Power demand side management optimization method based on electric-carbon coupling and dynamic carbon emission

Abstract

The application relates to the field of power market analysis, and discloses a power demand side management optimization method based on electricity-carbon coupling and dynamic carbon emission, which comprises the following steps: obtaining multiple demand response schemes and multiple energy generator unit data in a power market; constructing a multiple energy generator unit market model according to the multiple energy generator unit data in the power market; constructing a two-stage power market clearing model based on the multiple energy generator unit market model, wherein the two-stage power market clearing model comprises a first-stage model and a second-stage model, the first-stage model is a market planning and dispatching model, and the second-stage model is a demand response model; inputting each demand response scheme into the two-stage power market clearing model in sequence to obtain market operation parameter analysis results under each demand response scheme; and obtaining market performance analysis and demand response decision through comparison of all market operation parameter analysis results. The application can realize expansion of implementation comparison of different market-oriented demand response schemes.

Description

Technical Field

[0001] This invention relates to the field of power dispatching, and specifically to a power demand-side management optimization method based on electric carbon coupling and dynamic carbon emissions. Background Technology

[0002] The construction of sustainable power systems is crucial for achieving global decarbonization goals and meeting growing electricity demand. However, the integration of a high proportion of variable renewable energy sources has led to intermittent power supply issues, requiring greater flexibility in power systems. Demand response, as a key technology for unlocking the potential of demand-side resources, incentivizes consumers to adjust their electricity consumption patterns through technical and economic means, thereby improving system flexibility. With the coupling of electricity markets and carbon emission trading systems, competitive electricity markets are improving operational efficiency, reducing costs, and achieving environmental goals by introducing demand response. The introduction of demand response changes electricity demand patterns, thus affecting the dispatch of generation resources and market operation. Currently, there are two main market-based demand response mechanisms: price-based demand response and low-carbon demand response. Price-based demand response transmits price signals of market supply and demand changes to consumers through real-time pricing mechanisms or time-of-use pricing mechanisms, promoting optimal resource allocation. Low-carbon demand response, on the other hand, uses real-time dynamic electricity consumption carbon emission factors to guide users to proactively reduce carbon emissions, achieving direct emission reduction effects. Although existing research has recognized the differences between price-based and low-carbon demand response, a systematic comparative analysis of their specific impacts on the market has not yet been conducted.

[0003] Furthermore, the impact of demand response scheme types and implementation methods on market operation has not been fully studied. More and more scholars are beginning to focus on the role of demand response in the market, gradually shifting their research focus from demand-side response mechanisms and potential assessments to the impact of demand response itself on electricity market design and operation. However, these studies involve multiple market-based demand response schemes, with multi-dimensional but fragmented research perspectives, covering aspects such as power system investment planning, unit dispatch, market electricity prices, economic benefits, and environmental impact. Almost no research attempts to clearly distinguish the types of demand response schemes, let alone compare the differences in their impact on market operation. In addition, most studies still do not consider the new scenario of the gradual coupling of the electricity market and the carbon market under decarbonization goals, and most studies use similar macro-level methods to study the impact of demand response on market operation. For example, Mier et al. used a partial equilibrium electricity market model to assess the impact of implementing short-term demand response and energy efficiency on the transition of the European electricity market under climate goals. However, macro-level methods cannot fully reflect how these market-based demand response schemes affect unit commitment processes and power dispatch processes, thereby affecting market operation. They also cannot simulate changes in market operation under different scenarios, nor can they clarify and compare the implementation effects of market-based demand response.

[0004] Due to various problems with existing models, electricity market clearing models have emerged. These models are widely used in electricity market design, operation, and regulation, potentially providing an open and unified framework and mechanism for model construction. Electricity market clearing models mainly consist of two parts: unit allocation and economic dispatch. The consistency, constraints, and flexibility in objective construction of clearing rules give market clearing models good scalability, and they can support multi-scenario simulations through parameter control. Numerous model variations are widely used in electricity market planning and dispatch, unit allocation, scenario analysis, and policy simulation. For example, Xinyu et al. introduced a cross-sectoral high-resolution electricity allocation model to simulate the optimal national power system structure in China under different renewable energy investment levels from 2030 to 2050.

[0005] However, no model currently exists that can extend the comparison of different market-based demand response schemes under the scenario of electricity-carbon coupling. This involves many challenges that have not been previously considered, including the coupling mode and strength of the carbon market, the quantitative transmission of carbon costs, the design of participation mechanisms for different demand response schemes, the dynamic emission characteristics of generating units, and market policies. These all need to be modeled as components of objective functions and constraints according to real-world scenarios to characterize the optimal strategies for power system operators, power plants, and the demand side. Furthermore, in most studies on generating unit modeling, the carbon emission factor of the unit is assumed to be static. However, extensive actual monitoring and theoretical reasoning have demonstrated that during generator operation, the carbon emission factor is dynamic rather than static and closely related to its power output, or more precisely, inversely proportional to power.

[0006] To address the aforementioned problems, it is essential to research and design a novel power demand-side management optimization method based on electricity-carbon coupling and dynamic carbon emissions, thereby overcoming the existing issues in power market demand-side management optimization. Summary of the Invention

[0007] To address the limitations of existing power market demand-side management optimization methods, such as the inability to expand the comparison of different market-based demand response schemes under the scenario of electricity-carbon coupling, and the lack of consideration for the dynamic nature of carbon emission factors in existing power market demand response analysis, this invention provides a power demand-side management optimization method based on electricity-carbon coupling and dynamic carbon emissions.

[0008] The technical solution adopted by this invention to achieve the above objectives is: an optimization method for electricity demand-side management based on electric-carbon coupling and dynamic carbon emissions, comprising the following steps:

[0009] S1. Obtain multiple demand response solutions and multi-energy generator data in the electricity market;

[0010] S2. Construct a multi-energy generator set market model based on the multi-energy generator set data in the electricity market;

[0011] S3. Based on the multi-energy generator set market model, construct a two-stage electricity market clearing model. The two-stage electricity market clearing model includes a first-stage model and a second-stage model. The first-stage model is a market planning and scheduling model, and the second-stage model is a demand response model.

[0012] S4. Input each of the demand response schemes into the two-stage electricity market clearing model in sequence to obtain the market operation parameter analysis results under each demand response scheme;

[0013] S5. Market performance analysis and demand response decisions are obtained by comparing the analysis results of all the market operation parameters under each of the aforementioned demand response schemes.

[0014] According to some embodiments of the present invention, a method for optimizing electricity demand-side management based on electro-carbon coupling and dynamic carbon emissions, in step S1, the demand response scheme includes market operation and policy parameters and demand-side parameters. The market operation and policy parameters include coupled market dispatch configuration data, carbon price and quota data, renewable energy combination standard data, and energy storage configuration data. The demand-side parameters include user type data, baseline load data, regulation cost data, and regulation capacity data. The multi-energy generating unit data includes unit data and techno-economic parameters. The unit data includes new energy unit data, gas turbine unit data, coal-fired unit data, and electrochemical energy storage data. The techno-economic parameters include investment cost data, operating cost data, dynamic carbon emission factor data, and unit supply constraint data.

[0015] According to some embodiments of the present invention, a power demand-side management optimization method based on electricity-carbon coupling and dynamic carbon emissions, in step S2, the multi-energy generator market model includes a thermal power generator model, a renewable energy generator model, and an independent energy storage generator model. The thermal power generator model includes multiple thermal power units, the renewable energy generator model includes multiple renewable energy units, the renewable energy units include photovoltaic units and wind turbine units, and the independent energy storage generator model includes multiple independent energy storage units.

[0016] According to some embodiments of the present invention, an optimization method for electricity demand-side management based on electric carbon coupling and dynamic carbon emissions is provided.

[0017] The objective function of the thermal power generating unit model includes the marginal generation cost function of bidding under dynamic carbon emission intensity considering the coupling of the electricity carbon market, the carbon dioxide emission function of the thermal power unit within time interval t, and other cost functions of the thermal power unit participating in market bidding within time interval t.

[0018] The marginal generation cost function for bidding under the dynamic carbon emission intensity, considering the coupling of the electricity carbon market, is shown in formula (1):

[0019]

[0020] in, P represents the marginal cost of the thermal power unit Gi within the time interval t. Gi,t This represents the power generation of thermal power unit Gi during time interval t. This represents the binomial coefficient of the Gi fuel cost function for thermal power units. This represents the coefficient of the first-order term in the Gi fuel cost function for thermal power units. This indicates the proportion of fuel cost to total power generation cost for a thermal power unit (Gi). This represents the coefficient of the first term in the fitting curve of the dynamic carbon emission factor Gi of a thermal power unit versus its power output. This represents the constant of the fitting curve between the Gi dynamic carbon emission factor and power of a thermal power unit. This represents the baseline emission value for the Gi of a thermal power unit, pr c Indicates carbon price, Gi represents thermal power unit index, Ω G Ω represents the set of thermal power units, t represents the time interval, and Ω represents the total thermal power unit. T Represents the set of scheduling times.

[0021] The carbon dioxide emissions of the thermal power unit during the time interval t are shown in formula (2):

[0022]

[0023] in, This represents the carbon dioxide emissions of the thermal power unit Gi during the time interval t.

[0024] The other cost functions for the fire unit's participation in market declaration within the time interval t are shown in formula (3):

[0025]

[0026] in, This represents the other costs of the fire unit Gi during the time interval t. This indicates the no-load cost reported by the thermal power unit Gi within the time interval t. This represents the declared start-up cost of thermal power unit Gi within the time interval t. This represents the reported downtime cost of thermal power unit Gi within time interval t.

[0027] The constraints of the objective function of the thermal power generating unit model include the following constraints: the declared no-load cost of the thermal power generating unit within time interval t; the declared start-up cost of the thermal power generating unit within time interval t; the declared shutdown cost of the thermal power generating unit within time interval t; the output limit constraint of the thermal power generating unit; the ramp rate constraint of the thermal power generating unit; and the minimum switching time constraint of the thermal power generating unit.

[0028] The constraint on the declared no-load cost of the thermal power unit within the time interval t is shown in formula (4):

[0029]

[0030] Among them, L Gi This represents the no-load cost of the thermal power unit Gi, u Gi,t This represents the operating state of the thermal power unit Gi within the time interval t, and is a 0-1 variable.

[0031] The constraint on the application start-up cost of the thermal power unit within the time interval t is shown in formula (5):

[0032]

[0033] Among them, H Gi This represents the startup cost of the thermal power unit Gi, u Gi,t-1 This indicates the operating status of the thermal power unit Gi within the time interval t-1.

[0034] The constraint on the reported shutdown cost of the thermal power unit within the time interval t is shown in formula (6):

[0035]

[0036] Among them, J Gi This represents the downtime cost of the Gi power unit.

[0037] The power output limit constraint of the thermal power unit is shown in formula (7):

[0038]

[0039] Among them, P Gi,min This indicates the minimum generating capacity limit of the thermal power unit Gi, P Gi,max This indicates the maximum generating capacity limit of the Gi thermal power unit.

[0040] The ramp rate constraint of the thermal power unit is shown in formula (8):

[0041]

[0042] Among them, P Gi,t-1 This represents the power generation of thermal power unit Gi within the time interval t-1. This indicates the ramp rate limit for the Gi power unit. This indicates the downhill rate limit for the Gi power unit.

[0043] The minimum switching time constraint of the thermal power unit is shown in formula (9):

[0044]

[0045] Where T represents the scheduling period, h represents the time interval recording point, and u Gi,h This indicates the operating status of the thermal power unit Gi at the time interval recording point. This indicates the minimum startup time for the thermal power unit Gi. This indicates the minimum downtime of the fire unit Gi.

[0046] According to some embodiments of the present invention, an optimization method for electricity demand-side management based on electric carbon coupling and dynamic carbon emissions is provided.

