A two-tiered optimization strategy for power systems based on demand response guided by nodal carbon intensity and time-of-use pricing.

Abstract

This invention discloses a two-layer optimization strategy for power systems based on nodal carbon intensity and time-of-use pricing to guide demand response, belonging to the field of low-carbon demand response in power systems. This invention utilizes a carbon emission flow tracking model to transfer carbon emission responsibility to the load side in the form of nodal carbon intensity, tracking the changes in nodal carbon intensity with generator output at both temporal and spatial levels, thus more accurately calculating the carbon emissions generated by electricity consumption at load nodes. By adding constraints, the change in load after demand response guided by nodal carbon intensity and time-of-use pricing is obtained. Compared to before the response, after the response, flexible loads under the jurisdiction of each load aggregator shift from periods with higher carbon potential to periods with lower carbon potential, reducing total costs and significantly decreasing carbon emissions, thus alleviating the pressure on electricity consumption during periods of high carbon potential.

Description

Technical Field

[0001] This invention relates to a two-layer optimization strategy for power systems based on nodal carbon intensity and time-of-use pricing to guide demand response, belonging to the field of low-carbon demand response in power systems. Background Technology

[0002] With global carbon emissions steadily increasing, energy conservation, emission reduction, and green, low-carbon development have become a consensus among nations. In power systems, traditional carbon emission metering methods start from the source side. However, the "source follows load" characteristic of power systems leads to a growing contradiction in the separation of responsibilities for carbon emissions. Although the source side is the primary generator of carbon emissions, the electricity demand on the load side is the driving force behind these emissions. Furthermore, optimization methods for reducing direct carbon emissions from the system fail to reflect the direction and magnitude of carbon flow and do not consider the impact of load-side demand response on the power system's low-carbon emission reduction. Therefore, it is necessary to consider scientifically calculating nodal carbon intensity through carbon emission flow tracking and using this nodal carbon intensity to guide load aggregators in calling adjustable loads for demand response, thereby reducing the total cost for load aggregators while improving the system's low-carbon performance. Summary of the Invention

[0003] This invention provides a two-layer optimization strategy and system for power systems based on nodal carbon intensity and time-of-use pricing to guide demand response. It uses a carbon emission flow tracking model to transfer carbon emission responsibility to the load side in the form of nodal carbon intensity, and tracks the changes in nodal carbon intensity with generator output in terms of time and space, so as to more accurately measure the carbon emissions generated by electricity consumption at load nodes.

[0004] The technical solution of this invention is:

[0005] According to one aspect of the present invention, a two-layer optimization strategy for power systems based on nodal carbon intensity and time-of-use pricing to guide demand response is provided, comprising: step S1, determining the nodal carbon intensity of each load aggregator based on an established carbon emission flow tracking model; step S2, establishing a demand response model for flexible loads within the jurisdiction of the load aggregators; flexible loads include electric vehicles, loads that can be reduced, and loads that can be transferred; step S3, determining the actual carbon emissions of each load aggregator based on nodal carbon intensity, and allocating initial carbon emission quotas to each load aggregator based on purchased electricity, and establishing a carbon trading model; step S4, establishing a two-layer optimization scheduling model based on nodal carbon intensity and time-of-use pricing to guide demand response; wherein, the upper layer of the two-layer optimization scheduling model is the grid operator, and the lower layer is the load aggregator, both with the goal of minimizing total cost.

[0006] Step S1 includes: S1.1: calculating the distribution matrix in the node system; S1.2: constructing a carbon emission flow tracking model based on the proportional sharing principle and calculating the carbon intensity of each node.

[0007] Step S2 includes: S2.1: Based on the incentive contract signed between the electric vehicle and the load aggregator, and considering the charging and discharging characteristics of the electric vehicle, calculate the carbon emissions generated after the electric vehicle is charged and discharged based on the nodal carbon intensity; S2.2: Based on the incentive contract for the load that can be reduced and its response characteristics, calculate the carbon emissions after the load that can be reduced performs demand response based on the nodal carbon intensity; S2.3: Based on the incentive contract for the load that can be transferred and its response characteristics, calculate the carbon emissions after the load that can be transferred performs demand response based on the nodal carbon intensity.

[0008] The carbon emissions generated by the electric vehicle during charging and discharging The expression is:

[0009]

[0010] In the formula: P represents the node carbon intensity at time t after the load aggregator is connected to node m; t EV N represents the total charging and discharging amount of all connected electric vehicles at time t; ev For the number of electric vehicles; These represent the charging and discharging power of the nth electric vehicle at time t;

[0011] The carbon emissions after the load reduction is used for demand response The expression is:

[0012]

[0013] In the formula: P t cut The load after reduction at time t;

[0014] The carbon emissions of the transferable load after demand response The expression is:

[0015]

[0016] In the formula: P t tra The load that can be transferred at time t to participate in the demand response load.

[0017] Step S3 includes: S3.1, calculating the actual carbon emissions of each load aggregator using nodal carbon intensity; wherein the actual carbon emissions of each load aggregator include the original load carbon emissions and the carbon emissions after the flexible load performs demand response; S3.2, setting an initial carbon emission quota coefficient, allocating initial carbon emission quotas according to the amount of electricity purchased by the load aggregator from the upstream grid operator; S3.3, establishing a tiered carbon trading model, whereby if the actual carbon emissions are greater than the initial carbon emission quota, the load aggregator needs to pay the excess amount; conversely, if the actual carbon emissions are less than the initial carbon emission quota, the load aggregator can sell the excess carbon emission quotas to obtain revenue.

