Novel electric power system day-ahead robust scheduling optimization method considering carbon transaction

Abstract

The invention belongs to the technical field of low-carbon power systems, and particularly relates to a novel power system day-ahead robust scheduling optimization method considering carbon transaction, which establishes a robust optimization scheduling model considering a carbon transaction mechanism and scheduling potential of schedulable resources on both sides of a source and a load. The method comprises the following steps: firstly, analyzing scheduling potentials of schedulable resources on two sides of a source load, including deep peak regulation of a thermal power generating unit, a reducible load and a translational load; then, a carbon emission calculation model of a power supply side unit is constructed according to physical characteristics of the unit and components of gas generated by power generation, a unit carbon emission quota model is constructed according to average carbon emission factors published by the Anhui Province Energy Bureau, and a carbon transaction cost model of the system is provided; secondly, constructing a novel electric power system day-ahead robust optimization model considering carbon transaction based on the proposed multi-class flexible resources of the electric power system; and finally, arranging the model into a matrix form for description, and solving the robust optimization model.

Description

Technical Field

[0001] The present invention belongs to the technical field of low-carbon power systems, and more specifically, relates to a novel day-ahead robust dispatch optimization method for power systems taking carbon trading into account. Background Art

[0002] The consumption of fossil fuels has led to the gradual warming of the global climate. In order to alleviate the problem of global warming and overcome the long-term predicament of relying on fossil energy, it has established a positive image for global climate governance.

[0003] However, in the new power system, the large-scale access of new energy sources to reduce the consumption of fossil energy has brought new challenges to the safety and stability of the power grid. Since there are errors in the forecast of wind power output, photovoltaic output and load in the day-ahead dispatching, and the output of new energy sources is highly volatile, the traditional power system dispatching center performs deterministic dispatching based on the predicted values ​​and configures a certain proportion of spinning reserve capacity for the units on the basis of the normal dispatching plan. However, this method often causes a series of problems such as redundant or insufficient spinning reserve and reduced unit operating efficiency leading to additional CO2 emissions due to oversimplification. Therefore, the reasonable introduction of carbon trading costs into the dispatching model can effectively promote energy conservation and emission reduction of thermal power plants and the absorption of new energy.

[0004] In summary, the present invention proposes a new method for day-ahead robust dispatch optimization of power systems taking into account carbon trading. The method constructs a carbon quota and carbon trading cost model based on the basic concepts of carbon emission flow theory, and introduces it into the power system dispatch model; secondly, a robust optimization method is introduced considering the uncertainty of energy output, and the flexibility resources of the power system are fully utilized to construct a new type of day-ahead robust dispatch model of the power system taking into account multiple types of flexibility resources and carbon trading; through this method, while ensuring the stable operation of the system, the carbon emissions and operating costs of the system can be reduced, and the social and economic benefits can be improved. Summary of the invention

[0005] In view of the shortcomings of the existing technology, the present invention proposes a new method for robust scheduling optimization of power system day-ahead taking into account carbon trading. By using the proposed robust optimization method and making full use of the carbon trading mechanism, a new two-stage day-ahead robust scheduling model for power system considering multiple types of flexible resources and carbon trading is constructed; then, the objective function and constraints of the two-stage day-ahead scheduling model are established, and finally they are integrated to form a complete model. And it is organized into a matrix form for description. This method can effectively deal with the extreme situations that may occur in wind power output, photovoltaic output and load demand, thereby avoiding the problem of overly conservative scheduling strategies. And it can reduce the carbon emissions and operating costs of the system while ensuring the stable operation of the system.

[0006] To achieve the above object, the present invention adopts the following technical solutions:

[0007] A novel day-ahead robust dispatch optimization method for power systems taking carbon trading into account is characterized in that the method comprises the following steps:

[0008] Step 1: Modeling multiple types of flexibility resources in the new power system;

[0009] Step 2: A carbon emission calculation model for the power supply side units was constructed based on the physical characteristics of the units and the gas composition generated by power generation. A carbon emission quota model for the units was constructed based on the average carbon emission factor released by the Energy Bureau, and a carbon trading cost model for the system was further proposed.

[0010] Step 3: Using the power system flexibility resources, a new power system day-ahead two-stage robust dispatch model is constructed that considers multiple types of flexibility resources and carbon trading;

[0011] Step 4: Organize the objective function and constraints of the two-stage robust optimization dispatch model of the new power system taking into account carbon trading into matrix form, and solve it to achieve low-carbon economic operation of the new power system.

[0012] This technical solution is further optimized. The new power system day-ahead robust dispatch optimization method taking carbon trading into account is described. The objective function in the pre-dispatching stage dispatching scheme is to minimize the day-ahead dispatching scheme cost. The optimization decision variables include the start-stop plan δ of the thermal power unit thermal power unit, the power generation plan P gen , Upward rotation spare capacity plan R + and the downward rotation reserve capacity scheme R - It also includes the reduction plan ΔP for Class A load that can be reduced A-re , Compensation unit price for Class A load reduction during period t Class B load-shedding reserve capacity Compensation unit price for class B load that can be reduced during period t Dispatch plan for shiftable loads sh* , compensation price of movable load ρ sh and system carbon emissions In addition, k,t is a 0-1 variable, δ k,t =1 means that unit k is running during period t, δ k,t =0 means unit k is shut down during period t; is the power generation power of unit k during period t; and Upward and downward spare capacity reserved for unit k during period t; is the compensation price for the load that can be shifted during period t; the goal of the rescheduling adjustment plan is to minimize the risk cost of regulation. The specific adjustment includes the power adjustment plan ΔP of the thermal power unit gen , Reduction plan ΔP for class B load that can be reduced B-re , abandoned air volume ΔP cw , Amount of abandoned light ΔP cpv , Involuntary load loss of users on the load side ΔP cl .