[0047] The objective function of the renewable energy unit model includes the operating cost function of the renewable energy unit within the time interval t.

[0048] The operating cost function of the renewable energy unit within the time interval t is shown in formula (10):

[0049]

[0050] Among them, C RE,t This represents the operating cost of a renewable energy unit within a time interval t. This represents the unit operating cost of the wind turbine, WTi, where WTi represents the index of the wind turbine, and P... WTi,t This represents the power generation of the wind turbine WTi within the time interval t. This represents the unit operating cost of the photovoltaic (PV) unit, PVi, where PVi is the index of the PV unit. PVi,t PVi represents the power generation of the photovoltaic unit during the time interval t, Ω. WT Ω represents a set of wind turbine units. PV Indicates a collection of photovoltaic units.

[0051] The constraints on the operating cost function of the renewable energy units within the time interval t include the output constraints of the renewable energy units.

[0052] The output constraint of the renewable energy unit is shown in formula (11):

[0053]

[0054] Where, γ WTi,tQ represents the capacity factor of the wind turbine WTi within the time interval t. WTi Indicates the installed capacity of wind turbine WTi, γ PVi,t Q represents the capacity factor of the photovoltaic unit PVi within the time interval t. PVi This indicates the installed capacity of the photovoltaic unit, PVi.

[0055] According to some embodiments of the present invention, an optimization method for electricity demand-side management based on electric-carbon coupling and dynamic carbon emissions includes, in which the objective function of the independent energy storage unit model includes the operating cost function of the independent energy storage unit within a time interval t.

[0056] The operating cost function of the independent energy storage unit within the time interval t is shown in formula (12):

[0057]

[0058] Among them, C IESi,t This represents the operating cost of the independent energy storage unit IESi within the time interval t. This represents the unit operating cost of the independent energy storage unit IESi. This represents the charging power of the independent energy storage unit IESi within the time interval t. This represents the discharge power of the independent energy storage unit IESi within the time interval t, where IESi represents the index of the independent energy storage unit, Ω. IES It represents a collection of independent energy storage units.

[0059] The constraints of the objective function of the independent energy storage unit model include: charging rate constraint of the independent energy storage unit within time interval t; discharging rate constraint of the independent energy storage unit within time interval t; constraint that the independent energy storage unit will not charge and discharge simultaneously; maximum charging amount constraint of the independent energy storage unit model per unit scheduling time; maximum discharging amount constraint of the independent energy storage unit model per unit scheduling time; state of charge constraint of the independent energy storage unit model; and constraint that the state of charge of the independent energy storage unit model remains unchanged within the scheduling period.

[0060] The charging rate constraint of the independent energy storage unit within the time interval t is shown in Equation (13):

[0061]

[0062] in, Q represents the charging state of the independent energy storage unit IESi within the time interval t, and is a 0-1 variable. IESi This indicates the installed capacity of the independent energy storage unit IESi.

[0063] The discharge rate constraint of the independent energy storage unit within the time interval t is shown in formula (14):

[0064]

[0065] in, This represents the discharge state of the independent energy storage unit IESi within the time interval t, and is a 0-1 variable.

[0066] The constraint that the independent energy storage unit cannot charge and discharge simultaneously is shown in formula (15):

[0067]

[0068] The maximum charging amount constraint of the independent energy storage unit model within a unit scheduling time is shown in formula (16):

[0069]

[0070] Where Δt represents the duration of the time interval, Qs IESi This indicates the maximum energy storage capacity of the independent energy storage unit IESi. This represents the state of charge of the independent energy storage unit IESi within the time interval t-1. This indicates the charging efficiency of the independent energy storage unit IESi.

[0071] The maximum discharge limit of the independent energy storage unit within a unit scheduling time is shown in formula (17):

[0072]

[0073] in, This indicates the discharge efficiency of the independent energy storage unit IESi.

[0074] The maximum energy storage capacity of the independent energy storage unit is Qs. IESi As shown in formula (18):

[0075]

[0076] Among them, h IESi This indicates the maximum discharge duration of the independent energy storage unit IESi.

[0077] The state of charge constraint of the independent energy storage unit model is shown in formula (19):

[0078]

[0079] in, This represents the state of charge of the independent energy storage unit IESi during the time interval t.

[0080] The state of charge of the independent energy storage unit within the time interval t As shown in formula (20):

[0081]

[0082] The constraint that the state of charge of the independent energy storage unit model remains unchanged during the scheduling period is shown in Equation (21):

[0083]

[0084] in, This indicates the initial state of charge of the independent energy storage unit IESi during the dispatch cycle. This indicates the state of charge of the independent energy storage unit IESi at the end of the dispatch cycle. This indicates the initial energy storage level of the independent energy storage unit IESi.

[0085] According to some embodiments of the present invention, in a power demand-side management optimization method based on electricity-carbon coupling and dynamic carbon emissions, in step S3, the objective function of the market planning and scheduling model includes a cost objective function.

[0086] The cost objective function of the market planning and scheduling model is shown in formula (22):

[0087] f1=min(C Inv +w tr C Ope )(twenty two)

[0088] Where f1 represents the cost objective function of the market planning and scheduling model, and C Inv This indicates the annual amortization of investment costs, w tr C represents the conversion factor for operating costs. Ope This represents typical daily operating costs, multiplied by an annual conversion factor w. tr This can be converted into annual operating costs.

[0089] The annual amortization investment cost C Inv As shown in formula (23):

[0090]

[0091] in, This represents the annual amortization cost of the thermal power generating unit model. This represents the annual amortization cost of the renewable energy unit model. This represents the annual amortization cost of the stand-alone energy storage unit model.

[0092] Annual amortization cost of the thermal power generating unit model As shown in formula (24):

[0093]

[0094] Among them, κ g This represents the conversion factor for the annual investment cost of the thermal power generating unit model. N represents the unit investment cost of the thermal power unit Gi. Gi This indicates the installed capacity of the thermal power unit Gi, r Gi,t The variable is 0-1, representing the investment status of the thermal power unit Gi during the time interval t.

[0095] Annual amortization cost of the renewable energy unit model As shown in formula (25):

[0096]

[0097] Among them, κ wt This represents the conversion factor for the annual investment cost of wind turbine units. κ represents the unit investment cost of wind turbine WTi. pv This represents the conversion factor for the annual investment cost of photovoltaic units. This represents the unit investment cost of a photovoltaic (PV) unit, PVi.

[0098] Annual amortization cost of the independent energy storage unit model As shown in formula (26):

[0099]

[0100] Among them, κ ies This represents the conversion factor for the annual investment cost of the stand-alone energy storage unit model. This represents the unit investment cost of the independent energy storage unit IESi.

[0101] Typical daily operating cost C Ope As shown in formula (27):

[0102]

[0103] The cost conversion coefficients of the market planning and scheduling model are shown in formula (28):

[0104]

[0105] Where, μ Ope κ represents the annual operating equivalence factor, a represents the unit discount rate, y represents the equipment's service life, and κ represents the investment cost conversion factor.

[0106] The constraints of the cost objective function include constraints on the type of generating unit, the unit capacity, the proportion of renewable energy generation, reliability, and reserve requirements.

[0107] The unit construction type constraint of the market planning and scheduling model is shown in formula (29):

[0108]

[0109] Among them, U G This indicates the number of firepower aircraft assembled.

[0110] The unit capacity constraint is shown in formula (30):

[0111]

[0112] in, REi represents the maximum permitted installed capacity of thermal power units in the regional power grid system, and Ω represents the renewable energy unit index. RE It represents a collection of renewable energy units. Q represents the maximum permitted installed capacity of renewable energy units within the regional power grid. IESi This represents the maximum permissible independent energy storage unit capacity within the regional power grid, where ρ represents the minimum proportion of energy storage units that can be configured.

[0113] The constraint on the proportion of renewable energy generation is shown in formula (31):

[0114]

[0115] Where, ρ New This indicates the power generation penetration rate of renewable energy units. This indicates the system load after demand response.

[0116] The reliability constraints are shown in formula (32):

[0117]

[0118] The backup constraint is shown in formula (33):

[0119]

[0120] Where ε represents the load reserve ratio. Pr represents the proportion of renewable energy generation reserves. REi,t This represents the power generation of the renewable energy unit REi within the time interval t.

[0121] According to some embodiments of the present invention, in a power demand-side management optimization method based on electric carbon coupling and dynamic carbon emissions, in step S3,

[0122] The demand response model includes a second-stage function, which comprises a second-stage objective function for a price-based demand response model, a second-stage objective function for a marginal emission factor-based demand response model, and a second-stage objective function for a mean emission factor-based demand response model.

[0123] The second-stage objective function of the price-based demand response model is shown in formula (34):

[0124]

[0125] Where f2 represents the cost of electricity, λ t P represents the unit cost of electricity used within time interval t. l,t Indicates the initial system load. This indicates an increase in load. This indicates a reduction in load.

[0126] The second-stage objective function of the demand response model based on marginal emission factors is shown in Equation (35):

[0127]

[0128] Where f3 represents the emission reduction calculated using the marginal emission factor (MEF). t This represents the marginal emission factor within the time interval t.

[0129] The marginal emission factor (MEF) within the time interval t t As shown in formula (36):

[0130]

[0131] The second-stage objective function of the demand response model based on the average emission factor is shown in Equation (37):

[0132]

[0133] Where f4 represents the carbon dioxide emission reduction calculated using the average emission factor, XEF t This represents the average carbon emission factor over time t.

[0134] The average carbon emission factor XEF over time t t As shown in formula (38):

[0135]

[0136] in, This represents the system load after demand response. As shown in formula (39):

[0137]

[0138] The constraints of the second-stage function include the constraints on the maximum and minimum adjustment amounts of the demand response model, the constraints on the load increase amount of the demand response model within time interval t, the constraints on the load reduction amount of the demand response model within time interval t, the constraints on the demand response model that the load cannot be increased and decreased simultaneously within the same time interval, and the constraints on the demand response model that the total daily load before and after the electricity consumption behavior adjustment remains basically unchanged within the dispatch cycle.

[0139] The constraints on the maximum and minimum adjustment amounts of the demand response model are shown in formula (40):

[0140]

[0141] in, These are 0-1 variables, representing the logic for determining whether a user should increase their workload. This indicates that the load limit can be adjusted within the time interval t. This is a 0-1 variable representing the logic for determining whether a user should reduce their load; δ represents the adjustable load ratio.

[0142] The constraint on the load increase within the time interval t of the demand response model is shown in formula (41):

[0143]

[0144] in, Indicates the system's maximum rated load.

[0145] The constraint on the load reduction amount within the time interval t of the demand response model is shown in formula (42):

[0146]

[0147] The constraint in the demand response model that the load cannot be increased and decreased simultaneously within the same time interval is shown in formula (43):

[0148]

[0149] The constraint that the total daily load remains basically unchanged before and after the electricity consumption behavior adjustment during the scheduling period in the demand response model is shown in formula (44):

[0150]

[0151] in, This indicates the daily change in electricity consumption.

[0152] According to some embodiments of the present invention, a method for optimizing electricity demand-side management based on electric carbon coupling and dynamic carbon emissions, step S4 includes the following steps:

[0153] S401. Input the parameters of the demand response scheme into the first stage model;

[0154] S402. Linearize the first-stage model with parameters from the input demand response scheme to obtain a linear first-stage model;

[0155] S403. Obtain the minimum system cost through the linear first-stage model, and determine whether to perform demand response based on the minimum system cost. If demand response is performed, proceed to step S404; otherwise, proceed to step S406.

