Low-carbon economic dispatch method for cogen microgrid considering two-stage demand response

Abstract

The application discloses a kind of considering two-stage demand response's heat and power cogeneration microgrid low-carbon economic dispatching method.First, the equipment in heat and power cogeneration type microgrid (CHP-MG) system is modeled, then the predicted load of system is optimized multiple times in the two-stage demand response model introduced to system demand side, with the sum of system operation cost, carbon emission cost and environmental cost minimum as optimization target, then the opportunity constraint programming model is introduced to the whole system to cope with the uncertainty of wind and light and electric-thermal transfer load, finally, the optimal scheduling scheme is obtained by solving the system model using commercial solving software Gurobi.The method proposed in the application can effectively improve the energy utilization rate of the system and reduce the economic cost and environmental pollution under the premise of meeting the user's electricity demand.

Description

Technical Field

[0001] This invention belongs to the field of power system technology, specifically relating to a low-carbon economic dispatch method for cogeneration microgrids that considers two-stage demand response. Background Technology

[0002] Currently, with widespread attention focused on distributed energy and environmental issues, Combined Heat and Power Microgrids (CHP-MGs) can efficiently utilize renewable energy, thus their economic dispatch operation has attracted considerable attention. Demand-side management (DSM), proposed by the Electric Power Research Institute (EPI) in the United States, has evolved towards demand response (DR), with DR in the electricity market primarily including price-based DR and incentive-based DR. Introducing a two-stage demand response model into the operation and dispatch of CHP-MG systems can effectively reduce system operating costs. Simultaneously, with the implementation of national policies aimed at "carbon peaking and carbon neutrality," carbon emissions and environmental pollution must be considered during system operation. Most existing system dispatching methods lack carbon trading models and environmental cost models to balance system economics and low-carbon environmental friendliness. Furthermore, current technologies for CHP-MG operation largely neglect system uncertainties. Summary of the Invention

[0003] To address the aforementioned issues, this invention proposes a low-carbon economic dispatch method for CHP-MG microgrids that considers two-stage demand response. This method incorporates carbon trading and environmental cost models into the system dispatching process to balance economic efficiency with low-carbon environmental friendliness. Finally, a chance-constrained programming model is added to the operation of the CHP-MG to account for system uncertainties. This approach improves the system's energy efficiency, effectively reduces operating costs and carbon emissions, and achieves the optimal dispatching scheme while meeting power supply reliability and user demand.

[0004] This invention proposes a low-carbon economic dispatch method for cogeneration microgrids that considers two-stage demand response. The specific design scheme is as follows:

[0005] (1) Equipment modeling of the CHP-MG low-carbon economic system;

[0006] (2) Construct a two-stage demand response to obtain the optimal time-of-use electricity price;

[0007] (3) Establish a systematic carbon tiered trading model and cost function;

[0008] (4) Use opportunity-constrained programming to deal with uncertainties within the system;

[0009] (5) Use the Gurobi software to solve for the optimal system scheduling scheme.

[0010] Furthermore, the CHP-MG system equipment in step (1) includes photovoltaic equipment, wind power equipment, micro gas turbines, batteries, thermal storage tanks, and electric heating equipment, specifically:

[0011] Micro gas turbine:

[0012]

[0013] Q MT_h (t)=Q MT (t)η h COP ho

[0014] Storage battery:

[0015]

[0016] 0.2E b.min ≤E b (t)≤0.8E b.max

[0017] Heat storage tank:

[0018]

[0019] 0.2H TST.min ≤H TST (t)≤0.8H TST.max

[0020] Electric heating equipment:

[0021] Q EB (t)=P EB (t)η eb

[0022] In the formula: Q MT (t), P MT (t), η MT (t) represents the exhaust waste heat, power generation, and power generation efficiency of MT at time t; η L Q is the heat dissipation loss coefficient; MT_h (t) represents the heat generated by the bromine refrigeration mechanism at time t; COP ho It is the coefficient of performance (COP) of the bromine chiller, η h This represents the flue gas recovery rate of the bromine chiller; Δt is the scheduling period, set to 1 hour; E b (t) and τ are the BT capacity and self-discharge rate at time t, respectively; P b.ch (t), P b.dis (t) and η bch ηbdis These are the charging and discharging power and efficiency of BT at time t, respectively; E b.min E b.max These are the minimum and maximum capacities of BT, respectively; H TST (t), μ T These are the capacity and heat loss rate of the TST at time t, respectively; Q ch (t), Q dis (t) and η hch η hdis These are the charge / discharge heat power and efficiency of TST at time t; H TST.min H TST.max These are the minimum and maximum capacities of TST, respectively; Q EB (t), P EB (t) represents the heating power and electricity consumption of EB at time t; η eb This refers to the electrothermal conversion efficiency of EB.

