Optimization method for virtual power plant modeling considering distributed resource electricity carbon characteristics

Abstract

The invention provides a virtual power plant modeling optimization method considering distributed resource power carbon characteristics. The method comprises the following steps: S1, constructing a virtual power plant basic framework of distributed energy aggregation; s2, mathematical modeling is carried out on the distributed resource electric carbon characteristics in the virtual power plant; s3, mathematical modeling is carried out on constraint conditions of distributed resources under the virtual power plant; s4, establishing an electricity-carbon coupling price model based on load side electricity-carbon characteristics; s5, determining an objective function of the distributed energy aggregation virtual power plant; s6, solving and obtaining a virtual power plant optimization result in combination with the distributed resource mathematical model, the power-carbon characteristics and an objective function, and performing optimization scheduling; the electricity and carbon characteristics of the distributed resources are analyzed through a calculation method and a carbon emission flow method, and the enthusiasm of the new energy unit to participate in power supply is effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of virtual power plants, and in particular to an optimization method for virtual power plant modeling considering the electrical and carbon characteristics of distributed resources. Background Art

[0002] With the increasing global climate change and energy demand, the traditional centralized power system can no longer meet the needs of modern society. As a new form of energy, distributed resources have the characteristics of being decentralized, flexible, and renewable, and are gradually becoming an important part of the future power system. However, the randomness and uncertainty of distributed resources bring many challenges to the operation and management of the power system. How to effectively manage and utilize distributed resources, improve the operation efficiency of the power system and reduce carbon emissions, has become a hot issue in current research. At present, most of the modeling and optimization of virtual power plants only consider participating in the electricity market. In the electricity market, the dispatching of virtual power plants mainly focuses on the balance between power supply and demand and economic benefits, while less considering the carbon emission problem. This may lead to high carbon emissions during the operation of virtual power plants and cannot meet the low-carbon and environmental protection energy transformation goals. And less consideration is given to the optimal allocation and sharing of resources, which may hinder the development of new energy and cannot give full play to its potential. Therefore, in the modeling and optimization of virtual power plants, the electrical and carbon characteristics of distributed resources should be considered to promote virtual power plants to participate in the electricity-carbon coupling market.

[0003] In the research of distributed resources, the analysis of electrical and carbon characteristics and the modeling of distributed resources under virtual power plants are two important research directions. The analysis of electrical and carbon characteristics can help us understand the key characteristics such as the power generation and carbon emissions of distributed resources, and provide a reference basis for the planning and design of the power system. As a new energy management method, virtual power plants can integrate dispersed distributed resources to achieve the optimal allocation and sharing of resources. However, there are still some problems in the current research on the analysis of electrical and carbon characteristics of distributed resources and the modeling of distributed resources under virtual power plants. First, the data sources for the analysis of electrical and carbon characteristics are limited, lacking a comprehensive understanding of the electrical and carbon characteristics of distributed resources. Second, most of the current analysis of the electrical and carbon characteristics of distributed resources is only carried out on the power generation side, while the load side, as the consumer of electric energy, rarely analyzes the responsibility of load carbon emissions, greatly reducing the enthusiasm and flexibility of demand response to participate in reducing carbon emissions and power generation costs. Summary of the Invention

[0004] To overcome the deficiencies of the above-mentioned prior art, an optimization method for virtual power plant modeling considering the electrical and carbon characteristics of distributed resources. Based on this, the present invention models and optimizes the virtual power plant, aiming to improve the relationship of the electrical and carbon coupling characteristics of distributed resources, improve the aggregation and utilization rate of distributed resources, and promote virtual power plants to participate in the electricity-carbon coupling market, so as to provide strong support for realizing low-carbon and environmental protection energy transformation.

[0005] The technical solution provided by the present invention is: an optimization method for virtual power plant modeling considering the electrical and carbon characteristics of distributed resources, including the following steps:

[0006] S1: Construct the basic framework of the virtual power plant for distributed energy aggregation;

[0007] S2: Conduct mathematical modeling on the electrical and carbon characteristics of distributed resources in the virtual power plant;

[0008] S3: Conduct mathematical modeling on the constraint conditions of distributed resources under the virtual power plant;

[0009] S4: Establish an electricity-carbon coupling price model based on the electrical and carbon characteristics on the load side;

[0010] S5: Determine the objective function of the distributed energy aggregation virtual power plant;

[0011] S6: Combine the distributed resource mathematical model, electrical and carbon characteristics, and objective function to solve for the optimization result of the virtual power plant and perform optimal scheduling.

