Power distribution network expansion planning method considering spatio-temporal load transfer and multi-energy flow coordination

Abstract

The application discloses a power distribution network expansion planning method considering space-time transfer of loads and multi-energy flow coordination, comprising the following steps: (I) using SOP to perform flexible interconnection reconstruction on a network frame; (II) establishing a DHM-PCS operation model; (III) configuring DHM-PCS in the power distribution network according to a time sequence voltage sensitivity index; and (IV) constructing a planning model. The photovoltaic consumption method can consume energy in three dimensions of time energy translation, space energy transfer and energy category energy conversion; the dimension of consumption is improved from optimization of a point to coordination between networks; the DHM-PCS configuration model of limited key nodes is constructed by considering the coordination characteristics between feeders, so that the maximum global consumption benefit is realized.

Description

Technical Field

[0001] This invention belongs to the field of distribution network expansion planning technology, specifically relating to a distribution network expansion planning method that takes into account load spatiotemporal transfer and multi-energy flow coordination. Background Technology

[0002] Distributed photovoltaic power is mainly connected at the distribution network level. The reasons for its difficulty in absorption can be divided into three aspects: 1) Insufficient development of source-side resources: insufficient integration of the diverse needs of other energy systems such as heating and cooling with the distribution network; 2) The grid topology restricts the collaborative absorption capacity between feeders: energy interaction between feeders cannot be achieved; 3) Limited regulation resources: scarcity of various heterogeneous resources such as energy storage.

[0003] The future consumption model of photovoltaic energy is no longer simply large-scale grid connection, but requires deep integration with other industries. Current research mainly focuses on the rapid development of multi-type heterogeneous resource and energy storage technologies, which have become a core component of multi-energy complementarity. For example, the essence is to establish a multi-energy flow coupling system for cooling, heating, and electricity to meet energy demands. Traditional multi-energy flow coupling systems mostly utilize natural gas for coupling, but their safe and economical operation still presents challenges. Furthermore, regions lacking natural gas resources face more limitations, and the combustion of natural gas still emits greenhouse gases.

[0004] Hydrogen energy, as a clean energy source, is currently attracting significant attention. Existing research primarily focuses on proposing planning methods for distributed hydrogen-based microgrids. In particular, while the proposed absorption strategies are effective when these microgrids are integrated into most feeders, they are highly dependent on land availability. In reality, the scarcity and high cost of land together limit its widespread application. Therefore, it is necessary to consider photovoltaic absorption through the energy synergy between multiple feeders, rather than focusing on the configuration of absorption devices at the single feeder level. This requires precise resource allocation at limited key nodes to alleviate the absorption challenges faced by multi-feeder systems, with a key focus on how to correlate the grid structure between feeders.

[0005] The use of AC-DC technology for flexible interconnection of power grids has been widely studied. Current research combines energy storage with feeder-to-feeder coordinated operation for energy absorption. However, the lack of diverse heterogeneous regulation devices limits absorption methods and operational flexibility. In fact, traditional distribution networks are transforming into new types with massive resource access and increasingly complex spatiotemporal characteristics. These new distribution networks require both resource coupling within a single feeder and resource coordination between multiple feeders. Therefore, the planning challenges for energy absorption have shifted from local optimization of single feeders and single regulation devices to global coordinated configuration of various heterogeneous resources considering the energy coordination characteristics between multiple feeders. However, systematic research on this transformation is very limited. Summary of the Invention

[0006] This invention is proposed to overcome the shortcomings of the prior art, and its purpose is to provide a distribution network expansion planning method that takes into account load spatiotemporal transfer and multi-energy flow coordination.

[0007] This invention is achieved through the following technical solution:

[0008] A distribution network expansion planning method that considers load spatiotemporal transfer and multi-energy flow coordination includes the following steps:

[0009] (I) Modify the location of the tie switch in the distribution network into a SOP device and establish an SOP operation model;

[0010] Flexible interconnection transformation of the power grid using SOPs: The locations of tie switches in the distribution network are modified to SOP (soft open point) devices, and an SOP operation model is established to enable bidirectional power transmission between feeders. The method for SOP integration into the medium-voltage distribution network is as follows: Figure 1 As shown;

[0011] After installing the SOP equipment, the conversion and interconnection between AC and DC are realized through back-to-back voltage source converters (B2B VSC), thereby decoupling the electrical connection on both sides of the feeder and realizing independent control and regulation. The SOP equipment can realize bidirectional continuous power flow regulation between different feeders, improving the operating efficiency and system stability of the distribution network.

[0012] The SOP operating model includes active power constraints and capacity constraints;

[0013] The SOP operating model includes active power constraints and capacity constraints;

[0014] The active power and capacity constraints of the SOP are expressed as follows:

[0015]

[0016] Where: Where: Let be the three-phase power of node i connected through the s-th SOP at time t, in kVar; Let be the three-phase power of node j connected through the s-th SOP at time t, in kVar; and Both are equal; Ω SOP This is a collection of SOP devices; φ represents the three phase sequences; Let be the reactive power of the s-th SOP at node i at time t, in kVar; The capacity corresponds to the SOP, and the unit is kVA.

[0017] (II) A distributed hydrogen-based multi-energy photovoltaic (DHM-PCS) system is selected as the photovoltaic energy consumption device. The distribution network is linked to the cooling and heating network, and the cooling and heating loads of each node are distributed according to the proportion of electrical load. The conversion relationship between cooling, heating, electricity, and hydrogen in the distributed hydrogen-based multi-energy photovoltaic (DHM-PCS) system is established. Figure 2 );

[0018] The equipment includes: an alkaline electrolysis cell (AEC), a liponide absorption chiller (LBAC), an electric heat pump (EHP), a thermal storage tank (TST), a hydrogen storage tank (HST), a solid oxide fuel cell (SOFC), and an electric chiller (EC).

[0019] The energy balance equations for the multi-energy flow conversion relationship of cold, heat, electricity, and hydrogen include the electrical energy balance equation, the thermal energy balance equation, and the cold energy balance equation.

[0020] The DHM-PCS's electrical energy comes from the distribution network (including PV (Photovoltaic)) and SOFC, and its electrical load includes AEC, EHP, and EC. Therefore, the power balance equation is:

[0021] P PDN (t)=P SOFC,E (t)-P EHP,in (t)-P AEC,in (t)-P EC,in (t) (3)

[0022] In the formula: P PDN (t) represents the power consumption of the distributed hydrogen-based multi-energy photovoltaic system (DHM-PCS) at time t, in kW; P SOFC,E (t), P EHP,in (t), P AEC,in (t), P EC,in (t) represents the power emitted or absorbed at time t by the fuel cell SOFC, electric heat pump EHP, water electrolysis equipment AEC, and electric chiller EC, all in kW.

