Microgrid cluster planning method and device for improving reliability of low-voltage power distribution system

Abstract

The application discloses a micro-grid group planning method and device for improving reliability of low-voltage power distribution systems, and improves distribution network reliability, comprising: establishing an AC-DC hybrid micro-grid model, a micro-grid minimum cost objective function and a micro-grid operation constraint condition according to source and load AC-DC characteristics, and solving to obtain a first micro-grid group investment cost; constructing a node division objective function and a node division constraint as an upper layer division model to perform node division, and obtaining a node division result; taking the micro-grid minimum cost objective function and the micro-grid operation constraint as a lower layer planning model, and solving under the constraint of an islandable probability calculation model to obtain power demand of each micro-grid; the upper layer division model re-performs node division according to the power demand, and obtains a node division result; when the two division results are consistent, a second micro-grid group investment cost is obtained according to the power demand; and when it is judged that the first micro-grid group investment cost is greater than the second micro-grid group investment cost, it is determined that the micro-grid group has effective gain, and the solving is ended.

Description

Technical Field

[0001] This invention relates to the field of power distribution system technology, and in particular to a microgrid group planning method and apparatus for improving the reliability of low-voltage power distribution systems. Background Technology

[0002] Driven by the development of new elements and new business models in power generation and load, a large number of distributed power sources have developed rapidly, and the demand for different types of loads is also growing rapidly. Distributed power sources, represented by wind and solar power, are characterized by volatility and intermittency, which, together with the rapidly growing load demand, bring a dual challenge to the safe operation of the distribution network.

[0003] Currently, microgrids effectively aggregate large-load users in close proximity and connect them to the end of the distribution network by integrating geographically adjacent distributed power sources, achieving cascaded energy utilization and improving the local consumption rate of new energy. Simultaneously, optimized microgrid configuration can improve the self-sufficiency rate of electricity within its jurisdiction, reduce the overall electricity demand of the microgrid, and thus alleviate the pressure on the distribution network. However, the growth in load is accompanied by an increase in the proportion of DC load, with a large number of DC loads, represented by electric vehicles, being connected to new distribution networks. Traditional microgrids, using AC microgrid structures, require significant investment in converters to cope with the large increase in DC load, resulting in a substantial decrease in their economic efficiency and reliability. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a microgrid group planning method and device for improving the reliability of low-voltage power distribution systems, thereby improving the reliability of the power distribution network.

[0005] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows:

[0006] A microgrid group planning method for improving the reliability of low-voltage power distribution systems, characterized by comprising:

[0007] S1. Establish an AC / DC hybrid microgrid model based on the AC / DC characteristics of the source and load;

[0008] S2. Based on the AC / DC hybrid microgrid model, construct the microgrid minimum cost objective function and microgrid operation constraints, and solve for the investment cost of the first microgrid group;

[0009] S3. Construct a node partitioning objective function and node partitioning constraints as an upper-level partitioning model, and perform node partitioning on the AC / DC hybrid microgrid model according to the upper-level partitioning model to obtain the first node partitioning result;

[0010] S4. Construct a microgrid cluster islanding probability calculation model based on the Monte Carlo method;

[0011] S5. The minimum cost objective function of the microgrid and the microgrid operation constraints are used as the lower-level planning model, and the islanding probability calculation model is used as the constraint condition of the lower-level planning model to solve the problem and obtain the power demand of each microgrid.

[0012] S6. The upper-layer partitioning model re-partitions nodes according to the power requirements of each microgrid to obtain a second node partitioning result, and determines whether the second node partitioning result is consistent with the first node partitioning result. If not, it returns to step S3; if so, it obtains the second microgrid group investment cost according to the power requirements of each microgrid.

[0013] S7. Determine whether the investment cost of the first microgrid group is greater than the investment cost of the second microgrid group. If yes, the microgrid group is considered to have effective gains and the solution ends. If no, return to step S5 and use the investment cost of the first microgrid group as the upper limit constraint.

[0014] To solve the above-mentioned technical problems, another technical solution adopted by the present invention is as follows:

[0015] A microgrid group planning device for improving the reliability of low-voltage power distribution systems includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the various steps of the microgrid group planning method for improving the reliability of low-voltage power distribution systems as described above.

[0016] The beneficial effects of this invention are as follows: A hybrid AC / DC microgrid model is established considering the AC / DC characteristics of the source and load, which effectively reduces the investment in converters. While ensuring the economic efficiency of microgrid planning, it meets the growing demand for DC loads. Furthermore, by constructing a minimum cost objective function for the microgrid and solving for microgrid operation constraints, the investment cost of the first microgrid group without considering cluster partitioning is obtained. Then, by introducing an islanding probability threshold constraint during the microgrid group optimization configuration process, the power demand of the microgrid group on the distribution network is reduced, and the investment cost of the second microgrid group considering cluster partitioning is calculated. The solution ends only when the investment cost of the first microgrid group without considering cluster partitioning is greater than the investment cost of the second microgrid group considering cluster partitioning, i.e., only when the partitioned microgrid group is deemed to have effective gains, thereby improving the reliability of the distribution network. Attached Figure Description

[0017] Figure 1 This invention provides a microgrid group planning method for improving the reliability of low-voltage power distribution systems.