[0013] This technical solution is further optimized, and the step 1 specifically includes:

[0014] Step 1.1: Establish a cost model for deep peak load regulation units:

[0015] The operating cost of the deep peak load unit is as follows:

[0016]

[0017] In the formula, They are respectively the operating cost, fuel consumption cost, start-up and shutdown cost, standby cost, additional coal consumption cost and life loss cost of conventional unit k.

[0018] Step 1.2: Build a curtailable load model:

[0019] For the load that can be reduced, the load aggregator receives the compensation price adjusted by the power grid to evaluate the response potential of the load users, and then re-declare the maximum load reduction that can be provided in the next 4 hours. The dispatch center makes a plan within the adjustable range according to the demand of the specific period, and then after reaching a cooperation, the load that can be reduced responds to the power grid dispatch at the correct time. Therefore, the load model that can be reduced can be shown as follows:

[0020]

[0021] Where: ΔP t A-re is the reduction amount of class A load that can be reduced during period t; is the elastic coefficient of Class A load that can be reduced and participate in dispatch, The elastic modulus of the class A load that can be reduced; and They represent the compensation price when the Class A load-reducible users can just obtain benefits and the maximum response amount respectively; The maximum elastic coefficient that can reduce the load; is the compensation sensitivity of the load, The smaller the value, the more sensitive the user is to the compensation price, and the greater the impact of price changes on the user. The cost of load curtailment in period t. The load that can be curtailed is divided into two types, type A and type B. It refers to the part of the load that can be reduced without affecting daily requirements. Its power consumption time can be interrupted, and the power consumption duration can be reduced or increased. Among them, type A is similar to the shiftable load, and its response speed is also relatively slow. A calling plan needs to be formulated in advance, and its calling scheme cannot be decided temporarily. Moreover, the formulated calling plan needs to be strictly implemented and cannot be changed arbitrarily. Type B has the ability of rapid adjustment and can quickly respond to the dispatching requirements of the power grid. Therefore, this paper formulates its curtailment plan during the intraday rolling dispatch. The calling methods of type A and type B load that can be curtailed are basically the same, and the calling method of type B load that can be curtailed will not be introduced in detail here;

[0022] Step 1.3: Establish a shiftable load model:

[0023] When the compensation price issued by the dispatching center is relatively low, since the benefits obtained by users from changing the power consumption time are difficult to make up for the inconvenience and losses brought by shifting the power consumption amount, the enthusiasm of users to participate in the dispatch will be very low, making it difficult to meet the requirements of the dispatching center. According to the principles of consumer psychology, the shiftable load model can be expressed as:

[0024]

[0025] In the formula: t sh* is the starting period after the power consumption curve is shifted within a certain period; t sh*- (·) and t sh*+ (·) are the earliest and latest starting periods after the shiftable load shifts the power consumption curve;

[0026] For further optimization of this technical solution, step 2 specifically includes

[0027] Step 2.1: Establish a calculation method for carbon emissions in the power system:

[0028] The carbon emission intensity of coal-fired power generation unit k in period t can be calculated based on the coal consumption per unit of electricity. The formula is as follows:

[0029]

[0030] In the formula: μ k is the carbon content rate of the coal used by coal-fired unit k; is the molar mass of carbon dioxide; M C is the molar mass of carbon. The coal loss of coal-fired unit k for each degree of electricity generated Define the column vector of carbon emission intensity of the unit in period t as Its k-th element is

[0031] After the carbon emission intensity of the unit is calculated, the carbon emission of the unit can be calculated through the power generation of each unit. The cumulative carbon emission of the power supply unit during period t is The calculation formula is

[0032]

[0033] Step 2.2: Establish a method for allocating carbon emission quotas for the power system:

[0034] The baseline method first calculates the carbon emission factor based on the relationship between energy consumption and greenhouse gas emissions by the National Energy Administration, and then allocates it based on the carbon emission factor. This method controls the overall situation and is more conducive to encouraging various resources in the power grid to participate in emission reduction;

[0035] This method uses the baseline method to allocate the free quota of carbon emissions. The free quota of carbon emissions of the units in the system during period t is

[0036]

[0037] Where: is the free quota of the system; carb is the free quota per unit power, and its value is determined by the average emission factor of Anhui Province calculated by the state, η carb =0.5703tCO 2 / MWh;

[0038] Step 2.3: Establish a method for calculating the carbon trading cost of the power system:

[0039] This method adopts a gradient pricing strategy to set the carbon trading price. This strategy ensures that CO 2 A reasonable gradient relationship is formed between the transaction price and carbon emissions, so as to more effectively guide enterprises to reduce carbon emissions and achieve sustainable development. The carbon trading cost model is as follows;

[0040]

[0041] Where: It is a tiered carbon trading price; is the market benchmark price; α is the step price growth rate; d is the interval length of the step price.

[0042]

[0043] Where: is the carbon trading cost of the system in period t.

