Distributed hybrid cloud energy storage collaborative configuration method

Abstract

The invention relates to a distributed hybrid cloud energy storage collaborative configuration method. The method comprises the steps of grid cluster division and two-layer energy storage one-by-one planning optimization configuration. Through power system simulation, cluster division based on an electrical distance and a power balance condition is realized according to a grid structure, and double-layer one-by-one configuration of multiple energy storage is realized for an overall power distribution network according to a cluster division result, and the method comprises the following steps: step 1, establishing an energy storage equipment mathematical model; step 2, cluster division and model selection configuration; step 3, completing cooperative configuration of distributed hybrid cloud energy storage in the regional power distribution network based on a double-layer optimization structure; and step 4, optimization is carried out and an optimization result is obtained, loop optimization of nodes one by one is realized in a configuration process, node collaborative planning is realized by updating grid information, and finally an energy storage configuration result is output.

Description

Technical Field

[0001] The present invention relates to a technical route for distributed hybrid energy storage configuration and proposes a method for realizing distributed hybrid energy storage configuration. Background Art

[0002] With the rapid development of new power systems and the large-scale access of new power equipment to the power grid, the overall construction of the power system has undergone major changes. In order to ensure the safety, stability and economic benefits of the overall power network operation, the rational allocation of energy storage will be a top priority for future research. The hybrid coordinated application of multiple distributed energy storage such as electric energy storage, hydrogen storage, and thermal storage in various links of the distribution side is a current research hotspot.

[0003] At present, the research on energy storage configuration mainly includes: controlling the node voltage of the distribution system, adjusting the system frequency, optimizing the location and capacity of distributed electric energy storage, smoothing the power fluctuation of new energy, and improving the absorption capacity of new energy. Most of the research is to achieve optimization through basic objective functions and constraints. Few studies consider the structural characteristics of the grid to select and configure hybrid energy storage. Therefore, how to achieve the coordinated configuration of multi-regional hybrid energy storage is a problem to be solved. Summary of the invention

[0004] The present invention provides a distributed hybrid cloud energy storage collaborative configuration method. The technical solution is as follows:

[0005] A distributed hybrid cloud energy storage collaborative configuration method includes grid cluster division and double-layer energy storage one by one planning and optimization configuration; characterized in that: through power system simulation, cluster division based on electrical distance and power balance conditions is realized according to the grid structure, and double-layer one-by-one configuration of multiple energy storage is realized for the overall distribution network according to the cluster division result, including the following steps:

[0006] Step 1: Establish a mathematical model of energy storage equipment

[0007] For battery energy storage, the charging and discharging states are considered to reflect the changes in the battery storage capacity during charging and discharging; for thermal energy storage, the electric-to-heat equipment and the heat storage tank model are considered; for hydrogen energy storage, the mathematical models of the electrolyzer, hydrogen storage tank and fuel cell are established;

[0008] Step 2: Cluster division and selection configuration

[0009] The electrical distance is used as a structural indicator to measure the coupling tightness of the voltage amplitude changes between nodes; the power balance is used as a functional indicator to distinguish the regional electrical characteristics, active balance, and reactive balance, and divide the distribution network into clusters with active and reactive decoupling;

[0010] According to the cluster division results, the hierarchical analysis method is used to select the energy storage type of each cluster to achieve regional energy storage selection, and a hierarchical structure model is established, which is divided into the target layer, the criterion layer and the scheme layer; the target layer is the optimal solution for the user-side energy storage configuration results; the criterion layer uses network flexibility, economy and low-carbon benefits as decision indicators; the scheme layer configures different energy storage solutions for different types of users, with regional load characteristics as constraints;

[0011] Step 3: Complete the coordinated configuration of distributed hybrid cloud energy storage in the regional distribution network based on the two-layer optimization structure

[0012] According to the energy storage equipment mathematical model constructed in step 1 and the energy storage selection scheme selected in step 2, the coordinated configuration of distributed hybrid cloud energy storage in the regional distribution network is completed based on the double-layer optimization structure. The upper model aims at the optimal economic investment effect and constrains the power flow balance, node voltage, and branch capacity. The lower model is constructed based on the energy storage selection results. The objective function is to minimize the voltage offset while minimizing the annual operating cost of the system. The decision variable is the output power of the electric energy storage and hydrogen energy storage or thermal energy storage system per unit time.