[0023] Furthermore, the two-stage demand response introduced in step (2) is to enable multiple optimizations of the predicted load to reduce system operating costs, specifically:

[0024] The first phase of demand response constructs a price-based demand response model. The multi-objective function aims to minimize the peak-to-valley difference in predicted load and maximize user satisfaction. A multi-objective genetic algorithm is used to solve for the optimal time-of-use electricity price and, based on this, the first optimized load after peak shaving and valley filling. The objective function is as follows:

[0025]

[0026]

[0027]

[0028] In the formula: P l.max P represents the optimized maximum load value. l.min λ represents the optimized minimum load. l Weighting of user satisfaction with electricity usage habits; λ m Weighting of user satisfaction with electricity bills; S l It is the satisfaction level of users' electricity usage habits; S m It is the user's satisfaction with electricity payment fees; Δq i q is the change in user electricity consumption at time i; i ΔC represents the user's electricity consumption at time i; i C is the change in the user's electricity expenditure at time i; i It represents the user's electricity expenditure at time i;

[0029] The second stage of demand response is an incentive-based demand response model. Under the influence of incentives, users can readjust their load without affecting their energy comfort, allowing for the reduction of electrical load and the transfer of electrical and thermal loads. The model is constructed as follows:

[0030] P c (t)=P c.old (t)-ΔP c (t)

[0031] P t (t)=P t.old (t)-ΔP t (t)

[0032] Q t (t)=Q t.old (t)+η e ΔP t (t)

[0033] P l (t)=P n (t)+P c (t)+P t (t)

[0034] Incentive compensation expenses:

[0035]

[0036] In the formula: P c (t) represents the controllable load power consumption of residential users after they participate in the incentive response at time t; P c.old (t) represents the controllable load power consumption of residential users before they participate in the incentive response at time t; ΔP c (t) represents the amount of controllable load reduction for residential users when they participate in the response at time t; P t (t) represents the transferable load power after the residential user participates in the transfer response at time t; P t.old (t) represents the transferable load power of the residential user at time t before participating in the transfer response; ΔP t (t) represents the transferred electrical power when the residential user participates in the response at time t; η e Q is the conversion factor; t (t) represents the gas supply heat load demand; Q t.old (t) represents the gas heating load demand power before residential users participate in the response within time t; P l (t) represents the electrical load power at time t; P n (t) represents the fixed load for all users within time t; C d1 (t) represents the compensation cost that a user can obtain by reducing a unit of electrical power at time t; C d2(t) represents the compensation cost that a user can obtain by transferring a unit of electric heating power at time t.

[0037] The load is optimized a second time using incentive methods.

[0038] Furthermore, the carbon ladder trading model and cost function model established in step (3) are as follows:

[0039]

[0040]

[0041] F = F1 + F2 + F3 + η treat F4

[0042]

[0043] In the formula: E L (t) represents the system's carbon emission allowance at time t; E c (t) represents the actual carbon emissions of the system at time t; μ is the price per kg of carbon emissions traded; d is the length of the carbon emission range; h is the increase in μ for each step increase in carbon emissions; F2(t) is the carbon emission cost of the system at time t; when E c (t)<E L When F(t) is zero, F2(t) is negative, indicating that the system can sell allowances in the carbon trading market and obtain carbon revenue; F3 is the system's equipment operating cost; C Grid (t) represents the electricity purchase cost of the system at time t; C MT (t) represents the power generation cost of MT at time t; C b (t) represents the operating cost of GB at time t; C e (t) represents the operating cost of BT at time t; F4 represents the environmental cost of system operation; C treat (t) represents the cost of treating pollutants at time t; F represents the total operating cost of the system; η treat To assign weights to environmental costs.