[0044] The technical solution is further optimized. Step 3 establishes a two-stage robust optimization model for the day ahead. The model includes two stages: pre-scheduling and re-scheduling. Specifically, the following steps are included:

[0045] Step 3.1: Establish the objective function of the robust optimization model:

[0046] The goal of the day-ahead robust optimization model with multiple types of flexible resources taking into account carbon trading is to minimize the sum of the costs of the two stages, that is, the operating cost C during pre-dispatch. oper and the regulatory risk cost C during redispatching risk sum;

[0047]

[0048] in is the day-ahead dispatching plan; U is the uncertain parameters of the new energy and load in the problem; is the optimization variable during rescheduling; the pre-scheduling operation cost C oper It includes various sub-costs of the day-ahead dispatch plan, namely, unit operation cost The cost of reducing the load of Class A during period t, The cost of reducing Class B load and the dispatching cost of shiftable load in period t Carbon trading costs

[0049]

[0050] The control risk cost C during the rescheduling phase risk Including the cost of wind curtailment The cost of abandoning light and load loss cost

[0051]

[0052] Step 3.2: Establish constraints for the robust optimization model control phase:

[0053] Operation constraints during pre-scheduling:

[0054] The constraints of the pre-dispatch stage of the new power system day-ahead dispatch taking into account carbon trading include flexible load constraints, power balance constraints, line transmission restrictions, and system reserve;

[0055] The relevant constraints of the flexible load are

[0056]

[0057] Where: is the elastic coefficient of Class A load that can be reduced and participate in dispatch, The reserve capacity for class B load reduction during period t; is the elastic coefficient of Class B load that can be dispatched, P t A-re P is the load reduction amount of Class A in period t; t B-re is the load reduction amount of Class B in period t; t sh* It is the starting period after the power consumption curve moves within a certain period of time; t sh*- (·) and t sh*+ (·) is the earliest and latest starting time period after the power consumption curve is shifted by the shiftable load; To compensate the unit price;

[0058] The power balance constraint of the power system is

[0059]

[0060] Where: w∈{0,1,…,W-1}, W is the number of wind farms; v∈{0,1,…,V-1}, V is the number of photovoltaic power stations; is the power forecast value of wind farm w in period t; is the power prediction value of the photovoltaic power station v in the period t; P t load,pre is the predicted power in period t; P t sh is the predicted value of the load that can be translated during period t; P t sh* is the power value in period t after load shift;

[0061] The power system line transmission constraints are:

[0062]

[0063] Where: T l,k , T l,w , T l,v and T l,i are the power transmission allocation coefficients of unit k, wind farm w, photovoltaic power station v and load node i to line l in period t; F l max is the maximum power of branch l; is the predicted power of node i in period t, and

[0064] The spare capacity constraint is

[0065]

[0066] Where: and is the upper and lower reserve capacity value of the system in period t;

[0067] Operating constraints during rescheduling:

[0068] The constraint conditions during the day-ahead rescheduling of a new power system considering carbon trading include unit-related constraints, power balance constraints, and line transmission limits;

[0069] Upper and lower limits of unit output during rescheduling

[0070]

[0071] In the formula: is the power adjustment amount of unit k at time t; and are the maximum and minimum output values of unit k respectively;

[0072] The unit ramp rate constraint during rescheduling is

[0073]

[0074] The system power balance constraint during rescheduling is

[0075]

[0076] In the formula: and are the downward and upward ramp rates of unit k; is the power generation of unit k at time t-1; P w,t and P v,t are the actual wind power value and PV output value simulated during rescheduling respectively; P t is the actual load demand power value simulated during rescheduling, including flexible load and rigid load; is the curtailment power of the wth wind farm under the extreme scenario at time t; is the curtailment power of the vth PV power station under the extreme scenario at time t; ΔP t cl is the involuntary load shedding of users on the load side at time t;

[0077] The line transmission power constraint during rescheduling is:

[0078]

[0079] In the formula: P i,t is the actual load demand power value of node i at time t after scheduling;

[0080] This technical solution is further optimized. The objective function and constraint conditions of the day-ahead two-stage robust optimal scheduling model described in step 4 are sorted into matrix form as follows:

[0081] Step 4.1: Model solution algorithm

[0082] Based on the objective function and constraints of the two-stage robust optimization dispatch model of the new power system considering carbon trading constructed above, it is organized into a matrix form for description. The compact form of the proposed model can be expressed as

[0083]

[0084] Where: H oper (X) = 0 is the equality constraint condition in the pre-scheduling stage of the model; G oper (X)≤0 is the inequality constraint condition in the pre-scheduling stage of the model; H risk (X,U,Y)=0 represents the equality constraint condition of the rescheduling stage of the scheduling model; G risk (X,U,Y)≤0 represents the inequality constraint condition in the rescheduling phase of the scheduling model;

[0085] Since a min-max-min structure model is constructed, the first stage minimizes the operating cost, and the second stage minimizes the risk cost in severe scenarios. This structure is relatively complex to solve, and the model also contains uncertain variables, which makes the solution process more complicated. Therefore, the C&CG algorithm is used to solve the model. This method splits the actual problem into a main problem (MP) and a sub-problem (SP), and solves these two problems alternately to gradually approach the optimal solution. The mathematical expressions of the main problem MP and the sub-problem SP are as follows

[0086]

[0087] Where: θ is an auxiliary variable introduced to replace the subproblem, and then a temporary solution is directly obtained. Then, the structure of the subproblem is transformed through the duality theory, and the max-min is transformed into a max single-layer optimization problem;

[0088] Step 4.2: Solving Algorithm Flow

[0089] Through the above analysis, the overall solution process can be decomposed into the following steps.

[0090] Step 1: Initialize the upper bound UB = +∞ and the lower bound LB = -∞ of the objective function, the number of iterations n = 1, set the convergence gap between the upper and lower bounds to ε, and set ε to a small positive number;

[0091] Step 2: Let θ = 0 and find the initial solution X 0 , and then substitute it into the subproblem to solve the initial extreme scenario U 1 ;

[0092] Step 3: Set the extreme scenario U n Substitute into the main problem and get the optimal solution (X n,Y n ), update the lower bound LB to be equal to Y n ;

[0093] Step 4: The optimal solution X in step 3 n Substitute into the subproblem and get the optimal solution make Update the upper bound UB to With C oper (X n );

[0094] Step 5: Determine whether UB-LB≤ε is established. If the condition is established, the operation ends; otherwise, n=n+1 is set and the process returns to step 3.