[0023] The output heat energy is provided by EHP, TST, and SOFC; therefore, the heat balance equation is:

[0024] P h (t)=P EHP,out (t)+βP TST,out (t)+P SOFC,H (t)

[0025] -P LBAC,in (t)-(1-β)P TST,in (t) (4)

[0026] In the formula: P h (t) represents the thermal power emitted by the distributed hydrogen-based multi-energy photovoltaic system at time t, in kW; P EHP,out (t), P TST,out (t), P SOFC,H (t) represents the heat power emitted by EHP, TST, and SOFC at time t, respectively, in kW; P LBAC,in (t), P TST,in (t) represents the heat power absorbed by LBAC and TST at time t, respectively, in kW; β is a 0 / 1 variable, representing the operating state of TST;

[0027] The cold energy originates from EC and LBAC, therefore the cold energy balance equation is:

[0028] P c (t)=P EC,out (t)+P LBAC,out (t) (5)

[0029] In the formula: P c (t) represents the cooling power emitted by the distributed hydrogen-based multi-energy photovoltaic system at time t, in kW; P EC,out (t), P LBAC,out (t) represents the cooling power emitted by EC and LBAC at time t, both in kW.

[0030] The above-mentioned electrical energy balance equation, thermal energy balance equation, and cold energy balance equation constitute the cold-heat-electricity balance. The calculation methods for each variable involved in the balance formula are as follows:

[0031] P AEC,out (t)=α AEC P AEC,in (t) (6)

[0032] P SOFC,E (t)=α SOFC P HST,out (t) (7)

[0033] P EHP,out (t)=α EHP P EHP,in (t) (8)

[0034] P SOFC,H (t)=α SOFC,H θ SOFC P SOFC,E (t) (9)

[0035] P EC,out (t)=α EC P EC,in (t) (10)

[0036] P LBAC,out (t)=α LBAC P LBAC,in (11)

[0037]

[0038] E HST (0)=E HST (23) (13)

[0039] P HST,out (t) <= P HST,max (14)

[0040] P HST,in (t) <= P HST,max (15)

[0041]

[0042] E TST (0)=E TST (23) (17)

[0043] P TST,out (t)≤P TST,max (18)

[0044] P TST,in (t)≤P TST,max (19)

[0045] In the formula: α AEC α EHP α SOFC α EC α LBAC These are the energy conversion efficiencies of water electrolysis equipment (AEC), electric heat pump (EHP), fuel cell (SOFC), electric chiller (EC), and absorption chiller (LBAC); α SOFC,H For SOFC fuel cells; P AEC,in (t), P HST,out (t), P EHP,in (t), P SOFC,E (t), P EC,in (t), P LBAC,in(t) represents the power emitted or absorbed at time t by the water electrolysis unit AEC, hydrogen storage tank HST, electric heat pump EHP, fuel cell SOFC, electric chiller EC, and absorption chiller LBAC, all in kW; θ SOFC For fuel cell waste heat utilization rate; E HST (t), E TST (t) represents the energy stored in the thermal storage tank HST and the hydrogen storage tank TSTt at time t, both in kWh; η HST η TST For the charging and discharging efficiency of the heat storage tank HST and the hydrogen storage tank TST; β HST β TST P is a 0 / 1 variable reflecting the operating status of the thermal storage tank HST and the hydrogen storage tank TST; TST,in P HST,out P HST,in P TST,out Energy consumption for charging / discharging the thermal storage tank (HST) and the hydrogen storage tank (TST), both in kW; P HST,max P TST,max The maximum charging energy per unit time for the thermal storage tank HST and the hydrogen storage tank TST is in kW.

[0046] (III) Select typical days using the K-means clustering method and obtain three cooling, heating, electricity and photovoltaic curves for the typical days. Apply the cooling, heating, electricity and photovoltaic curve data of the typical days to steps (IV) and (V). The typical days represent typical electricity consumption scenarios in the transition season, summer and winter, respectively. This data can be used to simulate the operation of last year.

[0047] Step (IV) calculates sensitivity and step (V) applies typical daily data when solving the distribution network expansion planning model.

[0048] (IV) Determine the configuration location of the distributed hydrogen-based multi-energy flow photovoltaic (DHM-PCS) system in the distribution network based on the time-series voltage sensitivity index to reduce the land dependence and redundancy of the traditional single-feeder absorption method. The specific method includes the following steps:

[0049] (1) Calculate voltage sensitivity

[0050] Using voltage sensitivity analysis, the sensitivity of each node to voltage changes is calculated, and the node with the greatest impact on the system from voltage fluctuations is identified.

[0051] The specific method for calculating the sensitivity of the node to voltage changes is as follows:

[0052] A voltage offset weighting factor is introduced into the traditional sensitivity, and this factor can reflect the differentiated voltage regulation requirements of different nodes (upward and downward). The sensitivity S of the active power injection change at node n at time t to the voltage at node m is calculated. mn,t It can be represented as:

[0053] S mn,t =λ mn,t ×ΔV m,t (20)

[0054] ΔV m,t =V m,t -V eu,m,t (twenty one)

[0055] In the formula: S mn,t The sensitivity of the voltage at node m to the active power change injected into node n at time t; λ mn,t For traditional voltage sensitivity indicators; ΔV m,t V represents the deviation between the node voltage and the desired voltage, in kV. m,t V represents the current node voltage, in kV. eu,m,t The desired node voltage is expressed in kV.

[0056] The overall sensitivity S of node m to its feeder at time t m,t for:

[0057]

[0058] Where FL is the set of current feeder load nodes;

[0059] The voltage offset weighting factor in the above formula assigns weights to different nodes on the same time segment. However, in the time dimension, different weights should be assigned to different time periods based on the voltage exceedance situation on different time segments, introducing a comprehensive sensitivity S of the time weight. op,m,t as follows:

[0060]

[0061] In the formula: S op,m,t For overall sensitivity; S m,t N represents the overall sensitivity of node m to its feeder at time t. exceed,t V represents the number of voltage-over-limit nodes at time t. exceed,t The node voltage exceeds the limit at time t, in kV.

[0062] Based on the number of days under different operating conditions (e.g., different seasons), different weights are assigned to different operating conditions, and a comprehensive sensitivity S of the operating state weights is introduced. opw,m as follows:

[0063]

[0064] In the formula: S opw,m The overall sensitivity of the operating state weights; S op,m,t For overall sensitivity; Days opThe number of days is the number of days in different operating states; op represents different operating states, and M is the type of operating state.

[0065] For a distribution network containing R coordination units, the internal nodes of each coordination unit are sorted, and the sorted index I is... r As follows:

[0066]

[0067] Among them, Ω Lnode Let r be the set of load nodes in all feeders, and r be a certain coordination unit.

[0068] (2) Select key nodes

[0069] Based on the voltage change sensitivity results calculated in step (1), several key nodes with high voltage sensitivity are selected as the configuration locations of the distributed hydrogen-based multi-energy flow photovoltaic system.

[0070] For situations involving large coordination units, following the principle of non-clustered deployment, multiple nodes with high sensitivity and non-clustered (nearby) are selected and deployed in a distributed manner for DHM-PCS.