[0012] Preferably, the specific content of step S1 includes:

[0013] In the implementation of this step, model distributed wind energy, photovoltaic, thermal power units, and energy storage:

[0014] Limited by the technical parameters of wind turbines, the output of wind power shows a phased characteristic with the change of wind speed:

[0015]

[0016]

[0017] In the formula, C W is the performance parameter of the wind turbine, ρ is the air density; A WT is the projection of the swept area of the wind turbine blades on the vertical plane of the wind speed; v t represents the real-time wind speed of the wind turbine at time t; v in , v τ and v out respectively represent the cut-in wind speed, rated wind speed, and cut-out wind speed; is the rated power of the wind turbine;

[0018] Photovoltaic power generation also has characteristics such as randomness, intermittency, and fluctuation. Its output intensity is related to the light radiation intensity. The Beta distribution function is commonly used to describe the solar photovoltaic radiation intensity. The specific formula is as follows:

[0019]

[0020] In the formula, γ, γ maxrespectively represent the light intensity and the maximum light intensity, and α and β are two parameters in the Beta distribution. The specific formula for the relationship between the output of the photovoltaic power generation unit and the light intensity is as follows:

[0021]

[0022] In the formula, γ e is the rated light intensity of the photovoltaic power generation unit; is the rated power of the photovoltaic power generation unit;

[0023] The adjustable renewable power grid connection of the thermal power unit realizes the coordinated operation inside the virtual power plant, and its output model is as follows:

[0024]

[0025]

[0026] In the formula and respectively represent the power generation fuel cost and the power generation start-stop cost; a TG , b TG and c TG are the power generation energy consumption coefficients, g TG,t is the power generation power; u TG,t is the power generation state variable, a 0-1 variable, 1 represents operation, and 0 represents shutdown; D TG,t represents the start-up cost;

[0027] The output of the energy storage system can be expressed as:

[0028]

[0029]

[0030] In the formula, SOC(t) represents the state of charge of the energy storage device at time t, η chr and η disc respectively represent the charging and discharging efficiencies of the energy storage system, P chr (t) and P disc (t) represent the charging and discharging powers obtained by the energy storage system at time t, and E n is the rated capacity of the energy storage system.

[0031] Further preferably, the specific content in the step S2 includes:

[0032] Analyze the electrical carbon characteristics of the generator set based on the calculation method and carbon balance;

[0033]

[0034] Power supply carbon emission intensity, g / (KW·h); K is the CO2 generation coefficient per unit standard coal; b cp,n Coal consumption for power supply to the unit, g / (KW·h); η cp is the generating efficiency of the unit, %; ξ cp is the power consumption rate of the plant;

[0035] K=K1+K2 (10)

[0036]

[0037]

[0038] K1 is the CO2 generation coefficient per standard coal for coal combustion, K2 is the CO2 generation coefficient per standard coal for desulfurization; C bm is the actual carbon mass corresponding to the unit standard coal, g; OF M is the carbon oxidation rate, 44 / 12 is the molecular weight ratio of CO2 to C; S bm is the mass of sulfur element corresponding to the unit standard coal, g; OF S is the sulfur oxidation rate, is the flue gas desulfurization rate, and 44 / 12 is the molecular mass ratio of CO2 to S element.

[0039] Further preferably, the step S2 specifically includes:

[0040] Analyze the electric-carbon characteristics of the load based on the carbon emission flow theory;

[0041] The carbon emission flow is calculated as follows:

[0042]

[0043] In the formula is the carbon emission flow rate of branch j, P j,i is the portion of outflow branch j that contains inflow branch i, Carbon emission flow density of node i;

[0044]

[0045]

[0046] Where P j is the active power of node j, is the sum of active powers flowing into all branches;

[0047] Combining the above formulas, we can get:

[0048]

[0049] Further preferably, in the step S2, the carbon emission flow density of all the out-flow branches of a given node is the same, and the carbon emission density of the out-flow branch is the same as the carbon emission flow density of the in-flow node. Therefore, the node carbon density and the branch carbon density can be expressed as follows:

[0050]

[0051] In the formula and respectively represent the set of generator units connected to node i and the set of branches. P Gk represents the generated power of the k-th unit, and e Gk represents the carbon emission density of the k-th unit. represents the carbon emission density of branch i-j. represents the branch power from i to j at time t.

[0052] Further preferably, in the step S2, when studying the relationship between the power flow and the carbon emission flow in the power system, the correlation matrix between the generator units - branches - loads can be established by using the electric network knowledge, which can correspond the carbon emission flow flowing out of the generator to the carbon emission flow flowing through the branch or node. The corresponding transfer matrix is expressed as follows:

[0053] Branch - node power transfer matrix:

[0054]

[0055] Each element in the branch - node power matrix represents the power transfer state of a branch. If the transfer power from node i to node j is p, then P ij = p, otherwise P ij = 0;

[0056] Generator unit - node power transfer matrix:

[0057]

[0058] In the formula, P Gij represents the power flow of the power generated by generator unit i injected into node i;

[0059] Node active power flow matrix:

[0060]

[0061] This node active power flow matrix is used to describe the influence of each generator on the power of the nodes and between the nodes. The elements of this matrix can be obtained from P G and P B as follows:

[0062] P N = diag(ξ N+K ·PZ ) (21)

[0063] In the formula: ξ N+K is a column vector with all elements being 1;

[0064]

[0065] In the formula, e G is the carbon emission intensity of power supply of the generator set, which can be obtained by calculating the carbon emission intensity of power supply of the thermal power unit in step S2, and the carbon emission intensities of photovoltaic and wind power are 0. is the carbon emission density of each node.

