A method for calculating the optimal benchmark carbon price

Abstract

The present invention discloses a method for calculating an optimal benchmark carbon price, which includes: constructing a two-layer model of an optimal benchmark carbon price for an electric power system; generating a vertex set of a parameter space according to the constraints of the two-layer model, and constructing a boundary of the parameter space; solving the interior points of the parameter space according to a benchmark carbon price that meets the requirements, and constructing a simplex space; constructing a first decision mapping function of a lower-layer model, and solving a maximum error function; using a first decision mapping function whose maximum error function value is less than a first threshold as a second decision mapping function, and if the second decision mapping function meets a preset space set condition, using the second decision mapping function as a constraint, solving the minimization of the benchmark carbon price as an objective function, and generating an optimal benchmark carbon price, while ensuring the calculation accuracy, improving the calculation efficiency of the optimal benchmark carbon price, so as to achieve an improvement in the accuracy and efficiency of dispatching of an electric power system that meets carbon emission reduction.

Description

Technical Field

[0001] The present invention relates to the technical field of optimal benchmark carbon price calculation, and in particular to a method, device and storage medium for calculating the optimal benchmark carbon price. Background Art

[0002] With the low-carbon transformation of energy, the dispatch of power systems must meet the requirements of carbon emission reduction. If the price of carbon trading is too high, it will excessively increase the operating costs of the energy system and affect the enthusiasm of power generation companies; if the price of carbon trading is too low, the actual carbon emissions will not meet the corresponding carbon emission control requirements. Therefore, it is necessary to measure and determine the optimal benchmark carbon price under the total carbon emission control requirements, and then control the dispatch of power systems so that the dispatch of power systems meets the requirements of carbon emission reduction.

[0003] The existing methods for solving the optimal benchmark carbon price mainly include analytical methods and heuristic methods. Among them, the analytical method converts the two-layer model into a single-layer model through the Karush-Kuhn-Tucker (KKT) condition, and then further converts the single-layer model into a mixed integer linear programming problem based on the strong duality theorem and the binary extension method, and finally calls the solver for solving. However, since the converted model contains more integer variables, this method requires a long calculation time and has low calculation efficiency. Heuristic methods such as genetic algorithms and particle swarm algorithms update the benchmark carbon price through interactive iterations of upper and lower models until the convergence conditions are met. However, this type of method depends on the setting of parameters such as initial values. If the parameters are set improperly, it may fall into a local optimum, resulting in the inability to obtain the optimal benchmark carbon price and low accuracy of the calculation results. Therefore, the existing methods for solving the optimal benchmark carbon price have low calculation efficiency and low calculation result accuracy, which in turn reduces the efficiency and accuracy of the dispatch control of the power system based on the optimal benchmark carbon price. Summary of the invention

[0004] The present invention provides a method, device and storage medium for calculating an optimal benchmark carbon price, which improves the calculation efficiency of the optimal benchmark carbon price while ensuring the calculation accuracy, so as to improve the accuracy and efficiency of power system scheduling that meets carbon emission reduction.

[0005] The present invention provides a method for calculating an optimal benchmark carbon price, comprising: constructing a two-layer model of an optimal benchmark carbon price for an electric power system according to carbon emission data and generator set data of an electric power system; wherein the first objective function of the upper model is to minimize the benchmark carbon price; the first constraint of the upper model includes: the actual total carbon emission is less than or equal to the preset total carbon emission; the second objective function of the lower model is the minimum comprehensive operating cost of the regional electric power system under the benchmark carbon price condition, and the minimum comprehensive operating cost includes: the minimum operating cost of the generator set and the carbon trading cost; the second constraint of the lower model includes: the upper and lower limit constraints of the unit output, the climbing constraint, the system flow constraint and the interconnection line transmission power constraint;

[0006] According to the first constraint and the second constraint, a vertex set of the parameter space is generated, and the boundary of the parameter space is constructed; according to the benchmark carbon price that meets the preset requirements, the interior point of the parameter space is solved; s vertices and 1 interior point in the parameter space are selected to form a simplex space; s is a positive integer;

[0007] Taking the benchmark carbon price of the lower model as the input variable and the output of the corresponding power generation unit as the output variable, constructing the first decision mapping function of the lower model; constructing the maximum error function according to the first decision mapping function; solving the maximum error function according to the second constraint and the system operation cost approximate value constraint to generate the maximum error function value;

[0008] The first decision mapping function whose maximum error function value is less than the first threshold is used as the second decision mapping function. If the second decision mapping function satisfies the preset spatial set condition, minimizing the benchmark carbon price is used as the third objective function of the single-layer optimization model. The first constraint and the second decision mapping function are used as the third constraints of the single-layer optimization model, and the single-layer optimization model is solved to generate the optimal benchmark carbon price.

[0009] As a preferred solution, the present invention applies the decision mapping method to the calculation of the optimal benchmark carbon price, simplifies the complex two-layer optimization problem into a single-layer optimization problem based on the decision mapping function, and realizes the accurate calculation of the optimal benchmark carbon price. In the process of constructing the decision mapping function, the existing decision mapping method constructs a simplex space by randomly selecting points in the parameter space, resulting in a series of spaces that may have overlapping parts, reducing the solution efficiency of the algorithm. The present invention utilizes the constraints of the two-layer model to generate a set of vertex points in the parameter space, combines the benchmark carbon price of the lower model to solve the internal points of the parameter space, and constructs a simplex space, thereby avoiding the mutual overlap of the simplex space, improving the calculation efficiency and accuracy of the optimal benchmark carbon price, and improving the efficiency of dispatching and controlling the power system based on the optimal benchmark carbon price.

[0010] As a preferred solution, according to the first constraint and the second constraint, a vertex set of the parameter space is generated, and a boundary of the parameter space is constructed, specifically:

[0011] Convert all inequality constraints in the first constraint and the second constraint into equality constraints, iteratively solve the first equality equation group consisting of the equality constraints according to a preset rule, and construct a vertex set from solutions that satisfy all inequality constraints before conversion;

[0012] Select s points in the vertex set to form the s-1 dimensional boundary of the parameter space:

[0013] Among them, Θ v is the vertex set, Θ is the parameter space, and s points in the parameter space is a solution that satisfies all inequality constraints in the first and second constraints.