[0156] S404. Based on the characteristics of the demand scheme, input the parameters of the demand response scheme into the second-stage model, and obtain the minimized electricity cost or minimized carbon emissions through the second-stage model;

[0157] S405. Adjust the load curve by minimizing electricity costs or carbon emissions and update the load reduction and load increase parameters. Determine whether the updated load reduction and load increase parameters meet the convergence condition. If they do, proceed to step S406. If they do not, use the updated load reduction and load increase parameters as parameters in the updated demand response scheme and then execute step S401. The convergence condition is that the user load can only change once within any time interval t, as shown in equation (45):

[0158]

[0159] Among them, Ω n A set representing the number of iterations;

[0160] S406. Output the market operation parameter analysis results of the demand response plan;

[0161] S407. Repeat steps S401-S406 until the market operation parameter analysis results of all demand response schemes are obtained.

[0162] According to some embodiments of the present invention, an optimization method for electricity demand-side management based on electricity-carbon coupling and dynamic carbon emissions, in step S402, linearizing the first-stage model with parameters from the demand response scheme as input includes using a continuous variable Z. t Replace Q IESi With s IESi,t The product of is shown in formula (46):

[0163]

[0164] Among them, Z t This indicates the substitution of a continuous variable, s IESi,t These are 0-1 variables, representing the charge and discharge states of independent energy storage.

[0165] Introduce variable M to constrain continuous variable Z t As shown in formula (47):

[0166]

[0167] Where M is set to 10000,

[0168] The annual operating cost of the thermal power unit (Equation 27) can be rewritten as a quadratic function in general form, as shown in formula (48):

[0169]

[0170] Among them, Qu Gi Co represents the coefficient of the quadratic term. Gi Denotes the coefficient of the linear term.

[0171] The quadratic coefficient Qu Gi As shown in formula (49):

[0172]

[0173] The coefficient of the first term Co Gi As shown in formula (50):

[0174]

[0175] The annual operating cost of the aforementioned thermal power unit is processed by piecewise linearization, as shown in formula (51):

[0176]

[0177] Among them, P Gi,t,k This represents the power generation of thermal power unit Gi in the k-th segment during time interval t, where k represents the k-th linear segment, ΩK represents the set of segments, and η Gi,k This represents the slope of the Gi piecewise linearized operating cost function for the thermal power unit in the k-th segment.

[0178] The output of the fire unit in sections is shown in formula (52):

[0179]

[0180] The constraint on the output of the fire unit in sections is shown in formula (53):

[0181]

[0182] Where K represents the number of linear segments.

[0183] This invention presents an optimization method for electricity demand-side management based on electricity-carbon coupling and dynamic carbon emissions. By constructing a multi-energy generator market model, it exhibits good scalability and is suitable for multi-scenario simulation. It quantitatively characterizes electricity-carbon coupling, the dynamic carbon emission characteristics of generator units, and the demand response mechanism. By coupling thermal power generator models, renewable energy generator models, and independent energy storage generator models, it not only quantifies the cross-market transmission and coupling strength of carbon costs but also enables better analysis of demand response schemes. Furthermore, it introduces reliability constraints into the market planning and scheduling model to ensure real-time system source-load balance and introduces reserve constraints to address load forecasting and renewable energy output. This method reduces prediction errors and enhances the overall framework's flexibility. It allows for a deeper analysis of the specific impacts of demand response schemes on market operations, providing a scientific basis for deploying flexible strategies for sustainable power systems. It expands the comparison of different market-based demand response schemes under electricity-carbon coupling scenarios and considers the dynamic nature of carbon emission factors, avoiding the "black box" drawbacks of using macroscopic parameters to study the impact of demand response. Based on a model-driven approach, it fully demonstrates the internal flows of power generation, emissions, and economic variables in the market under the influence of demand response, elucidating the impact of demand response on market operations from a bottom-up perspective and providing decision-makers with more realistic information. Attached Figure Description

[0184] Figure 1 This is a flowchart illustrating the power demand-side management optimization method based on electric carbon coupling and dynamic carbon emissions according to an embodiment of the present invention.

[0185] Figure 2 This is a schematic diagram illustrating the impact of dynamic carbon emission intensity of thermal power units and carbon market coupling on the marginal cost of unit bidding in Embodiment 2 of the present invention.

[0186] Figure 3 These are the market planning results under different demand response schemes in Embodiment 3 of the present invention;

[0187] Figure 4 This is the power generation and load transfer performance under different demand response schemes in Embodiment 3 of the present invention, wherein (a) is the power generation situation of the units and the initial load in Scenario 1; (b) is the difference in power generation of the units from Scenario 1 to Scenario 2 and the load optimization scheduling based on electricity price; (c) is the difference in power generation of the units from Scenario 1 to Scenario 3 and the load optimization scheduling based on marginal carbon emission factor; and (d) is the difference in power generation of the units from Scenario 1 to Scenario 4 and the load optimization scheduling based on average carbon emission factor.

[0188] Figure 5This describes the load optimization scheduling under different demand response schemes in Embodiment 3 of the present invention;

[0189] Figure 6 This refers to the difference in power generation performance between dynamic carbon emission intensity and static carbon emission intensity in scenarios 1-4 of Embodiment 3 of the present invention; wherein (a) is the difference in power generation performance between dynamic carbon emission intensity and static carbon emission intensity in scenario 1; (b) is the difference in power generation performance between dynamic carbon emission intensity and static carbon emission intensity in scenario 2; (c) is the difference in power generation performance between dynamic carbon emission intensity and static carbon emission intensity in scenario 3; and (d) is the difference in power generation performance between dynamic carbon emission intensity and static carbon emission intensity in scenario 4.

[0190] Figure 7 This refers to the marginal carbon emission cost of the generator set under dynamic and static carbon emission intensity scenarios in Embodiment 3 of the present invention; where (a) is the marginal carbon emission cost of the generator set under the dynamic carbon emission intensity scenario; and (b) is the marginal carbon emission cost of the generator set under the static carbon emission intensity scenario.

[0191] Figure 8 This is a comparison of market operating parameters for different demand response schemes under different carbon market coupling strengths in Embodiment 3 of the present invention; wherein (a) is a comparison of market electricity prices for different demand response schemes under different carbon market coupling strengths; (b) is a comparison of power generation carbon emissions for different demand response schemes under different carbon market coupling strengths; (c) is a comparison of total system costs for different demand response schemes under different carbon market coupling strengths; and (d) is a comparison of industry profits for different demand response schemes under different carbon market coupling strengths.

[0192] Figure 9 This is a schematic diagram illustrating the correlation between marginal cost and emission priority under different carbon market coupling strengths in Embodiment 3 of the present invention. Detailed Implementation

[0193] The embodiments of the present invention will be described in further detail below with reference to the accompanying drawings and examples. The following examples are for illustrative purposes only and should not be construed as limiting the scope of the invention.

[0194] Example 1

[0195] This embodiment of a power demand-side management optimization method based on electric carbon coupling and dynamic carbon emissions includes the following steps:

[0196] S1. Obtain multiple demand response solutions and multi-energy generator data in the electricity market;

[0197] S2. Construct a multi-energy generator market model based on multi-energy generator data in the electricity market;

[0198] S3. Construct a two-stage electricity market clearing model based on a multi-energy generator market model. The two-stage electricity market clearing model includes a first-stage model and a second-stage model. The first-stage model is a market planning and dispatching model, and the second-stage model is a demand response model.

[0199] S4. Input each demand response scheme into the two-stage electricity market clearing model obtained in step S3 in sequence, and obtain the market operation parameter analysis results under each demand response scheme;

[0200] S5. Market performance analysis and demand response decisions are obtained by comparing the analysis results of all market operation parameters under each demand response scheme.

[0201] Example 2

[0202] This embodiment presents an optimization method for electricity demand-side management based on electro-carbon coupling and dynamic carbon emissions, such as... Figure 1 As shown, it includes the following steps:

[0203] S1. Obtain multiple demand response solutions and multi-energy generator data in the electricity market;

[0204] Specifically, the demand response plan includes market operation and policy parameters, and demand-side parameters. Market operation and policy parameters include coupled market dispatch and allocation data, carbon price and quota data, renewable energy combination standard data, and energy storage configuration data. Demand-side parameters include user type data, baseline load data, regulation cost data, and regulation capacity data. Multi-energy generator set data includes unit data and techno-economic parameters. Unit data includes new energy unit data, gas unit data, coal-fired unit data, and electrochemical energy storage data. Techno-economic parameters include investment cost data, operating cost data, dynamic carbon emission factor data, and unit supply constraint data.

[0205] S2. Construct a multi-energy generator market model based on multi-energy generator data in the electricity market;

[0206] Specifically, the multi-energy generator set market model includes a thermal power generator set model, a renewable energy generator set model, and an independent energy storage generator set model. The thermal power generator set model includes multiple thermal power units, the renewable energy generator set model includes multiple renewable energy units, renewable energy units include photovoltaic units and wind turbine units, and the independent energy storage generator set model includes multiple independent energy storage units.

[0207] Thermal power units generally include coal-fired power units and gas-fired power units. The fuel cost of thermal power units is generally characterized by a quadratic function of power generation output, as shown in formula (1):

[0208]

[0209] Among them, Ce Gi,t P represents the fuel cost of a thermal power unit within a time interval t. Gi,t This represents the power generation capacity of the thermal power unit within the time interval t. This represents the binomial coefficient of the fuel cost function for thermal power units. This represents the coefficient of the first-order term in the fuel cost function of a thermal power unit. Ω represents a constant in the fuel cost function of a thermal power unit, Gi represents the thermal power unit index, and Ω represents the thermal power unit index. G Ω represents the set of thermal power units, t represents the time interval, and Ω represents the total thermal power unit. T This represents the set of scheduling times.

[0210] In a benchmark-based carbon market, the carbon cost of a thermal power unit is mainly determined by the emission benchmark value of the category to which the thermal power unit belongs and the unit's carbon emission factor. In this case, the unit's carbon emission factor and carbon cost are constants, as shown in formula (2):

[0211]

[0212] Among them, Cc Gi,t E represents the carbon cost of a thermal power unit over a time interval t. Gi Indicates the carbon emission factor of the unit. This indicates the emission baseline value for thermal power units, pr c Indicates carbon price.

[0213] In actual operation of thermal power units, the unit's carbon emission factor is dynamic, and the dynamic carbon emission factor E Gi,t It is inversely related to carbon prices, such as Figure 2 As shown, the dynamic carbon emission factor gradually approaches or even falls below the emission benchmark value of thermal power units as power output increases. This means that, compared to a static scenario, thermal power units with stronger dynamic characteristics can change from buyers to sellers by increasing output, thereby lowering bid prices to obtain a larger share of output. The mathematical relationship between the dynamic carbon emission factor and the dynamic carbon price can be simplified to an approximation of a linear function, thereby quantitatively modifying the real-time carbon cost of thermal power units, as shown in formula (3):

[0214]

[0215] In the thermal power generating unit model, the generator's bid is an affine transformation of the marginal cost function. By differentiating and proportionally transforming the fuel cost and carbon cost curves of the thermal power unit, the first-order linear marginal generation cost can be obtained. The objective function of the thermal power generating unit model includes the bidding marginal generation cost function considering the coupling of the electricity and carbon markets under dynamic carbon emission intensity, the carbon dioxide emission function of the thermal power unit within time interval t, and other cost functions of the thermal power unit participating in market bidding within time interval t.

[0216] The marginal generation cost function for bidding considering the coupling of the electricity carbon market under dynamic carbon emission intensity is shown in Equation (4):

[0217]

[0218] in, P represents the marginal cost of the thermal power unit Gi within the time interval t. Gi,t This represents the power generation of thermal power unit Gi during time interval t. This represents the binomial coefficient of the Gi fuel cost function for thermal power units. This represents the coefficient of the first-order term in the Gi fuel cost function for thermal power units. This indicates the proportion of fuel cost to total power generation cost for a thermal power unit (Gi). This represents the coefficient of the first term in the fitting curve of the dynamic carbon emission factor Gi of a thermal power unit versus its power output. This represents the constant of the fitting curve between the Gi dynamic carbon emission factor and power of a thermal power unit. This represents the baseline emission value for the Gi of a thermal power unit, pr c Indicates carbon price, Gi represents thermal power unit index, Ω G Ω represents the set of thermal power units, t represents the time interval, and Ω represents the total thermal power unit. T Represents the set of scheduling times.