[0066] Further preferably, step S3 specifically includes:

[0067] Capacity and ramp constraints of the thermal power unit:

[0068] P G，min ≤P G,t ≤P G，man

[0069] R down ≤P G,t -P G,t-1 ≤R up (t≥2) (23)

[0070] In the formula, P G,min and P G,max respectively represent the maximum and minimum active power outputs of the thermal power unit, and R up and R down respectively represent the maximum and minimum ramp powers;

[0071] Capacity constraints of the wind power and photovoltaic units:

[0072] P W,min ≤P W,t ≤P W,max

[0073] P S,min ≤P S,t ≤P S,max (24)

[0074] In the formula, P W,min and P W,max respectively represent the maximum and minimum active power outputs of the wind power unit, and P S,min and P S,max respectively represent the maximum and minimum active power outputs of the photovoltaic unit;

[0075] Operating constraints of the energy storage:

[0076] SOCmin SOC(t) ≤ SOC max

[0077]

[0078]

[0079] where SOC min and SOC max are the minimum capacity and the maximum capacity of the energy storage system respectively, and are the maximum charge-discharge efficiency and the minimum efficiency;

[0080] The upper and lower limits of the load change are generally 20% of the node load, and the total load before and after is unchanged within one cycle. Demand response constraint:

[0081] 0.8D exp,i,t ≤ D tr,i,t ≤ 1.2D exp,i,t

[0082]

[0083] where D tr,i,t is the post-response load, and D exp,i,t is the pre-response load;

[0084] Line power flow upper and lower limit constraint:

[0085] Pline min,t ≤ Pline l,t ≤ Pline max,t (27)

[0086] where Pline l,t is the active power flow of line l at time t, and Pline min,t 、Pline max,t are the upper and lower limits of the transmission power between lines.

[0087] Further preferably, the specific content in step S4 includes:

[0088] Dividing the low, medium, and high carbon responsibility intervals according to the carbon potential, and establishing a time-of-use electricity-carbon coupling price model within different carbon responsibility intervals;

[0089] The responsibility interval division is as shown in the formula:

[0090]

[0091] where e min 、e max are the minimum and maximum load node carbon potentials at time t, e aveis the average carbon potential of the load node, e j (t) is the carbon potential of load node j;

[0092] Furthermore, the electricity-carbon coupling price can be established by combining the electricity price and the load carbon emission characteristics, which is expressed as:

[0093]

[0094] In the formula, q0 is the basic electricity price, and λ is the load carbon price difference in different responsibility intervals.

[0095] For further optimization, the specific steps in step S5 include:

[0096] The objective function of the upper-layer scheduling model is expressed as:

[0097] f1 = min(C H + C K + C W + C S + F1) (30)

[0098] In the formula, C H is the power generation cost of thermal power units, C K is the start-stop cost of thermal power units, C W is the power generation cost of wind power units, C S is the power generation cost of photovoltaic units. The unit operation and maintenance cost is also included in the unit power generation cost. F1 is the carbon trading cost;

[0099]

[0100] In the formula, N is the number of thermal power units, U i,t , U i,t-1 are the on-off states of thermal power unit i at time t and t-1. 1 represents on and 0 represents off; a i , b i , c i are the power generation cost coefficients of unit i; C i is the start-stop cost of unit i, Q and D are the numbers of wind power units and photovoltaic units respectively, P w,t , P s,t are the outputs of wind power and photovoltaic units at time t, f w , f s are the unit power generation costs of wind power and photovoltaic units; λ CM is the carbon trading market price, are the carbon emissions and carbon quotas of the VPP respectively, where where K i is the carbon emission intensity of the i-th unit, P G,i is the power generation power of the i-th unit, η k is the carbon emission quota coefficient;

[0101] The objective function of the lower-layer virtual power plant low-carbon scheduling model is to minimize the sum of the user price cost and the multi-type demand response cost:

[0102] f2 = min(C Q + C E )(32)

[0103] Where C Q is the demand response cost, and C E is the user price cost;

[0104]

[0105]

[0106] Where c dis is the cost coefficient per unit of power response, D dis,j,t is the load response amount of node j at time t; q j,t is the electricity-carbon coupling price of node j at time t, and P Q,j,t is the electricity consumption at node j at time t.

[0107] Further preferably, the specific content in step S6 includes:

[0108] First, input the carbon emission intensity coefficient of the thermal power unit obtained according to step S2, the wind, light, and load prediction data obtained in step S1, and basic data such as the carbon trading price; then solve the upper-layer virtual power plant low-carbon scheduling model to obtain the output power during the unit scheduling period and the active power data transmitted by each line, transmit the data to the lower-layer demand response model, and use the carbon emission flow method to calculate the carbon potential of each node and the electricity-carbon coupling price on the load side; further enter the lower-layer demand response low-carbon scheduling model to solve the optimized load amount; finally, substitute the optimized load amount back into the upper-layer model to perform day-ahead scheduling again and output the optimized scheduling plan.

[0109] The beneficial effects of the present invention are:

[0110] 1) When analyzing the electricity-carbon characteristics of thermal power units, the present invention uses the calculation method and carbon balance, connects the electricity-carbon characteristics of the units with the unit operation efficiency, and the calculated carbon emission intensity coefficient is more in line with the actual power generation and carbon emission relationship of the units;

[0111] 2) The present invention uses the carbon emission flow method to analyze the electricity-carbon characteristics on the load side, realizes the tracking of carbon emission components from the power generation side to the load side through the power system carbon emission flow theory, provides ideas for the low-carbon scheduling strategy of demand response loads, and increases the flexibility of demand response to participate in the scheduling of virtual power plants.