[0014] The inequality constraints in the first constraint and the second constraint are converted into equality constraints, and the first equality equation group consisting of the equality constraints is iteratively solved according to a preset rule, and the solutions satisfying all the inequality constraints before the conversion are constructed into a vertex set, specifically:

[0015] The total number of iterations is set according to the number of the inequality constraints and the dimension of the parameter variables of the inequality constraints; each time, a non-completely identical preset number of inequality constraints is selected in the first set of equations as the second set of equations to be solved to obtain a solution to the second set of equations until the number of iterations reaches the total number of iterations; wherein the number of the non-completely identical preset number of inequality constraints is the dimension of the parameter variables in the inequality constraints; the dimension of the parameter variables is not greater than the number of all inequality constraints; and the solution of the second set of equations that satisfies all inequality constraints is constructed into a vertex set.

[0016] According to the benchmark carbon price that meets the preset requirements, the interior points of the parameter space are solved, specifically:

[0017] Select an arbitrary first benchmark carbon price to solve the lower model and obtain the corresponding optimal output of the first generator set; substitute the first benchmark carbon price and the optimal output of the first generator set into all inequality constraints as the first inequality constraints, and determine whether all first inequality constraints meet the preset conditions; if so, take the first benchmark carbon price as the interior point;

[0018] If not, the first inequality constraint is updated by the second benchmark carbon price and the optimal output of the second generator set until the updated first inequality constraints meet the preset conditions; the second benchmark carbon price and the optimal output of the second generator set are obtained by solving the obstacle function in a preset search area; the obstacle function is constructed according to different obstacle factors, the first benchmark carbon price and the optimal output of the first generator set.

[0019] Substitute the first benchmark carbon price and the first generator set optimal output into all inequality constraints, and determine whether all first inequality constraints meet preset conditions, specifically:

[0020] Convert all inequality constraints to standard form: g j (x)≥0,1≤j≤t;

[0021] In the formula, x represents the variable related to the inequality constraint; t is the number of all inequality constraints; j is the number of the jth inequality constraint;

[0022] The first benchmark carbon price and the optimal output of the first generator set are substituted into all inequality constraints, and the subscripts of the standard form of the inequality constraints whose values ​​are not greater than zero are constructed into a first subscript set; if the first subscript set is an empty set, all first inequality constraints meet the preset conditions.

[0023] The first inequality constraint is updated by the second benchmark carbon price and the optimal output of the second generator set until the updated first inequality constraints meet the preset conditions, specifically:

[0024] Constructing a barrier function according to the barrier factor, the first benchmark carbon price and the optimal output of the first generator set; solving the barrier function in a preset search area to obtain the second benchmark carbon price and the optimal output of the second generator set;

[0025] Substitute the second benchmark carbon price and the optimal output of the second generator set into all inequality constraints to update the first inequality constraint, and determine whether all updated first inequality constraints meet preset conditions; if not, update the barrier factor and re-solve the second benchmark carbon price and the optimal output of the second generator set.

[0026] As a preferred solution, in the process of constructing the decision mapping function, the present invention obtains the vertex set of the parameter space by iteratively solving the constraints of the double-layer model to achieve vertex enumeration; the interior point segmentation is achieved by taking the benchmark carbon price that meets the conditions as the interior point of the parameter space; the decision mapping method is further improved based on the vertex enumeration and interior point segmentation technology to achieve complete and accurate segmentation of the parameter space, improve the calculation efficiency of the optimal benchmark carbon price, and then improve the efficiency of the dispatching control of the power system based on the optimal benchmark carbon price.

[0027] As a preferred solution, the benchmark carbon price of the lower model is used as the input variable, and the output of the corresponding generator set is used as the output variable to construct the first decision mapping function of the lower model, which is specifically:

[0028] Among them, H i,t is the first decision mapping function, is the benchmark carbon price, P i,t is the output of traditional generator set i in period t.

[0029] According to the first decision mapping function, a maximum error function is constructed, and according to the second constraint and the system operation cost approximate value constraint, the maximum error function is solved to generate a maximum error function value, specifically:

[0030] Obtain the benchmark carbon price in the simplex space to construct an interpolation matrix, and construct an optimal decision matrix according to the output of the power generation unit corresponding to the benchmark carbon price; use linear interpolation to establish an approximate expression of the first decision mapping function:

[0031] Among them, M is the interpolation matrix, X i,t is the optimal decision matrix is the first decision mapping function, is the approximate expression of the first decision mapping function; G q is a simplex space;

[0032] Using the approximate expression of the first decision mapping function as the first decision mapping function;

[0033] Construct the maximum error function:

[0034]

[0035] Among them, CP is the maximum error function, Represents the mapping function based on the first decision The calculated approximate value of the system operation cost; T represents an economic dispatch cycle; N c Indicates the number of generator sets in the system; a i b represents the quadratic coefficient of the operating cost of traditional generator set i; i represents the linear coefficient of the operating cost of the traditional generator set i; c i The constant term representing the operating cost of traditional generator set i; μ i represents the carbon emission coefficient per unit of electricity generated by generator set i, and ε represents the free carbon emission coefficient per unit of electricity generated; represents the carbon trading price of power generation unit i;

[0036] The approximate value constraint of the system operation cost is:

[0037]

[0038] According to the second constraint and the approximate value constraint of the system operating cost, the maximum error function is solved to obtain the maximum error value and the optimal solution of the benchmark carbon price.

[0039] The first decision mapping function whose maximum error function value is less than the first threshold is used as the second decision mapping function, and if the second decision mapping function meets the preset space set condition, it also includes:

[0040] If the solution of the third objective function is not less than the first threshold, then according to the optimal solution of the benchmark carbon price, s interior points are selected from the simplex space, and corresponding new first simplex spaces are respectively constructed; according to the new first simplex space, the interpolation matrix and the optimal decision matrix are updated; according to the updated interpolation matrix and the optimal decision matrix, the first decision mapping function is updated; s is a positive integer;

[0041] If all simplex spaces of the second decision mapping function do not satisfy the preset space set condition, then s+1 interior points are reselected to form a new second simplex space, and the first decision mapping function is reconstructed;

[0042] The preset space set condition is:

[0043] Among them, G1-G q are the 1st, 2nd, ..., qth simplex spaces respectively.