[0095] Different from the existing technology, the beneficial effects are mainly manifested in: In order to effectively promote energy conservation and emission reduction of thermal power plants and the consumption of new energy, and improve the flexible adjustment ability of the power system, a new power system day-ahead robust dispatch optimization method based on carbon trading is proposed. Compared with the general power system dispatch optimization method, through this method, the power system dispatch center can better deal with the uncertainty factors on both the source and load sides, and improve the dispatch efficiency and stability of the power grid. In addition, this method solves the model through the C&CG algorithm, which makes the solution rate faster, the performance better, and the structure more realistic, effectively improving the robust optimization method to construct a system day-ahead robust optimization dispatch model containing multiple types of flexible resources taking into account carbon emissions, thereby achieving efficient utilization of flexible resources in the power system and promoting the power system to achieve low-carbon economic operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0096] Figure 1 Provides a model framework for day-ahead robust optimization;

[0097] Figure 2 A flowchart for the day-ahead robust dispatch optimization of a new power system taking carbon trading into account;

[0098] Figure 3 It is the flow chart of C&CG algorithm;

[0099] Figure 4 This is the output plan diagram for each unit. DETAILED DESCRIPTION

[0100] In order to explain the technical content, structural features, achieved objectives and effects of the technical solution in detail, the following is a detailed description in conjunction with specific embodiments and accompanying drawings.

[0101] The present invention discloses a novel robust dispatch optimization method for power systems taking carbon trading into account. Considering that there are errors in the forecast of wind power output, photovoltaic output and load in the day-ahead dispatch, and the volatility of new energy output is large, the traditional power system dispatch center performs deterministic dispatch according to the predicted value, and configures a certain proportion of spinning reserve capacity of the unit on the basis of the normal dispatch plan. However, this method often causes redundant or insufficient spinning reserve and reduced unit operating efficiency due to oversimplification, resulting in additional CO 2 Therefore, the present invention introduces carbon trading costs into the dispatching model to effectively promote energy conservation and emission reduction of thermal power plants and the consumption of new energy. Then, according to the multiple types of flexible resources in the power system, a robust optimization method is used to construct a system-day-ahead robust optimization dispatching model containing multiple types of flexible resources taking into account carbon emissions, thereby ensuring that the flexible resources of the power system are fully utilized and promoting the power system to operate in a low-carbon economy.

[0102] Please refer to Figure 1 As shown in the figure, the framework diagram of the day-ahead robust optimization model is shown in Figure 2. In the pre-dispatch stage, the preliminary dispatching plan is mainly formulated based on the forecast values ​​of the day-ahead wind power output, photovoltaic output and load demand of the power system. The objective function of the dispatching plan in the pre-dispatch stage is to minimize the cost of the day-ahead dispatching plan. The optimization decision variables include the start-stop plan δ of the thermal power unit, the power generation plan P gen A series of variable parameters such as the power grid. The formulation of the pre-dispatch plan aims to meet the basic operating needs of the power grid and reserve a certain degree of flexibility to deal with uncertain factors. During the re-dispatch stage, the uncertainty of the forecast of renewable energy power generation and load demand is taken into account, and the power generation and load demand values ​​that may occur in the real scenario are simulated. Then, the pre-dispatch stage and the re-dispatch stage are iteratively solved to obtain the renewable energy power generation and load demand values ​​under extreme scenarios. Then, the plan formulated in the pre-dispatch stage is adjusted based on extreme scenarios. The goal of the adjustment plan is to minimize the risk cost of regulation. The specific adjustments include the power adjustment plan ΔP of the thermal power unit. gen , Reduction plan ΔP for class B load that can be reduced B-re Through this model, the power system dispatching center can better deal with the uncertain factors on both the source and load sides and improve the dispatching efficiency and stability of the power grid.

[0103] See Figure 2 As shown in the figure, the new power system day-ahead robust dispatch optimization flow chart taking into account carbon trading includes the following steps:

[0104] Step 1: Modeling multiple types of flexibility resources in the new power system;

[0105] Step 1.1: Establish a cost model for deep peak load regulation units:

[0106] The operating cost of the deep peak load unit is as follows:

[0107]

[0108] In the formula, They are respectively the operating cost, fuel consumption cost, start-up and shutdown cost, standby cost, additional coal consumption cost and life loss cost of conventional unit k.

[0109] Step 1.2: Build a curtailable load model:

[0110] For the load that can be reduced, the load aggregator receives the compensation price adjusted by the power grid to evaluate the response potential of the load users, and then re-declare the maximum load reduction that can be provided in the next 4 hours. The dispatch center makes a plan within the adjustable range according to the demand of the specific period, and then after reaching a cooperation, the load that can be reduced responds to the power grid dispatch at the correct time. Therefore, the load model that can be reduced can be shown as follows:

[0111]

[0112] Where: ΔP t A-re is the reduction amount of class A load that can be reduced during period t; is the elastic coefficient of Class A load that can be reduced and participate in dispatch, The elastic modulus of the class A load that can be reduced; and They represent the compensation price when the Class A load-reducible users can just obtain benefits and the maximum response amount respectively; The maximum elastic coefficient that can reduce the load; is the compensation sensitivity of the load, The smaller the value, the more sensitive the user is to the compensation price, and the greater the impact of price changes on the user. is the cost of reducing the load in time period t. The reducible load is divided into two types: Class A and Class B, which refers to the partial load that can be reduced without affecting daily needs. Its power consumption time can be interrupted, and the power consumption time can be reduced or increased. Among them, Class A is similar to the shiftable load, and the response speed is relatively slow. It is necessary to formulate a call plan in advance, and its call plan cannot be decided on temporarily. Moreover, the formulated call plan needs to be strictly implemented and cannot be changed at will. Class B has the ability to adjust quickly and can respond quickly to the dispatching needs of the power grid. Therefore, this embodiment formulates its reduction plan during intraday rolling dispatch. The calling methods of Class A and Class B reducible loads are basically the same, and the calling method of Class B reducible loads will not be introduced in detail here.