[0071] Specifically: When the number of nodes in the coordination unit is less than A, the node with the highest sensitivity in the coordination unit is selected as the configuration location of the distributed hydrogen-based multi-energy flow photovoltaic power consumption system;

[0072] When the number of nodes within a coordination unit is greater than A, the node with the highest sensitivity within each geographical area of ​​the coordination unit is selected as the configuration location for the distributed hydrogen-based multi-energy flow photovoltaic system.

[0073] The standard A for determining the number of nodes within a coordination unit is primarily based on several factors, including voltage sensitivity analysis, geographical environment and load density, system size and complexity, and historical operating data and experience. By calculating voltage sensitivity, the nodes most sensitive to voltage changes are identified, and these nodes are typically selected for initial evaluation. The number of nodes needs to be flexibly adjusted according to the regional geographical environment and load density; for example, the requirements may differ in urban, suburban, and rural areas. The standard for the number of nodes will also differ between large and small-to-medium-sized distribution network systems. In cases of high system complexity, the number of nodes can be appropriately increased. The determination of the number of nodes, based on the analysis of historical operating data and practical project experience, needs to be tailored to the specific geographical environment and discussed on a site-by-site basis.

[0074] The determination of the number of nodes within a coordination unit, according to standard A, must be negotiated on a case-by-case basis, taking into account factors such as geographical environment and load density, system size and complexity, and historical operational data and experience.

[0075] The coordination unit can be directly derived from the connection relationship between the distribution network lines. For example, if line a is connected to line b through an SOP device, then ab is a coordination unit.

[0076] The selection of the key nodes is specifically as follows:

[0077] S1, Filter candidate nodes

[0078] Based on the calculated timing voltage sensitivity of each node, nodes with timing voltage sensitivity in the top 5% are selected as candidate nodes.

[0079] S2. Determine the number of key nodes based on geographical environment and load density.

[0080] The geographical environment includes urban, suburban, and rural areas; the load density includes three levels: high, medium, and low.

[0081] When the geographical environment is urban, the number of key nodes is set to 5 to 10.

[0082] When the geographical environment is suburban, the number of key nodes should be set to 3 to 7.

[0083] When the geographical environment is rural, the number of key nodes is set to 1 to 5.

[0084] If the geographical environment is suburban but the load density is high, the number of critical nodes can be appropriately adjusted within the above range based on the actual situation. Specific adjustment methods are as follows:

[0085] Due to the high load density, the number of key nodes can be approached to the city's scale, that is, close to the upper limit, and can be set to 7.

[0086] If the load density is extremely high and close to urban levels, the number of key nodes can be further increased to 8 to 10 to ensure the stability and efficiency of the system.

[0087] S3. Adjust the number of key nodes based on historical operational data and actual project experience.

[0088] The historical operational data includes fault frequency and maintenance records;

[0089] Based on historical data, the number of critical nodes in areas with frequent failures increases by 10% to 20%, while the number of energy hub nodes in areas with good maintenance records decreases by 10% to 15%.

[0090] The definition of frequent failures can be quantitatively determined by the following criteria: if the number of failures or the duration of failures of a node per year exceeds 1.5 times the average of the entire network, then the area where that node is located can be considered to have frequent failures.

[0091] A good maintenance record is defined as: the number of maintenance operations per node per year is less than 0.5 times the network average.

[0092] In practical applications, the above indicators need to be evaluated comprehensively. The initial assessment determines candidate nodes based on voltage sensitivity analysis. Then, the number and location of energy hub nodes are refined based on geographical environment and load density. The specific number is then determined in combination with system scale and complexity. Finally, the rationality and practicality are ensured through verification and adjustment using historical operating data.

[0093] (V) Construct and solve the extended planning model of the distribution network.

[0094] The constructed distribution network expansion planning model covers network losses, curtailment of solar power, investment, maintenance costs, and the DHM-PCS revenue maximization operation strategy based on time-of-use pricing.

[0095] The distribution network expansion planning model takes into account both economic efficiency and absorption capacity. The objective function consists of the following components: annualized investment cost of each device, annual maintenance cost of each device, network loss cost, operating revenue, and curtailment cost.

[0096] The objective function of the power distribution network expansion planning model includes the annualized investment cost C of each device. inv Annual maintenance cost of each piece of equipment C opm Network loss fee C loss Operating revenue C m And the cost of wasted light C m ;

[0097] The objective function expression of the power distribution network expansion planning model is:

[0098] C = min(C inv +C opm +C loss +C m +C apv (26)

[0099] The annualized investment cost of each piece of equipment is expressed as follows:

[0100]

[0101] In the formula: C inv The annualized investment cost for each piece of equipment is in yuan. The investment cost for AEC is expressed in yuan. LBAC investment cost, in yuan; EHP investment cost, in yuan; The investment cost for TST is expressed in yuan. The investment cost for TST is expressed in yuan. The investment cost for SOFC is in yuan. EC investment cost, in yuan; SOP investment cost, in yuan; AEC rated power or capacity, in kW; The rated power or capacity of the LBAC is expressed in kW. The rated power or capacity of EHP is expressed in kW. This refers to the rated power or capacity of TST, in kW. The rated power or capacity of HST is expressed in kW. This refers to the rated power or capacity of SOFC, in kW. The rated power or capacity of the EC is expressed in kW. SOP rated power or capacity, in kW; τ AEC τ is the capital recovery factor for AEC. LBAC τ is the capital recovery factor for LBAC. EHP τ is the capital recovery factor for EHP. TST τ is the capital recovery factor for TST. HST τ is the capital recovery factor for HST. SOFC τ is the capital recovery factor for SOFC. EC τ is the capital recovery factor for EC. SOP The capital recovery factor for SOP; Ψ For the collection of various devices;

[0102] τ AEC τ LBAC τ EHP τ TST τ HST τ SOFC τ EC and τ SOP The calculation methods are the same, and the following formula is used for both:

[0103]

[0104] In the formula: τ is τ AEC τ LBAC τ EHP τ TST τ HST τ SOFC τ EC or τ SOP ; r is the interest rate; LT is the lifespan of AEC, LBAC, EHP, TST, HST, SOFC, EC, or SOP, in years;

[0105] The expressions for the annual maintenance costs of each piece of equipment are as follows:

[0106]

[0107] In the formula: C opm The annual maintenance cost for each piece of equipment is expressed in yuan. AEC unit capacity maintenance cost, in yuan; The unit cost of LBAC maintenance is in yuan. EHP unit capacity maintenance cost, in yuan; TST unit capacity maintenance cost, in yuan; HST unit capacity maintenance cost, in yuan; The unit maintenance cost for SOFC is in yuan. EC unit capacity maintenance cost, in yuan; The unit cost per unit capacity of SOP is expressed in yuan. AEC rated power or capacity, in kW; The rated power or capacity of the LBAC is expressed in kW. The rated power or capacity of EHP is expressed in kW. This refers to the rated power or capacity of TST, in kW. The rated power or capacity of HST is expressed in kW. This refers to the rated power or capacity of SOFC, in kW. The rated power or capacity of the EC is expressed in kW. The rated power or capacity of the SOP is in kW.

