related equipment, leading to increased operating costs for the site. On the other hand, the electric-hydrogen-electric coupling method results in excessive energy loss, causing unnecessary large amounts of energy loss.

[0207] Option 3 considers utilizing the configuration under the wind and solar power curtailment mode, with the following configuration scheme: Figure 11-13 As shown, regional grid A is equipped with a 0.8MW electrolyzer, regional grid B with a 0.6MW electrolyzer, and regional grid C with a 0.5MW electrolyzer. Based on the operational results of each region, for the photovoltaic output in region A, hydrogen energy storage mainly utilizes a small amount of surplus photovoltaic power generated during midday to produce hydrogen, which is then stored. Figure 12-13 In the middle, for regions B and C, which are mainly distributed power sources powered by wind power, the entire region can achieve long-term, small-scale hydrogen production throughout the day.

[0208] Table 1 shows the various operating costs of the three types of power station under the same site selection scheme. It can be seen that there are significant differences in operating costs among the three configuration schemes that meet the grid load regulation requirements. The configuration of hydrogen energy storage as the main method and electric energy storage as the auxiliary method can significantly reduce the operating costs over the life cycle of the power station.

[0209] Table 1 Comparison of Economic Indicator Results

[0210]

[0211] In summary:

[0212] (1) In order to solve the problem of the difficulty of new energy consumption in the active distribution network, this embodiment uses a hydrogen energy storage system that takes into account both energy conversion and flexibility to promote the green and low-carbon transformation of the power grid. This embodiment treats the hydrogen production and storage system as an adjustable load and determines the hydrogen energy storage site selection scheme based on the clustering hierarchical zoning theory, thereby realizing the precise configuration and autonomous adjustment of energy storage in each region.

[0213] (2) A two-stage optimization configuration model for the distribution network of the wind-solar-hydrogen energy storage system considering the optimal absorption of new energy is proposed. The upper layer of the model solves the partition model by optimizing the distribution network's active and reactive power flow distribution. The lower layer model aims to optimize the overall cost of distribution network optimization configuration and scheduling with distributed power sources and the regional power self-management index. It completes the optimal planning and layout of hydrogen energy storage stations with distributed power sources to achieve a site selection and capacity determination strategy that takes into account both flexibility and economic indicators.

[0214] (3) Based on the improved metaheuristic algorithm, the effectiveness of the proposed model and the superiority of the method in this embodiment were verified on the extended IEEE 33-node case. Iterative calculations were performed on the two configuration methods. Under the requirement of rapid solution of hydrogen energy storage configuration, it was determined that the configuration scheme with hydrogen energy storage as the main component and electric energy storage as the auxiliary component has better economic efficiency.

[0215] (4) Future research will further investigate the impact of geographic information on configuration schemes in real-world scenarios.

[0216] The technical solutions of the present invention are not limited to the specific embodiments described above. Any technical modifications made in accordance with the technical solutions of the present invention fall within the protection scope of the present invention.

Claims

1. A method for configuring a hydrogen energy storage system for a power grid under load-zone coordinated control, characterized in that, Includes the following steps: S1. Extract the combined active and reactive power sensitivity matrix of each node in the power grid network; S2. Each node to be partitioned in the power grid network is divided into its own category, that is, there are n initial groups, and each group has only one individual. S3. Calculate the electrical distance matrix between n groups; S4. Use the Ward method to merge two groups with the smallest electrical distance, i.e., calculate the increment of the sum of squared variances (ESS) within the groups before and after merging. ; S5. Recalculate the population increment before and after. This allows the two classes with the smallest increment to be merged; S6. Repeat steps S4-S5 until they are merged into one category; S7. Draw a hierarchical clustering diagram of the clustering process based on the ward distance between each category, and determine the number of classification groups, i.e., node partitions, based on the flat period that appears when the classification heights are compared and merged. S8. Solve the node partitioning and hydrogen energy storage system site selection with the objectives of minimizing distribution network loss, minimum voltage deviation, and minimum power fluctuation; S9. Using the initial investment cost, maintenance cost, and operation cost of a distribution network with distributed power sources as objective functions, solve for the capacity configuration and operating status. Finally, use metaheuristic algorithms to iteratively solve for the operating strategies of three configuration schemes: wind and solar power curtailment for hydrogen production, single hydrogen energy storage, and hydrogen energy storage as the main energy storage and power storage as the auxiliary energy storage.

2. The method for configuring a hydrogen energy storage system for a power grid based on load zoning coordinated control according to claim 1, characterized in that, Step S1 specifically includes: S11. Constructing the Jacobian matrix of the power grid network: （1） In the formula: , These are the active and reactive power imbalances injected into the nodes, respectively. , H is the correction amount for voltage phase angle and amplitude; Subarray, N is Jacobi subarray, J is Jacobi subarray, L is Sub-array; S12. Constructing a traditional sensitivity matrix using the Jacobian matrix: (2) In the formula: S is the sensitivity matrix constructed from the Jacobian matrix; S13, Through A Jacobian submatrix N is used to construct the comprehensive sensitivity matrix. To ensure the accuracy of load node partitioning, only the load node dimension of the N matrix is ​​considered. 1-th order matrix : （3） Considering the combined sensitivity matrix of active and reactive power for: （4）。 3. The method for configuring a hydrogen energy storage system for a power grid based on load zoning coordinated control according to claim 2, characterized in that, Step S3 specifically includes: S31. Calculate the electrical distance between load nodes using the combined active and reactive power sensitivity matrix. For the voltage coupling between the voltage phase angle changes at nodes i and j... Represented as: （5） In the formula: The sensitivity model integrates the weak coupling effect of reactive power on voltage amplitude and active power on voltage phase angle deviation, and its elements correspond to the elements of the S matrix. S32. To extract the electrical coupling relationship between nodes in the power network, the electrical distance between nodes i and j is further obtained as follows: (6).