[0018] Figure 2 This is a schematic diagram of an AC / DC hybrid microgrid model in a microgrid group planning method for improving the reliability of low-voltage power distribution systems, as described in an embodiment of the present invention.

[0019] Figure 3 This is a schematic diagram of a microgrid group planning device for improving the reliability of low-voltage power distribution systems, as described in an embodiment of the present invention. Detailed Implementation

[0020] To explain in detail the technical content, objectives, and effects of the present invention, the following description is provided in conjunction with the embodiments and accompanying drawings.

[0021] Please refer to Figure 1 A microgrid group planning method for improving the reliability of low-voltage power distribution systems includes:

[0022] S1. Establish an AC / DC hybrid microgrid model based on the AC / DC characteristics of the source and load;

[0023] S2. Based on the AC / DC hybrid microgrid model, construct the microgrid minimum cost objective function and microgrid operation constraints, and solve for the investment cost of the first microgrid group;

[0024] S3. Construct a node partitioning objective function and node partitioning constraints as an upper-level partitioning model, and perform node partitioning on the AC / DC hybrid microgrid model according to the upper-level partitioning model to obtain the first node partitioning result;

[0025] S4. Construct a microgrid cluster islanding probability calculation model based on the Monte Carlo method;

[0026] S5. The minimum cost objective function of the microgrid and the microgrid operation constraints are used as the lower-level planning model, and the islanding probability calculation model is used as the constraint condition of the lower-level planning model to solve the problem and obtain the power demand of each microgrid.

[0027] S6. The upper-layer partitioning model re-partitions nodes according to the power requirements of each microgrid to obtain a second node partitioning result, and determines whether the second node partitioning result is consistent with the first node partitioning result. If not, it returns to step S3; if so, it obtains the second microgrid group investment cost according to the power requirements of each microgrid.

[0028] S7. Determine whether the investment cost of the first microgrid group is greater than the investment cost of the second microgrid group. If yes, the microgrid group is considered to have effective gains and the solution ends. If no, return to step S5 and use the investment cost of the first microgrid group as the upper limit constraint.

[0029] As described above, the beneficial effects of this invention are as follows: A hybrid AC / DC microgrid model is established considering the AC / DC characteristics of the source and load, effectively reducing the investment in converters. While ensuring the economic efficiency of microgrid planning, it meets the growing demand for DC loads. Furthermore, by constructing a minimum cost objective function for the microgrid and solving for microgrid operation constraints, the investment cost of the first microgrid group without considering cluster partitioning is obtained. Then, by introducing an islanding probability threshold constraint during the microgrid group optimization configuration process, the power demand of the microgrid group on the distribution network is reduced, and the investment cost of the second microgrid group considering cluster partitioning is calculated. The solution ends only when the investment cost of the first microgrid group without considering cluster partitioning is greater than the investment cost of the second microgrid group considering cluster partitioning, i.e., only when the partitioned microgrid group is deemed to have effective gains, thereby improving the reliability of the distribution network.

[0030] Furthermore, the objective function for constructing the minimum cost of the microgrid based on the AC / DC hybrid microgrid model includes:

[0031] Acquire microgrid equipment data and operational data. The equipment data includes the number of devices, equipment capacity, equipment investment cost, and equipment maintenance cost. The operational data includes electricity purchase cost, electricity sales cost, and converter loss.

[0032] The annual investment cost and annual maintenance cost are obtained based on the number of equipment, equipment capacity, equipment investment cost, and equipment maintenance cost.

[0033] The annual operating cost is obtained based on the electricity purchase cost, electricity sales cost, and converter losses.

[0034] The minimum cost objective function for the microgrid is derived based on the annual investment cost, annual maintenance cost, and annual operating cost. Specifically:

[0035] The minimum cost objective function of the microgrid includes:

[0036]

[0037] In the formula: This is the total cost of a microgrid over m years, C AI Investment cost for that year, C AM Annual maintenance costs, C AO Operating costs for the year.

[0038] As described above, the annual investment cost and annual maintenance cost of equipment in a microgrid can be accurately described by the number of devices, equipment capacity, equipment investment cost, and equipment maintenance cost. In addition, the annual operating cost can be accurately described by the electricity purchase cost, electricity sales cost, and converter loss. Thus, an accurate minimum cost objective function for the microgrid can be constructed.

[0039] Furthermore, the annual investment cost includes:

[0040] C AI =(N WT C IWT +N PV C IPV +N BAT C IBAT +N TRANS C ITRANS CRF;

[0041]

[0042] Where: N WT N PV N BAT N TRANS These are the quantities of wind turbines, photovoltaic panels, batteries, and converters, respectively; C IWT C IPV C IBAT C ITRANS These are the unit costs of the wind turbine, photovoltaic panels, battery packs, and converter, respectively; CRF is the conversion factor for converting the initial investment cost into the annual investment cost, n is the number of years in the entire microgrid life cycle, and j is the annual interest rate;

[0043] The annual maintenance cost includes:

[0044] C AM =(Cap WT C MWT +Cap PV C MPV +Cap BAT C MBAT +Cap TRANS C MTRANS );

[0045] In the formula: Cap WT Cap PV Cap BAT Cap TRANS These are the capacities of the wind turbine, solar panels, batteries, and converters, respectively; C MWT C MPV C MBAT C MTRANS These are the unit annual maintenance costs for wind turbines, photovoltaic panels, battery packs, and converters, respectively.