[0018] The actual carbon emissions are expressed as follows:

[0019]

[0020] In the formula: The actual carbon emissions of the load polymerizer; P t load This is the initial load. Carbon emissions generated during the charging and discharging of electric vehicles; The node carbon intensity at time t after the load aggregator connects to node m; To reduce carbon emissions after demand response that can reduce load; Carbon emissions after demand response for transferable loads.

[0021] Step S4 includes:

[0022] The upper-level grid operator takes minimizing the total cost of the grid operator as the objective function, and establishes grid operator constraint conditions based on thermal power unit output constraints, thermal power unit ramping constraints, thermal power unit start-up and shutdown constraints, wind power output constraints, line transmission capacity constraints, slack node constraints, node power balance constraints, and line power flow equation constraints; the total cost of the grid operator includes thermal power generation cost, wind power generation cost, and grid operator carbon trading cost.

[0023] The lower-level load aggregator takes minimizing the total cost of the load aggregator as its objective function. Based on power balance constraints, electric vehicle charging and discharging constraints, and electric vehicle battery power constraints, the load aggregator's constraints are established. The total cost of the load aggregator includes the cost of purchasing electricity from the upstream, the load aggregator's carbon trading cost, the demand response subsidy cost of loads that can be reduced and transferred, and the electric vehicle discharge subsidy cost.

[0024] The objective function F1 of the power grid operator is:

[0025] min F1=min(C G +C W +C C1 )

[0026]

[0027] In the formula: C G For coal consumption cost; C W Cost of wind power generation; C C1 Carbon trading costs for grid operators; N G N represents the number of thermal power units. t For scheduling period; a i b i c i These are the coal consumption cost coefficients for the i-th thermal power unit; q W Cost of wind power generation; q Wq This refers to the cost coefficient for wind curtailment in wind power generation. The output of the i-th thermal power unit at time t; P t W To actually contribute to wind power; The predicted power output for the u-th wind power scenario; U is the total number of wind power scenarios; ε u Let e ​​be the probability of the u-th wind power scenario; w Indicates the carbon emission coefficient of wind turbine generators; e quote The carbon allowance corresponding to a unit of electricity generated by a generator set; Let represent the carbon emission intensity of the i-th thermal power unit.

[0028] The objective function F2 for the load aggregator is:

[0029] minF2=min(C buy +C c +C CT +C evd )

[0030]

[0031] In the formula: C buy To reduce the cost of purchasing electricity from higher levels, C c To cover the carbon trading costs of aggregators, C CT To reduce the cost of demand response subsidies for load reduction and load transfer, C evd Subsidizing the discharge cost of electric vehicles; N t The scheduling period is P; t buy Let q be the amount of electricity LA purchases from the grid operator at time t; t q cut q tra,in and q tra,out These are, respectively, the electricity purchase price, the unit compensation cost for load reduction, the unit compensation cost for load transfer in, and the unit compensation cost for load transfer out; P t cutP represents the load after reduction at time t. t tra,in P t tra,out These represent the transferable load input and output power at time t, respectively; N ev φ represents the number of electric vehicles; φ represents the electric vehicle discharge subsidy coefficient. Let represent the discharge power of the nth electric vehicle at time t.

[0032] According to another aspect of the present invention, a two-layer optimization system for a power system based on nodal carbon intensity and time-of-use pricing to guide demand response is provided, comprising: a determination module for determining the nodal carbon intensity of each load aggregator based on an established carbon emission flow tracking model; a first establishment module for establishing a demand response model for flexible loads within the jurisdiction of the load aggregators; the flexible loads include electric vehicles, loads that can be reduced, and loads that can be transferred; a second establishment module for determining the actual carbon emissions of each load aggregator based on nodal carbon intensity, and allocating initial carbon emission quotas to each load aggregator based on purchased electricity, thereby establishing a carbon trading model; and a third establishment module for establishing a two-layer optimization scheduling model based on nodal carbon intensity and time-of-use pricing to guide demand response; wherein the upper layer of the two-layer optimization scheduling model is the grid operator, and the lower layer is the load aggregator, both with the goal of minimizing total cost.

[0033] The beneficial effects of this invention are as follows: This invention utilizes a carbon emission flow tracking model to transfer carbon emission responsibility to the load side in the form of nodal carbon intensity, tracking the changes in nodal carbon intensity with generator output at both temporal and spatial levels, and more accurately calculating the carbon emissions generated by electricity consumption at load nodes; by adding constraints, the changes in load volume after demand response guided by nodal carbon intensity and time-of-use pricing are obtained. Compared with before the response, after the response, the flexible loads under the jurisdiction of each load aggregator shift from periods with higher carbon potential to periods with lower carbon potential, reducing total costs and significantly decreasing carbon emissions, thus alleviating the pressure on electricity consumption during periods of high carbon potential. Attached Figure Description

[0034] Figure 1 This is a flowchart of the method of the present invention;

[0035] Figure 2 A diagram illustrating the characteristics of carbon emission flows;

[0036] Figure 3 This invention improves the IEEE-30 node topology.

