Power distribution network double-layer extension planning method considering carbon transaction and demand side management

Abstract

The invention discloses a power distribution network double-layer expansion planning method considering carbon transaction and demand side management. The method comprises the following steps: acquiring data information of a power distribution network to be expanded; constructing a typical day scene based on a K-means clustering algorithm; the dynamic carbon emission intensity based on the carbon emission flow theory is adopted to calculate the carbon emission of the power distribution network; according to the time-of-use electricity price and the dynamic carbon emission intensity data of the power distribution network, the demand side electrical load is managed, and carbon transaction is carried out; establishing a power distribution network double-layer expansion planning model considering carbon transaction and demand side management; and upper and lower layer association of the power distribution network double-layer expansion planning model is converted into a single-layer model, linearization processing is carried out, and a solver is called for solving. According to the power distribution network double-layer extension planning method considering the carbon transaction and the demand side management, the carbon emission level of the power distribution network is effectively reduced, the system economy is improved, and a planning scheme which adapts to source load fluctuation and improves the operation stability of the power distribution network can be obtained.

Description

Technical Field

[0001] The present invention relates to the technical field of low-carbon distribution network expansion planning, and in particular to a distribution network double-layer expansion planning method considering carbon trading and demand-side management. Background Art

[0002] With the rapid development of distributed renewable energy in the distribution network, the uncertainty of renewable energy such as wind and solar has brought great impact on the planning of distribution network. Demand-side management in the distribution network can smooth the peak and fill the valley, promote the consumption of new energy, and reduce the adverse effects of distributed power sources. However, existing research on demand response is mostly carried out from the perspective of television to improve economic efficiency, and fails to consider the carbon emission reduction effect of demand-side management. Combining carbon trading with demand-side management can manage the load side from a carbon perspective and improve the low-carbon benefits of the distribution network. Therefore, it is of great significance to consider carbon trading and demand-side management in the process of distribution network expansion planning, and to formulate planning schemes and operation strategies that adapt to carbon emission reduction and new energy fluctuations.

[0003] In the prior art, the method of “considering carbon trading and demand-side management in the distribution network expansion planning process” is as follows:

[0004] 1) Chinese patent "A distribution network expansion planning method with coordinated multiple demand-side resources" (application number: 202010439620.0). The model designed by this invention coordinates and optimizes the expansion of existing power supply equipment in the distribution network and demand-side resources to obtain a distribution network planning scheme that minimizes the net cost of the distribution network. However, although this method takes into account the impact of demand-side resources on the distribution network during the planning process, the obtained planning scheme only considers economic efficiency and ignores the emission reduction effect of demand-side resources.

[0005] 2) Chinese patent "A distribution network planning method considering green certificate trading and carbon trading" (application number: 202211137124.5). Based on the three-stage green certificate trading model and the two-stage carbon trading model, this invention patent introduces the average unit power supply CO2 emission intensity of the distribution network, and obtains the overall new energy installed capacity proportion of the new distribution network, unit carbon emission intensity, and total revenue of the distribution network. However, the average unit power supply CO2 emission intensity of the distribution network cannot reflect the differences in carbon emissions generated by electricity consumption at different times, which will make it difficult to absorb new energy, and is not conducive to the scheduling of various flexible resources in the system, and cannot further reduce the carbon emissions of the distribution network.

[0006] To sum up, it is necessary to comprehensively consider the impact of carbon trading and demand-side management in the process of distribution network expansion planning, further improve the distribution network's ability to accept distributed power sources, and deeply explore the distribution network's potential for carbon reduction and emission reduction. Summary of the invention

[0007] In order to solve the above technical problems, the present invention provides a two-layer expansion planning method for distribution networks taking into account carbon trading and demand-side management, which can effectively reduce the carbon emission level of the distribution network, improve the system economy, and obtain a planning scheme that adapts to source and load fluctuations and improves the operating stability of the distribution network.

[0008] The technical solution adopted by the present invention is:

[0009] A two-layer expansion planning method for a distribution network considering carbon trading and demand side management includes the following steps:

[0010] Step 1: Obtain data information of the distribution network to be expanded;

[0011] Step 2: Construct a typical day scenario based on the K-means clustering algorithm;

[0012] Step 3: Calculate the carbon emissions of the distribution network using dynamic carbon emission intensity based on carbon emission flow theory;

[0013] Step 4: Manage the electricity load on the demand side and conduct carbon trading based on the time-of-use electricity price of the distribution network and dynamic carbon emission intensity data;

[0014] Step 5: Establish a two-layer expansion planning model for the distribution network considering carbon trading and demand-side management;

[0015] Step 6: Convert the upper and lower layers of the two-layer expansion planning model of the distribution network into a single-layer model and linearize it, and call the solver to solve it.

[0016] In the step 1, obtaining data information of the distribution network to be expanded specifically includes:

[0017] Obtain the topological structure of the distribution network to be expanded and equipment investment and operation data;

[0018] Obtain historical data on wind and solar loads of the distribution network to be expanded, and carbon emission data of the region;

[0019] Obtain time-of-use electricity price data, carbon trading data and demand-side management load data for the distribution network to be expanded.