[0108] The expression for the network loss cost is as follows:

[0109]

[0110] In the formula: C loss Network loss cost, in yuan; Days op The number of days under different operating conditions, in days; f loss,t The cost of network loss at time t is expressed in yuan; r mn This is the line resistance, measured in Ω. The square of the line current at time t in the op-th scenario is expressed in amperes (A). 2 ;

[0111] The expression for the operating revenue is as follows:

[0112]

[0113] In the formula: C mOperating revenue is expressed in yuan; EH represents the total number of distributed hydrogen-based multi-energy flow photovoltaic systems. The unit is yuan, representing the energy purchase cost of the distributed hydrogen-based multi-energy flow photovoltaic system at time t in the op-th scenario. The unit is the energy sales cost of the distributed hydrogen-based multi-energy flow photovoltaic system at time t under the OP scenario, expressed in yuan.

[0114] The objective function for the cost of wasted solar power is as follows:

[0115]

[0116] In the formula: C apv Cost of abandoned light, in yuan; f pv Cost of wasted light, in yuan; Days op The number of days under different operating conditions, in days; Let PV be the curtailment power of distributed photovoltaic power at time t in the op-th scenario, in kW; PV is the collection of distributed photovoltaic power.

[0117] ②The constraints of the distribution network extended planning model include distribution network operation power flow constraints, DHM-PCS equipment constraints, and SOP operation constraints;

[0118] The power flow constraints of the distribution network are as follows:

[0119]

[0120] In the formula: The square of the line current, in amperes (A). 2 ; The maximum current that the line can withstand is the square of the rated current, expressed in amperes (A). 2 Ω Line For the set of lines; Ω node A set of nodes;

[0121] The square of the upper limit of the node voltage, in kV. 2 ; The square of the lower limit of the node voltage, in kV. 2 ; The square of the current node voltage, in kV. 2 ;

[0122] P ij,op,t Q represents the active power flowing through branch ij at time t, in kW. ij,op,t Let be the reactive power flowing through branch ij at time t, in kVar; π(:,j) is the set of nodes with end node j; δ(j,:) is the set of nodes with beginning node j. The unit is the node's net load, expressed in kW.

[0123] r ij Line resistance, unit: Ω; x ij Line reactance, in Ω;

[0124] The unit is the node's net load, expressed in kW.

[0125] η EH This is a variable indicating whether a node is connected to a distributed hydrogen-based multi-energy flow photovoltaic consumption system.

[0126] η SOP This is a variable indicating whether a node is connected to the SOP.

[0127] EH input or absorbed power, in kW; Active power transmitted for SOP, all in kW; Net reactive power of the power grid, in kVar; Reactive power transmitted to SOP, in kVar; P jk,op,t and Q jk,op,t This represents the active and reactive power flowing from the current node to its child nodes, in kW. and The squares of the voltage magnitudes at nodes j and i, in kV. 2 ; and The active and reactive power of the line are squared, in kW. 2 .

[0128] The constraints for DHM-PCS equipment are: the power of each DHM-PCS device must be less than the rated power; the operating conditions of DHM-PCS equipment are given by formulas 6 to 9.

[0129] SOP execution constraints:

[0130] SOP constraints are given by Equations 1 to 2;

[0131] ③ Model solution method:

[0132] This application uses CPLEX for solving, but CPLEX cannot solve nonlinear equations, so Equation 2 and Equation 37 are processed by cone.

[0133] Design principle of the method of this invention:

[0134] (1) Power grid topology modification should consider energy coordination between feeders

[0135] Medium-voltage power distribution covers a large area, and with the integration of massive resources, the planning and operation of the distribution network is no longer a matter of local optimization of a single feeder, but requires coordination and cooperation among multiple feeders. However, traditional medium-voltage feeders cannot be interconnected, and feedback to the upstream power grid will affect its safe operation. Therefore, researching how to promote energy interaction among multiple feeders without compromising the stability of the upstream power grid will become a key path to improving energy absorption capacity in the future.

[0136] In the past, the construction of power distribution networks benefited from rapid urbanization. However, with the slowdown in urbanization, rebuilding or large-scale upgrading of power distribution networks to adapt to the continuous growth of power sources and loads is no longer realistic. Advances in power electronics technology have provided new solutions, among which AC / DC distribution technology has become an effective strategy. Currently, the most commonly used AC / DC upgrade schemes include SOP interconnection, AC-DC interconnection, AC-DC-AC interconnection, and "two AC-DC" interconnection. Table 1 shows the upgrade methods for each type.

[0137] Table 1. Methods for Upgrading Four Types of AC / DC Distribution Networks

[0138]

[0139] As can be seen from the comparison in Table 1, the SOP interconnection type is simpler to modify and offers more flexible scheduling. Therefore, this application adopts the SOP interconnection type for flexible interconnection modification of the grid structure.

[0140] By flexibly interconnecting different feeders using SOP (Standard Operating Procedure) technology, multiple feeders collectively form a coordination unit. This transforms the source-load balance of a single feeder into a balance within the coordination unit, and further into resource sharing within the unit. Specifically, this involves... Figure 1 As shown, the load balance is changed from single-feeder source-load balance to balance within the coordination unit. Unlike the traditional interconnection method of tie switches (which cannot control power flow accurately in real time), the distribution network based on SOP interconnection can start from the cooperative characteristics between multiple feeders, and can control the power flow between feeders accurately in real time, thereby optimizing the operating efficiency of the entire distribution network.

[0141] SOPs (Standard Operating Programs) are categorized into back-to-back voltage source converters (B2B VSCs), static synchronous series compensators (SSSCs), and unified power flow controllers (UPFCs). Among these, B2B VSCs are widely used due to their superior performance. This device connects two converters via a DC circuit, decoupling the AC feeders on both sides. Under steady-state operation, the SOP operates in PQ control mode, with variables involving the active and reactive power at both ports. Furthermore, this type of SOP has a transmission efficiency of 98%, therefore, transmission losses are negligible.

[0142] (2) Modeling of SOP operational characteristics

[0143] Modeling the operational characteristics of Standard Operating Procedures (SOPs) is crucial because SOPs possess the ability to precisely control power flow in real time, while traditional tie switches can only perform simple opening and closing operations and cannot regulate power flow. Therefore, SOP operational characteristic modeling can describe in detail the active and reactive power control and capacity limitations of the SOP, characteristics that form the basis for achieving refined power management. Compared to traditional tie switches, SOPs can not only connect feeders but also achieve energy sharing between multiple feeders through precise control, significantly improving the operating efficiency and overall performance of the distribution network.

[0144] (3) Energy coupling strategy and site selection of DHM-PCS

[0145] In addition to the flexible interconnection transformation of the grid structure, the configuration of the energy absorption device is equally crucial. This application selects the Distributed Hydrogen-Based Multi-Energy Flow Photovoltaic Absorption System (DHM-PCS) as the photovoltaic energy absorption device, which covers the supply and demand of multiple energy media such as cold, heat, electricity, and hydrogen.

[0146] The distributed hydrogen-based multi-energy photovoltaic system (DHM-PCS) includes an alkaline electrolysis cell (AEC), a lithium bromide absorption chiller (LBAC), an electric heat pump (EHP), a thermal storage tank (TST), a hydrogen storage tank (HST), a solid oxide fuel cell (SOFC), and an electric chiller (EC).