[0113] Step 1.3: Create a translatable load model:

[0114] When the compensation price released by the dispatch center is low, the benefits gained by users from changing their electricity consumption time cannot make up for the inconvenience and loss caused by shifting electricity consumption. Therefore, users will be less motivated to participate in the dispatch and it will be difficult to meet the needs of the dispatch center. According to the principles of consumer psychology, the shiftable load model can be expressed as:

[0115]

[0116] Where: t sh* It is the starting period after the power consumption curve moves within a certain period of time; t sh*- (·) and t sh*+ (·) is the earliest and latest starting time period after the power consumption curve is shifted by the shiftable load; is the compensation unit price for the translational load during period t.

[0117] Step 2: A carbon emission calculation model for the power supply side units was constructed based on the physical characteristics of the units and the gas composition generated by power generation, and a unit carbon emission quota model was constructed based on the average carbon emission factor released by the Energy Bureau, and a system carbon trading cost model was further proposed.

[0118] Step 2.1: Establish a method for calculating carbon emissions from the power system:

[0119] The carbon emission intensity of coal-fired power generation unit k in period t can be calculated based on the coal consumption per unit of electricity using the following formula:

[0120]

[0121] Where: μ k is the carbon content of coal used in coal-fired unit k; is the molar mass of carbon dioxide; M C is the molar mass of carbon. The coal loss of coal-fired unit k per kilowatt-hour of electricity produced The column vector of the carbon emission intensity of the unit in period t is defined as Its kth element is

[0122] After the carbon emission intensity of the unit is calculated, the carbon emission of the unit can be calculated through the power generation of each unit. The cumulative carbon emission of the power supply unit during period t is The calculation formula is

[0123]

[0124] Step 2.2: Establish a method for allocating carbon emission quotas for the power system:

[0125] The baseline method first calculates the carbon emission factor based on the relationship between energy consumption and greenhouse gas emissions by the National Energy Administration, and then allocates it based on the carbon emission factor. This method controls the overall situation and is more conducive to encouraging various resources in the power grid to participate in emission reduction;

[0126] This embodiment uses the baseline method to allocate the free quota of carbon emissions. The free quota of carbon emissions of the units in the system during period t is

[0127]

[0128] Where: is the free quota of the system; carb is the free quota per unit power, and its value is determined by the average emission factor of Anhui Province calculated by the state, η carb =0.5703tCO 2 / MWh.

[0129] Step 2.3: Establish a method for calculating the carbon trading cost of the power system:

[0130] This embodiment adopts a gradient pricing strategy to set the carbon trading price. This strategy ensures that CO 2 A reasonable gradient relationship is formed between the transaction price and carbon emissions, so as to more effectively guide enterprises to reduce carbon emissions and achieve sustainable development. The carbon trading cost model is as follows;

[0131]

[0132] Where: It is a tiered carbon trading price; is the market benchmark price; α is the step price growth rate; d is the interval length of the step price.

[0133]

[0134] Where: is the carbon trading cost of the system in period t.

[0135] Step 3: Utilize the power system flexibility resources to construct a new power system day-ahead two-stage robust dispatch model that considers multiple types of flexibility resources and carbon trading.

[0136] Step 3.1: Establish the objective function of the robust optimization model:

[0137] The goal of the day-ahead robust optimization model with multiple types of flexible resources taking into account carbon trading is to minimize the sum of the costs of the two stages, that is, the operating cost C during pre-dispatch. oper and the regulatory risk cost C during redispatching risk sum;

[0138]

[0139] in is the day-ahead dispatching plan; U is the uncertain parameters of the new energy and load in the problem; is the optimization variable during rescheduling; the pre-scheduling operation cost C oper It includes various sub-costs of the day-ahead dispatch plan, namely, unit operation cost The cost of reducing the load of Class A during period t, The cost of reducing Class B load and the dispatching cost of shiftable load in period t Carbon trading costs

[0140]

[0141] The control risk cost C during the rescheduling phase risk Including the cost of wind curtailment The cost of abandoning light and load loss cost

[0142]

[0143] Step 3.2: Establish constraints for the robust optimization model control phase:

[0144] Operation constraints during pre-scheduling:

[0145] The constraints in the day-ahead dispatch and pre-dispatch phase of the new power system taking carbon trading into account include flexible load constraints, power balance constraints, line transmission restrictions, and system reserves.

[0146] The relevant constraints of the flexible load are

[0147]

[0148] Where: is the elastic coefficient of Class A load that can be reduced and participate in dispatch, The reserve capacity for class B load reduction during period t; is the elastic coefficient of Class B load that can be dispatched, P t A-re P is the load reduction amount of Class A in period t; t B-re is the load reduction amount of Class B in period t; t sh* It is the starting period after the power consumption curve moves within a certain period of time; t sh*- (·) and t sh*+ (·) is the earliest and latest starting time period after the power consumption curve is shifted by the shiftable load; To compensate the unit price.