[0013] Step 4: Optimize and obtain the optimization results. During the configuration process, implement cyclic optimization of each node one by one, realize node collaborative planning by updating the grid information, and finally output the energy storage configuration results, including the capacity, power, and selection information of the energy storage nodes in the distribution network.

[0014] Further: In step 2, considering that the relationship between two nodes is not only related to themselves but also to other nodes in the network, assuming that the network has n nodes, the electrical distance between node i and node j is defined as:

[0015]

[0016] Furthermore, in step 2,

[0017] The reactive power balance is as follows:

[0018]

[0019]

[0020] Where: Q i is the reactive balance degree of cluster i; is the reactive power balance index; c is the number of clusters; Q sup The maximum value of reactive power supply within the cluster, including the reactive power provided by the node reactive power compensation device and the reactive power that can be provided by some inverters; Q needThe reactive power demand value within the cluster includes the normal reactive power demand of the node and the minimum reactive power required to regulate the node overvoltage when the renewable energy output penetration rate in the network is higher than the limit value;

[0021] The active power balance is an indicator based on typical time-varying scenarios of the network, as follows:

[0022]

[0023] Where: P i is the active power balance degree of cluster i; P clu (t) i is the net power characteristic of cluster i, which can be obtained based on the typical time-varying scenario of each node; T is the time length of the typical time-varying scenario; It is an indicator of active power balance.

[0024] Further: In step 2, the criterion layer includes 3 decision indicators, the solution layer includes 3 decision solutions, and a 3×3 judgment matrix is ​​constructed. The constructed hierarchical analysis matrix is ​​as follows:

[0025]

[0026] According to the hierarchical analysis matrix, the maximum eigenvalue of the judgment matrix A of the criterion layer is λ max , and its corresponding eigenvector is right After normalization, the weight coefficient ω of the corresponding decision index k is obtained: k , the selection of specific energy storage types for multi-element hybrid energy storage is realized according to the weights. According to the regional load characteristics, step 2 is repeated for areas without thermal load but with thermal energy storage selected for configuration until the ideal result is obtained.

[0027] Further: The upper model described in step 3 takes the optimal economic investment effect as the goal, and the annual cost of the investment in the battery within the life cycle is:

[0028]

[0029] Where: P B,R is the power capacity of the battery; E B,R is the rated storage capacity of the battery; k b , k e are the investment costs of battery unit power and unit storage capacity respectively; r is the discount rate; Y b is the service life of the battery;

[0030] Annual cost of investment in hydrogen energy storage system:

[0031]

[0032] Where: P E,R , P FC,R are the power capacities of the electrolyzer and fuel cell respectively; V R is the rated hydrogen storage volume; k h , k fc are the unit power investment costs of electrolyzer and fuel cell respectively; k v is the unit volume cost of the hydrogen storage tank; Y h is the service life of the hydrogen energy storage system;

[0033] Annual cost of investment in thermal energy storage system:

[0034]

[0035] Where: k T is the unit heat storage tank capacity cost, yuan / m 3 ; V T k is the configuration capacity of the heat storage tank; m V is the unit variable investment cost of electric boiler m, RMB / MW; m is the configuration capacity of electric boiler m; r is the discount rate; Y t The expected service life of the electric heating equipment;

[0036] The upper objective function is expressed as:

[0037] f up =C1+C2+C3 (22)

[0038] At the same time, the constraints are power flow balance, node voltage, and branch capacity.