[0147] Figure 2 The multi-energy flow conversion relationship of cold, heat, electricity and hydrogen in DHM-PCS is shown, and its energy balance equation is as described above;

[0148] (4) DHM-PCS Site Selection

[0149] In AC / DC hybrid distribution networks considering flexible energy exchange between multiple feeders, the system is automatically subdivided into multiple coordination units based on the interconnection between feeders. Within these coordination units, any node can exchange energy with other nodes within the same coordination unit via SOP devices (without going through the upper-level grid). Therefore, when deploying DHM-PCS, configuration only needs to be based on the coordination units, rather than configuring all or most feeders, reducing redundancy in the power absorption system and dependence on land resources.

[0150] This application utilizes time-series voltage sensitivity as a location selection criterion and simplifies its formula to reduce computational complexity. Furthermore, it extends the single-feeder location selection formula to power grid scenarios involving flexible interconnections of multiple feeders.

[0151] The beneficial effects of this invention are:

[0152] This invention provides a distribution network expansion planning method that considers load spatiotemporal transfer and multi-energy flow coordination. It proposes a photovoltaic (PV) distribution network planning model that considers integrated energy spatiotemporal mutual assistance and hydrogen-based multi-energy flow coupling. The proposed PV absorption method absorbs PV power from three dimensions: temporal energy transfer, spatial energy transfer, and energy category conversion. It proposes an operation strategy that considers energy coordination among multiple feeders, breaking the limitations of conventional "source-storage-load" models for distributed power absorption, and elevating the absorption dimension from "point" optimization to "network" coordination. Furthermore, it proposes a distributed hydrogen-based multi-energy coupled PV absorption system (DHM-PCS) operation mode, considering the coordination characteristics between feeders, and constructing a DHM-PCS configuration model with limited critical nodes to maximize the overall absorption benefits. Attached Figure Description

[0153] Figure 1 This is a schematic diagram showing the location of the SOP connected to the medium-voltage distribution network in this invention;

[0154] Figure 2 This is a schematic diagram of the DHM-PCS energy conversion in this invention;

[0155] Figure 3 This is an improved Portuguese 54-node system according to Embodiment 1 of the present invention;

[0156] Figure 4 This is a timing voltage sensitivity diagram of each node in Embodiment 1 of the present invention;

[0157] Figure 5 This is the thermal balance diagram of the coordination unit 1 in Case 1 of Embodiment 1 of the present invention;

[0158] Figure 6 This is the cold balance diagram of coordination unit 1 in Case 1 of Embodiment 1 of the present invention;

[0159] Figure 7 This is the HST state diagram of coordination unit 1 in Case 1 of Embodiment 1 of the present invention;

[0160] Figure 8 This is the state diagram of the coordination unit 1BESS in Case 2 of Embodiment 1 of the present invention.

[0161] For those skilled in the art, other related figures can be obtained from the above figures without any creative effort. Detailed Implementation

[0162] To enable those skilled in the art to better understand the technical solution of the present invention, the technical solution of the present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0163] Example 1

[0164] This embodiment applies the method of this application as follows: Figure 3 The improved Portuguese 54-node system shown was validated, which has 4 substations (S1 to S4), 10 feeders, and 5 tie switches.

[0165] The basic data sources for this embodiment are as follows: The load originates from the power distribution network of a city in China. The photovoltaic data is generated using the PVlib method of a refined physical model chain, and the required temperature, wind speed, GHI, DNI, and DHI data locations are the cities corresponding to the load.

[0166] The distribution network expansion planning method, which takes into account load spatiotemporal transfer and multi-energy flow coordination, specifically includes the following steps:

[0167] (I) Utilizing Standard Operating Procedures (SOPs) to flexibly interconnect the space frame

[0168] In the improved Portuguese 54-node system, all five tie switches were converted to Standard Operating Procedures (SOPs). With the SOPs configured, the 10 feeders were divided into three coordination units based on their electrical connections.

[0169] (II) Select DHM-PCS as the photovoltaic energy consumption device and establish a DHM-PCS operation model;

[0170] This embodiment only plans the distribution network, using DHM-PCS as a multi-energy flow coupling device to associate the distribution network with the heating and cooling network. The heating and cooling loads of each node are allocated according to the proportion of electrical load. Tables 2 and 3 show the equipment parameters and time-of-use electricity prices (electricity purchase and sale). The planning period is 20 years. The unit calorific value price is 0.223 yuan / kWh.

[0171] Table 2. Relevant parameters for each device

[0172]

[0173] Table 3 Time-of-use Electricity Prices

[0174]

[0175] In the study of planning problems, models are often extremely complex, and their computational time costs need to be considered. Therefore, most studies employ various techniques to reduce countless possible scenarios to a finite number. This application uses the K-means clustering method to select typical days. Table 4, through analysis of the SC coefficient and CH index, determines that the optimal number of clusters is three. Essentially, scenarios are divided according to seasonal operational characteristics, and typical scenarios are selected under different operational characteristics. Furthermore, the day with the highest correlation to other days within each cluster is selected as the typical day. In addition, the model proposed in this application exhibits scalability, meaning that typical days can be flexibly changed without requiring extensive modifications to the model.

[0176] Table 4. Determination of Cluster Numbers

[0177]

[0178] (III) Determine the location of DHM-PCS in the distribution network based on the timing voltage sensitivity index

[0179] according to Figure 4 The time-series voltage sensitivity indexes of different nodes at different times are displayed, and their average values ​​are connected. Time-series voltage sensitivity is an indicator that measures the sensitivity of voltage to power fluctuations; an increase in the index value indicates a decrease in the voltage stability of the corresponding node. Following the non-clustered configuration principle of DHM-PCS, nodes 8, 13, 25, 44, and 50 are selected as the configuration points for DHM-PCS.

[0180] pass Figure 3 It can be seen that it is divided into 3 coordination units.

[0181] In coordination unit 1, due to its small size (few nodes), only node 25 with the highest sensitivity is selected as the configuration node.

[0182] In Coordination Unit 2, due to its large size (many nodes), configuring only one DHM-PCS may be insufficient to cover the heating and cooling energy needs of all nodes. Based on the node sensitivity ranking, nodes 4-8, 12, 13, and 38 have relatively high sensitivity. However, the function of the DHM-PCS is to meet heating and cooling energy needs within a certain range; therefore, clustering them together would not align with its purpose. Observation shows that nodes 4-8 are within one geographical area, 12-13 within another, and 38 within its own geographical area. Therefore, we select the nodes with the highest sensitivity within each area as the configuration nodes: 8, 13, and 38.

[0183] In coordination unit 3, due to its small size (few nodes), only the node 50 with the highest sensitivity is selected as the configuration node.

[0184] (IV) Planning Results

[0185] Traditional single-feeder-level photovoltaic (PV) absorption configurations have been effectively demonstrated in numerous studies, but they suffer from high dependence on land resources and the inability to achieve energy interaction between feeders. Therefore, this application focuses on exploring the utilization of various heterogeneous resources for PV absorption under the premise of synergistic energy interaction between feeders.