[0044] Furthermore, the operational constraints of the equipment in step (1) and the constraints of the cost objective function in step (3) are as follows:

[0045] Electric power balance constraints:

[0046] P wt (t)+P pv (t)+P MT (t)+P Grid (t)+P b.dis (t)=P l (t)+P b.ch (t)+PEB (t)

[0047] Thermal power balance constraint:

[0048] Q MT_h (t)+Q dis (t)+Q GB (t)+Q EB (t)=Q l (t)+Q ch (t)

[0049] Power upper and lower limit constraints and equipment ramping constraints:

[0050]

[0051]

[0052] In addition, the state representations of BT and TST are also subject to constraints, as shown in the following formula:

[0053]

[0054] In the formula: P wt (t) represents the wind power output at time t; P pv (t) represents the photoelectric output at time t; P Grid (t) represents the purchased electricity at time t; P MT.min P MT.max and These are the minimum and maximum output power of the MT, and the lower and upper limits of its ramp rate, respectively; I1(t) represents the start / stop state of the MT at time t, where 0 indicates shutdown and 1 indicates operation; Q GB (t) represents the thermal output of GB at time t; Q l (t) represents the heat load at time t; Q GB.min Q GB.max and These are the minimum and maximum output power of GB, and the lower and upper limits of ramping, respectively; I2(t) represents the start / stop status of GB at time t, with 0 indicating shutdown and 1 indicating operation; P b.ch.min P b.ch.max and These are the minimum and maximum charging power of BT, and the lower and upper limits of the charging ramp, respectively; I3(t) represents the charging state of BT at time t, where 0 represents no charging and 1 represents charging; P b.dis.min P b.dis.max and These represent the minimum and maximum discharge power of BT, and the lower and upper limits of discharge ramp-up, respectively; I4(t) represents the discharge state of BT at time t, where 0 indicates no discharge and 1 indicates discharge; Q TST.ch.minQ TST.ch.max and These are the minimum and maximum values ​​of the charging power of the TST, and the lower and upper limits of the charging ramp-up, respectively; I5(t) represents the charging state of the TST at time t, where 0 represents the non-charging state and 1 represents the charging state; Q TST.dis.min Q TST.dis.max and These are the minimum and maximum values ​​of the heat release power of TST, and the lower and upper limits of the heat release ramp, respectively; I6(t) represents the heat release state of TST at time t, where 0 represents the non-heat release state and 1 represents the heat release state; T is 24.

[0055] Furthermore, opportunity-constrained programming is used to address uncertainties within the system. These uncertainties are expressed as follows:

[0056]

[0057]

[0058] The power balance constraint can be transformed into the following using the chance-constrained programming method:

[0059] P MT (t)+P Grid (t)+P b.dis (t)-P b.ch (t)-P l (t)-P EB (t)≥z s (t)

[0060]

[0061] Where: δ wt (t) represents the deviation between the actual and predicted wind power output; δ pv (t) represents the deviation between the actual and predicted power output of the photoelectric system; δ t (t) represents the deviation between the actual and predicted transfer of the electrothermal transfer load; P wt0 (t), P pv0 (t) and P t0 (t) represent the predicted values ​​for wind power, solar power, and electrothermal load transfer, respectively; σ wt (t), σ pv (t) and σ t (t) represent the standard deviations of wind power, solar power, and electrothermal transfer loads, respectively; P WTN P PVN These are the installed capacities of wind power and solar power, respectively; μ t (t), μ wt (t) and μ pv (t) represents the expected value; z αα is the quantile of the standard normal distribution.

[0062] Furthermore, the solution method for step (5) is as follows:

[0063] 1) Input the original predicted load into the demand response model of the first stage, and use a multi-objective genetic algorithm to solve for the time-of-use electricity price that minimizes the load peak-valley difference and maximizes user satisfaction, as well as an optimized predicted load obtained by peak shaving and valley filling of the original load under the time-of-use electricity price.

[0064] 2) Input the predicted load optimized in the first stage into the demand response model in the second stage, and reduce and transform the load according to the user's response under the stimulus to obtain a new predicted load.

[0065] 3) Incorporate a carbon tiered trading model and an environmental cost model into the system to consider carbon trading costs and environmental costs, assigning a certain weight to environmental costs. Construct an objective function that minimizes the sum of system operating costs, incentive costs, carbon trading costs, and environmental costs.

[0066] 4) Incorporate an opportunity-constrained programming model into the system to address the uncertainties of distributed energy resources and electricity-heat transfer loads.

[0067] 5) Under various constraints, the optimal running results are obtained by solving the problem using the commercial solution software Gurobi.