[0039] Further: The objective function of the lower model in step 3 related to the annual operating cost of the system is expressed as follows:

[0040]

[0041] Where: P B (t) is the output power of the battery at time t; P E (t) is the input power of the electrolytic cell at time t; P FC (t) is the output power of the hydrogen fuel cell at time t; P em (t) is the output power of the electric boiler at time t; P vm (t) is the output power of the heat storage tank at time t; c b , c e , c fc , c em , c vm are the unit power operating costs of batteries, electrolyzers, fuel cells, electric boilers, and heat storage tanks; kin is the active network loss penalty coefficient;

[0042] In the lower model, the accumulated voltage offset is:

[0043]

[0044] Where: N d is the number of distribution network nodes; U i.R is the ideal value of the voltage at the ith node; U i is the actual value of the voltage at each node; U imax , U imin are the upper and lower limits of the voltage of the i-th node respectively;

[0045] The system operation investment cost and voltage deviation minimum are expressed as:

[0046]

[0047] The present invention provides a distributed hybrid cloud energy storage collaborative configuration method, which first considers user needs to meet the user load needs of the distribution network; at the same time, it considers the network structure to ensure the safety and stability of the overall distribution network operation; in addition, it considers the economic benefits of investment and operation to achieve greater economic benefits for the investment entity, and ultimately realizes the site selection and capacity determination of the regional distribution network and the selection of energy storage types for diversified hybrid energy storage. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] The present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0049] Figure 1 It is a flow chart of the algorithm simulated by the present invention;

[0050] Figure 2 Divide the main performance indicators for the cluster;

[0051] Figure 3 It is a hierarchical analysis logic diagram;

[0052] Figure 4 This is a partial optimization result of the simulation of the present invention. DETAILED DESCRIPTION

[0053] Distributed hybrid energy storage has its application modes at all levels of the power system. The reasonable power, capacity, and location selection of distributed energy storage in different application modes are crucial to the safety and economy of the distribution network. In the process of considering the energy storage configuration at the distribution network level, considering the differentiation of energy storage needs in multiple scenarios, it is necessary to comprehensively consider the life cycle of energy storage equipment, investment and operation costs, line stability constraints, transmission losses, and user energy capacity and stability requirements.

[0054] Under the background of large-scale development of new power systems at this stage, a lot of research has been conducted at home and abroad on the optimal configuration of distributed hybrid energy storage, including various application scenarios such as the power supply side, distribution network side, and user side, taking into account many factors such as grid structure, load characteristics, and investment costs. However, few studies have comprehensively considered the demand impact relationship of various scenarios for the overall distribution network, and designed the power purchase and sales characteristics of commercial energy storage aggregation to realize the configuration of energy storage investment and operation characteristics in a flexible power market with elastic energy storage demand. In addition, most of the energy storage configuration issues remain at the configuration of energy storage of a single energy type, and few studies have proposed solutions for multi-hybrid energy storage.

[0055] The distributed hybrid cloud energy storage collaborative configuration method of the present invention includes grid cluster division and double-layer energy storage one by one planning and optimizing configuration; through power system simulation, cluster division based on electrical distance and power balance conditions is realized according to the grid structure, and double-layer one-by-one configuration of multiple energy storage is realized for the overall distribution network according to the cluster division result, including the following steps:

[0056] Step 1: Establish a reasonable mathematical model of energy storage equipment

[0057] (I) Battery

[0058] The battery model used in the current study takes into account the charging and discharging status, so it is necessary to model the changes in battery storage capacity during charging and discharging.

[0059] When the battery is charged, the energy stored is:

[0060] E bat (t) = E bat (t-1)(1-δ)+P bat_ch (t-1)·η ch ·η bat_DC-DC ·Δt (1)

[0061] When the battery is discharged, the stored energy is:

[0062]

[0063] Where: δ is the battery self-discharge rate; P bat_ch and P bat_dh is the charging and discharging efficiency of the battery; η bat_DC-DC is the conversion efficiency of the converter connected to the battery; t is the time step.