[0037] Figure 4 The response status of residential load aggregators in calling electric vehicles, load reduction, and load transfer;

[0038] Figure 5 The response status of commercial load aggregators in calling electric vehicles, load shedding, and load transfer;

[0039] Figure 6 The response status of industrial load aggregators in calling electric vehicles, load shedding, and load transfer;

[0040] Figure 7 Load curves before and after demand response for residential load aggregators;

[0041] Figure 8 A load curve before and after demand response for commercial load aggregators;

[0042] Figure 9 This is a load curve before and after the demand response of an industrial load aggregator. Detailed Implementation

[0043] The invention will be further described below with reference to the accompanying drawings and embodiments, but the scope of the invention is not limited to the description.

[0044] Example 1: As Figure 1-9 As shown, according to one aspect of the present invention, a two-layer optimization strategy for power systems based on node carbon intensity and time-of-use pricing to guide demand response is provided, including: Step S1, determining the node carbon intensity of each load aggregator based on an established carbon emission flow tracking model; Step S2, establishing a demand response model for flexible loads within the jurisdiction of the load aggregators; flexible loads include electric vehicles, loads that can be reduced, and loads that can be transferred; Step S3, determining the actual carbon emissions of each load aggregator based on node carbon intensity, and allocating initial carbon emission quotas to each load aggregator based on purchased electricity, and establishing a carbon trading model; Step S4, establishing a two-layer optimization scheduling model based on node carbon intensity and time-of-use pricing to guide demand response, with the upper layer being the grid operator and the lower layer being the load aggregator, both with the goal of minimizing total cost.

[0045] Further, step S1 includes: S1.1: calculating the distribution matrix in the node system; S1.2: constructing a carbon emission flow tracking model based on the proportional sharing principle and calculating the carbon intensity of each node.

[0046] Furthermore, the node system distribution matrix is ​​as follows:

[0047] In a power system with M nodes, V nodes have loads, K generator sets are connected, and the generator injection distribution matrix is ​​a K×M matrix. G =(P Gkj ) K×M Let k = 1, 2, ..., K, j ∈ [1, M] and be positive integers. If the power injected into node j by the k-th generator set is p1, then P Gkj =p1, otherwise P Gkj =0;

[0048] The load distribution matrix is ​​a V×M matrix, P L =(P Lvj ) V×M Let v = 1, 2, ..., V. If node j is the v-th node with load and the load is p2, then P Lvj =p2, otherwise P Lvj =0;

[0049] The branch power flow distribution matrix is ​​an M-order square matrix, P B =(P Bmj ) M×M Let m = 1, 2, ..., M. If a positive active power flow p3 flows between nodes m and j, then P... Bmj =p3, otherwise P Bmj =0, and the diagonal element P Bmm =0;

[0050] The carbon emission flow of a power system is essentially a quantitative determination of its direction and magnitude based on power flow tracking. The carbon emission flow is closely related to the power flow distribution within the power system. In carbon emission flow calculations, nodal carbon intensity is only affected by inflow power flows and not by outflow power flows.

[0051] like Figure 2 As shown, the sum of active power flow P at node m m Inflow power for the branches connected to the node:

[0052]

[0053] In the formula: P m This represents the sum of the inflow active power flow at node m; S represents the active power of branch S; + Let m be the set of branches into which power flows into node m. To obtain the output of generator k, the active power flux matrix P of the system nodes is obtained. M :

[0054] P M =diag(ξ M+K [P B P G ] T (2)

[0055] In the formula: ξ M+K P is an M+K order row vector, where all elements are 1; B Let P be the branch power flow distribution matrix. G Inject the distribution matrix into the generator set;

[0056]

[0057] In the formula: P is an M-dimensional unit row vector, with the m-th element being 1; Bsm P represents the element in the s-th row and m-th column of the branch power flow distribution matrix. Gkm This represents the element in the k-th row and m-th column of the generator injection distribution matrix;

[0058] Based on the principle of proportional sharing, the nodal carbon intensity of node m The carbon emissions are determined by the carbon emissions generated by the connected generators and the carbon emissions flowing in from other nodes:

[0059]

[0060] In the formula: ρ s Let be the carbon flux density of branch s; Let ρ be the carbon emission intensity of the k-th generator unit; s The carbon intensity of the branch starting node is used instead, and the node carbon intensity is written as:

[0061]

[0062] In the formula: E M The nodal carbon intensity vector. E G For the carbon emission vector of the generator set, P B P G transpose;

[0063] Further, step S2 includes: S2.1: Based on the incentive contract signed between the electric vehicle and the load aggregator, and considering the charging and discharging characteristics of the electric vehicle, calculate the carbon emissions generated by the electric vehicle after charging and discharging based on the nodal carbon intensity; S2.2: Based on the incentive contract for reducible loads and its response characteristics, calculate the carbon emissions after demand response by the reducible loads based on the nodal carbon intensity; S2.3: Based on the incentive contract for transferable loads and its response characteristics, calculate the carbon emissions after demand response by the transferable loads based on the nodal carbon intensity.