[0046] The annual operating costs include:

[0047]

[0048]

[0049]

[0050] In the formula: E(·) represents the expectation; The cost of running a microgrid in scenario i is given. Each scenario i lasts for 24 hours, and there are K scenarios in a year. These represent the energy purchased and sold by the microgrid to the main grid within time t in scenario i; C bprice C sprice It is the purchase / sale electricity price; σ is the converter efficiency; P trans C represents the converter power of the converter. loss Cost of converter losses; Let be a binary variable with values ​​of 0 and 1.

[0051] As described above, data on the quantity, capacity, and cost of microgrid equipment such as wind turbines, photovoltaic panels, batteries, and converters can describe the impact of different types of equipment on the annual investment and maintenance costs of the microgrid. Furthermore, by describing the operating scenarios of the microgrid, the costs incurred in different operating scenarios can be described, thereby accurately describing the annual investment, maintenance, and operating costs of the microgrid.

[0052] Furthermore, the microgrid operation constraints include AC sub-microgrid supply and demand balance constraints, DC sub-microgrid supply and demand balance constraints, and battery constraints;

[0053] The aforementioned supply and demand balance constraints include:

[0054] P W,t +P buy,t -P sell,t -β AC-DC,t P AC-DC,t +σβ DC-AC,t P DC-AC,t =P AC,t ;

[0055] β AC-DC,t +β DC-AC,t ≤1;

[0056] In the formula: P W,t P is the output power of the fan. buy,t and P sell,t These represent the power purchased from and sold to the upstream of the AC sub-microgrid, respectively; σ is the conversion efficiency of the rectifier; P AC-DC,t P represents the electrical power flowing from the AC subnetwork to the DC subnetwork. DC-AC,t P represents the electrical power flowing from the DC subgrid to the AC subgrid. AC,t For AC load; β AC-DC,t / β DC-AC,t It is a binary variable that takes values ​​of 0 and 1;

[0057] The supply and demand balance constraints of the DC sub-microgrid include:

[0058] P PV,t +μ d,t P d,t -μ c,t P c,t +σβ AC-DC,t P AC-DC,t -β DC-AC,t P DC-AC,t =P AC,t ;

[0059] μ d,t +μ c,t ≤1;

[0060] In the formula: P d,t and P c,t For the discharge and charging power of the battery pack; μ d,t and μ c,t It is a binary variable that takes values ​​of 0 and 1;

[0061] The battery constraints include:

[0062] E min ≤E ES,t ≤E max ;

[0063]

[0064] E min =SOC min Cap BAT E max =SOC max Cap BAT ;

[0065] In the formula: E ES,t It is the battery capacity, E min and E max These are the upper and lower limits of the battery capacity; η c and η d It refers to the battery's charging and discharging efficiency; SOC (State of Charge) min and SOC max These are the upper and lower limits of the battery charge ratio.

[0066] As described above, by constraining the supply and demand balance of the AC sub-microgrid, the DC sub-microgrid, and the battery, the objective function for solving the minimum cost of the microgrid is constrained, so that the solution can meet the actual operating conditions of the microgrid and improve the accuracy of the model solution.

[0067] Furthermore, the node partitioning objective function includes:

[0068] maxΩ=λ1ωL +λ2ω P ;

[0069] In the formula: λ1 and λ2 are weighting coefficients; ω L ω represents the node coupling degree. P This refers to the active power balance.

[0070] As described above, by constructing the node partitioning objective function through node coupling degree and active power balance degree, it is possible to partition nodes that are geographically close and have a clustering effect into microgrids in the same region, thereby realizing the partitioning of nodes.

[0071] Furthermore, the node coupling degree includes:

[0072]

[0073]

[0074] d ij =lg(S) VQ,jj / S VQ,ij );

[0075] ΔV jj =S VQ，jj ΔQ jj ;

[0076] ΔV ij =S VQ，ij ΔQ ij ;

[0077] In the formula: L ij Let μ be the electrical distance between node i and node j, and μ be a 0-1 variable; d ij The degree of closeness between node i and node j; d in The degree of closeness between node i and node n; d jn The degree of closeness between node j and node n; ΔV jj and ΔV ij ΔQ represents the change in voltage amplitude at distribution network nodes. jj and ΔQ ij S represents the change in reactive power at the node. VQ,jj The distribution network node sensitivity is obtained by calculating the ratio of the voltage change at node j to the reactive power change at node j; S VQ,ij The distribution network node sensitivity is obtained by calculating the ratio of the voltage change at node i to the reactive power change at node j.

[0078] The active power balance includes:

[0079]

[0080]

[0081] Where: N clu P represents the number of microgrids divided into groups; clu-i Let P(t) represent the active power balance of microgrid group i. i Let be the power demand of microgrid i at time t; T is the entire planning period.