[0064] (II) Thermal energy storage

[0065] Thermal storage tank storage capacity constraints:

[0066]

[0067] Where: Q t represents the energy of the energy storage system at the beginning of time period t; P t + , P t - Respectively represent the energy storage power and energy release power of the energy storage device in time period t; η c , η d Respectively represent the one-way energy storage and release efficiency (round-trip efficiency) of the energy storage system; δ t It indicates the self-consumption rate of the energy storage system per unit time period. The last term in the formula indicates the energy loss caused by the characteristics of the device. This formula can be understood as: the energy stored in the energy storage system is equal to the energy stored at the previous time point, minus (plus) the energy released (stored) in the time period, minus the energy loss caused by the characteristics of the device.

[0068] Upper and lower limits of heat storage tank capacity:

[0069] Q t,min ≤Q t ≤Q t,max (4)

[0070] Where Q t,max and Q t,min are the upper and lower limits of the energy storage capacity respectively.

[0071] in:

[0072] Q t,max =C s ·μ max (5)

[0073] Q t,min =C s ·μ min (6)

[0074] Where: μ max and μ min are the upper and lower limit coefficients of energy storage system respectively; C s is the rated energy storage capacity of the energy storage system.

[0075] In addition, in actual operation, the heat storage tank needs to constrain the energy storage state at the end of its optimization cycle.

[0076] Q T,min ≤Q T ≤Q T,max (7)

[0077] Where Q T,max and Q T,min They are respectively the upper and lower limits of the energy storage state at the end of the optimization cycle of the heat storage tank.

[0078] (III) Hydrogen Energy Storage

[0079] At present, most of the mathematical models used in the research on hydrogen energy storage systems are:

[0080] The electrolyzer converts and stores energy by electrolyzing water into hydrogen and oxygen. The output power can be expressed as:

[0081] P e-out =η e ·P e-in (8)

[0082] Where: η e is the electrolysis efficiency of the electrolytic cell, generally 60%; P e-in is the input power of the electrolyzer.

[0083] As a bidirectional device, the hydrogen storage tank can store the hydrogen produced by the electrolysis of water in the electrolyzer, and at the same time provide hydrogen for the fuel cell to generate electricity, thus improving the flexibility of the system. The mathematical model of its energy storage can be expressed as:

[0084]

[0085] Where: E store (t) is the energy stored in the hydrogen storage tank at time t; η e_DC-DC is the converter efficiency of the electrolyzer, η f_DC-DC is the efficiency of the fuel cell converter, which is generally 90%; and η store It is the working efficiency of the hydrogen storage tank, generally 50%.

[0086] The fuel cell model is a proton membrane exchange fuel cell. This technology uses hydrogen and oxygen as fuel, converts the chemical energy inside the fuel into electrical energy and stores it. Its output power is:

[0087] P f =P store-f ·η f (10)

[0089] Where: P store-f The power delivered from the hydrogen storage tank to the fuel cell; η f The efficiency of the fuel cell.

[0090] The maximum input power of the electrolyzer and the maximum output power of the fuel cell are limited by their own capacity and the remaining capacity of the hydrogen storage tank, and their limitations can be expressed as:

[0091]

[0092] Where: Ce and Cf are the capacities of the electrolyzer and fuel cell respectively; E store_max and E store_minThe upper and lower limits of the hydrogen storage tank’s energy storage capacity are generally taken as E store_max =0.85C store , E store_min =0.15C store ; C store is the capacity of the hydrogen storage tank.

[0093] Step 2: Cluster division and hierarchical analysis to determine network architecture and energy storage type

[0094] (1) Cluster division

[0095] According to the upper-level site selection and capacity determination results, the electrical distance is used as a structural indicator to measure the coupling tightness of the voltage amplitude changes between nodes; the power balance degree is used as a functional indicator. This division method organically divides the distribution network into active and reactive decoupled clusters, which has many advantages over the overall distribution network planning model: on the one hand, it can improve the supply and demand balance and voltage control capabilities of reactive power within the cluster and reduce the cross-cluster transmission of reactive power; on the other hand, it can promote the timing complementary matching of active power between nodes within the cluster and improve the autonomy of the cluster.