[0020] The data information of the distribution network to be expanded is shown in Table 1:

[0021] Table 1 Obtaining distribution network data information

[0022]

[0023] The step 2 comprises the following steps:

[0024] S2.1: Normalize the historical data of wind, solar and load, divide each historical data by the maximum value in the data, set the number of clusters k, and randomly select the initial k cluster centers;

[0025] S2.2: Calculate the distance between the data and the initial cluster center:

[0026]

[0027] In formula (1), d(X,Y) is the Euclidean distance between n-dimensional samples X and Y; i is the parameter in the sample; y ki is the kth cluster center; n is the total number of samples in the historical data;

[0028] S2.3: According to the sample division, enter the cluster where the nearest cluster center is located. At this time, there are k clusters. Recalculate the center of each cluster. The calculation formula is:

[0029]

[0030] In formula (2), x ki is the sample in the kth cluster; m is the number of samples in the kth cluster;

[0031] S2.4: Repeat steps S2.2-S2.3 until the cluster center no longer changes. At this time, the cluster center is a scene, and the probability of the scene is the ratio of the number of samples m in the cluster to the total number of samples.

[0032] In step 3, the dynamic carbon emission intensity based on the carbon emission flow theory is calculated, which specifically includes the following steps:

[0033] S3.1: Calculate the carbon potential in the upper power grid based on the carbon emission flow theory:

[0034]

[0035] In formula (3), e Ni,t represents the carbon potential of node i at time t; s represents the line number; B Sup is the set of lines in the system; P Bs,t represents the transmission power in line s at time t; R Bs,t represents the carbon flow density in line s at time t; P Gi,t is the power output of generator i at time t; e Gi,t is the carbon emission intensity of the corresponding generator.

[0036] S3.2: The node carbon potential obtained based on the carbon emission flow theory is spatially averaged to obtain the average carbon emission intensity of the area where the distribution network to be expanded is located:

[0037]

[0038] In formula (4), e m,tis the regional dynamic emission factor of the regional power grid at time t; P Li,t is the load of node q at time t; Z is the set of nodes in the regional power grid.

[0039] The step 4 comprises the following steps:

[0040] S4.1: Use dynamic carbon emission intensity to measure the carbon emissions of the distribution network; and calculate the carbon emission cost based on the system carbon quota:

[0041]

[0042] In formula (5), E em is the carbon emission of the distribution network; T is the number of time periods in a single operation cycle; P grid,t The amount of electricity purchased by the distribution network from the upper grid; C em is the carbon trading cost; C co2 is the unit carbon trading price; η is the free carbon emission quota per unit electricity. S4.2: Demand-side management requires load management based on time-of-use electricity price data and dynamic carbon emission intensity data, and transfers or reduces loads when electricity prices and carbon emission intensity are high to reduce operating costs. The calculation formula for the load that can be reduced and the load that can be transferred is as follows:

[0043]

[0044] In formula (6), represents the load reduction of node i at time t; In step 5, in the established two-layer expansion planning model for the distribution network, the upper planning layer objective function is to minimize the annual comprehensive cost, and the lower operation layer objective function is to minimize the operating cost, which specifically includes:

[0045] minC total =C INV +C OPE (7);

[0046] In the formula, C total is the annual comprehensive cost of the distribution network; C INV is the annual investment cost of the distribution network; C OPE The operating cost of the distribution network.

[0047] C INV =C line +C WTG +C PVG +C ESS

[0048] C OPE =C buy +C loss +C DG+C cut +C em +C com (8);

[0049] In the above formula, C line is the line investment cost; C WTG is the investment cost of the fan; C PVG is the photovoltaic investment cost; C ESS Overdraft cost for energy storage; C buy The cost of purchasing electricity for the main grid; C loss is the network loss cost; C DG is the operating cost of distributed power generation; C cut C is the cost of curtailing wind and solar power; em is the carbon trading cost; C com Compensate for demand-side management costs.

[0050] The above model meets the constraints of equipment investment, distribution network radial constraints, distribution network flow constraints, safety constraints, distributed power generation operation constraints, energy storage operation constraints, and demand side management constraints. Specifically, it includes:

[0051] 1) Equipment investment constraints:

[0052]

[0053] In formula (9), are the decision variables for line, wind power, photovoltaic and energy storage investment respectively; N wt 、N pv 、N ess is the upper limit of the number of wind power, photovoltaic and energy storage equipment installed at node j; Ω line is the set of newly created lines; Ω PVG is the set of photovoltaic nodes to be installed; Ω WTG is the set of wind turbine nodes to be installed; Ω ESS It is a collection of energy storage nodes waiting for dark transition.

[0054] 2) Network radial and connectivity constraints:

[0055] n=m+1(10);

[0056] In formula (10), n and m are the total number of nodes and lines in the planned distribution network, respectively.