[0044] Single-layer optimization model, specifically:

[0045] The third objective function of the single-layer optimization model is to minimize the benchmark carbon price:

[0046] The third constraint of the single-layer optimization model includes: a first constraint and a second decision mapping function;

[0047] The first constraint is: E S ≤(1-λ redu )E S0 ;

[0048]

[0049] Among them, E S represents the actual total carbon emissions of the generator set considering carbon trading; redu Represents the carbon reduction coefficient; E S0 represents the actual total carbon emissions of the power generation unit without carbon trading; T represents an economic dispatch cycle; N c Indicates the number of generator sets in the system; μi P represents the carbon emission coefficient of unit power generation of generator set i; i,t represents the output of traditional generator set i in period t;

[0050]

[0051] in, is the second decision mapping function on the nth simplex space, G n is the nth simplex space; n = 1,...,k,...q; k and q are both positive integers.

[0052] As a preferred solution, the present invention applies the decision mapping method to the calculation of the optimal benchmark carbon price, simplifies the complex two-layer optimization problem into a single-layer optimization problem based on the decision mapping function, reduces the instability of the solution results caused by the repeated iterations of the upper and lower layer models, and achieves accurate calculation of the optimal benchmark carbon price, thereby improving the accuracy of dispatching and control of the power system based on energy low-carbon transformation.

[0053] Accordingly, the present invention also provides an optimal benchmark carbon price calculation device, comprising: a double-layer model construction module, a simplex space construction module, a decision mapping function construction module and an optimal benchmark carbon price generation module;

[0054] The two-layer model building module is used to construct a two-layer model of the optimal benchmark carbon price of the power system based on the carbon emission data and the generator data of the power system; the first objective function of the upper model is to minimize the benchmark carbon price; the first constraint of the upper model includes: the actual total carbon emissions are less than or equal to the preset total carbon emissions; the second objective function of the lower model is the minimum comprehensive operating cost of the regional power system under the benchmark carbon price condition, and the minimum comprehensive operating cost includes: the minimum operating cost of the generator set and the carbon trading cost; the second constraint of the lower model includes: the upper and lower limit constraints of the unit output, the climbing constraint, the system flow constraint and the interconnection line transmission power constraint;

[0055] The simplex space construction module is used to generate a vertex set of the parameter space according to the first constraint and the second constraint, and construct the boundary of the parameter space; solve the interior point of the parameter space according to the benchmark carbon price that meets the preset requirements; select s vertices and 1 interior point in the parameter space to form a simplex space; s is a positive integer;

[0056] The decision mapping function construction module is used to take the benchmark carbon price of the lower model as an input variable and the output of the corresponding generator set as an output variable to construct a first decision mapping function of the lower model; construct a maximum error function according to the first decision mapping function; solve the maximum error function according to the second constraint and the system operation cost approximate value constraint to generate a maximum error function value;

[0057] The optimal benchmark carbon price generation module is used to use the first decision mapping function whose maximum error function value is less than the first threshold as the second decision mapping function. If the second decision mapping function satisfies the preset spatial set condition, the minimization of the benchmark carbon price is used as the third objective function of the single-layer optimization model, the first constraint and the second decision mapping function are used as the third constraints of the single-layer optimization model, and the single-layer optimization model is solved to generate the optimal benchmark carbon price.

[0058] As a preferred solution, the decision mapping function construction module of the device of the present invention applies the decision mapping method to the calculation of the optimal benchmark carbon price, simplifies the complex two-layer optimization problem into a single-layer optimization problem based on the decision mapping function, and uses the optimal benchmark carbon price generation module to achieve accurate calculation of the optimal benchmark carbon price. In the process of constructing the decision mapping function, the existing decision mapping method constructs a simplex space by randomly selecting points in the parameter space, resulting in a series of spaces that may have overlapping parts, reducing the solution efficiency of the algorithm. The simplex space construction module of the present invention generates a set of vertex points in the parameter space using the constraints of the two-layer model, solves the internal points of the parameter space in combination with the benchmark carbon price of the lower model, and constructs a simplex space, thereby avoiding the mutual overlap of the simplex space, improving the calculation efficiency and accuracy of the optimal benchmark carbon price, and improving the efficiency of dispatching and controlling the power system based on the optimal benchmark carbon price.

[0059] Correspondingly, the present invention also provides a computer-readable storage medium, which includes a stored computer program; wherein, when the computer program is running, it controls the device where the computer-readable storage medium is located to execute a method for calculating an optimal benchmark carbon price as described in the content of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 It is a flow chart of an embodiment of a method for calculating an optimal benchmark carbon price provided by the present invention;

[0061] Figure 2 It is a structural schematic diagram of an embodiment of an optimal benchmark carbon price calculation device provided by the present invention. DETAILED DESCRIPTION

[0062] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0063] Embodiment 1

[0064] Please refer to Figure 1 , a method for calculating an optimal benchmark carbon price provided by an embodiment of the present invention, comprising steps S101-S104:

[0065] Step S101: construct a two-layer model of the optimal benchmark carbon price of the power system based on the carbon emission data and the generator data of the power system; wherein the first objective function of the upper model is to minimize the benchmark carbon price; the first constraint of the upper model includes: the actual total carbon emissions are less than or equal to the preset total carbon emissions; the second objective function of the lower model is the minimum comprehensive operating cost of the regional power system under the benchmark carbon price condition, and the minimum comprehensive operating cost includes: the minimum operating cost of the generator set and the carbon trading cost; the second constraint of the lower model includes: the upper and lower limit constraints of the unit output, the climbing constraint, the system flow constraint and the interconnection line transmission power constraint;

[0066] In this embodiment, the first objective function of the upper model is to minimize the benchmark carbon price:

[0067]

[0068] The first constraint of the upper model is:

[0069] E S ≤(1-λ redu )E S0 (2)

[0070]

[0071] In the formula, represents the benchmark carbon price; E S represents the actual total carbon emissions of the generator set considering carbon trading; redu Represents the carbon reduction coefficient; E S0 represents the actual total carbon emissions of the power generation unit without carbon trading; T represents an economic dispatch cycle; N c Indicates the number of generator sets in the system; μ i P represents the carbon emission coefficient of unit power generation of generator set i; i,t Represents the output of traditional generator set i in time period t.