[0112] The present invention proposes a virtual power plant modeling and optimization method considering the electro-carbon characteristics of distributed resources. At the power generation side, carbon emissions mainly come from fossil fuel combustion. Therefore, through calculation methods and carbon balance, the relationship between the power generation and carbon emissions of thermal power units is calculated. At the load side, the electro-carbon characteristics of the load are analyzed through the carbon emission flow method. The carbon emission flow theory is based on power flow tracing and combines the proportional sharing principle to establish a carbon flow tracing model. This model can calculate the carbon emission flow indicators of each load node at each time period according to the known power flow distribution. Finally, through the analysis of the electro-carbon characteristics of distributed resources, a two-stage virtual power plant optimal scheduling model is established. This method can provide a reference basis for the planning and design of virtual power plants and new ideas and methods for the effective management and utilization of distributed resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0113] Figure 1 It is a flowchart of an optimization method for virtual power plant modeling considering the electro-carbon characteristics of distributed resources provided by the present invention;

[0114] Figure 2 It is a flowchart of the virtual power plant optimal scheduling model established in the optimization method for virtual power plant modeling considering the electro-carbon characteristics of distributed resources provided by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0115] In order to make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below in conjunction with embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0116] As Figure 1 shown, the method includes the following steps:

[0117] S1: Construct the basic framework of a virtual power plant for aggregating distributed energy;

[0118] In the implementation of this step, modeling is carried out for distributed wind energy, photovoltaic, thermal power units and energy storage;

[0119] Limited by the technical parameters of wind turbines, the output of wind power shows phased characteristics with the change of wind speed. There is,

[0120]

[0121]

[0122] In the formula C W is the performance parameter of the wind turbine, ρ is the air density; A WT is the projection of the swept area of the wind turbine blades on the vertical plane of the wind speed, v t represents the real-time wind speed of the wind turbine at time t; v in , vτ and v out represent the cut-in wind speed, rated wind speed and cut-out wind speed respectively; is the rated power of the wind turbine;

[0123] Photovoltaic power generation also has characteristics such as randomness, intermittency, and fluctuation. Its output intensity is related to the light radiation intensity. The Beta distribution function is commonly used to describe the solar photovoltaic radiation intensity. The specific formula is as follows:

[0124]

[0125] In the formula, γ, γ max represent the light intensity and the maximum light intensity respectively, and α, β are two parameters in the Beta distribution. The specific formula for the relationship between the output of the photovoltaic power generation unit and the light intensity is as follows:

[0126]

[0127] In the formula, γ e is the rated light intensity of the photovoltaic power generation unit; is the rated power of the photovoltaic power generation unit;

[0128] The adjustable renewable power grid connection of thermal power units realizes the coordinated operation inside the virtual power plant. Its output model is as follows:

[0129]

[0130]

[0131] In the formula and represent the power generation fuel cost and the power generation start-stop cost respectively; a TG , b TG and c TG are the power generation energy consumption coefficients, g TG,t is the power generation; u TG,t is the power generation state variable, a 0-1 variable, 1 means running, 0 means shutdown; D TG,t represents the start-up cost;

[0132] The output of the energy storage system can be expressed as:

[0133]

[0134]

[0135] In the formula, SOC(t) represents the state of charge of the energy storage device at time t, η chr and η disc represent the charging and discharging efficiencies of the energy storage system respectively, P chr (t) and Pdisc (t) represents the charging and discharging power obtained by the energy storage system at time t, and E n is the rated capacity of the energy storage system;

[0136] S2: Mathematically model the electro-carbon characteristics of distributed resources in the virtual power plant;

[0137] Among the distributed resources included in the virtual power plant built by the present invention, the carbon emissions generated during the power generation process of wind power and photovoltaic units are almost zero, so the carbon emissions can be ignored. On the power generation side, the electro-carbon characteristics of thermal power units can be analyzed through the calculation method and carbon balance. On the load side, through the carbon emission flow theory, the carbon emissions are traced from the power generation side to the power consumption side to analyze the electro-carbon characteristics of the load;

[0138] Analyze the electro-carbon characteristics of generator sets based on the calculation method and carbon balance:

[0139] The carbon emission process of coal-fired units mainly includes the coal combustion process and the desulfurization process. To link coal consumption and carbon emissions, the present invention calculates the relationship between coal consumption, power generation and carbon emissions during the operation of the unit through the calculation method;

[0140]

[0141] M CO2 Power supply carbon emission intensity, g / (KW·h); K is the CO2 generation coefficient per unit standard coal; b cp,n is the power supply coal consumption of the unit, g / (KW·h); η cp is the power generation efficiency of the unit, %; ξ cp is the auxiliary power consumption rate.