[0057] 3) Power flow constraints:

[0058]

[0059] In formula (11), δ j represents the set of lines with j as the starting point, and k is the end point of the corresponding line; φ jrepresents the set of lines with j as the end, and i is the beginning of the corresponding line; P ij,t is the active transmission power of branch ij; I ij,t is the current of branch ij; r ij is the resistance of branch ij; is the active power output of the wind turbine at node j; is the active power output of the photovoltaic power station at node j; Make contributions to the checkpoint; is the active load at node j; Q ij,t is the reactive transmission power of branch ij; x ij is the reactance of branch ij; is the reactive power output of wind turbine at node j; is the reactive power output of the photovoltaic power station at node j; Make a contribution to the checkpoint; is the reactive load at node j; V i,t 、V j,t Respectively represent the voltage of nodes i and j; Ω l is the collection of lines to be constructed; ij is the number in the collection of lines.

[0060] 4) Safety constraints:

[0061]

[0062] In formula (12), V min 、V max are the upper and lower limits of the node voltage respectively; is the upper limit of the current on line ij.

[0063] 5) Distributed power operation constraints:

[0064]

[0065] In formula (13), are the maximum active output of wind power and photovoltaic power at node j respectively; is the minimum power factor of the distributed power source; It is the reactive power output of distributed power source. They are the abandoned electricity of wind power and photovoltaic power respectively.

[0066] 6) Energy storage operation constraints:

[0067]

[0068] In formula (14), They are 0-1 state variables for energy storage charging and discharging respectively; The upper limit of the charging and discharging power of a single energy storage unit; is the state of charge of the energy storage; ηC , η D They are energy storage charging and discharging efficiency; the capacity allocated for energy storage; Charging power for energy storage; is the energy storage discharge power; is the upper limit of energy storage power of node j; is the upper limit of energy storage capacity of node j; It is the charge state of the energy storage at the start of an operation cycle; It is the charge state at the end of an energy storage operation cycle.

[0069] 7) Demand-side management constraints:

[0070]

[0071] In formula (15), λ cut is the maximum proportion of load that can be reduced in demand-side management; tran is the maximum proportion of load that can be transferred; is the original load power of node i at time t.

[0072] In step 6, the established two-layer expansion planning model for the distribution network separates planning and operation. The general solution is to iterate the upper and lower layers to obtain the optimal planning solution and operation strategy. The model can be quickly solved by using the upper and lower layer association method, that is, the upper and lower layer association of the model is converted into a single-layer model. In addition, there are nonlinear terms in the part of the model that considers the line and the flow, which are linearized here at the same time, specifically including the following steps:

[0073] 1) The model links the upper and lower layers, transforming variables between the planning and operation layers:

[0074]

[0075] In formula (16), S wt , S pv are the capacities of a single wind turbine and photovoltaic, respectively; They are the upper limits of charging and discharging power of a single energy storage unit respectively; The capacity configured for energy storage.

[0076] 2) Linearization processing:

[0077] The voltage drop and power transmission constraints in the power flow constraints are relaxed by the large M method, and the following calculation formula is obtained:

[0078]

[0079] -P max x line ≤P ij,t≤P max x line

[0080] -Q max x line ≤Q ij,t ≤Q max x line

[0081]

[0082] In formula (17), are the square equivalent values ​​of node i voltage and line ij current respectively, and the variables Instead of M is a sufficiently large positive number; P max , Q max I are the upper limits of active and reactive power of line transmission respectively; max is the upper limit of line current; x line is the line investment decision variable; Ω l A collection of line numbers.

[0083] Then, for the formula Performing second-order cone relaxation, we obtain Transpose both sides of the equation:

[0084]

[0085] Then it is further equivalent to:

[0086]

[0087] Its second-order cone constraint form is:

[0088]

[0089] The converted model is a single-layer mixed integer second-order cone programming model, which can be solved directly by calling the solver.

[0090] The present invention provides a distribution network double-layer expansion planning method considering carbon trading and demand side management, and the technical effects are as follows:

[0091] 1) The present invention adopts a dynamic carbon emission intensity based on carbon emission flow theory to measure the carbon emissions of the distribution network, which is more accurate. The intensity reflects the time difference of carbon emissions generated by electricity consumption at different times, which is conducive to the formulation of low-carbon operation strategies for distribution networks and promotes the consumption of new energy in the main grid;

[0092] 2) The present invention introduces carbon trading in the planning process, quantifies the benefits of carbon emission reduction, and increases the system's enthusiasm for carbon emission reduction. On this basis, demand-side management is considered, and load management considerations from a carbon perspective are added. The resulting planning scheme can not only achieve better economic efficiency, but also has good low-carbon properties;

[0093] 3) The double-layer expansion planning model of the distribution network adopted by the present invention processes planning and operation in layers, and can obtain the most reliable optimal planning scheme and operation strategy. The double-layer model is associated and linearized to reduce the difficulty of solving and improve the speed of solving. BRIEF DESCRIPTION OF THE DRAWINGS

[0094] The present invention will be further described below in conjunction with the accompanying drawings and embodiments:

[0095] Figure 1 Schematic diagram of the two-tier expansion planning method for distribution networks considering carbon trading and demand-side management.

[0096] Figure 2 This is the structural diagram of the two-level planning model for distribution networks considering carbon trading and demand-side management.

[0097] Figure 3 To improve the IEEE33 node system topology diagram.

[0098] Figure 4 This is a typical day scene of the generated scenery and water.