[0068] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0069] This invention targets CHP-MG (Combined Heat and Power Microgrid) with distributed energy resources. By considering two-stage demand response on the demand side and introducing carbon trading and opportunity-constrained programming models into the system operation, a low-carbon economic dispatch model for CHP-MG considering two-stage demand response is established.

[0070] 1) By guiding users through electricity price signals and incentive compensation, users spontaneously change their electricity and heat consumption habits, providing optimization space for the operation of CHP-MG and making the system scheduling more flexible and economical.

[0071] 2) At the same time, during the operation of CHP-MG, carbon trading and environmental costs are taken into account, which effectively limits the emission of carbon dioxide and pollutants from the system.

[0072] 3) In dealing with the uncertainty of source load, the opportunity constraint method adopted in this invention has a certain improvement in the absorption of wind and solar power and reduces the impact of uncertain variables on the system. Attached Figure Description

[0073] Figure 1 This is a diagram illustrating the energy flow structure of the CHP-MG scheduling system of the present invention.

[0074] Figure 2 This is the optimized load diagram obtained from the two-stage demand response in an example of the present invention;

[0075] Figure 3 The graphs show the carbon emission results for scenarios two and three in this invention. Detailed Implementation

[0076] The invention is further illustrated below with reference to specific embodiments and accompanying drawings. The invention proposes a low-carbon economic dispatch system structure for a two-stage demand response combined heat and power microgrid, as shown in the diagram below. Figure 1 As shown, to improve the energy efficiency of the system, a two-stage demand response model is established, and the specific implementation steps are as follows:

[0077] (1) Modeling the CHP-MG device

[0078] The CHP-MG system structure diagram is as follows: Figure 1 As shown, the system components include photovoltaic equipment, wind power equipment, micro gas turbines, batteries, thermal storage tanks, and electric heating equipment, specifically:

[0079] Micro gas turbine:

[0080]

[0081] Q MT_h (t)=Q MT (t)η h COP ho

[0082] Storage battery:

[0083]

[0084] 0.2E b.min ≤E b (t)≤0.8E b.max

[0085] Heat storage tank:

[0086]

[0087] 0.2H TST.min ≤H TST (t)≤0.8H TST.max

[0088] Electric heating equipment:

[0089] Q EB (t)=P EB (t)η eb

[0090] In the formula: QMT (t), P MT (t), η MT (t) represents the exhaust waste heat, power generation, and power generation efficiency of MT at time t; η L Q is the heat dissipation loss coefficient; MT_h (t) represents the heat generated by the bromine refrigeration mechanism at time t; COP ho It is the coefficient of performance (COP) of the bromine chiller, η h This represents the flue gas recovery rate of the bromine chiller; Δt is the scheduling period, set to 1 hour; E b (t) and τ are the BT capacity and self-discharge rate at time t, respectively; P b.ch (t), P b.dis (t) and η bch η bdis These are the charging and discharging power and efficiency of BT at time t, respectively; E b.min E b.max These are the minimum and maximum capacities of BT, respectively; H TST (t), μ T These are the capacity and heat loss rate of the TST at time t, respectively; Q ch (t), Q dis (t) and η hch η hdis These are the charge / discharge heat power and efficiency of TST at time t; H TST.min H TST.max These are the minimum and maximum capacities of TST, respectively; Q EB (t), P EB (t) represents the heating power and electricity consumption of EB at time t; η eb This refers to the electrothermal conversion efficiency of EB.

[0091] (2) Construct a two-stage demand response to obtain the optimal time-of-use electricity price.

[0092] The two-stage demand response is introduced to enable multiple optimizations of forecasted loads and reduce system operating costs. Specifically:

[0093] The first phase of demand response constructs a price-based demand response model. The multi-objective function aims to minimize the peak-to-valley difference in predicted load and maximize user satisfaction. A multi-objective genetic algorithm is used to solve for the optimal time-of-use electricity price and, based on this, the first optimized load after peak shaving and valley filling. The objective function is as follows:

[0094]

[0095]

[0096]

[0097] In the formula: Pl.max P represents the optimized maximum load value. l.min λ represents the optimized minimum load. l Weighting of user satisfaction with electricity usage habits; λ m Weighting of user satisfaction with electricity bills; S l It is the satisfaction level of users' electricity usage habits; S m It is the user's satisfaction with electricity payment fees; Δq i q is the change in user electricity consumption at time i; i ΔC represents the user's electricity consumption at time i; i C is the change in the user's electricity expenditure at time i; i It represents the user's electricity expenditure at time i;