[0142] K = K1 + K2 (10)

[0143]

[0144]

[0145] K1 is the CO2 generation coefficient per unit standard coal for coal combustion, and K2 is the CO2 generation coefficient per unit standard coal for desulfurization; C bm is the actual mass of carbon element corresponding to per unit standard coal, g; OF M is the carbon oxidation rate, 44 / 12 is the molecular formula mass ratio of CO2 to C; S bm is the mass of sulfur element corresponding to per unit standard coal, g; OF S is the sulfur oxidation rate, is the flue gas desulfurization rate, 44 / 12 is the molecular mass ratio of CO2 to S element;

[0146] Analyze the electro-carbon characteristics of the load based on the carbon emission flow theory:

[0147] The transmission characteristics of carbon emission flow in the network are similar to those of power flow because the carbon emission flow itself is a virtual carbon emission flow simulated based on the characteristics of power flow. Similar to power flow, the virtual carbon emission flow is generated at the power generation nodes, transmitted through the power network by the virtual carbon flow, and finally flows into the load nodes on the demand side. The calculation method of carbon emission flow is as follows:

[0148]

[0149] In the formula is the carbon emission flow rate of branch j, and P j,i is the component flowing out of branch j that contains the inflow from branch i, the carbon emission flow density of node i;

[0150]

[0151]

[0152] In the formula, P j is the active power of node j, is the sum of the active powers of all inflow branches;

[0153] Combining the above equations, we can get:

[0154]

[0155] Further, it is found that the carbon emission flow densities of all outflow branches of a given node are the same, and the carbon emission density of the outflow branch is the same as the carbon emission flow density of the inflow node. Therefore, the node carbon density and branch carbon density can be expressed as:

[0156]

[0157] In the formula and respectively represent the set of generator units and the set of branches connected to node i. P Gk represents the output power of the kth unit, and e Gk represents the carbon emission density of the kth unit, represents the carbon emission density of branch i-j, represents that the branch power is from i to j at time t;

[0158] When studying the relationship between power system power flow and carbon emission flow, by using the knowledge of the electrical network to establish the incidence matrix between generator units - branches - loads, the carbon emission flow flowing out of the generator can be corresponded to the carbon emission flow flowing through the branches or nodes. The corresponding transmission matrix is expressed as follows:

[0159] Branch - node power transmission matrix:

[0160]

[0161] Each element in the branch-node power matrix represents the power transmission state of a branch. If the transmission power from node i to node j is p, then P ij = p, otherwise P ij = 0;

[0162] Unit-node power transmission matrix:

[0163]

[0164] Where P Gij represents the power flow of the power generated by unit i injected into node i.

[0165] Node active power flow matrix:

[0166]

[0167] This node active power flow matrix is used to describe the influence of each generator on the power between nodes and nodes. The elements of this matrix can be obtained from P G and P B Obtain:

[0168] P N = diag(ξ N+K ·P Z ) (21)

[0169] Where: ξ N+K is a column vector with all elements being 1;

[0170]

[0171] Where e G is the power supply carbon emission intensity of the generator set, and the power supply carbon emission intensity of thermal power units can be obtained by the method in step 2.1, while the carbon emission intensity of photovoltaic and wind energy is 0, is the carbon emission density of each node,

[0172] S3: Mathematically model the constraint conditions of distributed resources in the virtual power plant;

[0173] Through the constraint conditions, the utilization mode of the internal resources of the virtual power plant can be restricted to ensure the rationality of its utilization and the authenticity of conforming to the actual operation conditions;

[0174] Capacity and ramp constraints of thermal power units:

[0175] P G，min ≤ P G,t ≤ P G，man

[0176] R down ≤P G,t -P G,t-1 ≤R up (t≥2) (23)

[0177] In the formula, P G,min and P G,max respectively represent the maximum and minimum active power outputs of the thermal power unit, and R up and R down respectively represent the maximum and minimum values of the ramping power;

[0178] Wind power and photovoltaic unit capacity constraint:

[0179] P W,min ≤P W,t ≤P W,max

[0180] P S,min ≤P S,t ≤P S,max (24)

[0181] In the formula, P W,min and P W,max respectively represent the maximum and minimum active power outputs of the wind power unit, and P S,min and P S,max respectively represent the maximum and minimum active power outputs of the photovoltaic unit;

[0182] Energy storage operation constraint:

[0183] SOC min ≤SOC(t)≤SOC max

[0184]

[0185]

[0186] In the formula, SOC min and SOC max respectively represent the minimum and maximum capacities of the energy storage system, and represent the maximum and minimum charge-discharge efficiencies;

[0187] The upper and lower limits of the load change are generally 20% of the node load, and the total load before and after is the same within one cycle. Demand response constraint:

[0188] 0.8D exp,i,t ≤D tr,i,t ≤1.2D exp ,i,t

[0189]

[0190] where D tr,i,t is the post-response load, and D exp,i,t is the pre-response load;

[0191] Line power flow upper and lower limit constraints:

[0192] Pline min,t ≤Pline l,t ≤Pline max,t (27)

[0193] where Pline l,t is the active power flow of line l at time t, and Pline min,t , Pline max,t are the upper and lower limits of the transmission power between lines.