[0082] As described above, the node coupling degree can be accurately described by the geographical distance between each node and the type of load connected to the node. The active power balance degree can be calculated by the ratio of the overall power demand of the microgrid cluster to the maximum power demand during the entire planning period. Based on the node coupling degree and the active power balance degree, an accurate node partitioning objective function can be constructed.

[0083] Furthermore, the node partitioning constraints include:

[0084]

[0085] In the formula: V AC.i For the i-th type of AC voltage, V represents the maximum and minimum values ​​of the AC voltage of type i, respectively. DC.j For the j-th type of DC voltage, These are the maximum and minimum values ​​of the DC voltage of type j, respectively.

[0086] As described above, the AC and DC voltages of the nodes serve as constraints for node partitioning, ensuring that the voltage fluctuations of each node within the distribution network cannot exceed the upper and lower limits when partitioning node clusters.

[0087] Furthermore, the model for calculating the probability of microgrid clusters becoming islanded based on the Monte Carlo method includes:

[0088]

[0089] Where: 8760 is the equivalent total duration per year; N is the total number of users in the system load; U l denoted as the number of users experiencing power outages during the l-th model run; T represents the duration of the fault.

[0090] As described above, by establishing a microgrid cluster islanding probability calculation model, it can be used to verify and analyze whether the power supply of the microgrid cluster formed after dividing the nodes is reliable. By simulating the operation scenario of the microgrid cluster, the operation status of the microgrid cluster is constructed, thereby calculating the microgrid cluster islanding probability.

[0091] Another embodiment of the present invention provides a microgrid group planning device for improving the reliability of low-voltage power distribution systems, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the various steps of the microgrid group planning method for improving the reliability of low-voltage power distribution systems as described above.

[0092] The microgrid group planning method and apparatus for improving the reliability of low-voltage power distribution systems provided by this invention can be applied to power distribution network system planning scenarios and can provide a valid reference for the planning and design of new power systems with multiple microgrids. The following is a detailed description of the implementation methods:

[0093] Example 1

[0094] Please refer to Figure 1 A microgrid group planning method for improving the reliability of low-voltage power distribution systems includes:

[0095] S1. Establish an AC / DC hybrid microgrid model based on the AC / DC characteristics of the source and load; please refer to... Figure 2 The model takes into account the AC and DC characteristics of the source load and connects it to different buses in different areas, which can effectively reduce the investment in converters and meet the growing demand of DC load while ensuring the economic efficiency of microgrid planning.

[0096] S2. Based on the AC / DC hybrid microgrid model, construct the microgrid minimum cost objective function and microgrid operation constraints, and solve for the investment cost of the first microgrid group; that is, considering the single microgrid configuration model without considering cluster partitioning, establish a microgrid planning model without considering microgrid group partitioning, aiming to minimize the investment cost of the microgrid itself. This investment cost will serve as the basis for judging the validity of the planning results in step S7. The model includes constructing the microgrid minimum cost objective function and microgrid operation constraints, specifically:

[0097] S21. Construct the minimum cost objective function for microgrids:

[0098] S211. Obtain equipment data and operation data of the microgrid. Equipment data includes the number of equipment, equipment capacity, equipment investment cost and equipment maintenance cost. Operation data includes electricity purchase cost, electricity sales cost and converter loss.

[0099] S212. Based on the number of equipment, equipment capacity, equipment investment cost, and equipment maintenance cost, the annual investment cost and annual maintenance cost are obtained;

[0100] Annual investment costs include:

[0101] C AI =(N WT C IWT +N PV C IPV+N BAT C IBAT +N TRANS C ITRANS CRF;

[0102]

[0103] Where: N WT N PV N BAT N TRANS These are the quantities of wind turbines, photovoltaic panels, batteries, and converters, respectively; C IWT C IPV C IBAT C ITRANS These are the unit costs of the wind turbine, photovoltaic panels, battery packs, and converter, respectively; CRF is the conversion factor for converting the initial investment cost into the annual investment cost, n is the number of years in the entire microgrid life cycle, and j is the annual interest rate.

[0104] Annual maintenance costs include:

[0105] C AM =(Cap WT C MWT +Cap PV C MPV +Cap BAT C MBAT +Cap TRANS C MTRANS );

[0106] In the formula: Cap WT Cap PV Cap BAT Cap TRANS These are the capacities of the wind turbine, solar panels, batteries, and converters, respectively; C MWT C MPV C MBAT C MTRANS These are the unit annual maintenance costs for wind turbines, photovoltaic panels, battery packs, and converters, respectively.

[0107] S213. The annual operating cost is obtained based on the electricity purchase cost, electricity sales cost, and converter losses.

[0108] Annual operating costs include:

[0109]

[0110]

[0111]

[0112] In the formula: E(·) represents the expectation; The cost of running the microgrid in scenario i is 24 hours per scenario i. There are K scenarios in a year. The scenario is a fixed parameter input by the planner. The number and differences of scenarios do not affect the planning model established. These represent the energy purchased and sold by the microgrid to the main grid within time t in scenario i; C bprice C sprice It is the purchase / sale electricity price; σ is the converter efficiency; P trans C represents the converter power of the converter. loss Cost of converter losses; It is a binary variable with values ​​of 0 and 1; used to prevent microgrids from purchasing and supplying electricity to the main grid. and To regulate the purchase and sale of electricity.