[0096] Considering that the relationship between two nodes is not only related to themselves, but also to other nodes in the network, assuming that the network has n nodes, the electrical distance between node i and node j is defined as:

[0097]

[0098] The reactive power balance index is as follows:

[0099]

[0100] Where: Q i is the reactive balance degree of cluster i; is the reactive power balance index; c is the number of clusters; Q sup The maximum value of reactive power supply within the cluster, including the reactive power provided by the node reactive power compensation device and the reactive power that can be provided by some inverters; Q need is the reactive power demand value within the cluster, which not only refers to the normal reactive power demand of the node, but also includes the minimum reactive power required to regulate the node overvoltage when the penetration rate of renewable energy output in the network is too high. It is expressed as

[0101]

[0102] Where: Q V To adjust cluster C k Minimum reactive power required at the overvoltage node; ΔV i is the voltage change at node i; S VQ,iiis the reactive voltage sensitivity of node i with respect to itself. It is obvious that the node has the highest sensitivity to itself.

[0103] The active power balance index is an index based on typical time-varying scenarios of the network, as follows:

[0104]

[0105] Where: P i is the active power balance degree of cluster i; P clu (t) i is the net power characteristic of cluster i, which can be obtained based on the typical time-varying scenario of each node; T is the time length of the typical time-varying scenario; It is an indicator of active power balance.

[0106] (2) Hierarchical analysis of energy storage selection

[0107] The hierarchical analysis method is applied to the cluster division results to realize the regional energy storage selection, and a hierarchical structure model is established, which is divided into the target layer, the criterion layer and the scheme layer. The target layer is the comprehensive optimization of the user-side energy storage configuration results; the criterion layer uses network flexibility, economy and low-carbon benefits as decision indicators; the scheme layer configures different energy storage solutions for different types of users, with regional load characteristics as constraints.

[0108] There are 3 decision indicators at the criterion level and 3 decision plans at the plan level, so a 3×3 judgment matrix needs to be constructed. The constructed hierarchical analysis matrix is ​​as follows:

[0109]

[0110] According to the matrix, let the maximum eigenvalue of the judgment matrix A of the criterion layer be λ max , and its corresponding eigenvector is right After normalization, the weight coefficient ω of the corresponding decision index k is obtained: k , and the selection of specific energy storage types for multi-element hybrid energy storage is realized based on the weights. According to the regional load characteristics, step 2 is repeated for areas without thermal load but with thermal energy storage selected for configuration until the ideal result is obtained.

[0111] Step 3: Complete the coordinated configuration of distributed hybrid cloud energy storage in the regional distribution network based on the two-layer optimization structure

[0112] In order to achieve more accurate energy storage configuration calculations and improve computing efficiency, a two-layer optimization approach is adopted to design a distributed hybrid cloud energy storage regional distribution network collaborative configuration system. This process mainly includes the following models:

[0113] (I) Upper model

[0114] Taking the optimal economic investment effect as the goal, the annual cost of the battery investment during the life cycle is:

[0115]

[0116] Where: P B,R is the power capacity of the battery; E B,R is the rated storage capacity of the battery; k b , k e are the investment costs of battery unit power and unit storage capacity respectively; r is the discount rate; Y b The service life of the battery.

[0117] Annual cost of investment in hydrogen energy storage system:

[0118]

[0119] Where: P E,R , P FC,R are the power capacities of the electrolyzer and fuel cell respectively; V R is the rated hydrogen storage volume; k h , k fc are the unit power investment costs of electrolyzer and fuel cell respectively; k v is the unit volume cost of the hydrogen storage tank; Y h is the service life of the hydrogen energy storage system.

[0120] Investment and annual cost of thermal energy storage system:

[0121]

[0122] Where: k T is the unit heat storage tank capacity cost, yuan / m 3 ; V T k is the configuration capacity of the heat storage tank; m V is the unit variable investment cost of electric boiler m, RMB / MW; m is the configuration capacity of electric boiler m; r is the discount rate; Y t The expected service life of the electric heating equipment.