[0099] Figure 5(a) shows the time-of-use electricity price and dynamic carbon emission factor curve (wind power scenario);

[0100] Figure 5(b) shows the time-of-use electricity price and dynamic carbon emission factor curve (load power scenario);

[0101] Figure 5(c) shows the time-of-use electricity price and dynamic carbon emission factor curve (photovoltaic power scenario).

[0102] Figure 6 Load balancing diagram for demand side management.

[0103] Figure 7 This is the energy storage charging and discharging power diagram. DETAILED DESCRIPTION

[0104] like Figure 2 As shown in the figure, the two-layer expansion planning method of the distribution network considering carbon trading and demand-side management is to achieve economic and low-carbon planning of the distribution network, which specifically includes the following steps:

[0105] Step 1: Obtain information about the distribution network to be expanded:

[0106] Obtain the topological structure of the distribution network to be expanded and equipment investment and operation data;

[0107] Obtain historical data on wind and solar loads of the distribution network to be expanded, and carbon emission data of the region;

[0108] Obtain time-of-use electricity price data, carbon trading data and demand-side management load data for the distribution network to be expanded.

[0109] Step 2: Construct a typical day scenario based on the K-means clustering algorithm:

[0110] By clustering the annual time series data of load, wind power, and photovoltaic power, we can construct a typical daily scenario that can more accurately characterize the uncertainty change characteristics of wind and photovoltaic load power throughout the year, using the more common K-means clustering algorithm.

[0111] 1) Normalize the historical data of wind, solar and load, divide each historical data by the maximum value in the data, set the number of clusters k, and randomly select the initial k cluster centers;

[0112] 2) Calculate the distance between the data and the initial cluster center:

[0113]

[0114] In formula (1), d(X,Y) is the Euclidean distance between n-dimensional samples X and Y; i is the parameter in the sample; y ki is the kth cluster center; n is the total number of samples in the historical data;

[0115] 3) According to the sample division, enter the cluster where the nearest cluster center is located. At this time, there are k clusters. Recalculate the center of each cluster. The calculation formula is:

[0116]

[0117] In formula (2), x ki is the sample in the kth cluster; m is the number of samples in the kth cluster;

[0118] 4) Repeat steps S2.2-S2.3 until the cluster center no longer changes. At this time, the cluster center is the scene, and the probability of the scene is the ratio of the number of samples m in the cluster to the total number of samples.

[0119] Step 3: Calculate the dynamic carbon emission intensity based on the carbon emission flow theory:

[0120] Carbon emission intensity is often measured by average carbon emission intensity, which is calculated based on the annual power generation and annual electricity consumption in a certain region. It is difficult to reflect the differences in carbon emissions in the context of new energy. The dynamic carbon emission intensity based on carbon emission flow theory allocates the responsibility for carbon emissions according to the power system flow distribution, which can reflect the differences in carbon emissions from electricity consumption in different regions and at different times. The calculation process is as follows:

[0121] 1) Based on the carbon emission flow theory, the carbon potential in the upper power grid is calculated:

[0122]

[0123] In formula (3), e Ni,t represents the carbon potential of node i at time t; s represents the line number; B Sup is the set of lines in the system; P Bs,t represents the transmission power in line s at time t; R Bs,t represents the carbon flow density in line s at time t; P Gi,t is the power output of generator i at time t; e Gi,t is the carbon emission intensity of the corresponding generator.

[0124] 2) The node carbon potential obtained based on the carbon emission flow theory is spatially averaged to obtain the average carbon emission intensity of the area where the distribution network to be expanded is located:

[0125]

[0126] In formula (4), e m,t is the regional dynamic emission factor of the regional power grid at time t; P Li,t is the load of node q at time t; Z is the set of nodes in the regional power grid.

[0127] Step 4: Carry out carbon trading and demand-side management based on time-of-use electricity price data and dynamic carbon emission intensity, including:

[0128] Both carbon trading and demand-side management strategies need to consider costs. Carbon trading is based on system emissions. When emissions exceed the system's free carbon emission rating, carbon quotas need to be purchased from the market. Demand-side management requires users to be compensated when shifting and reducing loads, so it gives priority to times when electricity prices and carbon emissions are higher to reduce operating costs.

[0129] 1) Use dynamic carbon emission intensity to measure the carbon emissions of the distribution network and calculate the carbon emission cost based on the system carbon quota:

[0130]

[0131] In formula (5), E em is the carbon emission of the distribution network; T is the number of time periods in a single operation cycle; P grid,t The amount of electricity purchased by the distribution network from the upper grid; C em For carbon trading costs; is the unit carbon trading price; η is the free carbon emission quota per unit electricity.

[0132] 2) Demand-side management requires load management based on time-of-use electricity price data and dynamic carbon emission intensity data, and load transfer or reduction at times when electricity prices and carbon emission intensity are high to reduce operating costs. The calculation formula for load that can be reduced and load that can be transferred is as follows:

[0133]

[0134] In formula (6), represents the load reduction of node i at time t; represents the load transfer amount of node i at time t.

[0135] Step 5: Establish a two-level planning model for distribution network expansion considering carbon trading and demand-side management:

[0136] The objective function of the upper planning layer of the model is to minimize the annual comprehensive cost, and the objective function of the lower operation layer is to minimize the operating cost. The two-layer expansion planning structure of the distribution network is as follows: Figure 2 shown.