[0113] S214. Based on the annual investment cost, annual maintenance cost, and annual operating cost, the microgrid minimum cost objective function is obtained, aiming to minimize the annual investment cost, annual maintenance cost, and annual operating cost. Specifically:

[0114] The minimum cost objective function for microgrids includes:

[0115]

[0116] In the formula: This is the total cost of a microgrid over m years, C AI Investment cost for that year, C AM Annual maintenance costs, C AO Operating costs for the year.

[0117] S22. Construct microgrid operation constraints: Among them, the operation constraints of a single microgrid include AC sub-microgrid supply and demand balance constraints, DC sub-microgrid supply and demand balance constraints, and battery constraints.

[0118] Exchange of supply and demand balance constraints:

[0119] The energy supply side of the AC sub-microgrid includes the electrical power transmitted from the grid, wind turbine output, photovoltaic output, AC load, and the net power transmitted from the AC sub-microgrid to the DC sub-microgrid, which is considered as the total load of the AC sub-microgrid. Its supply and demand balance constraints are as follows:

[0120] P W,t +P buy,t -P sell,t -β AC-DC,t P AC-DC,t +σβ DC-AC,t P DC-AC,t =P AC,t ;

[0121] β AC-DC,t +βDC-AC,t ≤1;

[0122] In the formula: P W,t P is the output power of the fan. buy,t and P sell,t These represent the power purchased from and sold to the upstream of the AC sub-microgrid, respectively; σ is the conversion efficiency of the rectifier; P AC-DC,t P represents the electrical power flowing from the AC subnetwork to the DC subnetwork. DC-AC,t P represents the electrical power flowing from the DC subgrid to the AC subgrid. AC,t For AC load; β AC-DC,t / β DC-AC,t It is a binary variable with values ​​of 0 and 1, used to prevent the AC subnet and DC subnet from transmitting power to the other side at the same time.

[0123] DC sub-microgrid supply and demand balance constraints:

[0124] The energy sources for the DC sub-microgrid include battery banks and photovoltaic power. Its total load includes the net power transmitted to the AC sub-grid and the DC load it is responsible for. The supply and demand balance constraint expression is as follows:

[0125] P PV,t +μ d,t P d,t -μ c,t P c,t +σβ AC-DC,t P AC-DC,t -β DC-AC,t P DC-AC,t =P AC,t ;

[0126] μ d,t +μ c,t ≤1;

[0127] In the formula: P d,t and P c,t For the discharge and charging power of the battery pack; μ d,t and μ c,t It is a binary variable with values ​​of 0 and 1, used to prevent the battery from charging and discharging simultaneously.

[0128] Battery constraints include:

[0129] E min ≤E ES,t ≤E max ;

[0130]

[0131] E min =SOC min Cap BAT E max=SOC max Cap BAT ;

[0132] In the formula: E ES,t It is the battery capacity, E min and E max These are the upper and lower limits of the battery capacity; η c and η d It refers to the battery's charging and discharging efficiency; SOC (State of Charge) min and SOC max These are the upper and lower limits of the battery charge ratio.

[0133] S3. Construct the node partitioning objective function and node partitioning constraints as the upper-level partitioning model, and perform node partitioning on the AC / DC hybrid microgrid model based on the upper-level partitioning model to obtain the first node partitioning result; the goal of this step is to partition nodes with geographically proximate locations and clustering effects into microgrid groups in the same region through node coupling degree and active power balance degree. This step is divided into two sub-steps: constructing the node partitioning objective function and node partitioning constraints, specifically:

[0134] S31. Construct the node partitioning objective function:

[0135] The objective function of the upper-level planning model is to maximize a comprehensive index composed of node coupling degree and power balance degree, as expressed below:

[0136] maxΩ=λ1ω L +λ2ω P ;

[0137] In the formula: λ1 and λ2 are weighting coefficients; ω L ω represents the node coupling degree. P For active power balance;

[0138] The node coupling degree depends on factors such as the geographical distance between nodes and the types of loads connected to the nodes, as shown in the following expression:

[0139]

[0140] In the formula: L ij Let μ be the electrical distance between node i and node j. μ is a 0-1 variable used to describe the source load type correlation of the nodes. If the source load types connected to node i and node j are related, either both are DC source load resources or both are AC source load resources, then μ is 1; otherwise, it is 0.

[0141] If a power distribution network system contains n nodes, then the electrical distance between nodes i and j can be defined as:

[0142]

[0143] d ij =lg(S) VQ,jj / S VQ,ij );

[0144] ΔV jj =S VQ，jj ΔQ jj ;

[0145] ΔV ij =S VQ，ij ΔQ ij ;

[0146] In the formula: d ij The degree of closeness between node i and node j; d in The degree of closeness between node i and node n; d jn The degree of closeness between node j and node n; ΔV jj and ΔV ij ΔQ represents the change in voltage amplitude at distribution network nodes. jj and ΔQ ij S represents the change in reactive power at the node. VQ,jj The distribution network node sensitivity is obtained by calculating the ratio of the voltage change at node j to the reactive power change at node j; S VQ,ij The distribution network node sensitivity is obtained by calculating the ratio of the voltage change at node i to the reactive power change at node j. The ratio of reactive power change to voltage change reflects the node coupling degree. The larger the ratio, the smaller the coupling degree between the two nodes; conversely, the larger the coupling degree, the smaller the electrical distance.