[0149] The power balance constraint of the power system is

[0150]

[0151] Where: w∈{0,1,…,W-1}, W is the number of wind farms; v∈{0,1,…,V-1}, V is the number of photovoltaic power stations; is the power forecast value of wind farm w in period t; is the power prediction value of the photovoltaic power station v in the period t; P t load,pre is the predicted power in period t; P t sh is the predicted value of the load that can be translated during period t; P t sh* is the power value in period t after the load is shifted.

[0152] The power system line transmission constraints are:

[0153]

[0154] Where: T l,k , T l,w , T l,v and T l,i are the power transmission allocation coefficients of unit k, wind farm w, photovoltaic power station v and load node i to line l in period t; F l max is the maximum power of branch l; is the predicted power of node i in period t, and

[0155] The spare capacity constraint is

[0156]

[0157] Where: and are the upper and lower spare capacity values ​​of the system in period t.

[0158] Operational constraints during rescheduling:

[0159] The constraints for day-ahead dispatch and re-dispatching of the new power system taking carbon trading into account include unit-related constraints, power balance constraints, and line transmission restrictions.

[0160] The upper and lower limits of unit output during the rescheduling phase:

[0161]

[0162] Where: is the power adjustment of unit k during period t; and are the maximum and minimum output values ​​of unit k respectively;

[0163] The ramp constraint of the unit during the re-dispatch phase is:

[0164]

[0165] The system power balance constraint in the rescheduling phase is:

[0166]

[0167] Where: and is the downward and upward climbing rate of unit k; is the power generation power of unit k during period t-1; P w,t and P v,t are the actual wind power value and photovoltaic output value simulated in the redispatch phase; P t The actual load demand power value simulated in the redispatch phase, including flexible load and rigid load; is the abandoned wind power of the w-th wind farm in the extreme scenario during period t; is the abandoned light power of the vth PV power station in the extreme scenario during period t; ΔP t cl It is the involuntary load loss of users on the load side during period t.

[0168] The line transmission power constraint in the rescheduling phase is:

[0169]

[0170] Where: P i,t is the actual load demand power value of node i in period t after scheduling.

[0171] Step 4: Organize the objective function and constraints of the two-stage robust optimization dispatch model of the new power system taking into account carbon trading into matrix form and solve it to achieve low-carbon economic operation of the new power system.

[0172] Step 4.1: Model solving algorithm

[0173] Based on the objective function and constraints of the two-stage robust optimization dispatch model of the new power system considering carbon trading constructed above, it is organized into a matrix form for description. The compact form of the proposed model can be expressed as

[0174]

[0175] Where: H oper(X) = 0 is the equality constraint condition in the pre-scheduling stage of the model; G oper (X)≤0 is the inequality constraint condition in the pre-scheduling stage of the model; H risk (X,U,Y)=0 represents the equality constraint condition in the rescheduling stage of the scheduling model; G risk (X,U,Y)≤0 represents the inequality constraint in the rescheduling phase of the scheduling model.

[0176] Since a min-max-min structure model is constructed, the first stage minimizes the operating cost, and the second stage minimizes the risk cost in severe scenarios. This structure is relatively complex to solve, and the model also contains uncertain variables, which makes the solution process more complicated. Therefore, the C&CG algorithm is used to solve the model. This method splits the actual problem into a main problem (MP) and a sub-problem (SP), and solves these two problems alternately to gradually approach the optimal solution. The mathematical expressions of the main problem MP and the sub-problem SP are as follows

[0177]

[0178] Where: θ is an auxiliary variable introduced to replace the subproblem, and then a temporary solution is directly obtained. Then, the structure of the subproblem is transformed through the duality theory, and the max-min is transformed into a max single-layer optimization problem;

[0179] Step 4.2: Solve the C&CG algorithm flow

[0180] Through the above analysis, the overall solution process can be decomposed into the following steps. Figure 3 As shown in the figure, it is the C&CG algorithm flow chart, which is as follows:

[0181] Step 1: Initialize the upper bound UB = +∞ and the lower bound LB = -∞ of the objective function, the number of iterations n = 1, set the convergence gap between the upper and lower bounds to ε, and set ε to a small positive number;

[0182] Step 2: Let θ = 0 and find the initial solution X 0 , and then substitute it into the subproblem to solve the initial extreme scenario U 1 ;

[0183] Step 3: Set the extreme scenario U n Substitute into the main problem and get the optimal solution (X n ,Y n ), update the lower bound LB to be equal to Y n ;

[0184] Step 4: The optimal solution X in step 3 n Substitute into the subproblem and get the optimal solution make Update the upper bound UB to the sum of C oper (X n );

[0185] Step 5: Determine whether UB - LB ≤ ε holds. If the condition holds, the operation ends; otherwise, let n = n + 1 and return to Step 3.

[0186] Refer to as Figure 4 shown, Figure 4 (a) is the system day-ahead robust scheduling scheme without considering the carbon trading mechanism when there is no flexible resource participating in power grid scheduling. Figure 4 (b) is the system day-ahead robust scheduling scheme considering the carbon trading mechanism when there is no flexible resource participating in power grid scheduling. Figure 4 (c) is the system day-ahead robust scheduling scheme without considering the carbon trading mechanism when there are flexible resources participating in power grid scheduling. Figure 4 (d) is the system day-ahead robust scheduling scheme considering the carbon trading mechanism when there are flexible resources participating in power grid scheduling. G1, G2, G3, G4, G5, and G6 are the unit numbers with different parameters respectively.