[0137] minC total =C INV +C OPE (7);

[0138] In formula (7), C total is the annual comprehensive cost of the distribution network; C INV is the annual investment cost of the distribution network; C OPE The operating cost of the distribution network.

[0139] C INV =C line +C WTG +C PVG +C ESS ;

[0140] In the above formula, C line is the line investment cost equivalent to one year and τ is the planned annual value coefficient of the equipment r is the discount rate, T is the fixed investment payback period of the equipment, c line is the investment cost of a single line, Ψ L is the set of lines to be selected, is the line investment decision variable; C WTG is the wind power investment cost and c wt is the wind power investment cost per unit capacity, is the wind power investment decision variable, S wt is the installed capacity of distributed wind power, Ω WTG is the set of distributed wind power nodes to be connected; C PVG is the photovoltaic investment cost and cpv is the photovoltaic investment cost per unit capacity, is the photovoltaic investment decision variable, S pv is the installed capacity of distributed photovoltaics, Ω PVG is the set of distributed photovoltaic nodes to be connected; C ESS The investment cost of energy storage and c ess is the energy storage investment cost per unit capacity, is the energy storage investment decision variable, S ess is the installation capacity of a single unit, Ω ESS A set of energy storage nodes to be connected;

[0141] C OPE =C buy +C loss +C DG +C cut +C em +C com ;

[0142] In the above formula, C buy The cost of purchasing electricity from the main grid and N is the number of days in a year, which is 365. S is the number of scenes constructed. s is the probability of the corresponding scene, Time-of-use electricity price, is the amount of electricity purchased at time t in scenario s; C loss is the network loss cost and f loss is the unit network loss cost, is the network power consumption at time t in scenario s; C DG The operating cost of distributed power generation and f DG is the unit operating cost of distributed power supply, The distributed power output at time t in scenario s;

[0143] C cut The cost of curtailing wind and solar power f CUT is the unit wind power curtailment cost, is the output of distributed power generation at time t in scenario s; C em For carbon trading costs;

[0144] C com To compensate for the costs of demand-side management and Reduce costs per unit load, is the load reduction amount at time t in scenario s, Compensate for load shifting costs, is the load transfer amount.

[0145] The model meets the constraints of equipment investment, distribution network radial constraints, distribution network flow constraints, safety constraints, distributed power generation operation constraints, energy storage operation constraints, and demand side management constraints. Specifically, it includes:

[0146] 1) Equipment investment constraints: The installation quantity constraints are considered during the construction process of each equipment.

[0147]

[0148] In formula (9), are the decision variables for line, wind power, photovoltaic and energy storage investment respectively; N wt 、N pv 、N ess is the upper limit of the number of devices installed on node j; Ω line is the set of newly created lines; Ω PVG is the set of photovoltaic nodes to be installed; Ω WTG is the set of wind turbine nodes to be installed; Ω ESS It is a collection of energy storage nodes waiting for dark transition.

[0149] 2) Network radial and connectivity constraints:

[0150] n=m+1(10);

[0151] In formula (10), n and m are the total number of nodes and lines in the planned distribution network, respectively.

[0152] 3) Power flow constraints:

[0153]

[0154] In formula (11), δ j represents the set of lines with j as the starting point, and k is the end point of the corresponding line; φ j represents the set of lines with j as the end, and i is the beginning of the corresponding line; P ij,t is the active transmission power of branch ij; I ij,t is the current of branch ij; r ij is the resistance of branch ij; is the active power output of the wind turbine at node j; is the active power output of the photovoltaic power station at node j; P t sp Make contributions to the checkpoint; is the active load at node j; Q ij,t is the reactive transmission power of branch ij; x ij is the reactance of branch ij; is the reactive power output of wind turbine at node j; is the reactive power output of the photovoltaic power station at node j; Make a contribution to the checkpoint; is the reactive load at node j; V i,t 、V j,t Respectively represent the voltage of nodes i and j; Ω l is the collection of lines to be constructed; ij is the number in the collection of lines.

[0155] 4) Safety constraints:

[0156]

[0157] In formula (12), V min 、V max are the upper and lower limits of the node voltage respectively; is the upper limit of the current on line ij.

[0158] 5) Distributed power operation constraints:

[0159] The output of distributed power generation is within the range of its installed capacity, and distributed power generation can generate reactive power. In addition, the abandonment of wind and solar power by distributed power generation must also be within a certain range.

[0160]

[0161] In formula (13), are the maximum active output of wind power or photovoltaic power at node j respectively; is the minimum power factor of the distributed power source; It is the reactive power output of distributed power source. It is the abandoned electricity of wind power and photovoltaic power.

[0162] 6) Energy storage operation constraints: Energy storage operation must not be charged or discharged at the same time and the capacity must be within the state of charge.

[0163]

[0164] In formula (14), They are 0-1 state variables for energy storage charging and discharging respectively; The upper limit of the charging and discharging power of a single energy storage unit; is the state of charge of the energy storage; η C , η D They are energy storage charging and discharging efficiency; The capacity configured for energy storage.