[0098] The second stage of demand response is an incentive-based demand response model. Under the influence of incentives, users can readjust their load without affecting their energy comfort, allowing for the reduction of electrical load and the transfer of electrical and thermal loads. The model is constructed as follows:

[0099] P c (t)=P c.old (t)-ΔP c (t)

[0100] P t (t)=P t.old (t)-ΔP t (t)

[0101] Q t (t)=Q t.old (t)+η e ΔP t (t)

[0102] P l (t)=P n (t)+P c (t)+P t (t)

[0103] Incentive compensation expenses:

[0104]

[0105] In the formula: P c (t) represents the controllable load power consumption of residential users after they participate in the incentive response at time t; P c.old (t) represents the controllable load power consumption of residential users before they participate in the incentive response at time t; ΔP c (t) represents the amount of controllable load reduction for residential users when they participate in the response at time t; P t (t) represents the transferable load power after the residential user participates in the transfer response at time t; Pt.old (t) represents the transferable load power of the residential user at time t before participating in the transfer response; ΔP t (t) represents the transferred electrical power when the residential user participates in the response at time t; η e Q is the conversion factor; t (t) represents the gas supply heat load demand; Q t.old (t) represents the gas heating load demand power before residential users participate in the response within time t; P l (t) represents the electrical load power at time t; P n (t) represents the fixed load for all users within time t; C d1 (t) represents the compensation cost that a user can obtain by reducing a unit of electrical power at time t; C d2 (t) represents the compensation cost that a user can obtain by transferring a unit of electric heating power at time t.

[0106] The load is optimized a second time using incentive methods.

[0107] (3) Establish a systematic carbon tiered trading model and cost function model.

[0108]

[0109]

[0110] F = F1 + F2 + F3 + η treat F4

[0111]

[0112] In the formula: E L (t) represents the system's carbon emission allowance at time t; E c (t) represents the actual carbon emissions of the system at time t; μ is the price per kg of carbon emissions traded; d is the length of the carbon emission range; h is the increase in μ for each step increase in carbon emissions; F2(t) is the carbon emission cost of the system at time t; when E c (t)<E L When F(t) is zero, F2(t) is negative, indicating that the system can sell allowances in the carbon trading market and obtain carbon revenue; F3 is the system's equipment operating cost; C Grid (t) represents the electricity purchase cost of the system at time t; C MT (t) represents the power generation cost of MT at time t; C b (t) represents the operating cost of GB at time t; C e (t) represents the operating cost of BT at time t; F4 represents the environmental cost of system operation; C treat (t) represents the cost of treating pollutants at time t; F represents the total operating cost of the system; η treatTo assign weights to environmental costs.

[0113] Electric power balance constraints:

[0114] P wt (t)+P pv (t)+P MT (t)+P Grid (t)+P b.dis (t)=P l (t)+P b.ch (t)+P EB (t)

[0115] Thermal power balance constraint:

[0116] Q MT_h (t)+Q dis (t)+Q GB (t)+Q EB (t)=Q l (t)+Q ch (t)

[0117] Power upper and lower limit constraints and equipment ramping constraints:

[0118]

[0119]

[0120] In addition, the state representations of BT and TST are also subject to constraints, as shown in the following formula:

[0121]

[0122] In the formula: P wt (t) represents the wind power output at time t; P pv (t) represents the photoelectric output at time t; P Grid (t) represents the purchased electricity at time t; P MT.min P MT.max and These are the minimum and maximum output power of the MT, and the lower and upper limits of its ramp rate, respectively; I1(t) represents the start / stop state of the MT at time t, where 0 indicates shutdown and 1 indicates operation; Q GB (t) represents the thermal output of GB at time t; Q l (t) represents the heat load at time t; Q GB.min Q GB.max and These are the minimum and maximum output power of GB, and the lower and upper limits of ramping, respectively; I2(t) represents the start / stop status of GB at time t, with 0 indicating shutdown and 1 indicating operation; P b.ch.min P b.ch.max and These are the minimum and maximum charging power of BT, and the lower and upper limits of the charging ramp, respectively; I3(t) represents the charging state of BT at time t, where 0 represents no charging and 1 represents charging; P b.dis.min P b.dis.max and These represent the minimum and maximum discharge power of BT, and the lower and upper limits of discharge ramp-up, respectively; I4(t) represents the discharge state of BT at time t, where 0 indicates no discharge and 1 indicates discharge; Q TST.ch.min Q TST.ch.max and These are the minimum and maximum values ​​of the charging power of the TST, and the lower and upper limits of the charging ramp-up, respectively; I5(t) represents the charging state of the TST at time t, where 0 represents the non-charging state and 1 represents the charging state; Q TST.dis.min Q TST.dis.max and These are the minimum and maximum values ​​of the heat release power of TST, and the lower and upper limits of the heat release ramp, respectively; I6(t) represents the heat release state of TST at time t, where 0 represents the non-heat release state and 1 represents the heat release state; T is 24.