[0219] The carbon dioxide emissions of the thermal power unit during the time interval t are shown in formula (5):

[0220]

[0221] in, This represents the carbon dioxide emissions of the thermal power unit Gi during the time interval t.

[0222] The thermal power generating unit model will also include start-up and no-load costs reported to the Independent System Operator (ISO) for unit combination decisions, as the sum of start-up, shutdown, and no-load costs.

[0223] Other cost functions for the fire unit's participation in market declaration during the time interval t are shown in formula (6):

[0224]

[0225] in, This represents the other costs of the fire unit Gi during the time interval t. This indicates the no-load cost reported by the thermal power unit Gi within the time interval t. This represents the declared start-up cost of thermal power unit Gi within the time interval t. This represents the reported downtime cost of thermal power unit Gi within time interval t.

[0226] The constraints of the objective function of the thermal power generating unit model include the following constraints: the declared no-load cost of the thermal power generating unit within time interval t; the declared start-up cost of the thermal power generating unit within time interval t; the declared shutdown cost of the thermal power generating unit within time interval t; the output limit constraint of the thermal power generating unit; the ramp rate constraint of the thermal power generating unit; and the minimum switching time constraint of the thermal power generating unit.

[0227] The declared no-load cost of thermal power unit Gi within time interval t. The constraints are shown in formula (7):

[0228]

[0229] Among them, L Gi This represents the no-load cost of the thermal power unit Gi, u Gi,t The variable is 0-1, representing the operating status of the thermal power unit Gi within the time interval t.

[0230] The declared start-up cost of thermal power unit Gi within time interval t The constraints are shown in formula (8):

[0231]

[0232] Among them, H Gi This represents the startup cost of the thermal power unit Gi, u Gi,t-1 This indicates the operating status of the thermal power unit Gi within the time interval t-1.

[0233] The reported shutdown cost of thermal power unit Gi within time interval t. The constraints are shown in formula (9):

[0234]

[0235] Among them, J Gi This represents the downtime cost of the Gi power unit.

[0236] The output limit constraints for each thermal power unit in the thermal power generator model are shown in Equation (10):

[0237]

[0238] Among them, P Gi,min This indicates the minimum generating capacity limit of the thermal power unit Gi, P Gi,max This indicates the maximum generating capacity limit of the Gi thermal power unit.

[0239] The ramp rate constraint for each thermal power unit in the thermal power generator model is shown in Equation (11):

[0240]

[0241] Among them, P Gi,t-1 This represents the power generation of thermal power unit Gi within the time interval t-1. This indicates the ramp rate limit for the Gi power unit. This indicates the downhill rate limit for the Gi power unit.

[0242] The minimum switching time constraint for each thermal power unit in the thermal power generating unit model is shown in formula (12):

[0243]

[0244] Where T represents the scheduling period, h represents the time interval recording point, and u Gi,h This indicates the operating status of the thermal power unit Gi at the time interval recording point. This indicates the minimum startup time for the thermal power unit Gi. This indicates the minimum downtime of the fire unit Gi.

[0245] Renewable energy units include wind turbines that generate energy using wind speed and photovoltaic (PV) units that generate energy using sunlight. It's worth noting that the power output of renewable energy units is constrained by a capacity factor. Furthermore, when the reserve capacity or transmission capacity in the multi-energy generator market model is insufficient, wind and solar curtailment will be permitted.

[0246] The objective function of the renewable energy unit model includes the operating cost function of the renewable energy unit over time interval t.

[0247] The operating cost function of the renewable energy unit within the time interval t is shown in formula (13):

[0248]

[0249] Among them, C RE,t This represents the operating cost of a renewable energy unit within a time interval t. This represents the unit operating cost of the wind turbine, WTi, where WTi represents the index of the wind turbine, and P... WTi,t This represents the power generation of the wind turbine WTi within the time interval t. This represents the unit operating cost of the photovoltaic (PV) unit, PVi, where PVi is the index of the PV unit. PVi,t PVi represents the power generation of the photovoltaic unit during the time interval t, Ω. WT Ω represents a set of wind turbine units. PV Indicates a collection of photovoltaic units.

[0250] The constraints on the operating cost function of renewable energy units within the time interval t include the output constraints of the renewable energy units.

[0251] The output constraint of renewable energy units is shown in formula (14):

[0252]

[0253] Where, γ WTi,t Q represents the capacity factor of the wind turbine WTi within the time interval t. WTi Indicates the installed capacity of wind turbine WTi, γ PVi,t Q represents the capacity factor of the photovoltaic unit PVi within the time interval t. PVi This indicates the installed capacity of the photovoltaic unit, PVi.

[0254] The independent energy storage unit model will participate in the market as an independent entity, with centralized optimization and scheduling by ISO. It will charge when the multi-energy generator market model generates too much electricity and discharge when there is high power load. During the scheduling cycle, the independent energy storage unit model only needs to provide marginal bid parameters and the expected start and end states of charge.

[0255] The objective function of the stand-alone energy storage unit model includes the operating cost function of the stand-alone energy storage unit over time interval t.

[0256] The operating cost of the independent energy storage unit within the time interval t is shown in formula (15):

[0257]

[0258] Among them, C IESi,t This represents the operating cost of the independent energy storage unit IESi within the time interval t. This represents the unit operating cost of the independent energy storage unit IESi. This represents the charging power of the independent energy storage unit IESi within the time interval t. This represents the discharge power of the independent energy storage unit IESi within the time interval t, where IESi represents the index of the independent energy storage unit, Ω. IES It represents a collection of independent energy storage units.

[0259] The constraints of the objective function of the independent energy storage unit model include: the charging rate constraint of the independent energy storage unit within time interval t; the discharging rate constraint of the independent energy storage unit within time interval t; the constraint that the independent energy storage unit will not charge and discharge simultaneously; the constraint of the maximum charging amount of the independent energy storage unit model per unit scheduling time; the constraint of the maximum discharging amount of the independent energy storage unit model per unit scheduling time; the constraint of the state of charge of the independent energy storage unit model; and the constraint that the state of charge of the independent energy storage unit model remains unchanged during the scheduling period.

[0260] The charging rate constraint of the independent energy storage unit within the time interval t is shown in formula (16):

[0261]

[0262] in, Q represents the charging state of the independent energy storage unit IESi within the time interval t, and is a 0-1 variable. IESi This indicates the installed capacity of the independent energy storage unit IESi.

[0263] The discharge rate constraint of the independent energy storage unit within the time interval t is shown in formula (17):

[0264]

[0265] in, This represents the discharge state of the independent energy storage unit IESi within the time interval t, and is a 0-1 variable.

[0266] The constraint that independent energy storage units cannot charge and discharge simultaneously is shown in formula (18):

[0267]

[0268] The maximum charging amount constraint for the independent energy storage unit model within a unit scheduling time is shown in formula (19):

[0269]

[0270] Where Δt represents the duration of the time interval, Qs IESi This indicates the maximum energy storage capacity of the independent energy storage unit IESi. This represents the state of charge (SOC) of the independent energy storage unit IESi within the time interval t-1. This indicates the charging efficiency of the independent energy storage unit IESi.

[0271] The maximum discharge limit of an independent energy storage unit within a unit dispatch time is shown in formula (20):

[0272]

[0273] in, This indicates the discharge efficiency of the independent energy storage unit IESi.

[0274] Maximum energy storage capacity Qs of independent energy storage units IESi As shown in formula (21):

[0275]

[0276] Among them, hIESi This indicates the maximum discharge duration of the independent energy storage unit IESi.

[0277] The state-of-charge constraints for the stand-alone energy storage unit model are shown in equation (22):

[0278]

[0279] in, This represents the state of charge (SOC) of the independent energy storage unit IESi within time interval t.

[0280] The state of charge (SOC) of the independent energy storage unit remains unchanged during the time interval t. As shown in formula (23):

[0281]

[0282] The constraint that the state of charge of the independent energy storage unit model remains unchanged during the scheduling period is shown in Equation (24):

[0283]

[0284] in, This indicates the initial state of charge of the independent energy storage unit IESi during the dispatch cycle. This indicates the state of charge of the independent energy storage unit IESi at the end of the dispatch cycle. This indicates the initial energy storage level of the independent energy storage unit IESi;

[0285] S3. A two-stage electricity market clearing model is constructed based on a multi-energy generating unit market model. This model comprises a first-stage model and a second-stage model. The first-stage model is a market planning and dispatching model, which is the main part of market operation and aims to minimize the overall cost of the multi-energy generating unit market model. Operating costs are determined by the marginal price submitted by the multi-energy generating unit market model. To reduce computational load, subsequent analyses will simulate the entire year's operation on typical days. The decision variables in the first stage include the type of generating unit construction, unit capacity constraints, and the optimal dispatching plan to meet load requirements. The second-stage model is a demand response model. Depending on the stimulus factors, users will respond in three ways: dynamic electricity price, marginal carbon emission factor (MEF), and average carbon emission factor (XEF).

[0286] The objective function of the market planning and scheduling model includes the cost objective function.

[0287] Specifically, the cost objective function of the market planning and scheduling model is shown in formula (25):

[0288] f1=min(C Inv +w tr COpe (25)

[0289] Where f1 represents the cost objective function of the market planning and scheduling model, and C Inv This indicates the annual amortization of investment costs, w tr C represents the conversion factor for operating costs. Ope This represents typical daily operating costs, multiplied by an annual conversion factor w. tr This can be converted into annual operating costs.

[0290] Annual amortization of investment cost C Inv As shown in formula (26):

[0291]

[0292] in, This represents the annual amortization cost of the thermal power generating unit model. This represents the annual amortization cost of the renewable energy unit model. This represents the annual amortization cost of the stand-alone energy storage unit model.

[0293] Annual amortization cost of thermal power generator unit model As shown in formula (27):

[0294]

[0295] Among them, κ g This represents the conversion factor for the annual investment cost of the thermal power generating unit model. N represents the unit investment cost of the thermal power unit Gi. Gi This indicates the installed capacity of the thermal power unit Gi, r Gi,t The variable is 0-1, representing the investment status of the thermal power unit within the time interval t.

[0296] Annual amortization cost of renewable energy unit model As shown in formula (28):

[0297]

[0298] Among them, κ wt This represents the conversion factor for the annual investment cost of wind turbine units. κ represents the unit investment cost of wind turbine WTi. pv This represents the conversion factor for the annual investment cost of photovoltaic units. This represents the unit investment cost of a photovoltaic (PV) unit, PVi.

[0299] Annual amortization cost of stand-alone energy storage unit model As shown in formula (29):

[0300]

[0301] Among them, κ ies This represents the conversion factor for the annual investment cost of the stand-alone energy storage unit model. This represents the unit investment cost of the independent energy storage unit IESi.

[0302] Typical daily operating cost C Ope As shown in formula (30):

[0303]

[0304] The time scales of annual amortized investment costs and annual operating costs are unified through a cost conversion factor, which is shown in formula (31) for the market planning and scheduling model.

[0305]

[0306] Where, μ Ope κ represents the annual operating equivalence factor, a represents the unit discount rate, y represents the equipment's service life, and κ represents the investment cost conversion factor.

[0307] The constraints of the cost objective function include constraints on the type of generating unit, unit capacity, the proportion of renewable energy generation, reliability, and reserve requirements.

[0308] The unit construction type constraints of the market planning and scheduling model are shown in formula (32):

[0309]

[0310] Among them, U G This indicates the number of firepower aircraft assembled.