[0187] The present invention introduces a robust optimization method considering the uncertainties of wind power output, photovoltaic power output, and load forecasting, and fully utilizes the flexible resources of the power system to construct a new power system day-ahead two-stage robust scheduling model considering multiple types of flexible resources and carbon trading, and proposes a new power system day-ahead robust scheduling optimization method based on carbon trading. The present invention effectively improves the flexibility, cleanliness, and economy of the system by considering carbon trading and multiple types of flexible resources, and ensures the stable operation of the new power system under the grid connection of large-scale uncertain new energy.

[0188] It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article or terminal device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or also includes elements inherent to such process, method, article or terminal device. Without further limitation, the elements defined by the statement "include..." or "comprise..." do not exclude the existence of additional elements in the process, method, article or terminal device including the said elements. In addition, in this article, "greater than", "less than", "exceeding", etc. are understood not to include the present number; "above", "below", "within", etc. are understood to include the present number.

[0189] Although the above embodiments have been described, once those skilled in the art know the basic creative concepts, they can make additional changes and modifications to these embodiments. Therefore, the above description is only an embodiment of the present invention and does not limit the patent protection scope of the present invention. Any equivalent structure or equivalent process transformation made by using the contents of the specification and drawings of the present invention, or directly or indirectly used in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A new method for optimizing robust dispatch of power system taking carbon trading into account, characterized in that: The scheduling optimization method comprises the following steps: Step 1: Modeling multiple types of flexibility resources in the new power system; Step 2: A carbon emission calculation model for the power supply side units was constructed based on the physical characteristics of the units and the gas composition generated by power generation. A carbon emission quota model for the units was constructed based on the average carbon emission factor released by the Energy Bureau, and a carbon trading cost model for the system was further proposed. Step 3: Using the power system flexibility resources, a new power system day-ahead two-stage robust dispatch model is constructed that considers multiple types of flexibility resources and carbon trading; Step 4: Organize the objective function and constraints of the two-stage robust optimization dispatch model of the new power system taking into account carbon trading into matrix form and solve it to achieve low-carbon economic operation of the new power system.

2. The novel day-ahead robust dispatch optimization method for power systems taking carbon trading into account as claimed in claim 1 is characterized in that: Step 1 is to model multiple types of flexibility resources in the new power system. The specific model is as follows: Step 1.1: Establish a cost model for deep peak load regulation units: The operating cost of the deep peak load unit is as follows: In the formula, They are the operating cost, fuel consumption cost, start-up and shutdown cost, standby cost, additional coal consumption cost and life loss cost of conventional unit k; Step 1.2: Build a curtailable load model: For the load that can be reduced, the load aggregator receives the compensation price adjusted by the power grid to evaluate the response potential of the load users, and then re-declare the maximum load reduction that can be provided in the next 4 hours. The dispatch center makes a plan within the adjustable range according to the demand of the specific period, and then after reaching a cooperation, the load that can be reduced responds to the power grid dispatch at the correct time. Therefore, the load model that can be reduced can be shown as follows: Where: ΔP t A-re is the amount of load reduction that can be achieved during period t; is the elastic coefficient of the load that can be reduced and participate in the dispatch, is the elastic modulus that can reduce the load; and They represent the compensation price when the load-reducing users can just obtain benefits and the maximum response amount respectively; The maximum elastic coefficient that can reduce the load; is the compensation sensitivity of the load, The smaller the value of, the more sensitive the user is to the compensation price, and the greater the impact of price changes on the user; The cost of reducing load for period t; Step 1.3: Create a translatable load model: When the compensation price released by the dispatch center is low, the benefits gained by users from changing their electricity consumption time cannot make up for the inconvenience and loss caused by shifting electricity consumption. Therefore, users will be less motivated to participate in the dispatch and it will be difficult to meet the needs of the dispatch center. According to the principles of consumer psychology, the shiftable load model can be expressed as: Where: t sh* It is the starting period after the power consumption curve moves within a certain period of time; t sh*- (·) and t sh*+ (·) is the earliest and latest starting time period after the shiftable load shifts the power consumption curve.

3. The novel day-ahead robust dispatch optimization method for power systems taking carbon trading into account according to claim 1, characterized in that: Step 2 is to model the carbon trading cost of the system. The specific model is as follows: Step 2.1: Establish a method for calculating carbon emissions from the power system: The carbon emission intensity of coal-fired power generation unit k in period t can be calculated based on the coal consumption per unit of electricity using the following formula: Where: μ k is the carbon content of coal used in coal-fired unit k; is the molar mass of carbon dioxide; M C is the molar mass of carbon; the coal loss of coal-fired unit k per kilowatt-hour of electricity produced The column vector of the carbon emission intensity of the unit in period t is defined as Its kth element is After the carbon emission intensity of the unit is calculated, the carbon emission of the unit can be calculated through the power generation of each unit. The cumulative carbon emission of the power supply unit during period t is The calculation formula is: Step 2.2: Establish a method for allocating carbon emission quotas for the power system: The baseline method is used to allocate the free quota of carbon emissions. The free quota of carbon emissions of the units in the system during period t is Where: It is the free quota of the system; η carb is the free quota per unit power, and its value is determined by the average emission factor of Anhui Province calculated by the state, η carb =0.5703tCO2 / MWh; Step 2.3: Establish a method for calculating the carbon trading cost of the power system: A gradient price strategy is adopted to set the carbon trading price, and the carbon trading cost model is as follows; Where: It is a tiered carbon trading price; is the market benchmark price; α is the step price growth rate; d is the interval length of the step price; Where: is the carbon trading cost of the system in period t.