[0064] Specifically, after an electric vehicle (EV) signs an incentive contract with a load aggregator, the load aggregator has the right to utilize the EV for charging and discharging during the EV's on-grid period, but must meet the EV's electricity demand when it leaves. (EV carbon emissions) The carbon emissions generated during charging minus the carbon emissions reduced during discharging.

[0065]

[0066] In the formula: The node carbon intensity at time t after the load aggregator connects to node m; N represents the total charging and discharging amount of all connected electric vehicles at time t; ev For the number of electric vehicles; These represent the charging and discharging power of the nth electric vehicle at time t;

[0067] Load aggregators sign incentive contracts with electricity users, which stipulate the reduction period, maximum reduction period, reduction coefficient, and transfer period and transfer coefficient of the load that can be transferred.

[0068] Reduceable load refers to the load that can be reduced by a certain percentage within a specified time.

[0069]

[0070] In the formula: The load after reduction at time t; This is the initial load. Let t be a 0-1 state variable representing the load that can be reduced at time t. When the reduction occurs... otherwise β cut This is the reduction factor;

[0071]

[0072] Where: N t For scheduling periods; These are the upper and lower limits of the load that can be reduced; These are the start and end time periods for reducing load response time. To reduce the upper limit of load response time, This represents the maximum reduction coefficient.

[0073] Its carbon emissions can be reduced by cutting loads to respond to demand. for:

[0074]

[0075] In the formula: The node carbon intensity at time t after the load aggregator connects to node m;

[0076] Transferable loads are loads that allow for interruption and have flexible operating times:

[0077] P t tra =P t tra,in -P t tra,out (10)

[0078] In the formula: P represents the load that can be transferred at time t to participate in the demand response load;t tra,in P t tra,out These represent the transferable load input and output power at time t, respectively.

[0079] Considering the actual usage of the equipment, it is not advisable to start and stop the equipment frequently. The load power and transfer time during each transfer should be subject to certain constraints, and the total amount of load transferred in and out should remain unchanged.

[0080]

[0081] In the formula, Let t represent the response states of the transferable load at time t, indicating the load transfer-in and transfer-out responses respectively, expressed as 0-1 variables. When a transfer-in occurs... otherwise When a transfer occurs otherwise These are the upper and lower limits of transferable load, respectively; These represent the start and end time periods for the transfer of transferable load, respectively; Δt is the time interval, and N is the load transfer-in and transfer-out time periods. t The scheduling period is N. In this embodiment of the invention, N is... t Let 24 be the value, and let the time interval Δt be 1, where t = 1, 2, ..., 24.

[0082] Carbon emissions of transferable loads after demand response for:

[0083]

[0084] Further, step S3 includes: S3.1, calculating the actual carbon emissions of each load aggregator through nodal carbon intensity; wherein the actual carbon emissions of each load aggregator include the original load carbon emissions and the carbon emissions after the flexible load performs demand response; S3.2, setting an initial carbon emission quota coefficient, allocating the initial carbon emission quota according to the amount of electricity purchased by the load aggregator from the upstream grid operator; S3.3, establishing a tiered carbon trading model, if the actual carbon emissions are greater than the initial carbon emission quota, the load aggregator needs to pay the excess fee, conversely, if the actual carbon emissions are less than the initial carbon emission quota, the load aggregator can sell the excess carbon emission quota to obtain revenue.

[0085] Specifically:

[0086] The actual carbon emissions of each load aggregator are calculated by nodal carbon intensity, including the carbon emissions of the original load and the carbon emissions of the flexible load after demand response.

[0087]

[0088] In the formula: This represents the actual carbon emissions of the load aggregator.

[0089] The actual carbon emission rights trading amount D of load aggregator LA participating in the carbon trading market at time t. t T for:

[0090]

[0091] In the formula: D t Q E represents the initial carbon emission allowance obtained by the load aggregator LA at time t. quote Carbon emission allowance coefficient per unit of electricity purchased.

[0092] Set the length of the carbon emission range and establish a tiered carbon trading model, based on the actual carbon emissions. Greater than the initial carbon emission allowance Load aggregators are required to pay for any excess allowances. Conversely, if actual carbon emissions are less than the initial carbon allowance, load aggregators can sell the excess allowances to generate revenue.

[0093]

[0094] In the formula: C c The tiered carbon trading cost is represented by λ, which is the benchmark price for carbon trading; α is the price growth coefficient; and l is the length of the carbon emission range.

[0095] Further, step S4 includes:

[0096] The upper-level grid operator takes minimizing the total cost of the grid operator as the objective function, and establishes grid operator constraint conditions based on thermal power unit output constraints, thermal power unit ramping constraints, thermal power unit start-up and shutdown constraints, wind power output constraints, line transmission capacity constraints, slack node constraints, node power balance constraints, and line power flow equation constraints; the total cost of the grid operator includes thermal power generation cost, wind power generation cost, and grid operator carbon trading cost.

[0097] The lower-level load aggregator takes minimizing the total cost of the load aggregator as its objective function. Based on power balance constraints, electric vehicle charging and discharging constraints, and electric vehicle battery power constraints, the load aggregator's constraints are established. The total cost of the load aggregator includes the cost of purchasing electricity from the upstream, the load aggregator's carbon trading cost, the demand response subsidy cost of loads that can be reduced and transferred, and the electric vehicle discharge subsidy cost.