[0165] 7) Demand-side management constraints:

[0166]

[0167] In formula (15), λ cut is the maximum proportion of load that can be reduced in demand-side management;tran is the maximum proportion of load that can be transferred.

[0168] Step 6: Associate and linearize the upper and lower layers of the model:

[0169] The model established above separates planning and operation. The general solution is to iterate the upper and lower layers to obtain the optimal planning solution and operation strategy. The model can be quickly solved by using the upper and lower layer association method, that is, the upper and lower layer association of the model is converted into a single-layer model. In addition, there are nonlinear terms in the part of the model that considers the line and the flow, which are linearized here at the same time. The specific steps include the following:

[0170] 1) The model links the upper and lower layers, transforming variables between the planning and operation layers:

[0171]

[0172] In formula (16), S wt , S pv are the capacities of a single wind turbine and photovoltaic, respectively; The upper limit of the charging and discharging power of a single energy storage unit; The capacity configured for energy storage.

[0173] 2) Linearization processing:

[0174] The voltage drop and power transmission constraints in the power flow constraints are relaxed by the large M method, and the following formula is obtained:

[0175] -P max x line ≤P ij,t ≤P max x line

[0176] -Q max x line ≤Q ij,t ≤Q max x line

[0177]

[0178] In formula (17), are the square equivalent values ​​of node i voltage and line ij current respectively, and the variables Instead of M is a sufficiently large positive number; P max , Q max I are the upper limits of active and reactive power of line transmission respectively; max is the upper limit of line current;

[0179] Then, for the formula Performing second-order cone relaxation, we obtain Transpose both sides of the equation:

[0180]

[0181] Then it is further equivalent to:

[0182]

[0183] Its second-order cone constraint form is:

[0184]

[0185] The converted model is a single-layer mixed integer second-order cone programming model. The model is expressed using an optimization mathematical model in the MATLAB platform and solved by calling gurobi 11.0.0 in the yalmip extension package.

[0186] Verification example:

[0187] To verify the effectiveness of the method proposed in the present invention, the modified IEEE33 node system is taken as an example and verified using the matlab platform. The reference voltage of the system is 12.66 kV, and the system topology is as follows: Figure 3 As shown: the numbers represent the node numbers, the solid lines represent the constructed lines, and the dotted lines represent the lines to be newly constructed, with a total of 14 lines. The system model includes 46 lines and 43 nodes, of which the newly added load nodes are 34-43.

[0188] The unit cost of the line is 80,000 yuan / km, and the unit impedance is 0.6+j0.59Ω / km. The discount rate is 10%, and the planning period is 10 years. The investment cost of PVG per unit capacity is 7,500 yuan / kW, and the investment cost of WTG per unit capacity is 6,500 yuan / kW. The unit power generation cost of PVG and WTG is 0.25 yuan / kWh, and the capacity of a single PVG and WTG is 100kW. The main grid purchase electricity cost is 0.8 yuan / kWh, the line network loss cost is 0.6 yuan / kWh, and the penalty cost for abandoning wind and light is 0.6 yuan / kWh. The capacity of a single distributed power source is 200kWh, and the total installed capacity of wind power and photovoltaic power is capped at 3,000kW respectively. The nodes to be installed for PVG are 20 and 29, and the nodes to be installed for WTG are 8 and 31. The unit capacity energy storage investment cost is 1000 yuan / kWh, the single energy storage capacity is 100kWh, the power is 40kW, the upper limit of the ESS state of charge is 0.9, the lower limit is 0.2, the initial state of charge of each operation cycle is 0.5, and the ESS installation nodes are 17 and 24. The transferable load nodes are 8, 9, 16, 17, 23, 24, 25, 26, 30, 31, 32, 33, and the load reduction nodes are 5, 12, 20, 28, 35, 36, 40, 41. The carbon trading price is 300 yuan / t, and the carbon quota is 0.45kg / kWh. Time-of-use electricity prices and dynamic carbon emission intensity data are as follows: Figure 4 shown.

[0189] Based on the annual historical data of a distribution network in Northwest China, a typical daily scenario is constructed. Figure 5(a) to Figure 5(c) As shown in Table 2, the typical daily scenario construction results based on k-means clustering have good stratification characteristics, covering all levels of load and DG from large to small, effectively reflecting the randomness and fluctuation characteristics of source load. The distribution planning scheme is shown in Table 2. Due to the consideration of the impact of carbon trading, the configuration capacity of distributed generation is relatively high.

[0190] Table 2 Planning results

[0191]

[0192] The investment and operation are shown in Table 3. It can be seen from Table 3 that the operation cost has decreased after adding distributed power generation. If new energy is not considered, the electricity will all come from the main grid, which is expensive. In addition, when carbon trading is not considered, the annual carbon emissions of the distribution network are 12306.43t, and the carbon emissions under this scheme are 8994.87t, and the carbon emission reduction exceeds 3000t per year.