[0186] This embodiment sets up three cases to study the impact of different configurations on photovoltaic absorption capacity.

[0187] Case 1: Consider the energy coordination and interaction between multiple feeders, and configure a DHM-PCS containing various heterogeneous resources for absorption.

[0188] Case 2: Consider the energy coordination and interaction between multiple feeders, and configure a single energy storage device for absorption.

[0189] Case 3: Only consider the energy coordination and interaction between multiple feeders, and only configure the SOP device.

[0190] 1) Comparative analysis of planning results

[0191] Table 5. Indicators for the three cases

[0192]

[0193] Table 5 details the investment and maintenance costs, network loss costs, operating revenue, total annual cost, and photovoltaic energy integration status under three scenarios (Case 1, Case 2, and Case 3). Specifically, Case 1 and Case 2 both achieved 100% photovoltaic integration, proving that considering energy coordination among multiple feeders and installing integration devices at key nodes is an effective strategy to improve photovoltaic energy integration. However, Case 1's diversified energy sales revenue (electricity, cooling, and heating), coupled with lower investment and maintenance costs, resulted in higher total annual revenue than Case 2. However, the DHM-PCS equipment involved in Case 1 requires coupling electricity, cooling, heating, and hydrogen energy, resulting in a significantly higher electricity demand compared to Case 2, leading to higher network losses.

[0194] Considering the energy coordination and absorption methods among multiple feeders (i.e., configuring SOP devices), their maximum effect is often achieved by combining them with energy storage devices; configuring SOPs alone often results in low economic efficiency. In Case 3, due to the lack of absorption devices, its photovoltaic energy absorption capacity depends entirely on the feeders with absorption functions and their load conditions. Furthermore, considering both absorption rate and configuration cost, the objective function setting for Case 3 leans towards minimizing network losses to facilitate the reconfiguration of absorption devices in practice. Case 3 only utilizes SOP devices for flexible interconnection of the grid architecture, thus exhibiting a relatively low absorption capacity. In addition, due to the lack of regulation from energy storage facilities, the network losses in Case 3 are significantly higher than in Case 2, but lower than in Case 1. Tables 6 and 7 show the capacity optimization results of each coordination unit in Case 1.

[0195] Table 6 Equipment Configuration of Each Coordination Unit

[0196]

[0197]

[0198] Table 7 SOP device configuration in Case 1

[0199]

[0200] 2) Analysis of running results

[0201] Figures 5 to 8The diagrams illustrate the operating status of each device in Case 1. During the transition season, to reduce the operating costs of the DHM-PCS, the photovoltaic power is primarily consumed by the EHP and ER devices during peak photovoltaic power generation periods. During other periods, heating power is mainly provided by the thermal storage tank. Simultaneously, considering time-of-use pricing, the EHP device ceases operation during high-price periods at night. Due to the constraint of consistent capacity at both ends of the day, the thermal storage tank requires replenishment after heat release. Therefore, the EHP is recharged during low-price periods at the end of the day. In the cooling system, the absorption chiller is the primary cooling device. The electric chiller, acting as a supplementary cold source for the absorption chiller, only outputs cooling energy during peak photovoltaic power generation periods. In winter, due to centralized heating, heat load demand surges. During the low-price period in the early morning, the EHP provides heating and stores the waste heat in the thermal storage tank. During high-price periods outside of peak photovoltaic power generation periods, the heat load is primarily provided by the thermal storage tank. In cold balance, the cold source is mainly provided by the absorption chiller, with the electric chiller only providing cooling at the end of the day. This is because during peak electricity price periods at night, the EHP (Extended Heating Utilization) stops operating, and the thermal storage tank releases a large amount of heat energy to meet demand. The thermal storage tank needs to maintain consistent energy storage at the end of the day; therefore, during periods of low electricity price at the end of the day, the EHP recharges the storage tank. In summer, electricity and cooling loads surge (especially electricity). During peak solar power generation periods, electricity load surges, with some periods of peak solar power generation exceeding solar output, effectively alleviating grid connection issues. The EHP operates similarly to winter, primarily avoiding purchasing electricity during peak electricity price periods. Due to the surge in cooling load, relying mainly on the LBAC (Balanced Load AC) presents a supply bottleneck. Therefore, cooling power is primarily provided by the ER (Extended Heating Utilization) and the LBAC.

[0202] The operational characteristics reveal the following: Considering operating costs, the EHP (Energy Power Generation System), as the primary power-consuming equipment, operates mainly during periods of low electricity prices and peak solar power generation. Thermal energy storage tanks and hydrogen storage tanks, acting as energy storage devices, store energy during periods of low electricity prices or peak solar power generation, and supply energy during periods of high electricity prices, thus converting electrical energy across time. SOFC (Solar Energy Storage Unit), as a hydrogen-electricity-thermal coupling device, primarily operates during periods of high electricity prices and outside of peak solar power generation periods. Due to economic and efficiency factors, LBAC (Liquid Batteries and Heat Storage Unit) serves as the primary cooling equipment. The coordinated operation of these multiple devices, through the DHM-PCS (Hydrogen Power Management System-PCS) and SOP (Standard Operating Procedure) devices, enables cross-time and spatial regulation and energy type conversion of solar energy, significantly promoting the integration of distributed solar power.

[0203] 3) Comparison of DHM-PCS and Energy Storage

[0204] The DHM-PCS is compared with conventional electrochemical energy storage systems in terms of operating characteristics, cost, efficiency, and environmental friendliness.

[0205] Operating characteristics Figure 7 and Figure 8The capacity of the two energy storage devices in Coordination Unit 1 at various times is shown. They are similar in that they charge during periods of high photovoltaic power generation and sell energy during periods of high electricity prices. Internal energy fluctuations can be compared using the coefficient of variation (CV).

[0206] Table 8 Energy CV Indicators of Coordination Unit 1HST and BESS

[0207]

[0208] In the photovoltaic energy consumption process, hydrogen energy storage exhibits lower energy volatility compared to electrochemical energy storage systems due to the integration of other consumption and regulation devices. This advantage is quantified by comparing the coefficient of variation (CV), as shown in Table 8. The lower CV value of the hydrogen energy storage system in Case 1, due to the coupled and interactive operation of multiple types of devices, demonstrates its ability to effectively reduce the volatility risk caused by energy surplus or shortage, further enhancing the overall resilience of the consumption system. In contrast, the electrical energy storage in Case 2, as the sole consumption device, lacks the coupled interaction of multiple types of devices, resulting in greater internal energy volatility and a more singular operating characteristic.

[0209] Regarding cost, Table 5 summarizes the economic costs in detail, showing that DHM-PCS is more economical. Due to the influence of electrochemical characteristics, the lifespan of BESS equipment is generally shorter than that of DHM-PCS equipment. Furthermore, the unit capacity investment cost of HST and TST energy storage equipment is much lower than that of BESS. Only SOFC is a high-investment device. Therefore, in terms of cost, DHM-PCS is less expensive than traditional BESS.