[0098] Furthermore, a two-layer optimization scheduling model based on node carbon intensity-guided demand response:

[0099] The upper-level grid operator aims to minimize total cost by adjusting generator output and allocating carbon emissions to the load side using a carbon emission flow tracking model, thus calculating the load-side node carbon intensity. The grid operator's optimization objective is to reduce total cost, including thermal power generation costs, wind power generation costs, and the grid operator's carbon trading costs. The objective function is:

[0100] min F1=min(C G +C W +C C1 (16)

[0101]

[0102] In the formula: C G For coal consumption cost; C W Cost of wind power generation; C C1 Carbon trading costs for grid operators; N G This refers to the number of thermal power units; a i b i c i These are the coal consumption cost coefficients for the i-th thermal power unit; q W Cost of wind power generation; q Wq This refers to the cost coefficient for wind curtailment in wind power generation. The output of the i-th thermal power unit at time t; P t W To actually contribute to wind power; For the predicted power output of the u-th wind power scenario; ε u Let e ​​be the probability of the u-th wind power scenario; w Indicates the carbon emission coefficient of wind turbine generators; e quote The carbon allowance corresponding to a unit of electricity generated by a generator set; Indicates the carbon emission intensity of the i-th thermal power unit;

[0103] Constraints

[0104] Thermal power unit output constraints

[0105]

[0106] In the formula: These are the minimum and maximum outputs of the i-th thermal power unit, respectively;

[0107] Thermal power unit ramping constraints

[0108]

[0109] In the formula: The output of the i-th thermal power unit at time t-1; R i U R iD denoted as the maximum gradeability and maximum landslide rate of the i-th thermal power unit, respectively; Δt is the time interval.

[0110] Thermal power unit start-stop constraints

[0111]

[0112] In the formula: These represent the operating and shutdown times of the i-th thermal power unit during time period t-1, respectively. These represent the shortest operating and downtime times for the i-th thermal power unit, respectively; u i,t u i,t-1 These represent the start-up and stop states of the i-th thermal power unit at times t and t-1, respectively. When starting, the value is 1, otherwise it is 0.

[0113] Wind power output constraints

[0114]

[0115] In the formula: This represents the maximum output power of the wind power plant.

[0116] Line transmission capacity constraints

[0117] P f,min ≤P f,t ≤P f,max (twenty two)

[0118] In the formula: P f,t P represents the active power flow of line f at time t; f,max P f,min These are the upper and lower limits of the power transmitted between lines;

[0119] Balance node constraints

[0120]

[0121] In the formula: Let t be the phase angle of the equilibrium node voltage.

[0122] Node power balance constraints

[0123]

[0124] In the formula: Let be the inflow and outflow power of node m at time t; Let be the electrical load of node m at time t;

[0125] Line power flow equality constraints

[0126]

[0127] Where: βmj Let θ be the reactance of line mj; m,t θ j,t These are the voltage phase angles at nodes m and j, respectively.

[0128] The lower-level LA responds to the carbon intensity signal from the upper-level nodes with the goal of minimizing total cost, adjusting its flexible load power consumption strategy to reduce the LA's carbon emissions and total cost. Based on user electricity demand, the LA incentivizes users to participate in low-carbon responses through price incentives, providing users with certain economic subsidies. The LA's total cost includes the cost of purchasing electricity from the upper-level node (C). buy LA carbon trading costs C c Demand response subsidy cost for load reduction and load transfer C CT Electric vehicle discharge subsidy cost C evd The objective function is:

[0129] minF2=min(C buy +C c +C CT +C evd (26)

[0130]

[0131] In the formula: Let q be the amount of electricity LA purchases from the grid operator at time t; t q cut q tra,in and q tra,out These are the electricity purchase price, the unit compensation cost for load reduction, the unit compensation cost for load transfer to, and the unit compensation cost for load transfer out; φ is the electric vehicle discharge subsidy coefficient. P represents the discharge power of the nth electric vehicle at time t; t tra,in P t tra ,out These represent the transferable load input and output power at time t, respectively.

[0132] Constraints

[0133] Power balance constraints

[0134]

[0135] These represent the charging and discharging power of the nth electric vehicle at time t; This is the initial load. The load after reduction at time t;

[0136] Electric vehicle charge and discharge constraints

[0137]

[0138] In the formula: Let n be the charging and discharging state of the nth electric vehicle (EV) at time t. When the EV is charging, The value is 1; during EV discharge, =1; These represent the maximum charging and discharging power of the EV.

[0139] Electric vehicle battery power constraints

[0140]

[0141] In the formula: These represent the initial energy, target energy, and rated battery capacity of the nth electric vehicle, respectively; E n,t Let SOC be the battery capacity of the nth electric vehicle at time t. n,t SOC n,t-1 These represent the battery state of charge of the nth electric vehicle at times t and t-1, respectively. These are the upper and lower limits of the battery state of charge for the nth electric vehicle at times t and t-1, respectively; t in t out These refer to the arrival and departure times of the electric vehicle, respectively; η c η d This indicates the charge / discharge efficiency.