[0194] S4: Establish an electricity-carbon coupling price model based on the electricity-carbon characteristics on the load side;

[0195] The higher the carbon potential of each load node in the power system at the same moment, the greater the carbon emission value per unit of electricity consumed. Based on this, through the optimal dispatching of the load side, if it can promote the load to use more electricity when the node carbon potential is low and less electricity when the node carbon potential is high, the carbon emission value per unit of electric energy of the load within one cycle will be reduced after dispatching, achieving effective energy conservation and emission reduction. From the perspective of low carbon, this invention analyzes according to the carbon potential, divides the low, medium, and high carbon responsibility intervals, and establishes a time-sharing electricity-carbon coupling price model within different carbon responsibility intervals;

[0196] The responsibility interval division is as shown in the formula:

[0197]

[0198] where e min , e max are the minimum and maximum load node carbon potentials at time t, e ave is the average carbon potential of the load node, and e j (t) is the carbon potential of load node j;

[0199] Furthermore, the electricity-carbon coupling price can be established by combining the electricity price and the load carbon emission characteristics, expressed as:

[0200]

[0201] where q0 is the basic electricity price and λ is the load carbon price difference in different responsibility intervals;

[0202] S5: Determine the objective function of the distributed energy aggregation virtual power plant;

[0203] The traditional unit scheduling model mainly considers minimizing the system cost. After adding the carbon trading market to the model, the model also needs to consider the carbon trading cost. This invention considers the load-side electricity-carbon characteristics and load demand response based on the carbon emission flow theory, and constructs a two-stage virtual power plant low-carbon economic scheduling optimization model;

[0204] The objective function of the upper-layer scheduling model is expressed as:

[0205] f1 = min(C H + C K + C W + C S + F1) (30)

[0206] In the formula, C H is the power generation cost of thermal power units, C K is the start-stop cost of thermal power units, C W is the power generation cost of wind power units, C S is the power generation cost of photovoltaic units. The unit operation and maintenance cost is also included in the unit power generation cost. F1 is the carbon trading cost;

[0207]

[0208] In the formula, N is the number of thermal power units, U i,t , U i,t-1 are the on-off states of thermal power unit i at time t and t-1. 1 means on, and 0 means off; a i , b i , c i are the power generation cost coefficients of unit i; C i is the start-stop cost of unit i, Q and D are the numbers of wind power units and photovoltaic units respectively, P w,t , P s,t are the outputs of wind power and photovoltaic units at time t, f w , f s are the unit power generation costs of wind power and photovoltaic units; λ CM is the carbon trading market price, are the carbon emissions and carbon quotas of the VPP respectively. Among them Among them, K i is the carbon emission intensity of the i-th unit, P G,i is the power generation power of the i-th unit, η k is the carbon emission quota coefficient;

[0209] The present invention obtains the start-stop plan and unit output of the generating set in the carbon trading market through the upper-layer model and sends the data to the lower-layer model; calculates the carbon potential and carbon emissions of each node in the current power generation plan through the carbon emission flow method. In the lower layer, considering demand response with the electricity-carbon coupling price on the load side, substitutes the node load after response into the upper-layer scheduling model to re-optimize the output of each unit;

[0210] The objective function of the lower-layer virtual power plant low-carbon scheduling model is to minimize the sum of the user price cost and the multi-type demand response cost:

[0211] f2 = min(C Q + C E )(32)

[0212] In the formula, C Q is the demand response cost, and C E is the user price cost;

[0213]

[0214]

[0215] In the formula, c dis is the cost coefficient per unit of power response, D dis,j,t is the load response amount of node j at time t; q j,t is the electricity-carbon coupling price of node j at time t, and P Q,j,t is the electricity consumption at node j at time t;

[0216] Figure 2 is the flow chart of the virtual power plant optimal scheduling model built by the present invention. The following describes this method in detail according to Figure 2 This method is specifically as shown in step S6:

[0217] S6: Combining the distributed resource mathematical model, electro-carbon characteristics, and objective function to solve for the virtual power plant optimization result and perform optimal scheduling;

[0218] First, input the basic data such as the carbon emission intensity coefficient of thermal power units obtained according to step S2, the wind, light, and load prediction data obtained in step S1, and the carbon trading price; then solve the upper-layer virtual power plant low-carbon scheduling model to obtain the output within the unit scheduling period and the active power data transmitted by each line, transmit the data to the lower-layer demand response model, and calculate the carbon potential of each node and the electricity-carbon coupling price on the load side using the carbon emission flow method; further enter the lower-layer demand response low-carbon scheduling model to solve for the optimized load amount; finally, substitute the optimized load amount back into the upper-layer model to re-perform day-ahead scheduling and output the optimal scheduling plan.

[0219] Other embodiments of the present invention will be readily apparent to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the invention following the general principles of the invention and including known common general knowledge or conventional technical means in the technical field not disclosed herein. The specification and examples are only to be considered as exemplary, and the true scope and spirit of the invention are pointed out by the claims.

[0220] It should be understood that the present invention is not limited to the exact structures shown in the above description and various modifications and changes can be made without departing from its scope. The scope of the present invention is only limited by the appended claims.

Claims

1. An optimization method for virtual power plant modeling considering the electro-carbon characteristics of distributed resources, characterized in that It includes the following steps: S1: Construct the basic framework of a virtual power plant for distributed energy aggregation; S2: Mathematically model the electricity-carbon characteristics of distributed resources in the virtual power plant; S3: Mathematically model the constraint conditions of distributed resources under the virtual power plant; S4: Establish an electricity-carbon coupling price model based on the electricity-carbon characteristics on the load side; S5: Determine the objective function of the virtual power plant for distributed energy aggregation; S6: Combine the mathematical model of distributed resources, electricity-carbon characteristics, and the objective function to solve for the optimization results of the virtual power plant and perform optimal scheduling.