[0311] The unit capacity constraint is shown in formula (33):

[0312]

[0313] in, REi represents the maximum permitted installed capacity of thermal power units in the regional power grid system, and Ω represents the renewable energy unit index. RE It represents a collection of renewable energy units. Q represents the maximum permitted installed capacity of renewable energy units within the regional power grid. IESi This represents the maximum permissible independent energy storage unit capacity within the regional power grid, where ρ represents the minimum proportion of energy storage units that can be configured.

[0314] The constraint on the proportion of renewable energy generation is shown in formula (34):

[0315]

[0316] Where, ρ New This indicates the power generation penetration rate of renewable energy units. This indicates the system load after demand response.

[0317] Market clearing must meet necessary constraints. Reliability constraints ensure real-time source-load balance. The reliability constraints are shown in formula (35):

[0318]

[0319] The standby constraint is used to address errors in load forecasting and renewable energy output forecasting, thereby improving the flexibility of the model. The standby constraint is shown in formula (36):

[0320]

[0321] Where ε represents the load reserve ratio. Pr represents the proportion of renewable energy generation reserves. REi,t This represents the power generation of the renewable energy unit REi within the time interval t.

[0322] Price-based demand response uses dynamic electricity pricing to minimize electricity costs, but ignores carbon emissions in the optimization. Marginal emission factor and average carbon emission factor-based demand response are based on carbon emissions and aim to minimize carbon emissions, while costs are ignored. Formulas (37)-(39) describe the differences between these three approaches.

[0323] The demand response model includes a second-stage function, which can be categorized into three types: a second-stage objective function based on a price-based demand response model, a second-stage objective function based on a marginal emission factor-based demand response model, and a second-stage objective function based on an average emission factor-based demand response model.

[0324] Specifically, the objective function of the second stage of the price-based demand response model is shown in formula (37):

[0325]

[0326] Where f2 represents the cost of electricity, λ t P represents the unit cost of electricity used within time interval t. l,t Indicates the initial system load. This indicates an increase in load. This indicates a reduction in load.

[0327] The second-stage objective function of the demand response model based on the marginal emission factor is shown in equation (38):

[0328]

[0329] Where f3 represents the emission reduction calculated using the marginal emission factor (MEF). t This represents the marginal emission factor within the time interval t.

[0330] Marginal emission factor (MEF) within time interval t t As shown in formula (39):

[0331]

[0332] The second-stage objective function of the demand response model based on the average carbon emission factor is shown in equation (40):

[0333]

[0334] Where f4 represents the carbon dioxide emission reduction calculated using the average emission factor, XEF t This represents the average carbon emission factor over time t.

[0335] The average carbon emission factor XEF over time t t As shown in formula (41):

[0336]

[0337] in, This represents the system load after demand response. As shown in formula (42):

[0338]

[0339] The constraints of the second-stage function include the constraints on the maximum and minimum adjustment amounts of the demand response model, the constraints on the load increase amount of the demand response model within time interval t, the constraints on the load reduction amount of the demand response model within time interval t, the constraints on the demand response model that the load cannot be increased and decreased simultaneously within the same time interval, and the constraints on the demand response model that the total daily load before and after the electricity consumption behavior adjustment remains basically unchanged within the dispatch cycle.

[0340] The constraints on the maximum and minimum adjustment amounts of the demand response model are shown in equation (43):

[0341]

[0342] in, The logic for determining whether a user's workload has increased is represented by a 0-1 variable. This indicates that the load limit can be adjusted within the time interval t. The logic for determining whether a user should reduce their load is a 0-1 variable, where δ represents the adjustable load ratio.

[0343] The constraint on the load increase within the time interval t in the demand response model is shown in Equation (44), indicating that the load increase of users based on the baseline load cannot exceed their maximum rated load.

[0344]

[0345] in, Indicates the system's maximum rated load.

[0346] The constraint on load reduction within time interval t in the demand response model is shown in formula (45), indicating that the load reduction of users in a certain period cannot exceed the baseline load of the current period:

[0347]

[0348] The constraint in the demand response model that the load cannot be increased and decreased simultaneously within the same time interval is shown in formula (46):

[0349]

[0350] The constraint that the total daily load remains basically unchanged before and after the electricity consumption behavior adjustment during the scheduling period in the demand response model is shown in formula (47):

[0351]

[0352] in, This indicates the daily change in electricity consumption;

[0353] S4. Input each demand response scheme into the two-stage electricity market clearing model in sequence to obtain the market operation parameter analysis results under each demand response scheme;

[0354] Specifically, step S4 includes the following steps:

[0355] S401. Input the parameters from the demand response plan into the first-stage model;

[0356] S402. Linearize the first-stage model with parameters from the input demand response scheme to obtain a linear first-stage model;

[0357] In step S402, the first-stage model with parameters from the input demand response scheme is linearized, including using continuous variable Z. t Replace Q IESi With s IESi,t The product of is shown in formula (48):

[0358]

[0359] Among them, Z t This indicates the substitution of a continuous variable, s IESi,t This represents the charge / discharge state of independent energy storage, and is a 0-1 variable.

[0360] Introduce variable M to constrain continuous variable Z t As shown in formula (49):

[0361]

[0362] Where M is set to 10000,

[0363] The annual operating cost of the thermal power unit is rewritten as a quadratic function and then linearized, as shown in formula (50):

[0364]

[0365] Among them, Qu Gi Co represents the coefficient of the quadratic term. Gi Denotes the coefficient of the linear term.

[0366] Quadratic coefficient Qu Gi As shown in formula (51):

[0367]

[0368] coefficient Co of the first term Gi As shown in formula (52):

[0369]

[0370] The annual operating cost of the thermal power unit is processed by piecewise linearization, as shown in formula (53):

[0371]

[0372] Among them, P Gi,t,k This represents the power generation of thermal power unit Gi in the k-th segment during time interval t, where k represents the k-th linear segment, ΩK represents the set of segments, and η Gi,k This represents the slope of the Gi piecewise linearized operating cost function for the thermal power unit in the k-th segment.

[0373] The output of the fire unit in sections is shown in formula (54):

[0374]

[0375] The constraint on the output of the fire unit in sections is shown in formula (55):

[0376]

[0377] Where K represents the number of linear segments;

[0378] S403. Obtain the minimum system cost through the linear first-stage model. Determine whether to perform demand response based on the minimum system cost. If demand response is performed, proceed to step S404; otherwise, proceed to step S406.

[0379] S404. Based on the characteristics of the demand scheme, input the parameters of the demand response scheme into the second-stage model, and obtain the minimized electricity cost or minimized carbon emissions through the second-stage model;

[0380] S405. Adjust the load curve by minimizing electricity costs or carbon emissions and update the load reduction and load increase parameters. Determine whether the updated load reduction and load increase parameters meet the convergence condition. If they do, proceed to step S406. If they do not, use the updated load reduction and load increase parameters as parameters in the updated demand response scheme and then execute step S401. The convergence condition is that the user load can only change once within any time interval t, as shown in equation (56):

[0381]

[0382] Among them, Ω n A set representing the number of iterations;

[0383] S406. Output the market operation parameter analysis results of the demand response plan;

[0384] S407. Repeat steps S401-S406 until the market operation parameter analysis results of all demand response schemes are obtained;

[0385] S5. Market performance analysis and demand response decisions are obtained by comparing the results of all market operation parameter analysis obtained in step S4.

[0386] Example 3

[0387] In this embodiment, a case study was tested in a regional power grid system using the proposed power demand-side management optimization method based on electric-carbon coupling and dynamic carbon emissions. The results were compared across four scenarios:

[0388] Scenario 1: BAU scenario. In the first phase, the model optimizes the scheduling plan based on the lowest system cost. In the second phase, the model does not adopt any demand response scheme.

[0389] Scenario 2: The first-stage model optimizes the scheduling plan based on the lowest system cost, while the second-stage model runs a price-based demand response (PBDR) based on dynamic prices.

[0390] Scenario 3: The first-stage model optimizes the scheduling plan based on the lowest system cost, and the second-stage model runs a low-carbon demand response based on marginal carbon emission factors (MEF-based Demand Response, MBDR).

[0391] Scenario 4: The first-stage model optimizes the scheduling plan based on the lowest system cost, and the second-stage model runs a low-carbon demand response (XEF-based Demand Response, XBDR) based on the average carbon emission factor.

[0392] The Australian Energy Market Operator website provided the load data used in this embodiment, and the open-source research report provided data on the output of photovoltaic and wind turbine units. Detailed information on the characteristic parameters of the thermal power units in this embodiment is shown in Table 1.

[0393] Table 1 Detailed Information on Characteristic Parameters of Thermal Power Units

[0394]

[0395] The specifications of the thermal power units used in this embodiment represent the main operating units currently, as shown in the table. and The coefficients of the fitting curve between the unit's dynamic carbon emission factor and power are used to quantitatively describe the role of the unit's dynamic characteristics in market operation. The economic and technical parameters of renewable energy and stand-alone energy storage units are shown in Table 2.

[0396] Table 2 Economic and technical parameters of renewable energy units and independent energy storage units

[0397]

[0398] It should be noted that the investment costs in Table 1-2 are amortized over 5 years at a 7% interest rate.

[0399] Figure 3This shows the differences in the types and capacities of market unit investment under different demand response scenarios. Scenario 2 has the lowest installed capacity, decreasing by 3.56% compared to Scenario 1. Scenario 3-4, however, increase installed capacity by 3.04% and 2.30%, respectively. It can be observed that demand response does not change the composition of basic power generation units, such as Gen2 and Gen3, which consistently appear with the same capacity in each scenario. The changes in installed capacity mainly stem from changes in the investment capacity of Gen6 and gas turbine units, fluctuating between these two investment capacities in Scenario 1-4. This is related to their potential marginal position in unit output. Compared to Scenario 1, Scenario 2 shifts investment from Gen7 and Gen9 to Gen6, increasing the capacity share of coal-fired power units in thermal power units by 3.8%. Scenario 3-4, building on Scenario 1, further increase investment in Gen7, increasing the capacity share of gas turbine units in thermal power units by 2.51% and 1.56%, respectively.

[0400] The changes in investment capacity for renewable energy units and stand-alone energy storage units are not significant compared to those for thermal power units; the rate of change in investment capacity for scenarios 2-4 does not exceed 1%. Due to the low generation capacity factor, renewable energy units account for the highest proportion of installed capacity in each scenario. Furthermore, across all scenarios, the investment capacity of stand-alone energy storage units is on average 2.18% higher than the minimum energy storage ratio (set at 20% in these scenarios), which is particularly evident in scenario 2. This indicates that the electricity-carbon coupling market can increase the operational capacity of stand-alone energy storage units in the electricity market through market-based mechanisms, complementing thermal power units and enhancing the flexibility of the electricity market.

[0401] To further explain the changes in the unit portfolio and its response to market conditions, this embodiment will analyze the timing of the unit's operation within the scheduling cycle. Figure 4 This demonstrates the changes in unit output and load shifting under different demand response scenarios. In Scenario 1... Figure 4 (a) Under the electricity-carbon market coupled with a market mechanism, renewable energy units with near-zero marginal costs have priority in bidding and are fully absorbed by the market. Independent energy storage units charge and discharge appropriately during peak output and load periods of renewable energy units to make up for load gaps. Among them, coal-fired power units with lower overall marginal costs, such as Gen2 and Gen3, will be prioritized for elimination. Due to their inflexibility, they always provide stable output. Coal-fired power units Gen6 and gas-fired power units Gen7 and Gen9, with higher overall marginal costs, are used as marginal units during the dispatch cycle due to their better flexibility. Their output is adjusted more frequently to adapt to the constantly changing load curve. For ease of description later, the time period when gas-fired power units are used as marginal units is defined as the peak load period (6:00-8:00 and 16:00-24:00), and the remaining time period is defined as the valley load period.