[0147] Active power balance includes:

[0148] The active power balance index is calculated as the ratio of the overall power demand of the microgrid cluster to the maximum power demand during the entire planning period. The specific expression is as follows:

[0149]

[0150]

[0151] In the formula: Nclu is the number of microgrid groups; Pclu-i is the active power balance of microgrid group i; P(t) i Let be the power demand of microgrid i at time t; T is the entire planning period.

[0152] S32. Construct node partitioning constraints:

[0153] When dividing a distribution network area into node clusters, the voltage fluctuation of each node within the cluster must not exceed the upper or lower limits, as constrained as follows:

[0154]

[0155] In the formula: V AC.i For the i-th type of AC voltage, V represents the maximum and minimum values ​​of the AC voltage of type i, respectively. DC.j For the j-th type of DC voltage, These are the maximum and minimum values ​​of the DC voltage of type j, respectively.

[0156] S4. Construct a microgrid cluster islanding probability calculation model based on the Monte Carlo method. The purpose is to verify the reliability of the power supply to the microgrid cluster formed after node partitioning in step S3. To simulate the microgrid cluster's operating scenarios, this step uses the Monte Carlo method to construct the microgrid cluster's operating states and calculate the islanding probability. The normal and fault outage states of each distribution network device are simulated using the sequential Monte Carlo simulation method. The probabilities of each state are as follows:

[0157]

[0158] In the formula: P f and P o Let λ represent the probabilities of the fault state and the operating state of device i, respectively; i and μ i These are the component failure rate and repair rate, respectively.

[0159] The equipment is sampled, and the uptime (TTF) and fault recovery time (TTR) are obtained through sampling based on the two-state model of the equipment.

[0160]

[0161]

[0162] In the formula: U i and V i These are random numbers that follow a uniform distribution in (0,1).

[0163] The probability of a microgrid becoming islanded is calculated based on the equipment failure and normal operation times obtained from Monte Carlo sampling simulations, using the following formula:

[0164]

[0165] In the formula: 8760 is the equivalent total duration in one year, i.e., 365 × 24; N is the total number of users in the system load; U l denoted as the number of users experiencing power outages during the l-th model run; T represents the duration of the fault.

[0166] S5. The minimum cost objective function of the microgrid and the microgrid operation constraints are used as the lower-level planning model, and the islanding probability calculation model is used as the constraint condition for solving the lower-level planning model to obtain the power demand of each microgrid. Its objective function needs to pursue economy, but it needs to be constrained by the islanding probability threshold of the microgrid group. The islanding probability threshold constraint is used as the planning constraint to achieve a balance between the economy and reliability of microgrid group planning. Based on the planning results, the power demand of each microgrid group is calculated and passed to the upper-level model. Specifically:

[0167] Based on the microgrid minimum cost objective function established in step S2 and the microgrid operation constraints, the lower-level planning model is obtained, and an islanding probability threshold constraint is introduced. The planned microgrid group needs to satisfy the islanding probability threshold constraint as follows:

[0168] ASAI≥ASAI lower ;

[0169] In the formula: ASAI lower This is the lower bound for the probability of becoming an isolated island, set based on experience.

[0170] S6. The upper-layer partitioning model re-partitions nodes based on the power demand of each microgrid, obtaining a second node partitioning result. It then determines whether the second node partitioning result is consistent with the first node partitioning result. If not, it returns to step S3; if so, it obtains the investment cost of the second microgrid group based on the power demand of each microgrid. This step verifies the upper-layer model. After verifying the correctness of the partitioning, the upper-layer model calculates the node voltage based on the power demand transmitted from the lower layer and re-partitions the nodes. Specifically:

[0171] The upper-level partitioning model uses mathematical tools such as MATPOWER to perform power flow verification and calculate the voltage of each node based on the grid structure. It then re-partitions the nodes according to the method in step S3. If the result is consistent with the previous partitioning, step S7 is executed; otherwise, it returns to S3. This step realizes the interactive coupling between the upper-level partitioning model (node ​​partitioning model) and the lower-level partitioning model (microgrid group planning model). The upper-level partitioning model is the basis of the lower-level partitioning model, but the power demand obtained from the microgrid group planning result will be used as the boundary condition for the upper-level node partitioning. Therefore, after the lower-level model finishes solving, it passes the calculated active power demand to the upper level. Based on the power demand passed from the lower level, the upper-level model performs power flow calculation based on the distribution network structure to obtain the voltage of each node and re-partitions the nodes.