[0123] (4) Use opportunity-constrained programming to deal with uncertainties in the system.

[0124] The opportunity-constrained programming method expresses these three uncertainties as follows:

[0125]

[0126]

[0127] The power balance constraint can be transformed into the following using the chance-constrained programming method:

[0128] P MT (t)+P Grid (t)+P b.dis (t)-P b.ch (t)-P l (t)-P EB (t)≥z s (t)

[0129]

[0130] Where: δ wt (t) represents the deviation between the actual and predicted wind power output; δ pv (t) represents the deviation between the actual and predicted power output of the photoelectric system; δ t (t) represents the deviation between the actual and predicted transfer of the electrothermal transfer load; P wt0 (t), P pv0 (t) and Pt0 (t) represent the predicted values ​​for wind power, solar power, and electrothermal load transfer, respectively; σ wt (t), σ pv (t) and σ t (t) represent the standard deviations of wind power, solar power, and electrothermal transfer loads, respectively; P WTN P PVN These are the installed capacities of wind power and solar power, respectively; μ t (t), μ wt (t) and μ pv (t) represents the expected value; z α α is the quantile of the standard normal distribution.

[0131] (5) Use Gurobi to solve the entire CHP-MG scheduling system to obtain the optimal scheduling scheme. The specific scheduling process is as follows:

[0132] 1) Input the original predicted load into the demand response model of the first stage, and use a multi-objective genetic algorithm to solve for the time-of-use electricity price that minimizes the load peak-valley difference and maximizes user satisfaction, as well as an optimized predicted load obtained by peak shaving and valley filling of the original load under the time-of-use electricity price.

[0133] 2) Input the predicted load optimized in the first stage into the demand response model in the second stage, and reduce and transform the load according to the user's response under the stimulus to obtain a new predicted load.

[0134] 3) Incorporate a carbon tiered trading model and an environmental cost model into the system to consider carbon trading costs and environmental costs, assigning a certain weight to environmental costs. Construct an objective function that minimizes the sum of system operating costs, incentive costs, carbon trading costs, and environmental costs.

[0135] 4) In the system, an opportunity-constrained programming model is added to address the uncertainties of distributed energy resources and electricity-heat transfer loads.

[0136] 5) Under various constraints, the optimal running results are obtained by solving the problem using the commercial solution software Gurobi.

[0137] Four scenarios were set up for comparison and solution verification to verify that the proposed method is optimal:

[0138] Scenario 1: Without incorporating tiered carbon trading costs and two-stage demand response, and employing a deterministic typical planning model.

[0139] Scenario 2: Consider a two-stage demand response without incorporating tiered carbon trading costs and employing a deterministic planning model.

[0140] Scenario 3: Simultaneously considering two-stage demand response and the inclusion of tiered carbon trading costs, and employing a deterministic planning model.

[0141] Scenario 4: A planning model that simultaneously considers two-stage demand response and incorporates tiered carbon trading costs, and employs opportunity-constrained planning.

[0142] Table 1 shows the operating costs for the four scenarios. As can be seen from Table 1, compared to the traditional scheduling scenario 1, scenario 2, after introducing the two-stage demand response model, reduces the total operating cost by 2.3%. Scenario 3 reduces carbon emissions by 8.5% compared to scenario 2, and environmental costs are also reduced. Scenario 4, with the participation of the opportunity-constrained planning method in scheduling, improves the absorption of wind and solar power, and the total operating cost is reduced again.

[0143] Table 1. Operating costs for the four scenarios

[0144]

[0145] The following are examples of using two-stage demand response to optimize load: Figure 2 As shown in the figure, the total load remains unchanged after the first stage of optimization, but there is significant peak shaving and valley filling. In the second stage, the optimized load is optimized a second time, at which point the total load changes, and this change is transferred to user compensation incentives. The carbon emission results are shown in the figure below. Figure 3 As shown, the introduction of the carbon tiered trading model significantly limits carbon dioxide emissions.