[0402] In a price-based DR scenario, such as Figure 4 (b) The load curve adjusts along the market clearing price. It can be observed that the price trend is similar to the load curve; therefore, the price-based DR reduces the peak load (18:00-24:00) by an average of 267.92MW and shifts it to the valley load (10:00-15:00 and 1:00-5:00). The reduction in peak load leads to the cancellation of investments in the original marginal units Gen7 and Gen9, with the marginal units being transferred to the coal-fired unit Gen6. Simultaneously, the output of independent energy storage units and other units decreases to varying degrees, with the imbalance being met by Gen6, resulting in an 87.08% increase in Gen6's power generation. At the same time, the peak-shaving and valley-filling effect reduces the demand for independent energy storage units, ultimately reducing power generation by 1.12% compared to Scenario 1.

[0403] like Figure 4 As shown in (c), in contrast to Scenario 2, MEF-based DR will cause a reverse change in load, with electricity load shifting from off-peak to peak periods. The increased load during the MEF off-peak period is primarily balanced by the addition of larger-capacity, cleaner marginal units (Gen7) and independent energy storage units, while some Gen9 investments are cancelled due to insufficient capacity to meet the increased load. The decreased load during the MEF peak period is primarily balanced by reduced power generation from thermal power units and increased charging power from energy storage; any remaining imbalance is supplemented by increased output from renewable energy units. Unlike Scenario 2, the increased peak load of MEF-based DR increases the discharge demand of independent energy storage units, ultimately resulting in a 1.12% increase in power generation compared to Scenario 1.

[0404] like Figure 4 As shown in (d), XEF-based DR exhibits the dual effects of Scenario 2 and Scenario 3. Specifically, XEF-based DR will cause a load "peak shaving and valley filling" effect during periods of high renewable energy unit penetration, such as 9:00-15:00, and may cause even higher peak loads during peak load periods (17:00-20:00) when thermal power unit penetration is high. Therefore, the final operating condition of the units in XEF-based DR depends on the difference between the effects of Scenario 1 and Scenario 2, and the final power generation is basically unchanged compared to Scenario 1.

[0405] Figure 5This paper comprehensively summarizes the changes in load curves under different demand response schemes. The original load curves exhibit typical peak and valley periods. The peak-valley difference rate is used to quantify the implementation effect of different demand response schemes. As previously analyzed, Price-based DR has a significant peak-shaving and valley-filling effect, with the peak-valley difference rate decreasing by 7.08% compared to Scenario 1. Scenario 3 and Scenario 4 both improve the peak-valley difference rate by 6.41%, meaning that the load curves at this time have higher peak values ​​and lower peak-valley values. However, it is worth noting that Scenario 4 increases the load by an average of 232.68MW during some valley periods (9:00-15:00) to achieve a compromise between Scenario 2 and Scenario 3.

[0406] This embodiment also summarizes the overall impact of different demand response schemes on market operation under the electric-carbon coupling scenario, as shown in Table 3.

[0407] Table 3 Comparison of Market Operations under Different Demand Response Schemes

[0408]

[0409] Note: The percentages in parentheses in the table represent the growth rate of the current scenario relative to the baseline scenario.

[0410] As shown in Table 3, compared to Scenario 1, Price-based DR reduces both market clearing price and system cost, but increases carbon emissions by 3.32% and significantly reduces unit profitability. MEF-based DR is the complete opposite, increasing total system cost by 1.49% while reducing carbon emissions by 0.79%, with slight increases in market price and unit profitability. XEF-based DR continues to strike a balance between the two, with no significant changes in carbon emissions and system cost, but reducing market electricity price and unit profitability by 2.99% and 6.49% respectively, meaning users may need to bear lower electricity costs.

[0411] These results correspond to the aforementioned unit portfolio and timing of operation. According to the market clearing principle of electricity-carbon coupling, gas-fired units, which have higher marginal costs but are cleaner, will be dispatched after coal-fired units. This results in the clearing order of units being unrelated to emissions factors. In this example system, marginal gas-fired units such as Gen7 determine the electricity price level during peak load periods, while marginal coal-fired units such as Gen6 determine the electricity price level during off-peak periods. Therefore, the price curve and load curve trend in the same direction, but the MEF curve trends in the opposite direction. When Price-based DR is implemented, the peak-shaving and valley-filling effect eliminates investment in gas-fired units during peak load periods, while the transferred peak load is balanced by higher-emission coal-fired units, resulting in reduced power generation. Therefore, Price-based DR will lower market electricity prices, reduce costs, but increase carbon emissions. When MEF-based DR is implemented, the market operation is completely opposite. The load aggregated during peak load periods will lead to the construction of more gas-fired units and more power generation, while XEF-based DR, as a compromise, always results in a market operation outcome between the two.

[0412] It can be observed that:

[0413] (1) Under the scenario of electric carbon coupling, the impact of demand response schemes based on different stimuli on load transfer varies greatly. Price-based DR has the effect of peak shaving and valley filling, but MEF-based DR will lead to higher peak loads.

[0414] (2) In the scenario of electricity-carbon coupling, demand response schemes based on different stimulus drivers affect market operation by changing the marginal unit portfolio and reshaping the operating status of almost all units. Price-based DR and MEF-based DR have incompatible trade-offs between cost and emissions.

[0415] (3) XEF-based DR can be regarded as a compromise between Price-based DR and MEF-based DR. Its implementation effect is between the two in terms of load transfer, unit investment and operation, and economic environment impact.

[0416] (4) The above conclusions are robust until the value performance of units, especially marginal units, is fully correlated with emission factors, due to the clearing rules coupled to the carbon electricity market.

[0417] This embodiment also discusses the impact of dynamic carbon emission characteristics of the unit. As a comparison with the dynamic carbon emission scenario, this embodiment sets the unit carbon emission parameters to static during model operation, while keeping other parameter settings unchanged. Table 4 shows the market operation performance of different demand response schemes under the static carbon emission factor.

[0418] Table 4. Comparison of Market Operations under Different Demand Response Schemes (Static Emission Factors)

[0419]

[0420] Note: The percentages in parentheses in the table represent the growth rate of the current scenario relative to its dynamic scenario.

[0421] Observations show that market operation is similar to the dynamic scenario in terms of both cost and emissions. Scenario 4 continues to achieve a compromise between Scenario 2 and Scenario 3, indicating that the unit portfolio and dispatch operation mode are similar to the dynamic scenario under the static carbon emission factor scenario. However, due to the switching of unit emission characteristics, the impact of different demand response schemes on the electricity market varies significantly. Specifically, compared to the dynamic carbon emission intensity scenario (DCEI), the static carbon emission intensity scenario (SCEI) increases market prices, system costs, and unit profitability by an average of 6.67%, 3.21%, and 7.74%, respectively, but also increases carbon emissions by an average of 0.41%, with the most significant increase in Scenario 1, reaching a growth rate of 2.03%. It is worth noting that the average investment cost decreased by -0.38% across all scenarios.

[0422] To better explain the changes in the above market operating parameters, Figure 6 This paper summarizes the impact of static emission factors on different demand response schemes from the perspective of unit portfolio and operation. Scenario 1 is used as an example. Figure 6 (a) For example, the Static Emission Intensity Scenario (SCEI) causes the investment and load of gas turbine Gen9 units to be entirely transferred to coal-fired power units Gen6. Since the ability to reduce the real-time dynamic carbon emission factor by increasing power load is ignored, the overall cost of Gen9 is no longer advantageous compared to Gen6. Figure 7 This further demonstrates the impact of DCEI on the marginal carbon emissions of generating units. Meanwhile, due to the inflexibility of fixed marginal carbon costs, more power generation will no longer lead to lower operating costs, and the market will tend to limit unnecessary investment in generating units and power generation, thus reducing the activity of independent energy storage units. Calculations show that the market installed capacity decreased by 115MW, and the total system power generation decreased by 1760MWh, corresponding to a reduction in IES discharge during peak load periods. The decrease in installed capacity and the shift of load to higher carbon emission units resulted in lower investment costs but increased carbon emissions. On the other hand, the Static Emission Intensity Scenario (SCEI) increases market prices and operating costs, which is related to the transmission of more carbon costs. The carbon cost transmission rate is calculated using equation (57):

[0423] τ c =(λ c -λ a )÷pr c ×100% (57)

[0424] Where, τ c λ represents the carbon cost transmission rate. c λ represents the market price with a carbon market. a This indicates the market price without a carbon market.

[0425] It can be observed that the carbon cost transmission rate of scenarios 1-4 is 10.14% higher on average than that of the dynamic scenarios.

[0426] Similar patterns can also be found in scenarios 2-4, such as... Figure 6 bd, more load is shifting to units that are more advantageous under the static scenario, while the activity of energy storage decreases, indicating the universality of the Static Carbon Intensity Scenario (SCEI). For example, in Figure 6 In scenario b, the 883MW of power from Gen6 is more advantageous under the Static Emission Intensity Scenario (SCEI) than under Gen2 and Gen3. It is noteworthy that the investment and load shifts caused by the SCEI under different demand response schemes are not robust to unit profitability and carbon emissions. For example, scenarios 1 and 3 reduce unit profitability, while scenarios 2 and 4 have the opposite effect, indicating the uncertainty of the SCEI effect.

[0427] It can be observed that:

[0428] (1) The impact of different demand response methods on the electricity market under the static carbon emission intensity scenario (SCEI) is universal, but the degree of impact varies significantly, manifesting as higher costs and emissions in this regional system.

[0429] (2) The Static Emission Intensity Scenario (SCEI) specifically affects market investment and operation by ignoring the dynamic emission characteristics of generating units. Ignoring these effects may prevent decision-makers from seeing the true situation and leading to erroneous decisions. On the one hand, SCEI inhibits investment and operation of units with dynamic advantages rather than static advantages, potentially shifting investment and power generation to units with higher carbon emissions, thus triggering carbon emission risks. On the other hand, the inflexibility of marginal costs of generating units under SCEI may limit system power generation and installed capacity, and reduce the activity of independent energy storage (IES) devices. Finally, SCEI may pass on more ineffective carbon costs to the market, potentially increasing system operating costs and causing uncertainty in industry profits.

[0430] like Figure 7As shown, Dynamic Emission Intensity Scenario (DCEI) allows units to flexibly adjust their pricing by changing their output to gain greater competitiveness. However, with the application of Static Emission Intensity Scenario (SCEI), the marginal carbon cost of a unit becomes a fixed value. Units with relatively strong dynamic performance may lose competitiveness in the market, and their investment and power generation performance will be reshaped by units that are more advantageous under SCEI. For example, if a 600MW load is needed at a certain time, Gen3 is more advantageous than Gen2 under DCEI, but it is no longer advantageous under SCEI, and the load will be shifted.

[0431] This embodiment also discusses the potential impact of carbon market coupling strength on the effectiveness of demand response implementation. Carbon markets typically alter the coupling strength with the electricity market by regulating emission benchmarks for thermal power units and carbon prices. As energy transition progresses, future carbon prices will gradually rise above 900 yuan, and emission benchmarks for thermal power units will continue to tighten, leading to a continuous increase in coupling strength. Table 5 provides a detailed simulation of this process using 10 scenarios (C1-C10), where C1 represents the current carbon market situation.