[0142] According to another aspect of the present invention, a two-layer optimization system for a power system based on nodal carbon intensity and time-of-use pricing to guide demand response is provided, comprising: a determination module for determining the nodal carbon intensity of each load aggregator based on an established carbon emission flow tracking model; a first establishment module for establishing a demand response model for flexible loads within the jurisdiction of the load aggregators; flexible loads include electric vehicles, loads that can be reduced, and loads that can be transferred; a second establishment module for determining the actual carbon emissions of each load aggregator based on nodal carbon intensity, and allocating initial carbon emission quotas to each load aggregator based on purchased electricity, thereby establishing a carbon trading model; and a third establishment module for establishing a two-layer optimization scheduling model based on nodal carbon intensity and time-of-use pricing to guide demand response; wherein the upper layer of the two-layer optimization scheduling model is the grid operator, and the lower layer is the load aggregator, both with the goal of minimizing total cost. For parts of the modules not described in detail above, please refer to the relevant descriptions in the embodiments.

[0143] To demonstrate the effectiveness of this invention, it is now based on Figure 3 The improved IEEE-30 node was simulated and analyzed by sequentially executing steps S1-S4 of this invention. Detailed parameter settings for the simulation analysis are as follows:

[0144] (1) Residential load aggregator, commercial load aggregator and industrial load aggregator are connected to nodes 24, 26 and 29 respectively. A wind farm with a capacity of 600MW replaces the thermal power unit at node 13. The relevant parameters of the thermal power unit are shown in Table 1 below, and the time-of-use electricity price is shown in Table 2.

[0145] (2) Wind power generation cost q W The cost is 60 yuan / MW, and the wind curtailment penalty coefficient is q. Wq The carbon emission coefficient e of the wind turbine is 250 yuan / MW. w It is 0.043 tCO2 / MW.

[0146] (3) LA Initial Carbon Emission Quota Factor E quote The carbon emission rate is 0.728 tCO2 / MW, the carbon trading benchmark price λ is 252 yuan / t, the price growth coefficient α is 0.25, and the carbon emission range lengths for each load aggregator are 1t, 25t, and 90t, respectively.

[0147] (4) For ease of calculation, it is assumed that all electric vehicles are of the same model, and the relevant parameters of the electric vehicles are shown in Table 3 below. The number of electric vehicles in each load aggregator is set to 800, 300, and 500 respectively. The transferable load and load reduction contract parameters are shown in Tables 4 and 5 below.

[0148] Table 1 Parameters of Thermal Power Units

[0149]

[0150] Table 2 Time-of-use Electricity Price Table

[0151] Time period <![CDATA[Time-of-use electricity price / (yuan·(kW·h) -1 )]]> Peak hours 08:00-11:00、18:00-23:00 0.9164 Normal period 12:00-17:00 0.6164 Valley period 00:00-07:00、23:00-24:00 0.3113

[0152] Table 3. Electric Vehicle Scheduling Related Parameters

[0153]

[0154] Table 4 Transferable Load Incentive Contract Parameters

[0155]

[0156] Table 5 Parameters of Reduced Load Incentive Contracts

[0157]

[0158] With attachment Figures 4-6 The figure shows the demand response results for electric vehicles, load shedding, and load transferable, as examples. The results before and after demand response for each load aggregator are attached. Figures 7-9 As shown.

[0159] It is evident that load aggregators transfer transferable load out during periods of high carbon potential and transfer it in during periods of low carbon potential, while reducing load during periods of high carbon potential. During the time when electric vehicles (EVs) are connected, EV charging is concentrated during periods of relatively low carbon potential, while discharging is concentrated during periods of relatively high carbon potential. When carbon potential rises, load aggregators guide EVs to begin discharging, increasing the discharge volume to reduce regional electricity purchases. After incentivizing flexible loads, EVs are scheduled for orderly charging and discharging, load reduction and transfer shifting peak-hour loads to off-peak hours, reducing peak-hour electricity pressure. The peak-to-valley load difference for residential load aggregators decreased from 28.61MW to 23.16MW. The peak-to-valley load difference for commercial load aggregators increased from 43.10MW to 46.83MW, and for industrial load aggregator LA3, it increased from 62.40MW to 78.01MW. Although the peak-to-valley load differences for commercial and industrial load aggregators increased, the total load decreased during the peak hours of 4 PM to 9 PM, playing a positive role in alleviating peak-hour electricity shortages.

[0160] To verify the effectiveness of the proposed model and method, we analyze the demand response of flexible loads guided by time-of-use pricing and nodal carbon intensity, setting up the following five scenarios:

[0161] Scenario 1: Considering fixed electricity prices, loads not participating in optimized dispatch, and carbon trading not considered;

[0162] Scenario 2: Considering a two-tiered optimization dispatch strategy that uses time-of-use pricing to guide flexible load adjustments, without considering carbon trading;

[0163] Scenario 3: Considering time-of-use pricing and tiered carbon trading, flexible loads are not included in optimized scheduling;

[0164] Scenario 4: Considering a two-tiered optimization scheduling strategy that combines fixed electricity prices, tiered carbon trading, and nodal carbon intensity-guided flexible load adjustments;

[0165] Scenario 5: Consider a two-tiered optimization scheduling strategy that combines time-of-use pricing, tiered carbon trading, and flexible load adjustment guided by nodal carbon intensity.

[0166] The carbon emissions and costs of each load aggregator are shown in Table 6 below:

[0167] Table 6. Carbon emissions and costs of each load aggregator before and after demand response.