[0146] The above embodiments are used to explain the present invention, but not to limit the present invention. Any modifications and changes made to the present invention within the spirit and scope of the claims shall fall within the protection scope of the present invention.

Claims

1. A low-carbon economic dispatch method for cogeneration microgrids considering two-stage demand response, characterized in that: Includes the following steps: Step 1: Model the equipment of the CHP-MG low-carbon economic system; Step 2 involves constructing a two-stage demand response model to obtain the optimal time-of-use (TOU) price. This includes: The first stage constructs a price-based demand response model with multiple objective functions: minimizing the peak-valley difference in predicted load and maximizing user satisfaction. A multi-objective genetic algorithm is used to solve for the optimal TOU price, and the original predicted load is then optimized based on this optimal TOU price. The second stage constructs an incentive-based demand response model. Using the first-stage optimized load as input, incentives are used to guide users to reduce their electricity load and shift their heat load without affecting their energy comfort. The multi-objective function of the price-based demand response model in the first stage is as follows: ， ， ， In the formula: This represents the optimized maximum load value. This represents the optimized minimum load. Weighting of user satisfaction with electricity usage habits; Weighting of user satisfaction with electricity bill expenditure; It is the satisfaction level of users' electricity usage habits; It is the user's satisfaction with electricity payment fees; It is the change in the user's electricity consumption at time i; This refers to the user's electricity consumption at time i. It represents the change in the user's electricity expenditure at time i; It represents the user's electricity expenditure at time i; The second-stage incentive-based demand response model is as follows: ， ， ， ， Incentive compensation expenses: ， In the formula: The controllable load power consumption of residential users after participating in the incentive response at time t; The controllable load power consumption of residential users before they participate in the incentive response at time t; The amount of controllable load reduction for residential users when they participate in the response at time t; Let t be the transferable load power after residential users participate in the transfer response; Let t be the transferable load power of residential users before they participate in the transfer response; The transferred electrical power at time t when residential users participate in the response; This is the conversion factor; To meet the heat load demand for gas supply; The gas heating load demand power before residential users participate in the response within time t; Let be the electrical load power at time t; For all users, the fixed load within time t; The user can receive compensation for reducing the unit power output at time t. The compensation cost that a user can obtain for transferring a unit of electrical heating power at time t; Step 3: With the goal of minimizing the sum of incentive compensation costs, system carbon emission costs, system equipment operating costs, and system environmental operating costs, establish a carbon tiered trading model and a cost function model for the system. Step 4: Use opportunity-constrained programming to address uncertainties within the system; Step 5: Use the Gurobi software to solve for the optimal system scheduling scheme.

2. The low-carbon economic dispatch method for cogeneration microgrids considering two-stage demand response as described in claim 1, characterized in that: In step 1, the CHP-MG system equipment includes photovoltaic equipment, wind power equipment, micro gas turbine (MT), battery (BT), thermal storage tank (TST), and electric heating equipment (EB). The modeling of the micro gas turbine (MT), battery (BT), thermal storage tank (TST), and electric heating equipment (EB) is specifically as follows: Micro gas turbine: ， ， Storage battery: ， ， Heat storage tank: ， ， Electric heating equipment: ， In the formula: , , Let t represent the exhaust waste heat, power generation, and power generation efficiency of MT at time t. This is the heat dissipation loss coefficient; The heat generated by the bromine refrigeration mechanism at time t; It is the coefficient of performance (COP) of the bromine chiller. It is the flue gas recovery rate of the bromine chiller; The scheduling period is set to 1 hour. , These are the BT capacity and self-discharge rate at time t, respectively. , and , These are the charging and discharging power and efficiency of BT at time t, respectively. , These are the minimum and maximum capacities of BT, respectively. , These are the capacity and heat loss rate of the TST at time t, respectively. , and , These are the heat dissipation power and efficiency of TST at time t, respectively. , These are the minimum and maximum capacities of TST, respectively. , These represent the heating power and electricity consumption of EB at time t, respectively. This refers to the electrothermal conversion efficiency of EB.