[0432] Table 5 Scenario Setting for Carbon Market Coupling Strength

[0433]

[0434] This embodiment comprehensively evaluates the impact of carbon market coupling strength on the effectiveness of different demand response schemes, such as market prices, carbon emissions, system costs, and unit profitability, and other market operating parameters. Figure 8 As shown, all curves generally go through three typical stages: a plateau period, a period of change, and a convergence period, but the duration of each stage varies. With increasing coupling strength, price-based DR can ultimately achieve a unified goal of reducing carbon emissions, lowering system costs, and improving unit profitability while ensuring that market prices do not fluctuate significantly. This may suggest that the order in which units exit the power market is decoupled from their performance is becoming directly related to emissions, influenced by the increasing coupling strength of the carbon market. Figure 9The correlation coefficient trends calculated under different carbon market coupling intensities validate this observation. This trend will lead to more carbon-intensive units being moved to the margins of power supply, where they can then be replaced by Price-based DR, thus enhancing the overall benefits of Price-based DR implementation. However, the carbon emission reduction and unit profitability improvement effects of MEF-based DR and XEF-based DR will gradually decrease, while system costs will not decrease significantly during the intensity increase period, potentially indicating limitations in the scope of MEF-based DR and XEF-based DR implementation. But this process is gradual, and the curves will undergo multiple changes and even reverse fluctuations.

[0435] It can be observed that:

[0436] The strength of carbon market coupling significantly impacts the effectiveness of different demand response schemes. It improves the effectiveness of PBDRs by enhancing the correlation between marginal costs and emissions in the electricity market clearing sequence, but may simultaneously limit the applicability of MBDRs and XBDRs. However, these changes are gradual and slow, not instantaneous.

[0437] Market-based demand response schemes are crucial for building sustainable power systems. They enable better control over system flexibility resources, thus contributing to more efficient, flexible, and sustainable system operation. To implement more effective market-based demand response schemes, this study proposes an extended two-stage market clearing model. It also compares and analyzes the quantitative impacts of three market-based demand response schemes—Price-based DR, MEF-based DR, and XEF-based DR—on the operation of the electricity-carbon coupling market, drawing the following important conclusions:

[0438] (1) Under the current simulation scenario, Price-based DR simultaneously reduces peak market load, electricity costs, and unit profits, but leads to higher carbon emissions. MEF-based DR is the opposite, while XEF-based DR is a compromise between the two. However, as the carbon market coupling strength increases, Price-based DR can eventually achieve multiple objectives of decarbonization, economics, and unit profitability simultaneously, while the carbon emission reduction effects of MEF-based DR and XEF-based DR will gradually decrease, limiting their application. These changes can ultimately be explained by the fact that the value performance of units in the electricity market is gradually becoming fully correlated with emissions, but this process is slow and fluctuating.

[0439] (2) Sensitivity analysis of the dynamic emission characteristics of generating units shows that ignoring the dynamic carbon emission characteristics of generating units in the electricity-carbon coupling market may affect the global optimality of demand-side flexible resource scheduling. On the one hand, ignoring this will reduce the economic efficiency of market operation and weaken the activity of energy storage units. On the other hand, a lack of understanding of dynamic carbon emission factors may lead to erroneous investments, triggering carbon emission risks and uncertainty in unit profitability.

[0440] It is evident that the effectiveness of demand response schemes is closely related to the coupling strength with the carbon market and the market planning and dispatching mechanism. The construction of a sustainable power system should make greater efforts to optimize the deployment of market-based demand response schemes and achieve better demand-side flexibility resource management. Therefore, the following recommendations are proposed:

[0441] (1) Steadily advance the reform of the electricity market and the construction of a carbon quota allocation mechanism. From a fundamental and long-term perspective, only when the construction of the electricity market and the regulation of reasonable emission benchmarks are coordinated to unify the market clearing sequence of generating units with the emission sequence, making their costs related to emissions, will carbon-intensive generating units be completely transferred to the margin of supply and replaced by price-based DR, thereby simultaneously achieving multiple objectives. However, this process is lengthy and volatile, which also explains the current instability in the assessment of the environmental benefits of demand response.

[0442] (2) Dynamic emission characteristics of generating units should be considered in electricity market planning and dispatching. Dynamic emission characteristics allow thermal power units to flexibly participate in the coupled market by adjusting their output, enhancing the market activity of energy storage units while reducing market carbon emission risks and improving economic efficiency. Therefore, it is recommended that dynamic emission characteristics of generating units be considered in market planning and dispatching to maximize the global optimality of demand-side flexible resource dispatching.

[0443] (3) Strengthen the digitalization of the power system and establish reliable and stable two-way communication capabilities between the market and the demand side. The implementation of market-based demand response schemes relies on real-time communication transmission and information feedback, such as market prices, carbon emission factor releases, and load adjustment feedback. It is recommended to strengthen the development of smart grids and communication infrastructure to ensure that two-way communication is always available.

[0444] This invention establishes a unified analytical framework for demand response under the coupling of the electricity carbon market and designs a two-stage clearing model for real-time coupling of electricity carbon with integrated demand response procedures. Corresponding to real-world scenarios, it first quantifies the cross-market transmission and coupling strength of carbon costs, clarifies the ways in which emerging players participate in the market, and uses regression functions to quantitatively express the dynamic emission characteristics of generating units. Secondly, based on the market clearing model, it extends the model to consider how different demand response schemes affect the objectives and constraints of market operation. Finally, to improve solution efficiency, the model is transformed into a mixed-integer linear programming problem. Using real-world data, three typical market-based demand response schemes—Price-based DR, MEF-based DR, and XEF-based DR—are simulated and compared. The comparison results are multifaceted, involving market investment and operation, peak-valley ratio, cost, and emissions. We found that different demand response schemes have significantly different impacts on coupled market operation, and there are incompatible trade-offs between market operation parameters such as cost and emissions. Simultaneously, a sensitivity analysis experiment was designed to clarify the impact of dynamic emission characteristics of generating units and the coupling strength of the carbon market on the effectiveness of demand response implementation, providing support for guiding demand response implementation. This invention avoids the "black box" drawbacks of using macroscopic parameters to study the impact of demand response. The method presented in this invention comprehensively reveals the internal flows of power generation, emissions, and economic variables in the market under the influence of demand response, elucidating the impact of demand response on market operation from a bottom-up perspective and providing policymakers with more realistic information. Furthermore, this invention theoretically analyzes how demand response interacts with the market under an electricity-carbon coupling scenario, and simultaneously compares different market-based demand response schemes. The established general model may pave the way for many future directions, such as energy storage dispatch and policy simulation.

[0445] The embodiments of the present invention are given for illustrative and descriptive purposes only, and are not intended to be exhaustive or to limit the invention to the forms disclosed. Many modifications and variations will be apparent to those skilled in the art. The embodiments were chosen and described in order to better illustrate the principles and practical application of the invention, and to enable those skilled in the art to understand the invention and to design various embodiments with various modifications suitable for a particular purpose.

Claims

1. A method for optimizing electricity demand-side management based on electricity-carbon coupling and dynamic carbon emissions, characterized in that, Includes the following steps: S1. Obtain multiple demand response solutions and multi-energy generator data in the electricity market; S2. Construct a multi-energy generator set market model based on the multi-energy generator set data in the electricity market; S3. Construct a two-stage electricity market clearing model based on the multi-energy generator market model. The two-stage electricity market clearing model includes a first-stage model and a second-stage model. The first-stage model is a market planning and scheduling model, and the second-stage model is a demand response model. S4. Input each of the demand response schemes into the two-stage electricity market clearing model in sequence to obtain the market operation parameter analysis results under each demand response scheme; S5. By comparing the analysis results of all the market operation parameters under each of the aforementioned demand response schemes, market performance analysis and demand response decisions are obtained; Among them, the multi-energy generator set market model includes thermal power generator set model, renewable energy generator set model and independent energy storage generator set model; The objective function of the thermal power generating unit model includes the marginal generation cost function of bidding considering the coupling of the electricity carbon market under dynamic carbon emission intensity, the carbon dioxide emission function of the thermal power unit within time interval t, and other cost functions of the thermal power unit participating in market bidding within time interval t. The constraints of the objective function of the thermal power generating unit model include the constraints of the no-load cost of the thermal power unit within time interval t, the constraints of the start-up cost of the thermal power unit within time interval t, the constraints of the shutdown cost of the thermal power unit within time interval t, the output limit constraint of the thermal power unit, the ramp rate constraint of the thermal power unit, and the minimum switching time constraint of the thermal power unit. The objective function of the renewable energy unit model includes the operating cost function of the renewable energy unit within the time interval t; the constraints of the objective function of the renewable energy unit model include the output constraints of the renewable energy unit. The objective function of the independent energy storage unit model includes the operating cost function of the independent energy storage unit within time interval t; the constraints of the objective function of the independent energy storage unit model include the charging rate constraint of the independent energy storage unit within time interval t, the discharging rate constraint of the independent energy storage unit within time interval t, the constraint that the independent energy storage unit will not charge and discharge simultaneously, the constraint of the maximum charging amount of the independent energy storage unit model per unit scheduling time, the constraint of the maximum discharging amount of the independent energy storage unit within unit scheduling time, the state of charge constraint of the independent energy storage unit model, and the constraint that the state of charge of the independent energy storage unit model remains unchanged within the scheduling period. Wherein: the objective function of the market planning and scheduling model includes a cost objective function; the constraints of the objective function of the market planning and scheduling model include constraints on the type of unit construction, unit capacity, the proportion of renewable energy generation, reliability, and reserve. Wherein: the demand response model includes a second-stage function, which includes a second-stage objective function based on a price-based demand response model, a second-stage objective function based on a marginal emission factor-based demand response model, and a second-stage objective function based on an average emission factor-based demand response model; the constraints of the second-stage function include the constraints of the maximum and minimum adjustment amounts of the demand response model, the constraints of the load increase amount of the demand response model within the time interval t, the constraints of the load reduction amount of the demand response model within the time interval t, the constraints that the demand response model cannot simultaneously increase and decrease the load within the same time interval, and the constraints that the total daily load before and after the electricity consumption behavior adjustment remains basically unchanged within the dispatch cycle.

2. The method for optimizing electricity demand-side management based on electricity-carbon coupling and dynamic carbon emissions according to claim 1, characterized in that, In step S1, the demand response scheme includes market operation and policy parameters and demand-side parameters. The market operation and policy parameters include coupled market dispatch configuration data, carbon price and quota data, renewable energy combination standard data, and energy storage configuration data. The demand-side parameters include user type data, baseline load data, regulation cost data, and regulation capacity data. The multi-energy generator set data includes unit data and techno-economic parameters. The unit data includes new energy unit data, gas unit data, coal-fired unit data, and electrochemical energy storage data. The techno-economic parameters include investment cost data, operating cost data, dynamic carbon emission factor data, and unit supply constraint data.

3. The method for optimizing electricity demand-side management based on electricity-carbon coupling and dynamic carbon emissions according to claim 1, characterized in that, In step S2, the thermal power generator model includes multiple thermal power units, the renewable energy unit model includes multiple renewable energy units, the renewable energy units include photovoltaic units and wind turbine units, and the independent energy storage unit model includes multiple independent energy storage units.