2. The optimization method for virtual power plant modeling considering the electro-carbon characteristics of distributed resources according to claim 1, characterized in that, Specifically included in step S1 are: In the implementation of this step, model distributed wind energy, photovoltaic, thermal power units, and energy storage: Subject to the technical parameters of wind turbines, the output of wind power shows a phased characteristic with the change of wind speed: Where C W is the performance parameter of the wind turbine, ρ is the air density; A WT is the projection of the swept area of the wind turbine blades on the vertical plane of the wind speed; v t represents the real-time wind speed of the wind turbine at time t; v in , v τ and v out represent the cut-in wind speed, rated wind speed and cut-out wind speed respectively; is the rated power of the wind turbine; Photovoltaic power generation also has characteristics such as randomness, intermittency, and fluctuation. Its output intensity is related to the intensity of light radiation. The Beta distribution function is commonly used to describe the solar photovoltaic radiation intensity. The specific formula is as follows: where γ and γ max represent the light intensity and the maximum light intensity respectively, and α and β are two parameters in the Beta distribution. The specific formula for the relationship between the output of the photovoltaic power generation unit and the light intensity is as follows: where γ e is the rated light intensity of the photovoltaic power generation unit; is the rated power of the photovoltaic power generation unit; The adjustable renewable power grid connection of thermal power units enables coordinated operation within the virtual power plant. Its output model is as follows: where and represent the power generation fuel cost and the power generation start-stop cost respectively; a TG , b TG and c TG are the power generation energy consumption coefficients, g TG,t is the power generation power; u TG,t is the power generation state variable, a 0-1 variable, 1 represents running, 0 represents shutdown; D TG,t represents the start-up cost. The output of the energy storage system can be expressed as: where SOC(t) represents the state of charge of the energy storage device at time t, η chr and η disc represent the charging and discharging efficiencies of the energy storage system respectively, P chr (t) and P disc (t) represent the charging and discharging powers obtained by the energy storage system at time t, and E n is the rated capacity of the energy storage system.

3. The optimization method for virtual power plant modeling considering the electro-carbon characteristics of distributed resources according to claim 1, characterized in that, Specifically included in step S2 are: Analyze the electricity-carbon characteristics of generator sets based on the calculation method and carbon balance; Power supply carbon emission intensity, g / (kW·h); K is the CO2 generation coefficient of unit standard coal; b cp,n is the power supply coal consumption of the unit, g / (kW·h); η cp is the power generation efficiency of the unit, %; ξ cp is the auxiliary power consumption rate; K = K1 + K2 (10) K1 is the CO2 generation coefficient of coal combustion per unit standard coal, and K2 is the CO2 generation coefficient of desulfurization per unit standard coal; C bm is the actual mass of carbon element corresponding to per unit standard coal, g; OF M is the carbon oxidation rate, and 44 / 12 is the molecular formula mass ratio of CO2 to C; S bm is the mass of sulfur element corresponding to per unit standard coal, g; OF S is the sulfur oxidation rate, is the flue gas desulfurization rate, and 44 / 12 is the molecular mass ratio of CO2 to S element.

4. An optimization method for virtual power plant modeling considering the electro-carbon characteristics of distributed resources according to claim 1, characterized in that, Specifically included in step S2 are: Analyze the electricity-carbon characteristics of the load based on the theory of carbon emission flow; The calculation method of carbon emission flow is as follows: where is the carbon emission flow rate of branch j, and P j,i is the component of the inflow branch i contained in the outflow branch j, is the carbon emission flow density of node i; where P j is the active power of node j, is the sum of the active powers of all incoming branches; The simultaneous solution of the above equations gives:

5. An optimization method for virtual power plant modeling considering the electro-carbon characteristics of distributed resources according to claim 1, characterized in that, In step S2, the carbon emission flow density of all outflow branches of a given node is the same, and the carbon emission density of the outflow branch is the same as the carbon emission flow density of the inflow node. Therefore, the node carbon density and branch carbon density can be expressed as: where and represent the set of generator units and the set of branches connected to node i respectively, P Gk represents the power output of the k-th generator unit, e Gk represents the carbon emission density of the k-th generator unit, represents the carbon emission density of branch i-j, represents the branch power from i to j at time t.