[0172] S7. Determine if the investment cost of the first microgrid group is greater than the investment cost of the second microgrid group. If yes, the microgrid group is considered to have effective gains and the solution ends; otherwise, return to step S5 and use the investment cost of the first microgrid group as an upper limit constraint. That is, this step proposes a standard for calculating the effectiveness of microgrid group enhancement, which is used to analyze and compare the configuration costs of each microgrid in the scenario in step S2 without considering microgrid group division with the microgrid group configuration costs in step S6. After effectiveness determination, if the microgrid group has effective gains, the planning is considered to have obtained the optimal solution. The specific effectiveness calculation standard is as follows:

[0173]

[0174] In the formula: f MG-clu(s) The planned investment cost of microgrid cluster s is the investment cost of the first microgrid cluster. The investment cost of microgrid m, which belongs to microgrid cluster s, is the sum of the investment costs of the second microgrid cluster. This formula shows that microgrids can reduce the overall planning cost after forming a microgrid cluster.

[0175] Example 2

[0176] Please refer to Figure 3 A microgrid group planning device for improving the reliability of low-voltage power distribution systems includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the various steps in the microgrid group planning method for improving the reliability of low-voltage power distribution systems as described in Embodiment 1.

[0177] In summary, this invention provides a microgrid group planning method and apparatus for improving the reliability of low-voltage power distribution systems. Its core idea is to establish a two-layer optimization planning model for microgrid groups to enhance the reliability of low-voltage power distribution systems: the upper layer constructs a distribution network node partitioning model, and the lower layer contains a microgrid group optimization configuration model. The two layers interact through the power demand of the microgrid group. Specifically, the upper distribution network layer considers node coupling and active power balance, establishing a node partitioning model; the lower microgrid group layer, in order to improve the energy self-consumption rate within the microgrid and maximize microgrid autonomy, constructs a microgrid islanding probability and a microgrid group gain effectiveness criterion based on the Monte Carlo method as planning constraints, and considers the AC / DC characteristics of the source load during the optimization planning process; then, an effective gain judgment criterion for microgrid group planning is established, and the effectiveness of the microgrid group planning scheme is judged after the upper and lower layers have converged iteratively. The microgrid cluster partitioning model established by the two-layer planning method of this invention can achieve a high degree of autonomy for microgrid clusters, thereby effectively improving the reliability of low-voltage power distribution systems. Compared with existing AC microgrids, it can effectively reduce investment costs and provide a valid reference for the planning and design of new power distribution networks with multiple microgrids. Furthermore, traditional AC microgrids require significant investment in converters to meet load growth demands, while the AC / DC hybrid planning model proposed in this patent improves the economic efficiency of microgrid planning while also meeting load growth requirements.

[0178] The above description is merely an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent modifications made based on the content of the present invention's specification and drawings, or direct or indirect applications in related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A microgrid group planning method for improving the reliability of low-voltage power distribution systems, characterized in that, include: S1. Establish an AC / DC hybrid microgrid model based on the AC / DC characteristics of the source and load; S2. Based on the AC / DC hybrid microgrid model, construct the microgrid minimum cost objective function and microgrid operation constraints, and solve for the investment cost of the first microgrid group; S3. Construct a node partitioning objective function and node partitioning constraints as an upper-level partitioning model, and perform node partitioning on the AC / DC hybrid microgrid model according to the upper-level partitioning model to obtain the first node partitioning result; S4. Construct a microgrid cluster islanding probability calculation model based on the Monte Carlo method; S5. The minimum cost objective function of the microgrid and the microgrid operation constraints are used as the lower-level planning model, and the islanding probability calculation model is used as the constraint condition of the lower-level planning model to solve the problem and obtain the power demand of each microgrid. S6. The upper-layer partitioning model re-partitions nodes according to the power requirements of each microgrid to obtain a second node partitioning result, and determines whether the second node partitioning result is consistent with the first node partitioning result. If not, it returns to step S3. If so, the investment cost of the second microgrid group is obtained based on the power requirements of each microgrid. S7. Determine whether the investment cost of the first microgrid group is greater than the investment cost of the second microgrid group. If yes, the microgrid group is considered to have effective gains and the solution ends. If no, return to step S5 and use the investment cost of the first microgrid group as the upper limit constraint. The model for calculating the probability of microgrid clusters becoming isolated is constructed based on the Monte Carlo method; The purpose is to verify whether the power supply of the microgrid group formed after node partitioning in analysis step S3 is reliable. In order to simulate the operation scenario of the microgrid group, this step uses the Monte Carlo method to construct the operation state of the microgrid group and calculate the probability of the microgrid group becoming islanded. The normal and fault outage states of each equipment in the distribution network are simulated using the sequential Monte Carlo simulation method, and the probabilities of each state are as follows: ； In the formula: and These represent the probabilities of the fault state and the operating state of device i, respectively. and These are the component failure rate and repair rate, respectively. The equipment is sampled, and the uptime (TTF) and fault recovery time (TTR) are obtained through sampling based on the two-state model of the equipment. ； ； In the formula: and These are random numbers that follow a uniform distribution in (0,1). The probability of a microgrid becoming islanded is calculated based on the equipment failure and normal operation times obtained from Monte Carlo sampling simulations, using the following formula: ； In the formula: 8760 is the equivalent total duration of one year, i.e., 365 × 24; N is the total number of users in the system load; denoted as the number of users experiencing power outages during the l-th model run; T represents the duration of the fault.