[0072] In the carbon trading market, each carbon emission source (such as a traditional power generation unit) receives a limited free carbon quota. If the carbon emissions of a carbon emission source exceed the quota, it must purchase carbon emission rights on the market. On the contrary, if the carbon emissions of a carbon emission source are lower than the quota, it can sell carbon emission rights on the market. The free carbon quota and actual carbon emissions of each unit can be calculated using equations (4)-(5), as shown below:

[0073]

[0074] In the formula, E i represents the actual carbon emissions of generator set i; represents the free carbon quota of generator set i; ε represents the free carbon emission coefficient per unit of electricity generation;

[0075] Under the reward-penalty ladder carbon trading mechanism, the carbon trading cost of each power generation unit can be calculated by equations (6)-(7), namely:

[0076]

[0077] In the formula, C(E i ) represents the carbon trading cost of power generation unit i; represents the carbon trading price of generator set i; δ represents the reward coefficient in the ladder-type carbon trading mechanism; σ represents the penalty coefficient in the ladder-type carbon trading mechanism. η represents the carbon emission interval division constant.

[0078] The second objective function of the lower model is to calculate the minimum comprehensive operating cost of the regional power system under a given benchmark carbon price:

[0079] The comprehensive operating cost of the system consists of the operating cost of traditional power generation units and the carbon trading cost, namely:

[0080]

[0081] In the formula, F sod Represents the comprehensive operating cost of the system; a i b represents the quadratic coefficient of the operating cost of traditional generator set i; i represents the linear coefficient of the operating cost of the traditional generator set i; c i The constant term representing the operating cost of the traditional generator set i;

[0082] At the same time, the generator sets in the system must also meet certain physical constraints, including upper and lower limits of the unit output and climbing constraints, as shown in equations (9)-(10):

[0083] P i,min ≤P i,t ≤P i,max ; 1≤i≤Nc (9)

[0084]

[0085] Where P i,min represents the minimum output of generator set i; P i,max represents the maximum output of generator set i; Indicates the upper limit of the ramp rate of generator set i; Indicates the lower limit of the ramp rate of generator set i;

[0086] The system power flow constraints and tie line transmission power constraints are shown in equations (11)-(14).

[0087]

[0088] P l,t =B m,n (θ m,t -θ n,t );1≤t≤T,1≤l≤N br (12)

[0089] -P l,max ≤P l,t ≤P l,max ; 1≤t≤T,1≤l≤N br (13)

[0090] θ m,min ≤θ m,t ≤θ m,max ; 1≤t≤T,1≤m≤N bus (14)

[0091] In the formula, Represents the set of generator sets at node n; represents the set of load components on node n; represents the set of all other nodes connected to node n; P d,t represents the power of load d in time period t; B n,m represents the conductance of the line between node n and node m; θ n,t represents the phase angle of node n in time period t; P l,t N represents the transmission power of branch l in time period t; br Indicates the number of branches; N bus represents the number of nodes; P l,max represents the maximum transmission power of the branch; θ m,min and θ m,max Indicates the minimum and maximum values ​​of the node phase angle.

[0092] Step S102: Generate a vertex set of the parameter space according to the first constraint and the second constraint, and construct the boundary of the parameter space; solve the interior point of the parameter space according to the benchmark carbon price that meets the preset requirements; select s vertices and 1 interior point in the parameter space to form a simplex space; s is a positive integer;

[0093] In this embodiment, according to the first constraint and the second constraint, a vertex set of the parameter space is generated, and a boundary of the parameter space is constructed, specifically:

[0094] Convert all inequality constraints in the first constraint and the second constraint into equality constraints, iteratively solve the first equality equation group consisting of the equality constraints according to a preset rule, and construct a vertex set from solutions that satisfy all inequality constraints before conversion;

[0095] Select s vertices from the vertex set to form the s-1 dimensional boundary of the parameter space:

[0096] Among them, Θ v is the vertex set, Θ is the parameter space, and s points in the parameter space is a solution that satisfies all inequality constraints in the first and second constraints.

[0097] In this embodiment, the inequality constraints in the first constraint and the second constraint are converted into equality constraints, and the first equation system consisting of the equality constraints is iteratively solved according to preset rules, and the solutions that satisfy all the inequality constraints before the conversion are constructed into a vertex set, specifically:

[0098] The total number of iterations is set according to the number of the inequality constraints and the dimension of the parameter variables of the inequality constraints; each time, a non-completely identical preset number of inequality constraints is selected in the first set of equations as the second set of equations to be solved to obtain a solution to the second set of equations until the number of iterations reaches the total number of iterations; wherein the number of the non-completely identical preset number of inequality constraints is the dimension of the parameter variables in the inequality constraints; the dimension of the parameter variables is not greater than the number of all inequality constraints; and the solution of the second set of equations that satisfies all inequality constraints is constructed into a vertex set.

[0099] As a preferred solution, a specific process for constructing a vertex set is provided: the inequality constraints in the feasible domain determined by equations (2)-(7) and equations (9)-(14) are converted into equality constraints, and all the equation systems in the model are solved; wherein the number of equation systems to be solved is determined by the number t of inequalities in the feasible domain and the dimension s of the parameter variables, specifically:

[0100] Initialization: number of iterations k = 0, vertex set is empty set Θ v=φ;

[0101] Determine whether the number of iterations satisfies the formula:

[0102] If yes, let k = k + 1; select s inequality constraints in the feasible domain and convert them into equality constraints; call the solver CONOPT to solve the equation system and obtain the solution

[0103] Determine the solution Do all the inequality constraints in the problem meet? If so, solve Stored in vertex set Θ v ; After the iteration is completed, output the vertex set Θ v .

[0104] In this embodiment, according to the benchmark carbon price that meets the preset requirements, the interior point of the parameter space is solved, specifically:

[0105] Select an arbitrary first benchmark carbon price to solve the lower model and obtain the corresponding optimal output of the first generator set; substitute the first benchmark carbon price and the optimal output of the first generator set into all inequality constraints as the first inequality constraints, and determine whether all first inequality constraints meet the preset conditions; if so, take the first benchmark carbon price as the interior point;

[0106] If not, the first inequality constraint is updated by the second benchmark carbon price and the optimal output of the second generator set until the updated first inequality constraints meet the preset conditions; the second benchmark carbon price and the optimal output of the second generator set are obtained by solving the obstacle function in a preset search area; the obstacle function is constructed according to different obstacle factors, the first benchmark carbon price and the optimal output of the first generator set.