3. The low-carbon economic dispatch method for cogeneration microgrids considering two-stage demand response as described in claim 2, characterized in that: The carbon tiered trading model and cost function model in step 3 are as follows: ， ， ， ， In the formula: Let be the system's carbon emission allowance at time t; The actual carbon emissions of the system at time t; The price per kilogram of carbon emissions for trading; The length of the carbon emission range; For each step up in carbon emissions The growth rate; Let be the carbon emission cost of the system at time t; when hour, A negative value indicates that the system can sell allowances in the carbon trading market, thereby generating carbon revenue for the system. For the system's equipment operating costs; The electricity purchase cost of the system at time t; Let MT be the cost of electricity generation at time t; Let GB be the operating cost at time t; Let be the operating cost of BT at time t; Environmental costs for system operation; The cost of treating pollutants at time t; This represents the total operating cost of the system. To assign weights to environmental costs.

4. The low-carbon economic dispatch method for cogeneration microgrids considering two-stage demand response as described in claim 3, characterized in that: The operational constraints of the equipment in step 1 and the constraints of the cost function in step 3 are as follows: Electric power balance constraints: ， Thermal power balance constraint: ， Power upper and lower limit constraints and equipment ramping constraints: ， ， In addition, the state representations of BT and TST are also subject to constraints, as shown in the following formula: ， In the formula: Let t be the wind power output at time t; The photoelectric output at time t; The electricity is purchased from outside at time t; , and , These are the minimum and maximum output power of MT, and the lower and upper limits of ramping. This represents the start / stop status of MT within time t, where 0 indicates shutdown and 1 indicates operation; Let GB be the heat output at time t; Let t be the heat load at time t; , and , These are the minimum and maximum output power of GB, and the lower and upper limits of ramping. This represents the start / stop status of GB within time t, where 0 indicates shutdown and 1 indicates operation; , and , These are the minimum and maximum charging power of BT, and the lower and upper limits of charging ramp. This represents the charging state of BT at time t, where 0 indicates no charging and 1 indicates charging. , and , These are the minimum and maximum discharge power of BT, and the lower and upper limits of discharge ramp-up, respectively. This represents the discharge state of BT at time t, where 0 represents the non-discharge state and 1 represents the discharge state. , and , These are the minimum and maximum values ​​of TST's charging power, and the lower and upper limits of its charging ramp-up. This represents the heated state of TST at time t, where 0 represents the unheated state and 1 represents the heated state. , and , These are the minimum and maximum values ​​of TST's heat release power, as well as the lower and upper limits of its heat release ramp-up. This represents the heat release state of TST at time t, where 0 represents the non-heat release state and 1 represents the heat release state; T is 24.

5. The low-carbon economic dispatch method for cogeneration microgrids considering two-stage demand response as described in claim 4, characterized in that: The opportunity-constrained programming method in step 4 is specifically as follows: The uncertainty is expressed as follows: ， ， The power balance constraint is transformed into: using the opportunity-constrained programming method. ， ， In the formula: The deviation between the actual output and the predicted output of wind power; The deviation between the actual output and the predicted output of the photoelectric system; This represents the deviation between the actual and predicted transfer of the electrothermal load. , and These are the predicted values ​​for wind power, solar power, and electrothermal load transfer, respectively. , and These are the standard deviations of wind power, solar power, and electrothermal transfer loads, respectively. , These are the installed capacities of wind power and solar power, respectively. , and This is the expected value; For standard normal distribution Quantiles.

6. The low-carbon economic dispatch method for cogeneration microgrids considering two-stage demand response as described in claim 1, characterized in that: The solution method for step 5 is as follows: 1) Input the original predicted load into the demand response model of the first stage, and use a multi-objective genetic algorithm to solve for the time-of-use electricity price that minimizes the load peak-valley difference and maximizes user satisfaction, as well as an optimized predicted load obtained by peak shaving and valley filling of the original load under the time-of-use electricity price. 2) Input the predicted load optimized in the first stage into the demand response model in the second stage, and reduce and transform the load according to the user's response under the stimulus to obtain a new predicted load; 3) Incorporate a carbon tiered trading model and an environmental cost model into the system to consider carbon trading costs and environmental costs, and assign a certain weight to environmental costs; construct an objective function that minimizes the sum of system operating costs, incentive costs, carbon trading costs, and environmental costs; 4) In the system, an opportunity-constrained programming model is incorporated to address the uncertainties of distributed energy resources and electricity-heat transfer loads; 5) Under various constraints, the optimal operating result is obtained by solving the problem using the commercial solution software Gurobi, which is the optimal scheduling scheme for the system.

Images