[0193] Table 3 Costs

[0194]

[0195]

[0196] Demand side management and energy storage operation in scenario 3 Figure 6 and Figure 7 As shown. Figure 6 The load that is transferred in the upper half of the coordinate axis is the load that is transferred out, which is concentrated in the time of high carbon emissions and high electricity prices, such as 9:00-13:00 and 19:00-22:00. The night time is the time of the highest carbon emission intensity, so a large amount of transfer is required. Compared with the transfer load, the load that can be reduced is generated at all times, because after the load is reduced, no carbon emissions will be generated, and the benefits are higher. Figure 7 The opposite of transferable load is that medium energy storage is used by transferring the electricity from other times to this time. Similarly, it is also dispatching electricity with more electricity prices and carbon emissions. The results show that flexible resources with dispatchable properties can improve economic and low-carbon benefits.

[0197] In summary, the present invention provides a two-layer expansion planning method for distribution networks taking into account carbon trading and demand-side management. It considers the uncertainty of distributed power sources and load power, the expansion of lines, the joint planning of distributed power sources and energy storage, and formulates the optimal operation strategy of the distribution network under carbon trading and demand-side management. The calculation results show that the obtained scheme has better economy and low carbon performance.

Claims

1. A two-layer expansion planning method for distribution network considering carbon trading and demand side management, characterized in that The following steps are involved: Step 1: Obtain data information of the distribution network to be expanded; Step 2: Construct a typical day scenario based on the K-means clustering algorithm; Step 3: Calculate the carbon emissions of the distribution network using dynamic carbon emission intensity based on carbon emission flow theory; Step 4: Manage the electricity load on the demand side and conduct carbon trading based on the time-of-use electricity price of the distribution network and dynamic carbon emission intensity data; Step 5: Establish a two-layer expansion planning model for the distribution network considering carbon trading and demand-side management; Step 6: Convert the upper and lower layers of the two-layer expansion planning model of the distribution network into a single-layer model and linearize it, and call the solver to solve it.

2. A distribution network double-layer expansion planning method considering carbon trading and demand side management according to claim 1, characterized in that: In the step 1, obtaining data information of the distribution network to be expanded specifically includes: Obtain the topological structure of the distribution network to be expanded and equipment investment and operation data; Obtain historical data on wind and solar loads of the distribution network to be expanded, and carbon emission data of the region; Obtain time-of-use electricity price data, carbon trading data and demand-side management load data for the distribution network to be expanded.

3. A distribution network double-layer expansion planning method considering carbon trading and demand side management according to claim 1, characterized in that: The step 2 comprises the following steps: S2.1: Normalize the historical data of wind, solar and load, divide each historical data by the maximum value in the data, set the number of clusters k, and randomly select the initial k cluster centers; S2.2: Calculate the distance between the data and the initial cluster center: In formula (1), d(X,Y) is the Euclidean distance between n-dimensional samples X and Y; i is the parameter in the sample; y ki is the kth cluster center; n is the total number of samples in the historical data; S2.3: According to the sample division, enter the cluster where the nearest cluster center is located. At this time, there are k clusters. Recalculate the center of each cluster. The calculation formula is: In formula (2), x ki is the sample in the kth cluster; m is the number of samples in the kth cluster; S2.4: Repeat steps S2.2-S2.3 until the cluster center no longer changes. At this time, the cluster center is a scene, and the probability of the scene is the ratio of the number of samples m in the cluster to the total number of samples.

4. A distribution network double-layer expansion planning method considering carbon trading and demand side management according to claim 1, characterized in that: In step 3, the dynamic carbon emission intensity based on the carbon emission flow theory is calculated, which specifically includes the following steps: S3.1: Calculate the carbon potential in the upper power grid based on the carbon emission flow theory: In formula (3), e Ni,t represents the carbon potential of node i at time t; s represents the line number; B Sup is the set of lines in the system; P Bs,t represents the transmission power in line s at time t; R Bs,t represents the carbon flow density in line s at time t; P Gi,t is the power output of generator i at time t; e Gi,t is the carbon emission intensity of the corresponding generator; S3.2: The node carbon potential obtained based on the carbon emission flow theory is spatially averaged to obtain the average carbon emission intensity of the area where the distribution network to be expanded is located: In formula (4), e m,t is the regional dynamic emission factor of the regional power grid at time t; P Li,t is the load of node q at time t; Z is the set of nodes in the regional power grid.

5. A distribution network double-layer expansion planning method considering carbon trading and demand side management according to claim 1, characterized in that: The step 4 comprises the following steps: S4.1: Use dynamic carbon emission intensity to measure the carbon emissions of the distribution network; and calculate the carbon emission cost based on the system carbon quota: In formula (5), E em is the carbon emission of the distribution network; T is the number of time periods in a single operation cycle; P grid,t The amount of electricity purchased by the distribution network from the upper grid; C em For carbon trading costs; is the unit carbon trading price; η is the free carbon emission quota per unit of electricity; S4.2: Demand-side management manages the load according to the time-of-use electricity price data and dynamic carbon emission intensity data, and transfers or reduces the load when the electricity price and carbon emission intensity are high to reduce the operating cost; the calculation formula for the load that can be reduced and the load that can be transferred is as follows: In formula (6), represents the load reduction of node i at time t; represents the load transfer amount of node i at time t.