6. The optimization method for virtual power plant modeling considering the electro-carbon characteristics of distributed resources according to claim 1, characterized in that In step S2, when studying the relationship between power flow and carbon emission flow in the power system, using electrical network knowledge to establish an incidence matrix between generator sets - branches - loads can correspond the carbon emission flow flowing out of the generator with the carbon emission flow flowing through the branch or node. The corresponding transfer matrix is expressed as follows: Branch - node power transfer matrix: Each element in the branch-node power matrix represents the power transmission state of a branch. If the transmission power from node i to node j is p, then P ij = p, otherwise P ij = 0; Unit - node power transfer matrix: where P Gij represents the power flow of the power injected by unit i into node i; Node active power flow matrix: The active power flow matrix of this node is used to describe the influence of each generator on the power of nodes and between nodes. The elements of this matrix can be obtained from P G and P B obtained as follows: P N = diag(ξ N+K ·P Z ) (21) In the formula: ξ N+K is a column vector with all elements being 1; where e G is the power supply carbon emission intensity of the generator set, which can be obtained by calculating the power supply carbon emission intensity of the thermal power unit in step S2, while the carbon emission intensities of photovoltaic and wind energy are 0, is the carbon emission density of each node, 7. An optimization method for virtual power plant modeling considering the electro-carbon characteristics of distributed resources according to claim 1, characterized in that, Specifically included in step S3 are: Capacity and ramp constraints of thermal power units: P G，min ≤P G,t ≤P G，man R down ≤P G,t -P G,t-1 ≤R up (t≥2) (23) where P G,min and P G,max represent the maximum and minimum active power outputs of the thermal power unit respectively, and R up and R down represent the maximum and minimum values of the ramping power respectively; Capacity constraints of wind power and photovoltaic units: P W,min ≤P W,t ≤P W,max P S,min ≤P S,t ≤P S,max (24) where P W,min , P W,max represent the maximum and minimum active power outputs of the wind turbine respectively, and P S,min , P S,max represent the maximum and minimum active power outputs of the photovoltaic unit respectively; Energy storage operation constraints: SOC min ≤SOC(t)≤SOC max where SOC min and SOC max are the minimum capacity and the maximum capacity of the energy storage system respectively, and are the maximum charge-discharge efficiency and the minimum efficiency; The upper and lower limits of the load change amount are generally 20% of the node load, and the total load before and after is the same within one cycle. Demand response constraint: 0.8D exp,i,t ≤D tr,i,t ≤1.2D exp,i,t where D tr,i,t is the post-response load, and D exp,i,t is the pre-response load; Line power flow upper and lower limit constraints: Pline min,t ≤ Pline l,t ≤ Pline max,t (27) where Pline l,t is the active power flow of line l at time t, Pline min,t , Pline max,t are the upper and lower limits of the transmission power between each line.

8. An optimization method for virtual power plant modeling considering the electro-carbon characteristics of distributed resources according to claim 1, characterized in that Specifically included in step S4 are: Divide the low, medium, and high carbon responsibility intervals according to the carbon potential, and establish a time-sharing electricity-carbon coupling price model within different carbon responsibility intervals; The interval division of responsibilities is shown in the formula: where e min , e max are the minimum and maximum carbon potentials of the load nodes at time t, e ave is the average carbon potential of the load nodes, and e j (t) is the carbon potential of load node j; Furthermore, in combination with electricity prices and load carbon emission characteristics, establish an electricity-carbon coupling price, expressed as: In the formula, q0 is the basic electricity price, and λ is the load carbon price difference in different responsibility intervals.

9. An optimization method for virtual power plant modeling considering the electro-carbon characteristics of distributed resources according to claim 1, characterized in that, Specifically included in step S5 are: The objective function of the upper-layer scheduling model is expressed as: f1 = min(C H + C K + C W + C S + F1) (30) Where C H is the power generation cost of thermal power units, C K is the start-up and shutdown cost of thermal power units, C W is the power generation cost of wind power units, C S is the power generation cost of photovoltaic units, where the unit operation and maintenance cost is also included in the unit power generation cost, and F1 is the carbon trading cost; where N is the number of thermal power units, U i,t , U i,t-1 are the on / off states of thermal power unit i at time t and t-1, 1 for on and 0 for off; a i , b i , c i are the power generation cost coefficients of unit i; C i is the start-up and shut-down cost of unit i, Q and D are the numbers of wind power units and photovoltaic power units respectively, P w,t , P s,t are the outputs of wind power and photovoltaic power units at time t, f w , f s are the unit power generation costs of wind power and photovoltaic power units; λ CM is the carbon trading market price, are the carbon emissions and carbon quotas of the VPP respectively, where where K i is the carbon emission intensity of the i-th unit, P G,i is the power generation power of the i-th unit, η k is the carbon emission quota coefficient; The objective function of the lower-layer virtual power plant low-carbon scheduling model is to minimize the sum of user price costs and multi-type demand response costs; f2 = min(C Q + C E ) (32) where C Q is the demand response cost, and C E is the user price cost; where c dis is the cost coefficient per unit power of the response, D dis,j,t is the load response amount of node j at time t; q j,t is the electricity-carbon coupling price of node j at time t, P Q,j,t is the electricity consumption at node j at time t.

10. An optimization method for virtual power plant modeling considering the electro-carbon characteristics of distributed resources according to claim 1, characterized in that, Specifically included in step S6 are: First, input the basic data such as the carbon emission intensity coefficient of the thermal power unit obtained according to step S2, the wind, light, and load prediction data obtained in step S1, and the carbon trading price. Then, solve the upper-layer virtual power plant low-carbon scheduling model to obtain the output power during the unit scheduling period and the active power data transmitted by each line, and transmit the data to the lower-layer demand response model. Use the carbon emission flow method to calculate the carbon potential of each node and the electricity-carbon coupling price on the load side. Further, enter the lower-layer demand response low-carbon scheduling model to solve the optimized load. Finally, substitute the optimized load back into the upper-layer model to perform day-ahead scheduling again, and output the optimized scheduling plan.

Images