[0168]

[0169] Based on Table 6, analysis of scenarios one through five reveals that scenarios one and three, which do not consider demand response, have the highest carbon emissions. Scenario three, which considers time-of-use pricing, has the highest total cost. Scenario two guides flexible loads to respond to demand through time-of-use pricing, and scenario four guides load-side demand response through nodal carbon intensity. Both scenarios reduce carbon emissions after response. Scenario five combines scenarios two and four, simultaneously considering time-of-use pricing and nodal carbon intensity to guide load aggregators to respond to demand. Compared to scenario one, the total cost for each LA increases, primarily due to carbon trading costs. However, the cost of electricity purchase for each LA decreases, and carbon emissions for each LA decrease by 19.99t, 12.56t, and 31.21t, respectively, improving the low-carbon nature of the LAs.

[0170] Scenario 5, building upon Scenario 2 by incorporating demand response, shows that flexible loads, influenced by time-of-use pricing and nodal carbon intensity, experience reduced carbon trading costs and electricity purchase costs for each LA (Local Area), resulting in significant reductions in carbon emissions of 3.04t, 0.16t, and 11.03t, respectively. This demonstrates that using time-of-use pricing and nodal carbon intensity to guide flexible loads effectively improves the system's low-carbon and economic efficiency. Furthermore, a comparison of Scenario 4 and Scenario 5 clearly shows that, compared to fixed pricing, Scenario 5, considering time-of-use pricing, enhances the responsiveness of flexible loads in demand response. Carbon emissions for each LA are reduced by 3.16t, 1.62t, and 12.16t, respectively, compared to Scenario 4, with a corresponding decrease in carbon trading costs. However, due to the limitations of electric vehicle access and departure times, charging may occur during peak hours of time-of-use pricing to meet electric vehicle charging needs, leading to a slight increase in electricity purchase costs in Scenario 5 compared to Scenario 4. Therefore, although the model proposed in this paper increases costs to some extent, it reduces dependence on the power generation side by improving the response of flexible loads, resulting in greater reductions in carbon emissions and improving the low-carbon nature of the system.

[0171] After grid operators consider carbon trading, their generating unit output plans are adjusted, thus affecting costs and carbon emissions. Table 7 analyzes the carbon emissions and costs before and after grid operators participate in carbon trading in Scenario 5.

[0172] Table 7 Comparison of carbon emissions and costs before and after grid operators participate in carbon trading.

[0173]

[0174] As shown in Table 7, after participating in carbon trading, the carbon emissions of grid operators decreased, but the total cost increased compared to before participating in carbon trading. The main increase was in carbon trading costs, while coal consumption costs increased slightly. This is because, when considering the total cost, grid operators, in order to reduce carbon emissions and minimize carbon trading costs, selected units with higher coal consumption costs but lower carbon emission coefficients to operate at certain times.

[0175] The specific embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited to the above embodiments. Within the scope of knowledge possessed by those skilled in the art, various changes can be made without departing from the spirit of the present invention.

Claims

1. A two-tier optimization strategy for power systems based on nodal carbon intensity and time-of-use pricing to guide demand response, characterized in that, include: Step S1: Based on the established carbon emission flow tracking model, determine the node carbon intensity of each load aggregator; Step S2: Establish a demand response model for flexible loads within the jurisdiction of the load aggregator; flexible loads include electric vehicles, loads that can be reduced, and loads that can be transferred. Step S3: Determine the actual carbon emissions of each load aggregator based on the node carbon intensity, and allocate the initial carbon emission quotas of each load aggregator based on the purchased electricity volume to establish a carbon trading model; Step S4: Establish a two-layer optimal scheduling model based on nodal carbon intensity and time-of-use pricing to guide demand response; wherein, the upper layer of the two-layer optimal scheduling model is the grid operator and the lower layer is the load aggregator, both with the goal of minimizing total cost; Step S4 includes: The upper-level grid operator takes minimizing the total cost of the grid operator as the objective function, and establishes grid operator constraint conditions based on thermal power unit output constraints, thermal power unit ramping constraints, thermal power unit start-up and shutdown constraints, wind power output constraints, line transmission capacity constraints, slack node constraints, node power balance constraints, and line power flow equation constraints; the total cost of the grid operator includes thermal power generation cost, wind power generation cost, and grid operator carbon trading cost. The lower-level load aggregator takes the minimum total cost of the load aggregator as the objective function. Based on the power balance constraint, electric vehicle charging and discharging constraint, and electric vehicle battery power constraint, the load aggregator's constraint conditions are established. The total cost of the load aggregator includes the cost of purchasing electricity from the upper level, the load aggregator's carbon trading cost, the demand response subsidy cost of loads that can be reduced and transferred, and the electric vehicle discharge subsidy cost. The objective function F1 of the power grid operator is: In the formula: For coal consumption costs; Cost of wind power generation; C C1 Carbon trading costs for grid operators; N G N represents the number of thermal power units. t For scheduling period; a i b i c i These are the coal consumption cost coefficients for the i-th thermal power unit; q W Cost of wind power generation; q Wq This refers to the cost coefficient for wind curtailment in wind power generation. The output of the i-th thermal power unit at time t; To actually contribute to wind power; The predicted power output for the u-th wind power scenario; U is the total number of wind power scenarios; ε u Let u be the probability of the u-th wind power scenario; Indicates the carbon emission coefficient of wind turbine generators; e quote The carbon allowance corresponding to a unit of electricity generated by a generator set; Let represent the carbon emission intensity of the i-th thermal power unit.