6. A distribution network double-layer expansion planning method considering carbon trading and demand side management according to claim 1, characterized in that: In the step 5, in the established two-layer expansion planning model for the distribution network, the upper planning layer objective function is to minimize the annual comprehensive cost, and the lower operation layer objective function is to minimize the operating cost, which specifically includes: my C total =C INV +C OPE (7); In the formula, C total is the annual comprehensive cost of the distribution network; C INV is the annual investment cost of the distribution network; C OPE The operating cost of the distribution network; C INV =C line +C WTG +C PVG +C ESS C OPE =C buy +C loss +C DG +C cut +C em +C com (8); In the above formula, C line is the line investment cost; C WTG is the investment cost of the fan; C PVG is the photovoltaic investment cost; C ESS is the overdraft cost of energy storage; C buy The cost of purchasing electricity for the main grid; C loss is the network loss cost; C DG is the operating cost of distributed power generation; C cut C is the cost of curtailing wind and solar power; em is the carbon trading cost; C com Compensate for demand-side management costs.

7. A distribution network double-layer expansion planning method considering carbon trading and demand side management according to claim 6, characterized in that: The two-layer expansion planning model of the distribution network meets the constraints of equipment investment, distribution network radial constraints, distribution network flow constraints, safety constraints, distributed power generation operation constraints, energy storage operation constraints, and demand side management constraints; specifically, it includes: 1) Equipment investment constraints: In formula (9), are the decision variables for line, wind power, photovoltaic and energy storage investment respectively; N wt 、N pv 、N ess is the upper limit of the number of wind power, photovoltaic and energy storage equipment installed at node j; Ω line is the set of newly created lines; Ω PVG is the set of photovoltaic nodes to be installed; Ω WTG is the set of wind turbine nodes to be installed; Ω ESS It is a collection of energy storage nodes waiting for dark transition; 2) Network radial and connectivity constraints: n=m+1(10); In formula (10), n and m are the total number of nodes and lines in the planned distribution network, respectively; 3) Power flow constraints: In formula (11), δ j represents the set of lines with j as the starting point, and k is the end point of the corresponding line; φ j represents the set of lines with j as the end, and i is the beginning of the corresponding line; P ij,t is the active transmission power of branch ij; I ij,t is the current of branch ij; r ij is the resistance of branch ij; is the active power output of the wind turbine at node j; is the active power output of the photovoltaic power station at node j; P t sp Make contributions to the checkpoint; is the active load at node j; Q ij,t is the reactive transmission power of branch ij; x ij is the reactance of branch ij; is the reactive power output of wind turbine at node j; is the reactive power output of the photovoltaic power station at node j; Make a contribution to the checkpoint; is the reactive load at node j; V i,t 、V j,t Respectively represent the voltage of nodes i and j; Ω l is the collection of construction lines; ij is the number in the collection of lines; 4) Safety constraints: In formula (12), V min 、V max are the upper and lower limits of the node voltage respectively; is the upper limit of the current on line ij; 5) Distributed power operation constraints: In formula (13), are the maximum active output of wind power and photovoltaic power at node j respectively; is the minimum power factor of the distributed power source; Reactive power output of distributed power sources; They are the abandoned electricity of wind power and photovoltaic power respectively; 6) Energy storage operation constraints: In formula (14), They are 0-1 state variables for energy storage charging and discharging respectively; The upper limit of the charging and discharging power of a single energy storage unit; is the state of charge of the energy storage; η C , η D They are energy storage charging and discharging efficiency; the capacity allocated for energy storage; Charging power for energy storage; is the energy storage discharge power; is the upper limit of energy storage power of node j; is the upper limit of energy storage capacity of node j; It is the charge state of the energy storage at the start of an operation cycle; It is the charge state of energy storage at the end of an operation cycle; 7) Demand-side management constraints: In formula (15), λ cut is the maximum proportion of load that can be reduced in demand-side management; tran is the maximum proportion of load that can be transferred; is the original load power of node i at time t.

8. A distribution network double-layer expansion planning method considering carbon trading and demand side management according to claim 1, characterized in that: In step 6, the established distribution network double-layer expansion planning model processes planning and operation separately, and the model can be quickly solved by using the upper and lower layer association method, that is, the upper and lower layer association of the model is converted into a single-layer model, and linearization is performed at the same time, which specifically includes the following steps: 1) The model links the upper and lower layers, transforming variables between the planning and operation layers: In formula (16), S wt , S pv are the capacities of a single wind turbine and photovoltaic, respectively; They are the upper limits of charging and discharging power of a single energy storage unit respectively; the capacity allocated for energy storage; 2) Linearization processing: The voltage drop and power transmission constraints in the power flow constraints are relaxed by the large M method, and the following calculation formula is obtained: -P max x line ≤P ij,t ≤P max x line -Q max x line ≤Q ij,t ≤Q max x line In formula (17), are the square equivalent values ​​of node i voltage and line ij current respectively, and the variables Instead of M is a sufficiently large positive number; P max , Q max They are respectively the upper limits of active power and reactive power transmitted by the line; I max is the upper limit of line current; x line is the line investment decision variable; Ω l is a line number set; Then, for the formula Performing second-order cone relaxation, we obtain Transpose both sides of the equation: Then it is further equivalent to: Its second-order cone constraint form is: The converted model is a single-layer mixed integer second-order cone programming model, which can be solved directly by calling the solver.

Images