[0123] Therefore, the upper objective function is expressed as:

[0124] f up =C1+C2+C3 (23)

[0125] At the same time, the constraints are power flow balance, node voltage, and branch capacity.

[0126] (II) Lower-level model

[0127] The lower optimization model is established based on the energy storage selection results. The objective function is to minimize the voltage offset while minimizing the annual operating cost of the system. The decision variables are the output power of the battery and the hydrogen energy storage or thermal energy storage system at each moment.

[0128] The objective function is expressed as follows:

[0129]

[0130] Where: P B (t) is the output power of the battery at time t; P E (t) is the input power of the electrolytic cell at time t; P FC (t) is the output power of the hydrogen fuel cell at time t; P em (t) is the output power of the electric boiler at time t; P vm (t) is the output power of the heat storage tank at time t; c b , c e , c fc , c em , c vm They are the unit power operating costs of batteries, electrolyzers, fuel cells, electric boilers, and heat storage tanks. in is the active network loss penalty coefficient.

[0131] The relative value of the accumulated node voltage offset is:

[0132]

[0133] Where: N d is the number of distribution network nodes; U i.R is the ideal value of the voltage at the ith node; U i is the actual value of the voltage at each node; U imax , U imin are the upper and lower limits of the voltage of the i-th node respectively.

[0134] The final optimization multi-objective function is to minimize the system's operating investment cost and voltage deviation:

[0135]

[0136] Step 4: Optimize and obtain optimization results

[0137] Optimize and obtain the optimization results. During the configuration process, implement cyclic optimization of each node one by one. Implement node collaborative planning by updating grid information. Finally, output the energy storage configuration results, including the capacity, power, and selection information of the energy storage nodes in the distribution network.

Claims

1. A distributed hybrid cloud energy storage collaborative configuration method, including grid cluster division and double-layer energy storage one by one planning and optimization configuration; characterized by: Through power system simulation, cluster division based on electrical distance and power balance conditions is realized according to the grid structure. Based on the cluster division results, multiple energy storages are configured one by one in two layers for the overall distribution network, including the following steps: Step 1: Establish a mathematical model of energy storage equipment For battery energy storage, the charging and discharging states are considered to reflect the changes in the battery storage capacity during charging and discharging; for thermal energy storage, the electric-to-heat equipment and the heat storage tank model are considered; for hydrogen energy storage, the mathematical models of the electrolyzer, hydrogen storage tank and fuel cell are established; Step 2: Cluster division and selection configuration The electrical distance is used as a structural indicator to measure the coupling tightness of the voltage amplitude changes between nodes; the power balance is used as a functional indicator to distinguish the regional electrical characteristics, active balance, and reactive balance, and divide the distribution network into clusters with active and reactive decoupling; According to the cluster division results, the hierarchical analysis method is used to select the energy storage type of each cluster to achieve regional energy storage selection, and a hierarchical structure model is established, which is divided into the target layer, the criterion layer and the scheme layer; the target layer is the optimal solution for the user-side energy storage configuration results; the criterion layer uses network flexibility, economy and low-carbon benefits as decision indicators; the scheme layer configures different energy storage solutions for different types of users, with regional load characteristics as constraints; Step 3: Complete the coordinated configuration of distributed hybrid cloud energy storage in the regional distribution network based on the two-layer optimization structure According to the energy storage equipment mathematical model constructed in step 1 and the energy storage selection scheme selected in step 2, the coordinated configuration of distributed hybrid cloud energy storage in the regional distribution network is completed based on the double-layer optimization structure. The upper model aims at the optimal economic investment effect and constrains the power flow balance, node voltage, and branch capacity. The lower model is constructed based on the energy storage selection results. The objective function is to minimize the voltage offset while minimizing the annual operating cost of the system. The decision variable is the output power of the electric energy storage and hydrogen energy storage or thermal energy storage system per unit time. Step 4: Optimize and obtain the optimization results. During the configuration process, implement cyclic optimization of each node one by one, realize node collaborative planning by updating the grid information, and finally output the energy storage configuration results, including the capacity, power, and selection information of the energy storage nodes in the distribution network.