2. The microgrid group planning method for improving the reliability of low-voltage power distribution systems according to claim 1, characterized in that, The objective function for constructing the minimum cost of the microgrid based on the AC / DC hybrid microgrid model includes: Acquire microgrid equipment data and operational data. The equipment data includes the number of devices, equipment capacity, equipment investment cost, and equipment maintenance cost. The operational data includes electricity purchase cost, electricity sales cost, and converter loss. The annual investment cost and annual maintenance cost are obtained based on the number of equipment, equipment capacity, equipment investment cost, and equipment maintenance cost. The annual operating cost is obtained based on the electricity purchase cost, electricity sales cost, and converter losses. The minimum cost objective function of the microgrid is obtained based on the annual investment cost, annual maintenance cost, and annual operating cost.

3. A microgrid group planning method for improving the reliability of low-voltage power distribution systems according to claim 1 or 2, characterized in that, The minimum cost objective function of the microgrid includes: ； In the formula: This is the total cost of a microgrid over m years. Investment costs for that year, Maintenance costs for that year, Operating costs for the year.

4. The microgrid group planning method for improving the reliability of low-voltage power distribution systems according to claim 3, characterized in that, The annual investment cost includes: ； ； In the formula: , , , These are the quantities of wind turbines, photovoltaic panels, batteries, and converters, respectively. , , , These are the unit costs of the wind turbine, solar panels, battery packs, and converter, respectively; CRF is the conversion factor for converting the initial investment cost into the annual investment cost. n It is the number of years in the entire microgrid lifespan, and j is the annual interest rate; The annual maintenance cost includes: ； In the formula: , , , These refer to the capacity of the wind turbine, photovoltaic panel, battery, and converter, respectively. , , , These are the unit annual maintenance costs for wind turbines, photovoltaic panels, battery packs, and converters, respectively. The annual operating costs include: ； ； ； In the formula: E(·) represents the expectation; The cost of running a microgrid in scenario i is given. Each scenario i lasts for 24 hours, and there are K scenarios in a year. , These represent the energy that the microgrid purchases and sells to the main grid within time t in scenario i; , It is the purchase / sale price of electricity; The converter efficiency; The converter power; Cost of converter losses; , Let be a binary variable with values ​​of 0 and 1.

5. A microgrid group planning method for improving the reliability of low-voltage power distribution systems according to claim 1, characterized in that, The microgrid operation constraints include AC sub-microgrid supply and demand balance constraints, DC sub-microgrid supply and demand balance constraints, and battery constraints. The supply and demand balance constraints of the AC sub-microgrid include: ； ； In the formula: This refers to the output power of the fan. and These refer to the power purchased by the AC sub-microgrid from its superior and the power sold to its superior. The conversion efficiency of the rectifier; The electrical power flowing from the AC subnetwork to the DC subnetwork; This represents the electrical power flowing from the DC subnetwork to the AC subnetwork. For AC load; / It is a binary variable that takes values ​​of 0 and 1; The supply and demand balance constraints of the DC sub-microgrid include: ； ； In the formula: and For the discharge power and charging power of the battery pack; and It is a binary variable that takes values ​​of 0 and 1; The battery constraints include: ； ； ； In the formula: It's the battery charge. and These are the upper and lower limits of the battery capacity; and It refers to the charging and discharging efficiency of the battery; and These are the upper and lower limits of the battery charge ratio.

6. The microgrid group planning method for improving the reliability of low-voltage power distribution systems according to claim 1, characterized in that, The node partitioning objective function includes: ； In the formula: These are the weighting coefficients; The degree of node coupling; This refers to the active power balance.

7. A microgrid group planning method for improving the reliability of low-voltage power distribution systems according to claim 6, characterized in that, The node coupling degree includes: ； ； ； ； ； In the formula: Let i be the electrical distance between node i and node j. 0-1 variables; The degree of closeness between node i and node j; The degree of closeness between node i and node n; The degree of closeness between node j and node n; as well as This indicates the change in voltage amplitude at distribution network nodes; as well as This represents the change in reactive power at the node. The distribution network node sensitivity is obtained by calculating the ratio of the voltage change at node j to the reactive power change at node j. The distribution network node sensitivity is obtained by calculating the ratio of the voltage change at node i to the reactive power change at node j. The active power balance includes: ； ； In the formula: The number of micro-network groups; The active power balance of microgrid group i; Let be the power demand of microgrid i at time t; T is the entire planning period.

8. A microgrid group planning method for improving the reliability of low-voltage power distribution systems according to claim 1, characterized in that, The node partitioning constraints include: ； In the formula: For the i-th type of AC voltage, , These are the maximum and minimum values ​​of the i-th type of AC voltage, respectively; For the j-th type of DC voltage, , These are the maximum and minimum values ​​of the DC voltage of type j, respectively.

9. A microgrid group planning device for improving the reliability of low-voltage power distribution systems, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements each step of the microgrid group planning method for improving the reliability of low-voltage power distribution systems as described in any one of claims 1-8.

Images