[0210] In terms of energy density and application flexibility, hydrogen storage significantly outperforms electrochemical storage. Its high volumetric energy density allows it to store more energy in a smaller footprint, making it particularly suitable for large-scale energy storage needs where land resources are scarce. Furthermore, in situations of power shortages, hydrogen energy can be transported to compensate for the power outage. In contrast, electrochemical storage systems have limitations in terms of flexibility and energy storage density.

[0211] In terms of environment and stability, hydrogen storage, especially hydrogen produced from renewable energy sources, has low greenhouse gas emissions over its life cycle, making it key to the clean energy transition. While it is flammable, appropriate technologies can ensure safety. Electrical energy storage systems, particularly lithium batteries, may impact the environment during production and recycling and pose safety risks in the event of damage, despite continuous advancements in safety technologies.

[0212] In terms of energy coupling, traditional BESS can only interact from an electrical energy perspective, but hydrogen energy storage can couple with multiple energy flows such as cooling, heating, and electricity, improving the system's operational flexibility from the perspective of energy medium conversion and enabling effective docking and integration with other energy systems.

[0213] In summary, the synergistic operation of hydrogen storage with multiple types of energy coupling devices can greatly enhance the flexibility of energy storage devices and the overall resilience of the energy consumption system. Driven by policies promoting the rapid development of clean energy, hydrogen-based multi-energy flow coupling generalized energy storage and consumption systems will become a key focus for future development.

[0214] 4) Analysis of the effect of energy coordination among multiple feeders

[0215] Traditional grid integration methods often deploy integration devices on each feeder. While this method effectively alleviates grid integration challenges, it suffers from low efficiency and a huge demand for land resources. In practice, the demand for large amounts of land is often difficult to meet, especially in areas with scarce land resources. Upgrading the grid structure by considering energy coordination among multiple feeders and interconnecting them using Standard Operating Procedures (SOPs) can alleviate the land resource constraints of traditional methods. For example, traditional distributed grid integration methods require eight integration devices, while this method only requires five. However, considering only energy coordination among multiple feeders without integrating devices results in limited integration capacity. For instance, in Case 3, the integration rate is only 69.29%. In contrast, Cases 1 and 2, by configuring integration devices in addition to SOPs, can maximize the utilization rate of renewable energy. Although Case 3 suffers from insufficient integration capacity, the power control between feeders using SOPs can stabilize the voltage. This is similar to Cases 1 and 2. In Cases 1-3, the voltage of all nodes remained stable between 0.95 and 1.05.

[0216] The method of this invention utilizes SOP to flexibly interconnect the grid structure, enabling feeders to have bidirectional continuous power flow regulation capability; based on the time-series voltage sensitivity differences, a limited number of key nodes in the distribution network are selected to deploy distributed hydrogen-based multi-energy flow photovoltaic consumption systems (DHM-PCS); the constructed planning model covers grid loss, curtailment, investment, maintenance costs, and a DHM-PCS revenue maximization operation strategy based on time-of-use pricing.

[0217] The applicant declares that the above description is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Those skilled in the art should understand that any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention fall within the protection and disclosure scope of the present invention.

Claims

1. A distribution network expansion planning method considering load spatiotemporal transfer and multi-energy flow coordination, characterized in that: Includes the following steps: (I) Modify the location of the tie switch in the distribution network into a SOP device and establish an SOP operation model; (II) Select a distributed hydrogen-based multi-energy flow photovoltaic consumption system as the photovoltaic energy consumption device, link the distribution network with the cold and heat network, distribute the cold and heat loads of each node according to the proportion of electrical load, and establish the cold, heat, electricity and hydrogen multi-energy flow conversion relationship in the distributed hydrogen-based multi-energy flow photovoltaic consumption system. (III) Select typical days using the K-means clustering method, obtain three cooling, heating, power and photovoltaic curves for the typical days, and apply the cooling, heating, power and photovoltaic curve data of the typical days to step (IV) to calculate the time-series voltage sensitivity and (V) to solve the distribution network extended planning model; (IV) Determine the configuration location of distributed hydrogen-based multi-energy flow photovoltaic power consumption system in the distribution network based on the time-series voltage sensitivity index; (V) Construct and solve the extended planning model of the distribution network; The objective function C of the power distribution network expansion planning model includes the annualized investment cost C of each device. inv Annual maintenance cost of each piece of equipment C opm Network loss cost C loss Operating revenue C m And the cost of abandoned light C apv ; C=min(C inv +C opm +C loss +C m +C apv ) (26); The constraints of the distribution network expansion planning model include distribution network operation flow constraints, distributed hydrogen-based multi-energy flow photovoltaic consumption system constraints, and SOP operation constraints.

2. The distribution network expansion planning method considering load spatiotemporal transfer and multi-energy flow coordination according to claim 1, characterized in that: The SOP operating model includes active power constraints and capacity constraints; The active power constraint of the SOP is as follows (1): In the formula: This represents the three-phase power of nodes i and j connected through the s-th SOP at time t. and Both are equal; Ω SOP This is a collection of SOP devices; φ represents the three phase sequences; The capacity constraint of the SOP is as follows (2): In the formula: Let be the active and reactive power of the s-th SOP at node i at time t, in kVar; The capacity corresponds to the SOP, and the unit is kVA.

3. The distribution network expansion planning method considering load spatiotemporal transfer and multi-energy flow coordination according to claim 1, characterized in that: The energy balance equations for the multi-energy flow conversion relationship of cold, heat, electricity, and hydrogen include the electrical energy balance equation, the thermal energy balance equation, and the cold energy balance equation. The energy balance equation is as follows: P PDN (t)=P SOFC,E (t)-P EHP,in (t)-P AEC,in (t)-P EC,in (t) (3) In the formula: P PDN (t) represents the electricity consumption of the distributed hydrogen-based multi-energy photovoltaic system at time t, in kW; P SOFC,E (t) represents the power emitted by the SOFC at time t, in kW; P EHP,in (t), P AEC,in (t), P EC,in (t) represents the power absorbed by EHP, AEC, and EC at time t, respectively, in kW; The heat energy balance equation is: P h (t)=P EHP,out (t)+βP TST,out (t)+P SOFC,H (t)-P LBAC,in (t)-(1-β)P TST,in (t) (4) In the formula: P h (t) represents the thermal power emitted by the distributed hydrogen-based multi-energy photovoltaic system at time t, in kW; P EHP,out (t), P TST,out (t), P SOFC,H (t) represents the heat power emitted by EHP, TST, and SOFC at time t, respectively, in kW; P LBAC,in (t), P TST,in (t) represents the heat power absorbed by LBAC and TST at time t, respectively, in kW; β is a 0 / 1 variable, representing the operating state of TST; The cold energy balance equation is: P c (t)=P EC,out (t)+P LBAC,out (t) (5) In the formula: P c (t) represents the cooling power emitted by the distributed hydrogen-based multi-energy photovoltaic system at time t, in kW; P EC,out (t), P LBAC,out (t) represents the cooling power emitted by EC and LBAC at time t, both in kW.