[0107] In this embodiment, the first benchmark carbon price and the optimal output of the first generator set are substituted into all inequality constraints, and it is determined whether all first inequality constraints meet preset conditions, specifically:

[0108] Convert all inequality constraints to standard form: g j (x)≥0,1≤j≤t;

[0109] In the formula, x represents the variable related to the inequality constraint; t is the number of all inequality constraints; j is the number of the jth inequality constraint;

[0110] The first benchmark carbon price and the optimal output of the first generator set are substituted into all inequality constraints, and the subscripts of the standard form of the inequality constraints whose values ​​are not greater than zero are constructed into a first subscript set; if the first subscript set is an empty set, all first inequality constraints meet the preset conditions.

[0111] In this embodiment, the first inequality constraint is updated by the second benchmark carbon price and the second generator set optimal output until the updated first inequality constraint satisfies the preset conditions, specifically:

[0112] Constructing a barrier function according to the barrier factor, the first benchmark carbon price and the optimal output of the first generator set; solving the barrier function in a preset search area to obtain the second benchmark carbon price and the optimal output of the second generator set;

[0113] Substitute the second benchmark carbon price and the optimal output of the second generator set into all inequality constraints to update the first inequality constraint, and determine whether all updated first inequality constraints meet preset conditions; if not, update the barrier factor and re-solve the second benchmark carbon price and the optimal output of the second generator set.

[0114] As a preferred solution, a specific process of using a sequence unconstrained minimization method to generate interior points in the parameter space is provided:

[0115] 1) Any given initial point Solve the lower model to obtain the optimal output of the generator set and other related variables, such as Set the initial barrier factor r (1) >0, reduce the coefficient β∈(0,1), and set k=1;

[0116] 2) Convert all inequality constraints in the optimal benchmark carbon price model into the following standard form:

[0117] g j (x)≥0,1≤j≤t (15)

[0118] In the formula, x represents the variable related to the inequality constraint, such as P i,t wait.

[0119] 3) For Determine the subscript set T of the inequality constraints (k) and S (k) ,Right now:

[0120]

[0121] 4) If S (k) =φ, then is the interior point,

[0122] 5) If S (k) ≠φ, construct barrier function And solve for the minimum point

[0123] Determine the search area R (k)

[0124]

[0125] Constructing a barrier function

[0126]

[0127] by As the starting point, at R (k) Internal solution barrier function The minimum point

[0128] 6) Let r (k+1) =βr (k) , set k=k+1, and return to step 2).

[0129] In this embodiment, the vertex set θ is selected v s points in Constructing parameter space Θ The s-1 dimensional boundary:

[0130]

[0131] Select Θ bound and the interior points of the parameter space Θ Constructing the simplex space G q , which can be specifically expressed as:

[0132]

[0133] Step S103: Taking the benchmark carbon price of the lower model as an input variable and the output of the corresponding generator set as an output variable, constructing a first decision mapping function of the lower model; constructing a maximum error function according to the first decision mapping function; solving the maximum error function according to the second constraint and the system operation cost approximate value constraint to generate a maximum error function value;

[0134] In this embodiment, the benchmark carbon price of the lower model is used as an input variable, and the output of the corresponding generator set is used as an output variable to construct the first decision mapping function of the lower model, which is specifically:

[0135] Among them, H i,t is the first decision mapping function, is the benchmark carbon price, P i,t is the output of generator set i in period t.

[0136] In this embodiment, in the lower optimization problem, the decision variable is the output P of the generator set. i,t , the parameter variable is the benchmark carbon price (Input from the upper model to the lower model). The parameter space is represented by Θ and is defined as: There is an optimal solution to the underlying optimization problem.

[0137] Randomly select a set of points in the parameter space Θ The simplex space G formed by this set of points q It can be expressed as:

[0138]

[0139] Where q is the simplex space G q The serial number; s represents the parameter variable The dimension of k Represents a rational number between 0 and 1.

[0140] Furthermore, in the simplex space G q The interpolation matrix M can be expressed as:

[0141]

[0142] In the simplex space G q On the optimal decision matrix X i,t It can be expressed as:

[0143]

[0144] Where P i,t,1 ,P i,t,2 ,…P i,t,s+1 At each benchmark carbon price The lower-level optimization problems are solved separately to obtain the optimal decision for generator set i and time period t.

[0145] The first decision mapping function of the lower model is H i,t It is represented and defined as follows:

[0146]

[0147] The first decision mapping function H i,t The input variable is the optimal benchmark carbon price The output variable is the optimal decision P for generator set i and time period t i,t However, the decision mapping function H i,t It is difficult to solve directly, so this embodiment uses linear interpolation to establish its approximate expression

[0148] In this embodiment, according to the first decision mapping function, a maximum error function is constructed, and according to the second constraint and the system operation cost approximate value constraint, the maximum error function is solved to generate a maximum error function value, specifically:

[0149] Obtain the benchmark carbon price in the simplex space to construct an interpolation matrix, and construct an optimal decision matrix according to the output of the power generation unit corresponding to the benchmark carbon price; use linear interpolation to establish an approximate expression of the first decision mapping function:

[0150]

[0151] Among them, M is the interpolation matrix, X i,t is the optimal decision matrix; is the first decision mapping function, is the approximate expression of the first decision mapping function; G q is a simplex space;

[0152] Using the approximate expression of the first decision mapping function as the first decision mapping function;

[0153] Construct the maximum error function:

[0154]

[0155] Among them, CP is the maximum error function, Represents the mapping function based on the first decision The calculated approximate value of the system operation cost; T represents an economic dispatch cycle; N c Indicates the number of generator sets in the system; a i b represents the quadratic coefficient of the operating cost of traditional generator set i; i represents the linear coefficient of the operating cost of the traditional generator set i; c i The constant term representing the operating cost of traditional generator set i; μ i represents the carbon emission coefficient per unit of electricity generated by generator set i, and ε represents the free carbon emission coefficient per unit of electricity generated; represents the carbon trading price of power generation unit i;

[0156] The approximate value constraint of the system operation cost is:

[0157]

[0158]

[0159] According to the second constraints (9)-(14) and the system operating cost approximation constraints (29)-(32), the maximum error function is solved to obtain the maximum error value and the optimal solution of the benchmark carbon price.

[0160] Step S104: The first decision mapping function whose maximum error function value is less than the first threshold is used as the second decision mapping function. If the second decision mapping function satisfies the preset spatial set condition, minimizing the benchmark carbon price is used as the third objective function of the single-layer optimization model. The first constraint and the second decision mapping function are used as the third constraint of the single-layer optimization model, and the single-layer optimization model is solved to generate the optimal benchmark carbon price.