4. The method for configuring a hydrogen energy storage system for a power grid based on load zoning coordinated control according to claim 3, characterized in that, In step S8, the hydrogen energy storage system includes an electrolyzer, a hydrogen storage tank, a fuel cell, and an electrical energy storage system. Hydrogen production capacity of electrolyzer: (7); In the formula: This represents the amount of hydrogen produced during time period t. For the efficiency of the electrolytic cell, The electrolytic cell power during time period t. The amount of hydrogen produced from 1 kW‧h of electricity; The mathematical model for fuel cell operation can be simplified as follows: (8); In the formula: Let t be the electrical power released by the combustion of hydrogen in the SOFC during time period t, in kW; m represents the amount of hydrogen consumed by the SOFC during time period t. 3 ; The energy conversion efficiency of SOFC; The amount of hydrogen required to produce 1 kWh of electricity, m 3 ; Hydrogen storage tank input working status: (9) Output working status: (10) In the formula: Let m be the volume of hydrogen gas in the storage tank. 3 ; The leakage rate of the hydrogen storage tank; For fuel cells The required volume of hydrogen gas, m 3 ; Energy storage system charging operation status: (11); Discharge operating status: (12); In the formula: The amount of energy stored in the electrical system during time period t; and The charging and discharging power of the energy storage during time period t; and To improve the charging and discharging efficiency of electrical energy storage; Solving for node partitioning and hydrogen energy storage system site selection specifically includes: Construct the objective function and constraints: (1) Minimum network loss (13) In the formula: n is the total number of nodes; k is the set of nodes that have a direct coupling relationship with node i; , Let be the voltage magnitudes at nodes i and j; Indicates the line conductance between nodes i and j; The voltage phase difference between nodes i and j; (2) Minimum voltage deviation (14) In the formula: Let be the expected voltage value at node i; (3) Minimum active power fluctuation (15) In the formula: , Let represent the maximum and minimum power that node i can achieve at time t; Let be the natural power of node i at time t; 1) Power Constraint In distribution network optimization, the active and reactive power balance constraints at each node are as follows: (16) （17） In the formula: , These represent the combined active and reactive power of the load at node i and the DG, respectively; , These represent the conductivity and susceptance of the circuit, respectively. The voltage phase difference between nodes i and j; 2) Node voltage constraints (18) In the formula: , These represent the upper and lower limits of the voltage, respectively.

5. The method for configuring a hydrogen energy storage system for a power grid based on load zoning coordinated control according to claim 4, characterized in that, Step S9 specifically includes: Construct the objective function and constraints: (1) Initial investment cost (19) In the formula: For initial investment costs, This is the capital recovery coefficient, where d is the discount rate; n is the equipment lifespan. The daily investment cost of a hydrogen energy storage system; The investment cost of the energy storage system; , , These are the unit costs of fuel cells, electrolyzers, and hydrogen storage tanks, respectively. , Each fuel cell is configured with power; Configure the capacity of the hydrogen storage tank; , These are the battery cost and the battery converter cost, respectively. Battery energy storage capacity; Rated power for battery energy storage; (2) Maintenance costs (20) In the formula: This covers the total maintenance cost of hydrogen energy storage stations; Maintenance costs for hydrogen energy storage systems; Maintenance costs for the energy storage system; These are the proportions of maintenance costs for fuel cells, electrolyzers, hydrogen storage tanks, and batteries and converters to the initial investment cost; (3) Operating costs (twenty one) In the formula: The total operating costs of hydrogen energy storage stations; For the operating costs of the energy storage system; Operating costs for hydrogen energy storage systems; Let be the grid electricity price at time t; , These represent the electricity purchased and sold by the energy storage system to the grid at time t, respectively. , These are the gas-to-electricity efficiency and the electricity-to-gas efficiency of the hydrogen energy storage system, respectively. This refers to the discharge power of the hydrogen energy storage system. Hydrogen energy storage system charging power; (1) Constraints related to hydrogen energy storage system: Constraints related to hydrogen energy storage system include upper and lower limits of hydrogen energy storage power, operating status constraints of hydrogen energy storage system, and capacity balance constraints of hydrogen energy storage system. Hydrogen energy storage power upper and lower limits constraints: (twenty two) Operating constraints of hydrogen energy storage systems: (twenty three) Hydrogen storage capacity balance constraints: (twenty four) In the formula: , , The charging and discharging power and upper and lower limits for hydrogen energy storage; (2) Constraints related to energy storage systems (25) (26) (3) Power balance constraint: The power balance constraint means that the actual grid power demand and the sum of the actual load and grid loss in the system are equal. Here, the constraint conditions are simplified and the influence of grid loss is ignored: (27) In the formula: For the real-time power demand of electrical load, The variables are 0-1, representing sufficient peak shaving and insufficient peak shaving, respectively. This represents the peak-shaving difference between the wind, solar, and hydrogen storage energy stations at time t.

Images