2. The two-layer optimization strategy for power systems based on nodal carbon intensity and time-of-use pricing to guide demand response, as described in claim 1, is characterized in that... Step S1 includes: S1.1: The distribution matrix in the computation node system; S1.2: Construct a carbon emission flow tracking model based on the proportional sharing principle to calculate the carbon intensity of each node.

3. The two-layer optimization strategy for power systems based on nodal carbon intensity and time-of-use pricing to guide demand response, as described in claim 1, is characterized in that... Step S2 includes: S2.1: Based on the incentive contract signed between the electric vehicle and the load aggregator, and considering the charging and discharging characteristics of the electric vehicle, calculate the carbon emissions generated by the electric vehicle after charging and discharging based on the nodal carbon intensity. S2.2: Based on the incentive contracts for reducible loads and their response characteristics, calculate the carbon emissions after demand response for reducible loads based on nodal carbon intensity; S2.3: Based on the transferable load incentive contracts and their response characteristics, calculate the carbon emissions after demand response by the transferable load based on the nodal carbon intensity.

4. The two-layer optimization strategy for power systems based on nodal carbon intensity and time-of-use pricing to guide demand response, as described in claim 3, is characterized in that... The carbon emissions generated by the electric vehicle during charging and discharging The expression is: In the formula: The node carbon intensity at time t after the load aggregator is connected to node m; N represents the total charging and discharging amount of all connected electric vehicles at time t; ev For the number of electric vehicles; These represent the charging and discharging power of the nth electric vehicle at time t; The carbon emissions after the load reduction is used for demand response The expression is: In the formula: The load after reduction at time t; The carbon emissions of the transferable load after demand response The expression is: In the formula: The load that can be transferred at time t to participate in the demand response load.

5. The two-layer optimization strategy for power systems based on nodal carbon intensity and time-of-use pricing to guide demand response, as described in claim 1, is characterized in that... Step S3 includes: S3.1 Solve for the actual carbon emissions of each load aggregator by nodal carbon intensity; where the actual carbon emissions of each load aggregator include the original load carbon emissions and the carbon emissions of the flexible load after demand response; S3.2 Set the initial carbon emission quota coefficient and allocate the initial carbon emission quota according to the amount of electricity purchased by the load aggregator from the upstream grid operator; S3.3 Establish a tiered carbon trading model. If the actual carbon emissions exceed the initial carbon emission allowance, the load aggregator needs to pay the excess amount. Conversely, if the actual carbon emissions are less than the initial carbon emission allowance, the load aggregator can sell the excess carbon emission allowance to generate revenue.

6. The two-layer optimization strategy for power systems based on nodal carbon intensity and time-of-use pricing to guide demand response, as described in claim 5, is characterized in that... The actual carbon emissions are expressed as follows: In the formula: This represents the actual carbon emissions of the load aggregator; This is the initial load. Carbon emissions generated during the charging and discharging of electric vehicles; The node carbon intensity at time t after the load aggregator is connected to node m; To reduce carbon emissions after demand response that can reduce load; Carbon emissions after demand response for transferable loads.

7. The two-layer optimization strategy for power systems based on nodal carbon intensity and time-of-use pricing to guide demand response, as described in claim 1, is characterized in that... The objective function F2 for the load aggregator is: In the formula: C buy To reduce the cost of purchasing electricity from higher levels, C c To cover the carbon trading costs of aggregators, C CT To reduce the cost of demand response subsidies for load reduction and load transfer, C evd Subsidizing the discharge cost of electric vehicles; N t The scheduling period; Let q be the amount of electricity LA purchases from the grid operator at time t; t q cut q tra,in and q tra,out These are the electricity purchase price, the unit compensation fee for load reduction, the unit compensation fee for load transfer to the receiving unit, and the unit compensation fee for load transfer to the sending unit. The load after reduction at time t; These represent the transferable load input and output power at time t, respectively; N ev For the number of electric vehicles; The electric vehicle discharge subsidy coefficient; Let represent the discharge power of the nth electric vehicle at time t.

8. A power system bi-level optimization system based on nodal carbon intensity and time-of-use pricing to guide demand response, implementing the bi-level optimization strategy of claim 1, characterized in that, include: The determination module is used to determine the node carbon intensity of each load aggregator based on the established carbon emission flow tracing model; The first module is used to establish a demand response model for flexible loads within the jurisdiction of the load aggregator; flexible loads include electric vehicles, loads that can be reduced, and loads that can be transferred. The second module is used to determine the actual carbon emissions of each load aggregator based on the node carbon intensity, and to allocate the initial carbon emission quotas of each load aggregator based on the purchased electricity, thereby establishing a carbon trading model. The third module is used to establish a two-layer optimal scheduling model based on nodal carbon intensity and time-of-use pricing to guide demand response. The upper layer of the two-layer optimal scheduling model consists of grid operators and load aggregators, both with the goal of minimizing total cost.

Images