[0161] In this embodiment, if the maximum error is CP Less than or equal to the allowed value ò, the record describes the simplex space G q Inequalities and simplex space G q The decision mapping function RH on i,t,q , that is, the second decision mapping function, RH i,t,q Should meet:

[0162]

[0163] In this embodiment, the first decision mapping function whose maximum error function value is less than the first threshold is used as the second decision mapping function. If the second decision mapping function satisfies the preset space set condition, it also includes:

[0164] If the solution of the third objective function is not less than the first threshold, then according to the optimal solution of the benchmark carbon price, s interior points are selected from the simplex space, and corresponding new first simplex spaces are respectively constructed; according to the new first simplex space, the interpolation matrix and the optimal decision matrix are updated; according to the updated interpolation matrix and the optimal decision matrix, the first decision mapping function is updated; s is a positive integer;

[0165] If all simplex spaces of the second decision mapping function do not satisfy the preset space set condition, then s+1 interior points are reselected to form a new second simplex space, and the first decision mapping function is reconstructed;

[0166] The preset space set condition is:

[0167] Among them, G1-G q are the 1st, 2nd, ..., qth simplex spaces respectively.

[0168] As a preferred solution, if the maximum error is CP Greater than or equal to the allowed value ò, optional There are s values ​​in the and the optimal solution of the benchmark carbon price obtained by solving the maximum error function Together, s new simplex spaces are established as follows:

[0169]

[0170] For each simplex space, the interpolation matrix and the optimal decision matrix are reconstructed, and a first decision mapping function is reconstructed.

[0171] As a preferred solution, the second decision mapping function RH is generated i,t,q After that, determine whether all the generated simplex spaces satisfy equation (35):

[0172]

[0173] If not, let q = q + 1 and generate a new benchmark carbon price The simplex space G q , reconstruct the interpolation matrix and the optimal decision matrix, and reconstruct a first decision mapping function.

[0174] If so, a single-layer optimization model for the benchmark carbon price is constructed.

[0175] In this embodiment, the single-layer optimization model is specifically:

[0176] The third objective function of the single-layer optimization model is to minimize the benchmark carbon price:

[0177] The third constraint of the single-layer optimization model includes: a first constraint and a second decision mapping function;

[0178] The first constraint is: E S ≤(1-λ redu )E S0 ;

[0179]

[0180] Among them, E S represents the actual total carbon emissions of the generator set considering carbon trading; redu Represents the carbon reduction coefficient; E S0 represents the actual total carbon emissions of the power generation unit without carbon trading; T represents an economic dispatch cycle; N c Indicates the number of generator sets in the system; μ i P represents the carbon emission coefficient of unit power generation of generator set i; i,t represents the output of generator set i in period t;

[0181]

[0182] in, is the second decision mapping function on the nth simplex space, G nis the nth simplex space; n = 1,...,k,...q; k and q are both positive integers.

[0183] In this embodiment, the solver CONOPT is called to solve the single-layer optimization model to obtain the optimal benchmark carbon price; the generated optimal benchmark carbon price is used to control the dispatch of the power system.

[0184] Implementing the embodiments of the present invention has the following effects:

[0185] The present invention applies the decision mapping method to the calculation of the optimal benchmark carbon price, simplifies the complex two-layer optimization problem into a single-layer optimization problem based on the decision mapping function, and realizes the accurate calculation of the optimal benchmark carbon price. In the process of constructing the decision mapping function, the existing decision mapping method constructs a simplex space by randomly selecting points in the parameter space, resulting in a series of spaces that may have overlapping parts, reducing the solution efficiency of the algorithm. The present invention utilizes the constraints of the two-layer model to generate a set of vertex points in the parameter space, combines the benchmark carbon price of the lower model to solve the internal points of the parameter space, constructs a simplex space, avoids the mutual overlap of the simplex space, improves the calculation efficiency and accuracy of the optimal benchmark carbon price, and improves the efficiency of dispatching and controlling the power system based on the optimal benchmark carbon price.

[0186] Embodiment 2

[0187] Please refer to Figure 2 , a device for calculating an optimal benchmark carbon price provided by an embodiment of the present invention, comprising: a double-layer model construction module 201, a simplex space construction module 202, a decision mapping function construction module 203 and an optimal benchmark carbon price generation module 204;

[0188] The two-layer model construction module 201 is used to construct a two-layer model of the optimal benchmark carbon price of the power system based on the carbon emission data and the generator data of the power system; the first objective function of the upper model is to minimize the benchmark carbon price; the first constraint of the upper model includes: the actual total carbon emissions are less than or equal to the preset total carbon emissions; the second objective function of the lower model is the minimum comprehensive operating cost of the regional power system under the benchmark carbon price condition, and the minimum comprehensive operating cost includes: the minimum operating cost of the generator set and the carbon trading cost; the second constraint of the lower model includes: the upper and lower limit constraints of the unit output, the climbing constraint, the system flow constraint and the interconnection line transmission power constraint;

[0189] The simplex space construction module 202 is used to generate a vertex set of the parameter space according to the first constraint and the second constraint, and construct the boundary of the parameter space; solve the interior point of the parameter space according to the benchmark carbon price that meets the preset requirements; select s vertices and 1 interior point in the parameter space to form a simplex space; s is a positive integer;

[0190] The decision mapping function construction module 203 is used to construct a first decision mapping function of the lower model by taking the benchmark carbon price of the lower model as an input variable and the output of the corresponding generator set as an output variable; construct a maximum error function according to the first decision mapping function; solve the maximum error function according to the second constraint and the system operation cost approximate value constraint to generate a maximum error function value;

[0191] The optimal benchmark carbon price generation module 204 is used to use the first decision mapping function whose maximum error function value is less than the first threshold as the second decision mapping function. If the second decision mapping function satisfies the preset spatial set condition, the minimization of the benchmark carbon price is used as the third objective function of the single-layer optimization model, the first constraint and the second decision mapping function are used as the third constraints of the single-layer optimization model, and the single-layer optimization model is solved to generate the optimal benchmark carbon price.