4. The method for optimizing electricity demand-side management based on electrical-carbon coupling and dynamic carbon emissions according to claim 3, characterized in that, in, The marginal generation cost function for bidding under the dynamic carbon emission intensity considering the coupling of the electricity carbon market is shown in formula (1): （1） in, This indicates the number of thermal power units within time interval t. marginal cost This indicates the number of firepower units within time interval t. Power generation capacity, Indicates thermal power unit The binomial coefficients of the fuel cost function, Indicates thermal power unit The coefficients of the first-order term of the fuel cost function, Indicates thermal power unit The proportion of fuel costs in the total cost of power generation, Indicates the fire unit The coefficients of the first-order term of the fitting curve between dynamic carbon emission factor and power. Indicates thermal power unit The constant of the fitting curve between dynamic carbon emission factor and power. Indicates thermal power unit The emission benchmark value, Indicates carbon price, Indicates the index of the thermal power unit. Indicates a collection of fire-fighting units. Indicates time interval, Represents the set of scheduling times. The carbon dioxide emissions of the thermal power unit during the time interval t are shown in formula (2): （2） in, This indicates the number of firepower units within time interval t. Carbon dioxide emissions. The other cost functions for the fire unit's participation in market declaration within the time interval t are shown in formula (3): （3） in, This indicates the number of firepower units within time interval t. Other costs, This indicates the number of firepower units within time interval t. Declare empty load costs, This indicates the number of firepower units within time interval t. The application start-up cost, This indicates the number of firepower units within time interval t. The cost of declaring shutdown The constraint on the declared no-load cost of the thermal power unit within the time interval t is shown in formula (4): （4） in, Indicates thermal power unit The cost of no-load operation This indicates the number of firepower units within time interval t. The running state is a 0-1 variable. The constraint on the application start-up cost of the thermal power unit within the time interval t is shown in formula (5): （5） in, Indicates thermal power unit Startup costs, This indicates the number of thermal power units within the time interval t-1. The running status, The constraint on the reported shutdown cost of the thermal power unit within the time interval t is shown in formula (6): （6） in, Indicates thermal power unit Downtime costs, The power output limit constraint of the thermal power unit is shown in formula (7): （7） in, Indicates thermal power unit Minimum power generation limit Indicates thermal power unit Maximum power generation limit The ramp rate constraint of the thermal power unit is shown in formula (8): （8） in, This indicates the number of thermal power units within the time interval t-1. Power generation capacity, Indicates thermal power unit Climbing speed limit, Indicates thermal power unit Downhill speed limit, The minimum switching time constraint of the thermal power unit is shown in formula (9): （9） in, Indicates the scheduling period. Indicates the time interval recording point. This indicates the fire unit at the time interval recording point. Running status Indicates thermal power unit Minimum startup time, Indicates thermal power unit The minimum stopping time.

5. The method for optimizing electricity demand-side management based on electricity-carbon coupling and dynamic carbon emissions according to claim 3, characterized in that, in, The operating cost function of the renewable energy unit within the time interval t is shown in formula (10): （10） in, This represents the operating cost of a renewable energy unit within a time interval t. Indicates wind turbine The unit operating cost Indicates the index of the wind turbine unit. This indicates the number of wind turbine units within the time interval t. Power generation capacity, Indicates photovoltaic unit The unit operating cost Indicates the index of the photovoltaic unit. This indicates the photovoltaic unit within the time interval t. Power generation capacity, This represents a collection of wind turbine units. Indicates a collection of photovoltaic units. The output constraint of the renewable energy unit is shown in formula (11): （11） in, This indicates the number of wind turbine units within the time interval t. capacity factor, Indicates wind turbine The installed capacity, This indicates the photovoltaic unit within the time interval t. capacity factor, Indicates photovoltaic unit The installed capacity.

6. The method for optimizing electricity demand-side management based on electro-carbon coupling and dynamic carbon emissions according to claim 3, characterized in that, in, The operating cost function of the independent energy storage unit within the time interval t is shown in formula (12): （12） in, This indicates the number of independent energy storage units within time interval t. Operating costs Indicates independent energy storage unit The unit operating cost This indicates the number of independent energy storage units within time interval t. The charging power, This indicates the number of independent energy storage units within time interval t. The discharge power, Indicates an index for an independent energy storage unit. It represents a collection of independent energy storage units. The charging rate constraint of the independent energy storage unit within the time interval t is shown in formula (13): （13） in, This indicates the number of independent energy storage units within time interval t. The charging state is a variable between 0 and 1. Indicates independent energy storage unit The installed capacity, The discharge rate constraint of the independent energy storage unit within the time interval t is shown in formula (14): （14） in, This indicates the number of independent energy storage units within time interval t. The discharge state is a 0-1 variable. The constraint that the independent energy storage unit cannot charge and discharge simultaneously is shown in formula (15): （15） The maximum charging amount constraint of the independent energy storage unit model within a unit scheduling time is shown in formula (16): （16） in, Indicates the duration of the time interval. Indicates independent energy storage unit Maximum energy storage capacity, This indicates the number of independent energy storage units within the time interval t-1. The state of charge, Indicates independent energy storage unit Charging efficiency, The maximum discharge limit of the independent energy storage unit within a unit scheduling time is shown in formula (17): （17） in, Indicates independent energy storage unit The discharge efficiency, The independent energy storage unit Maximum energy storage capacity As shown in formula (18): （18） in, Indicates independent energy storage unit Maximum discharge duration, The state of charge constraint of the independent energy storage unit model is shown in formula (19): （19） in, This indicates the number of independent energy storage units within time interval t. The state of charge, The state of charge of the independent energy storage unit within the time interval t As shown in formula (20): （20） The constraint that the state of charge of the independent energy storage unit model remains unchanged during the scheduling period is shown in Equation (21): （21） in, This indicates the initial independent energy storage unit during the scheduling cycle. The state of charge, This indicates that the independent energy storage unit will be in operation after the scheduling cycle ends. The state of charge, Indicates independent energy storage unit The initial energy storage level.

7. The method for optimizing electricity demand-side management based on electric carbon coupling and dynamic carbon emissions according to claim 6, characterized in that, in, The cost objective function of the market planning and scheduling model is shown in formula (22): （22） in, This represents the cost objective function of the market planning and scheduling model. This indicates the annual amortization of investment costs. This represents the conversion factor for operating costs. This represents typical daily operating costs. The annual amortization investment cost As shown in formula (23): （23） in, This represents the annual amortization cost of the thermal power generating unit model. This represents the annual amortization cost of the renewable energy unit model. This represents the annual amortization cost of the stand-alone energy storage unit model. Annual amortization cost of the thermal power generating unit model As shown in formula (24): （24） in, This represents the conversion factor for the annual investment cost of the thermal power generating unit model. Indicates thermal power unit The unit investment cost Indicates thermal power unit The installed capacity, The variable is 0-1, representing the number of thermal power units within the time interval t. The investment status, Annual amortization cost of the renewable energy unit model As shown in formula (25): （25） in, This represents the conversion factor for the annual investment cost of wind turbine units. Indicates wind turbine The unit investment cost This represents the conversion factor for the annual investment cost of photovoltaic units. Indicates photovoltaic unit The unit investment cost Annual amortization cost of the independent energy storage unit model As shown in formula (26): （26） in, This represents the conversion factor for the annual investment cost of the stand-alone energy storage unit model. Indicates independent energy storage unit The unit investment cost Typical daily operating costs As shown in formula (27): （27） in, This represents the operating cost of a renewable energy unit within a time interval t. The cost conversion coefficients of the market planning and scheduling model are shown in formula (28): （28） in, This represents the annual operating equivalence coefficient. This represents the unit discount rate. Indicates the lifespan of the equipment. This represents the investment cost conversion factor. The constraints on the type of unit construction are shown in formula (29): （29） in, This indicates the number of firepower aircraft assembled. The unit capacity constraint is shown in formula (30): （30） in, This indicates the maximum permissible installed capacity of thermal power units in the regional power grid system. Indicates an index of renewable energy units. It represents a collection of renewable energy units. This indicates the maximum permitted installed capacity of renewable energy units within the regional power grid. This indicates the maximum permissible independent energy storage unit capacity within the regional power grid. This indicates the minimum proportion of energy storage units required for configuration. The constraint on the proportion of renewable energy generation is shown in formula (31): （31） in, This indicates the power generation penetration rate of renewable energy units. This indicates the system load after demand response. The reliability constraints are shown in formula (32): （32） The backup constraint is shown in formula (33): （33） in, Indicates the load reserve ratio. Indicates the proportion of renewable energy generation reserves. This indicates the number of renewable energy units within the time interval t. Power generation capacity.

8. The method for optimizing electricity demand-side management based on electric carbon coupling and dynamic carbon emissions according to claim 1, characterized in that, The second-stage objective function of the price-based demand response model is shown in formula (34): （34） in, Indicates electricity cost, This represents the unit cost of electricity within the time interval t. Indicates the initial system load. This indicates an increase in load. This indicates a reduction in load. The second-stage objective function of the demand response model based on marginal emission factors is shown in formula (35): （35） in, This represents the emission reduction calculated using the marginal emission factor. This represents the marginal emission factor within the time interval t. The marginal emission coefficient within the time interval t As shown in formula (36): （36） The second-stage objective function of the demand response model based on the average emission factor is shown in formula (37): （37） in, This represents the carbon dioxide emission reduction calculated using the average emission factor. This represents the average carbon emission factor over time t. Average carbon emission factor over time t As shown in formula (38): （38） in, This indicates the system load after demand response. This indicates the number of independent energy storage units within time interval t. The charging power, the system load after demand response As shown in formula (39): （39） The constraints on the maximum and minimum adjustment amounts of the demand response model are shown in formula (40): （40） in, The logic for determining whether a user's workload has increased is represented by a 0-1 variable. This indicates that the load limit can be adjusted within the time interval t. The logic for determining whether a user should reduce their load is represented by a 0-1 variable. Indicates the adjustable load ratio. The constraint on the load increase within the time interval t of the demand response model is shown in formula (41): （41） in, Indicates the system's maximum rated load. The constraint on the load reduction amount within the time interval t of the demand response model is shown in formula (42): （42） The constraint that the demand response model cannot simultaneously increase and decrease the load within the same time interval is shown in formula (43): （43） The constraint that the total daily load remains basically unchanged before and after the electricity consumption behavior adjustment during the scheduling period in the demand response model is shown in formula (44): （44） in, This indicates the daily change in electricity consumption.

9. The method for optimizing electricity demand-side management based on electro-carbon coupling and dynamic carbon emissions according to claim 8, characterized in that, Step S4 includes the following steps: S401. Input the parameters of the demand response scheme into the first stage model; S402. Linearize the first-stage model with parameters from the input demand response scheme to obtain a linear first-stage model; S403. Obtain the minimum system cost through the linear first-stage model, and determine whether to perform demand response based on the minimum system cost. If demand response is performed, proceed to step S404; otherwise, proceed to step S406. S404. Based on the characteristics of the demand scheme, input the parameters of the demand response scheme into the second-stage model, and obtain the minimized electricity cost or minimized carbon emissions through the second-stage model; S405. Adjust the load curve by minimizing electricity costs or carbon emissions and update the load reduction and load increase parameters. Determine whether the updated load reduction and load increase parameters meet the convergence condition. If they do, proceed to step S406. If they do not, use the updated load reduction and load increase parameters as parameters in the updated demand response scheme and then execute step S401. The convergence condition is that the user load can only change once within any time interval t, as shown in equation (45): （45） in, A set representing the number of iterations; S406. Output the market operation parameter analysis results of the demand response plan; S407. Repeat steps S401-S406 until the market operation parameter analysis results for all demand response schemes are obtained.

10. The power demand-side management optimization method based on electric carbon coupling and dynamic carbon emissions according to claim 9, characterized in that, In step S402, the linearization transformation of the first-stage model, which contains parameters from the demand response scheme, includes the use of continuous variables. replace and The product of is shown in formula (46): （46） in, This indicates the substitution of continuous variables. This represents the charge / discharge state of independent energy storage, and is a 0-1 variable. Introduce variable M to constrain continuous variables. As shown in formula (47): （47） Where M is set to 10000, The annual operating cost of the thermal power unit (Equation 27) can be rewritten as a quadratic function in general form, as shown in formula (48): （48） in, Denotes the coefficient of the quadratic term. Denotes the coefficient of the linear term. The coefficient of the quadratic term As shown in formula (49): （49） The coefficient of the first term As shown in formula (50): （50） The annual operating cost of the aforementioned thermal power unit is processed by piecewise linearization, as shown in formula (51): （51） in, This indicates the number of firepower units within time interval t. The power generation in the k-th segment, where k represents the k-th linear segment. Represents a segmented set. Indicates thermal power unit Piecewise linearization of the slope of the running cost function in the k-th segment The output of the fire unit in sections is shown in formula (52): （52） The constraint on the output of the fire unit in sections is shown in formula (53): （53） in, This represents the number of linear segments.

Images