4. The novel day-ahead robust dispatch optimization method for power systems taking carbon trading into account according to claim 1, characterized in that: The objective function of the robust scheduling model before step 3 is as follows: Step 3.1: Establish the objective function of the robust optimization model: The goal of the day-ahead robust optimization model with multiple types of flexible resources taking into account carbon trading is to minimize the sum of the costs of the two stages, that is, the operating cost C during pre-dispatch. oper and the regulatory risk cost C during redispatching risk sum; in is the day-ahead scheduling plan; U is the uncertain parameter of the new energy and load in the problem; is the optimization variable during rescheduling; the pre-scheduling operation cost C oper It includes various sub-costs of the day-ahead dispatch plan, namely, unit operation cost The cost of reducing the load of Class A during period t, The cost of reducing Class B load and the dispatching cost of shiftable load in period t Carbon trading costs The control risk cost C during the rescheduling phase risk Including the cost of wind curtailment The cost of abandoning light and load loss cost Step 3.2: Establish constraints for the robust optimization model control phase: Operation constraints during pre-scheduling: The constraints of the pre-dispatch stage of the new power system day-ahead dispatch taking into account carbon trading include flexible load constraints, power balance constraints, line transmission restrictions, and system reserve; The relevant constraints of the flexible load are Where: is the elastic coefficient of Class A load that can be reduced and participate in dispatch, The reserve capacity for class B load reduction during period t; is the elastic coefficient of Class B load that can be dispatched, P t A-re P is the load reduction capacity of Class A in period t; t B-re is the load reduction amount of Class B in period t; t sh* It is the starting period after the power consumption curve moves within a certain period of time; t sh*- (·) and t sh*+ (·) is the earliest and latest starting time period after the power consumption curve is shifted by the shiftable load; To compensate for the unit price; The power balance constraint of the power system is Where: w∈{0,1,…,W-1}, W is the number of wind farms; v∈{0,1,…,V-1}, V is the number of photovoltaic power stations; is the power forecast value of wind farm w in period t; is the power prediction value of the photovoltaic power station v in the period t; P t load,pre is the predicted power in period t; P t sh is the predicted value of the load that can be translated during period t; P t sh* is the power value in period t after load shift; The power system line transmission constraints are: Where: T l,k , T l,w , T l,v and T l,i are the power transmission allocation coefficients of unit k, wind farm w, photovoltaic power station v and load node i to line l in period t; F l max is the maximum power of branch l; is the predicted power of node i in time period t, and The spare capacity constraint is Where: and is the upper and lower reserve capacity value of the system in period t; Operational constraints during rescheduling: The constraints for day-ahead dispatch and re-dispatch of the new power system taking carbon trading into account include unit-related constraints, power balance constraints, and line transmission restrictions; Upper and lower limits of unit output during the rescheduling phase Where: is the power adjustment of unit k during period t; and are the maximum and minimum output values ​​of unit k respectively; The ramp constraint of the unit in the re-dispatch phase is: The system power balance constraint in the rescheduling phase is: Where: and is the downward and upward climbing rate of unit k; is the power generation power of unit k during period t-1; P w,t and P v,t are the actual wind power value and photovoltaic output value simulated in the redispatch phase; P t The actual load demand power value simulated in the redispatch phase, including flexible load and rigid load; is the abandoned wind power of the w-th wind farm in the extreme scenario during period t; is the abandoned light power of the vth PV power station in the extreme scenario during period t; ΔP t cl is the involuntary load loss of users on the load side during period t; The line transmission power constraint in the rescheduling phase is: Where: P i,t is the actual load demand power value of node i in period t after scheduling.

5. The novel day-ahead robust dispatch optimization method for power systems taking carbon trading into account as claimed in claim 1, characterized in that: The objective function and constraint conditions of the two-stage robust optimization scheduling model in step 4 are organized into a matrix form, as shown below: Step 4.1: Model solving algorithm Based on the objective function and constraints of the two-stage robust optimization dispatch model of the new power system considering carbon trading constructed above, it is organized into a matrix form for description. The compact form of the proposed model can be expressed as Where: H oper (X) = 0 is the equality constraint condition in the pre-scheduling stage of the model; G oper (X)≤0 is the inequality constraint condition in the pre-scheduling stage of the model; H risk (X,U,Y)=0 represents the equality constraint condition in the rescheduling stage of the scheduling model; G risk (X,U,Y)≤0 represents the inequality constraint condition in the rescheduling phase of the scheduling model; The C&CG algorithm is used to solve the model. This method divides the actual problem into a main problem (MP) and a subproblem (SP), and solves these two problems alternately to gradually approach the optimal solution. The mathematical expressions of the main problem MP and the subproblem SP are as follows Where: θ is an auxiliary variable introduced to replace the subproblem, and then a temporary solution is directly obtained. Then, the structure of the subproblem is transformed through the duality theory, and the max-min is transformed into a max single-layer optimization problem; Step 4.2: Solve the C&CG algorithm flow Step 1: Initialize the upper bound UB = +∞ and the lower bound LB = -∞ of the objective function, the number of iterations n = 1, set the convergence gap between the upper and lower bounds to ε, and set ε to a small positive number; Step 2: Let θ = 0, find the initial solution X0, and then substitute it into the sub-problem to solve the initial extreme scenario U1; Step 3: Set the extreme scenario U n Substitute into the main problem and get the optimal solution (X n ,Y n ), update the lower bound LB to be equal to Y n ; Step 4: The optimal solution X in step 3 n Substitute into the subproblem and get the optimal solution make Update the upper bound UB to With C oper (X n ); Step 5: Determine whether UB-LB≤ε is established. If the condition is established, the operation ends; otherwise, n=n+1 is set and the process returns to step 3.

Images