[0192] The simplex space construction module 202 includes: a vertex set construction unit and an interior point solving unit;

[0193] The vertex set construction unit is used to convert all inequality constraints in the first constraint and the second constraint into equality constraints, and iteratively solve the first equation system consisting of the equality constraints according to a preset rule, and construct a vertex set with solutions that satisfy all inequality constraints before conversion;

[0194] Select s points in the vertex set to form the s-1 dimensional boundary of the parameter space:

[0195] Among them, Θ v is the vertex set, Θ is the parameter space, and s points in the parameter space is a solution that satisfies all inequality constraints in the first and second constraints.

[0196] The inequality constraints in the first constraint and the second constraint are converted into equality constraints, and the first equality equation group consisting of the equality constraints is iteratively solved according to a preset rule, and the solutions satisfying all the inequality constraints before the conversion are constructed into a vertex set, specifically:

[0197] The total number of iterations is set according to the number of the inequality constraints and the dimension of the parameter variables of the inequality constraints; each time, a non-completely identical preset number of inequality constraints is selected in the first set of equations as the second set of equations to be solved to obtain a solution to the second set of equations until the number of iterations reaches the total number of iterations; wherein the number of the non-completely identical preset number of inequality constraints is the dimension of the parameter variables in the inequality constraints; the dimension of the parameter variables is not greater than the number of all inequality constraints; and the solution of the second set of equations that satisfies all inequality constraints is constructed into a vertex set.

[0198] The interior point solving unit is used to solve the interior point of the parameter space according to the benchmark carbon price that meets the preset requirements, specifically:

[0199] Select an arbitrary first benchmark carbon price to solve the lower model and obtain the corresponding optimal output of the first generator set; substitute the first benchmark carbon price and the optimal output of the first generator set into all inequality constraints as the first inequality constraints, and determine whether all first inequality constraints meet the preset conditions; if so, take the first benchmark carbon price as the interior point;

[0200] If not, the first inequality constraint is updated by the second benchmark carbon price and the optimal output of the second generator set until the updated first inequality constraints all meet the preset conditions; the second benchmark carbon price and the optimal output of the second generator set are obtained by solving the barrier function in the preset search area; the barrier function is substituted into all inequality constraints by the first benchmark carbon price and the optimal output of the first generator set, and determines whether all first inequality constraints meet the preset conditions, specifically:

[0201] Convert all inequality constraints to standard form: g j (x)≥0,1≤j≤t;

[0202] In the formula, x represents the variable related to the inequality constraint; t is the number of all inequality constraints; j is the number of the jth inequality constraint;

[0203] The first benchmark carbon price and the optimal output of the first generator set are substituted into all inequality constraints, and the subscripts of the standard form of the inequality constraints whose values ​​are not greater than zero are constructed into a first subscript set; if the first subscript set is an empty set, all first inequality constraints meet the preset conditions.

[0204] It is constructed based on different barrier factors, the first benchmark carbon price and the optimal output of the first generating unit.

[0205] The first inequality constraint is updated by the second benchmark carbon price and the optimal output of the second generator set until the updated first inequality constraints meet the preset conditions, specifically:

[0206] Constructing a barrier function according to the barrier factor, the first benchmark carbon price and the optimal output of the first generator set; solving the barrier function in a preset search area to obtain the second benchmark carbon price and the optimal output of the second generator set;

[0207] Substitute the second benchmark carbon price and the optimal output of the second generator set into all inequality constraints to update the first inequality constraint, and determine whether all updated first inequality constraints meet preset conditions; if not, update the barrier factor and re-solve the second benchmark carbon price and the optimal output of the second generator set.

[0208] The decision mapping function construction module 203 includes: a function construction unit, an error calculation unit and a function update unit;

[0209] The function construction unit is used to take the benchmark carbon price of the lower model as the input variable and the output of the corresponding generator set as the output variable to construct the first decision mapping function of the lower model, specifically:

[0210] Among them, H i,t is the first decision mapping function, is the benchmark carbon price, P i,t is the output of traditional generator set i in period t.

[0211] The error calculation unit is used to construct a maximum error function according to the first decision mapping function, solve the maximum error function according to the second constraint and the system operation cost approximate value constraint, and generate a maximum error function value, specifically:

[0212] Obtain the benchmark carbon price in the simplex space to construct an interpolation matrix, and construct an optimal decision matrix according to the output of the power generation unit corresponding to the benchmark carbon price; use linear interpolation to establish an approximate expression of the first decision mapping function:

[0213] Among them, M is the interpolation matrix, X i,t is the optimal decision matrix is the first decision mapping function, is the approximate expression of the first decision mapping function; G q is a simplex space;

[0214] Using the approximate expression of the first decision mapping function as the first decision mapping function;

[0215] Construct the maximum error function:

[0216]

[0217] Among them, CPis the maximum error function, Represents the mapping function based on the first decision The calculated approximate value of the system operation cost; T represents an economic dispatch cycle; N c Indicates the number of generator sets in the system; a i b represents the quadratic coefficient of the operating cost of traditional generator set i; i represents the linear coefficient of the operating cost of the traditional generator set i; c i The constant term representing the operating cost of traditional generator set i; μ i represents the carbon emission coefficient per unit of electricity generated by generator set i, and ε represents the free carbon emission coefficient per unit of electricity generated; represents the carbon trading price of power generation unit i;

[0218] The approximate value constraint of the system operation cost is:

[0219]

[0220] According to the second constraint and the approximate value constraint of the system operating cost, the maximum error function is solved to obtain the maximum error value and the optimal solution of the benchmark carbon price.

[0221] The function updating unit is used for selecting s interior points in the simplex space according to the optimal solution of the benchmark carbon price, and respectively constructing corresponding new first simplex spaces if the solution of the third objective function is not less than the first threshold; updating the interpolation matrix and the optimal decision matrix according to the new first simplex space; and updating the first decision mapping function according to the updated interpolation matrix and the optimal decision matrix; s is a positive integer;

[0222] If all simplex spaces of the second decision mapping function do not satisfy the preset space set condition, then s+1 interior points are reselected to form a new second simplex space, and the first decision mapping function is reconstructed;

[0223] The preset space set condition is:

[0224] Among them, G1-G q are the 1st, 2nd, ..., qth simplex spaces respectively.