2. The energy storage planning method according to claim 1, characterized in that: In step 2, considering that the relationship between two nodes is not only related to themselves but also to other nodes in the network, assuming that the network has n nodes, the electrical distance between node i and node j is defined as:

3. The energy storage planning method according to claim 1, characterized in that: In step 2, The reactive power balance is as follows: Where: Q i is the reactive balance degree of cluster i; is the reactive power balance index; c is the number of clusters; Q sup The maximum value of reactive power supply within the cluster, including the reactive power provided by the node reactive power compensation device and the reactive power that can be provided by some inverters; Q need The reactive power demand value within the cluster includes the normal reactive power demand of the node and the minimum reactive power required to regulate the node overvoltage when the renewable energy output penetration rate in the network is higher than the limit value; The active power balance is an indicator based on typical time-varying scenarios of the network, as follows: Where: P i is the active power balance degree of cluster i; P clu (t) i is the net power characteristic of cluster i, which can be obtained based on the typical time-varying scenario of each node; T is the time length of the typical time-varying scenario; It is an indicator of active power balance.

4. The energy storage planning method according to claim 1, characterized in that: In step 2, the criterion layer includes 3 decision indicators, the solution layer includes 3 decision solutions, and a 3×3 judgment matrix is ​​constructed. The constructed hierarchical analysis matrix is ​​as follows: According to the hierarchical analysis matrix, the maximum eigenvalue of the judgment matrix A of the criterion layer is λ max , and its corresponding eigenvector is right After normalization, the weight coefficient ω of the corresponding decision index k is obtained: k , the selection of specific energy storage types for multi-element hybrid energy storage is realized according to the weights. According to the regional load characteristics, step 2 is repeated for areas without thermal load but with thermal energy storage selected for configuration until the ideal result is obtained.

5. The energy storage planning method according to claim 1, characterized in that: The upper model described in step 3 aims at the optimal economic investment effect. The annual cost of the investment in the battery during the life cycle is: Where: P B,R is the power capacity of the battery; E B,R is the rated storage capacity of the battery; k b , k e are the investment costs of battery unit power and unit storage capacity respectively; r is the discount rate; Y b is the service life of the battery; Annual cost of investment in hydrogen energy storage system: Where: P E,R , P FC,R are the power capacities of the electrolyzer and fuel cell respectively; V R is the rated hydrogen storage volume; k h , k fc are the unit power investment costs of electrolyzer and fuel cell respectively; k v is the unit volume cost of the hydrogen storage tank; Y h is the service life of the hydrogen energy storage system; Investment and annual cost of thermal energy storage system: Where: k T is the unit heat storage tank capacity cost, yuan / m 3 ; V T k is the configuration capacity of the heat storage tank; m V is the unit variable investment cost of electric boiler m, RMB / MW; m is the configuration capacity of electric boiler m; r is the discount rate; Y t The expected service life of the electric heating equipment; The upper objective function is expressed as: minF up =C1+C2+C3 (22) At the same time, the constraints are power flow balance, node voltage, and branch capacity.

6. The energy storage planning method according to claim 1, characterized in that: The objective function of the lower model described in step 3 related to the annual operating cost of the system is expressed as follows: Where: P B (t) is the output power of the battery at time t; P E (t) is the input power of the electrolytic cell at time t; P FC (t) is the output power of the hydrogen fuel cell at time t; P em (t) is the output power of the electric boiler at time t; P vm (t) is the output power of the heat storage tank at time t; c b , c e , c fc , c em , c vm are the unit power operating costs of batteries, electrolyzers, fuel cells, electric boilers, and heat storage tanks; k in is the active network loss penalty coefficient; In the lower model, the accumulated voltage offset is: Where: N d is the number of distribution network nodes; U i.R is the ideal value of the voltage at the ith node; U i is the actual value of the voltage at each node; U imax , U imin are the upper and lower limits of the voltage of the i-th node respectively; The system operation investment cost and voltage deviation minimum are expressed as:

Images