4. The distribution network expansion planning method considering load spatiotemporal transfer and multi-energy flow coordination according to claim 1, characterized in that: The method for determining the location of the distributed hydrogen-based multi-energy flow photovoltaic power grid in step (IV) includes the following steps: (1) Calculate voltage sensitivity The voltage sensitivity analysis method is used to calculate the sensitivity of each node to voltage changes, i.e., the time-series voltage sensitivity of each node. (2) Select key nodes Based on the time-series voltage sensitivity calculated in step (1), several key nodes with high time-series voltage sensitivity are selected as the configuration locations of the distributed hydrogen-based multi-energy flow photovoltaic system.

5. The distribution network expansion planning method considering load spatiotemporal transfer and multi-energy flow coordination according to claim 4, characterized in that: The selection of the key nodes is specifically as follows: S1, Filter candidate nodes Based on the calculated timing voltage sensitivity of each node, nodes with timing voltage sensitivity in the top 5% are selected as candidate nodes. S2. Determine the number of key nodes based on geographical environment and load density. The geographical environment includes urban, suburban, and rural areas; the load density includes three levels: high, medium, and low. When the geographical environment is urban, the number of key nodes is set to 5 to 10. When the geographical environment is suburban, the number of key nodes should be set to 3 to 7. When the geographical environment is rural, the number of key nodes is set to 1 to 5. S3. Adjust the number of key nodes based on historical operational data and actual project experience. The historical operational data includes fault frequency and maintenance records; Based on historical data, the number of critical nodes in areas with frequent failures increases by 10% to 20%, while the number of energy hub nodes in areas with good maintenance records decreases by 10% to 15%.

6. The distribution network expansion planning method considering load spatiotemporal transfer and multi-energy flow coordination according to claim 4, characterized in that: The specific method for calculating the sensitivity of the node to voltage changes is as follows: In the traditional sensitivity, a voltage offset weighting factor is introduced, and the sensitivity S of the active power change injected at node n at time t to the voltage at node m is calculated. mn,t The calculation formula is: S mn,t =λ mn,t ×ΔV m,t (20) ΔV m,t =V m,t -V eu,m,t (21) In the formula: λ mn,t Traditional voltage sensitivity index, ΔV m,t V represents the deviation between the node voltage and the desired voltage, in kV. m,t V represents the current node voltage, in kV. eu,m,t The desired node voltage is expressed in kV. The overall sensitivity S of node m to its feeder at time t m,t The calculation formula is: Where: FL is the set of current feeder load nodes; Comprehensive sensitivity S with time weighting op,m,t Overall sensitivity S op,m,t The calculation formula is: Where: N exceed,t The voltage over-limit node C at time t apv Number of dots, V exceed,t The node voltage exceeds the limit at time t, in kV. Based on the number of days under different operating conditions, different weights are assigned to different operating conditions, and a comprehensive sensitivity S of the operating state weight is introduced. opw,m The overall sensitivity S of the running state weights opw,m The calculation formula is as follows: In the formula: Days op The number of days represents different operating states, in days; op represents different operating states, and M is the type of operating state.

7. The distribution network expansion planning method considering load spatiotemporal transfer and multi-energy flow coordination according to claim 1, characterized in that: The annualized investment cost C of each piece of equipment inv The objective function is as follows: The formula for calculating the capital recovery factor τ of each piece of equipment in formula (27) is as follows: (28) In the formula: The investment cost for AEC is expressed in yuan. LBAC investment cost, in yuan; EHP investment cost, in yuan; The investment cost for TST is expressed in yuan. The investment cost for TST is expressed in yuan. The investment cost for SOFC is in yuan. EC investment cost, in yuan; SOP investment cost, in yuan; P k 设备 The rated power or capacity of each device is specified, where the device is designated as AEC, LBAC, EHP, TST, HST, SOFC, EC, or SOP, and the unit is kW; τ 设备 denoted as the capital recovery factor for the equipment, where the equipment includes AEC, LBAC, EHP, TST, HST, SOFC, EC, and SOP, and the capital recovery factor for all equipment is a constant; r is the interest rate; LT is the lifespan in years; and Ψ represents the set of equipment. Annual maintenance cost C for each piece of equipment opm The objective function is as follows: In the formula: c opm for AEC unit capacity maintenance cost, in yuan; The unit cost of LBAC maintenance is in yuan. EHP unit capacity maintenance cost, in yuan; TST unit capacity maintenance cost, in yuan; HST unit capacity maintenance cost, in yuan; The unit maintenance cost for SOFC is in yuan. EC unit capacity maintenance cost, in yuan; The unit cost per unit capacity of SOP is expressed in yuan. Network loss cost C loss The objective function is as follows: In the formula: f loss,t The cost of network loss at time t is expressed in yuan; r mn The resistance of the circuit is expressed in ohms. The square of the line current at time t in the op-th scenario is expressed in amperes (A). 2 ; Operating revenue C m The objective function is as follows: In the formula: EH represents the total number of distributed hydrogen-based multi-energy flow photovoltaic systems; The unit is yuan, representing the energy purchase and sales costs of the distributed hydrogen-based multi-energy flow photovoltaic system at time t under the OP scenario. Cost of abandoned light C apv The objective function is as follows: In the formula: f pv Cost of wasted light, in yuan; Let PV be the curtailment power of distributed photovoltaic power at time t in the op-th scenario, in kW; PV is the collection of distributed photovoltaic power.

8. The distribution network expansion planning method considering load spatiotemporal transfer and multi-energy flow coordination according to claim 1, characterized in that: The power flow constraints of the distribution network operation are specifically as follows: In the formula: The maximum current that the line can withstand is the square of the rated current, expressed in amperes (A). 2 Ω Line Ω node For the set of lines and the set of nodes, These are the squares of the upper and lower limits of the node voltage, in kV. 2 ; The square of the current node voltage, in kV. 2 ; π(:,j) is the set of nodes whose terminal node is j; δ(j,:) is the set of nodes whose head node is j; r ij x ij P represents line resistance and reactance, measured in ohms. ij,op,t Q ij,op,t These are the active power and reactive power flowing through branch ij at time t, respectively, in kW and kVar. The unit is the node's net load, expressed in kW. η EH η SOP The variables related to whether a node is connected to the distributed hydrogen-based multi-energy flow photovoltaic consumption system and the SOP (Start of Operation) are: These represent the power absorbed and emitted by the distributed hydrogen-based multi-energy photovoltaic system, respectively, in kW; P jk,op,t Q jk,op,t This represents the power flowing from the current node to its child nodes, in kW. The square of the node voltage amplitude, in kV. 2 ; The square of the line current, in amperes (A). 2 ; and The power squared is the line power, in kW. 2 .

9. The distribution network expansion planning method considering load spatiotemporal transfer and multi-energy flow coordination according to claim 1, characterized in that: The distribution network extended planning model is solved using CPLEX, and the capacity constraint of the SOP and the power flow constraint of the distribution network extended planning model are processed using cone processing.

Images