[0225] The optimal benchmark carbon price generation module 204 includes: a single-layer optimization model unit;

[0226] The single-layer optimization model unit is used to solve the single-layer optimization model;

[0227] The third objective function of the single-layer optimization model is to minimize the benchmark carbon price:

[0228] The third constraint of the single-layer optimization model includes: a first constraint and a second decision mapping function;

[0229] The first constraint is: E S ≤(1-λ redu )E S0 ;

[0230]

[0231] Among them, E S represents the actual total carbon emissions of the generator set considering carbon trading; redu Represents the carbon reduction coefficient; E S0 represents the actual total carbon emissions of the power generation unit without carbon trading; T represents an economic dispatch cycle; N c Indicates the number of generator sets in the system; μ i P represents the carbon emission coefficient of unit power generation of generator set i; i,t represents the output of traditional generator set i in period t;

[0232]

[0233] in, is the second decision mapping function on the nth simplex space, G n is the nth simplex space; n = 1,...,k,...q; k and q are both positive integers.

[0234] The above-mentioned device for calculating an optimal benchmark carbon price can implement a method for calculating an optimal benchmark carbon price in the above-mentioned method embodiment. The options in the above-mentioned method embodiment are also applicable to this embodiment and will not be described in detail here. The rest of the contents of the embodiment of this application can refer to the contents of the above-mentioned method embodiment, and will not be repeated in this embodiment.

[0235] Implementing the embodiments of the present invention has the following effects:

[0236] The decision mapping function construction module of the device of the present invention applies the decision mapping method to the calculation of the optimal benchmark carbon price, simplifies the complex two-layer optimization problem into a single-layer optimization problem based on the decision mapping function, and uses the optimal benchmark carbon price generation module to achieve accurate calculation of the optimal benchmark carbon price. In the process of constructing the decision mapping function, the existing decision mapping method constructs a simplex space by randomly selecting points in the parameter space, resulting in a series of spaces that may have overlapping parts, reducing the solution efficiency of the algorithm. The simplex space construction module of the present invention generates a set of vertex points in the parameter space using the constraints of the two-layer model, solves the internal points of the parameter space in combination with the benchmark carbon price of the lower model, and constructs a simplex space, thereby avoiding the mutual overlap of the simplex space, improving the calculation efficiency and accuracy of the optimal benchmark carbon price, and improving the efficiency of dispatching and controlling the power system based on the optimal benchmark carbon price.

[0237] Embodiment 3

[0238] Accordingly, the present invention also provides a computer-readable storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute a method for calculating an optimal benchmark carbon price as described in any one of the above embodiments.

[0239] Exemplarily, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to implement the present invention. The one or more modules / units may be a series of computer program instruction segments capable of implementing specific functions, which are used to describe the execution process of the computer program in the terminal device.

[0240] The terminal device may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The terminal device may include, but is not limited to, a processor and a memory.

[0241] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the terminal device, and uses various interfaces and lines to connect various parts of the entire terminal device.

[0242] The memory can be used to store the computer program and / or module, and the processor realizes various functions of the terminal device by running or executing the computer program and / or module stored in the memory, and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application required for a function, etc.; the data storage area can store data created according to the use of the mobile terminal, etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (SecureDigital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0243] Wherein, if the module / unit integrated in the terminal device is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Wherein, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium, etc.

[0244] The specific embodiments described above further illustrate the purpose, technical solutions and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. It is particularly pointed out that for those skilled in the art, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for calculating the optimal benchmark carbon price, characterized in that: include: According to the carbon emission data and the data of the power system's generator sets, a two-layer model of the optimal benchmark carbon price of the power system is constructed; wherein the first objective function of the upper model is to minimize the benchmark carbon price; the first constraint of the upper model includes: the actual total carbon emissions are less than or equal to the preset total carbon emissions; the second objective function of the lower model is the minimum comprehensive operating cost of the regional power system under the benchmark carbon price condition, and the minimum comprehensive operating cost includes: the minimum operating cost of the generator set and the carbon trading cost; the second constraint of the lower model includes: the upper and lower limit constraints of the unit output, the climbing constraint, the system flow constraint and the interconnection line transmission power constraint; According to the first constraint and the second constraint, a vertex set of the parameter space is generated, and the boundary of the parameter space is constructed; according to the benchmark carbon price that meets the preset requirements, the interior point of the parameter space is solved; s vertices and 1 interior point in the parameter space are selected to form a simplex space; s is a positive integer; Taking the benchmark carbon price of the lower model as the input variable and the output of the corresponding power generation unit as the output variable, constructing the first decision mapping function of the lower model; constructing the maximum error function according to the first decision mapping function; solving the maximum error function according to the second constraint and the system operation cost approximate value constraint to generate the maximum error function value; The first decision mapping function whose maximum error function value is less than the first threshold is used as the second decision mapping function. If the second decision mapping function satisfies the preset spatial set condition, minimizing the benchmark carbon price is used as the third objective function of the single-layer optimization model. The first constraint and the second decision mapping function are used as the third constraints of the single-layer optimization model, and the single-layer optimization model is solved to generate the optimal benchmark carbon price.

2. The method for calculating the optimal benchmark carbon price according to claim 1, characterized in that: The step of generating a vertex set of the parameter space according to the first constraint and the second constraint and constructing a boundary of the parameter space is specifically as follows: Convert all inequality constraints in the first constraint and the second constraint into equality constraints, iteratively solve the first equality equation group consisting of the equality constraints according to a preset rule, and construct a vertex set from solutions that satisfy all inequality constraints before conversion; Select s points in the vertex set to form the s-1 dimensional boundary of the parameter space: Among them, Θ v is the vertex set, Θ is the parameter space, and s points in the parameter space are s solutions that satisfy all inequality constraints in the first and second constraints.

3. The method for calculating the optimal benchmark carbon price according to claim 2, characterized in that: The inequality constraints in the first constraint and the second constraint are converted into equality constraints, and the first equation system consisting of the equality constraints is iteratively solved according to a preset rule, and the solutions that satisfy all the inequality constraints before the conversion are constructed into a vertex set, specifically: The total number of iterations is set according to the number of the inequality constraints and the dimension of the parameter variables of the inequality constraints; each time, a non-completely identical preset number of inequality constraints is selected in the first set of equations as the second set of equations to be solved to obtain a solution to the second set of equations until the number of iterations reaches the total number of iterations; wherein the number of the non-completely identical preset number of inequality constraints is the dimension of the parameter variables in the inequality constraints; the dimension of the parameter variables is not greater than the number of all inequality constraints; and the solution of the second set of equations that satisfies all inequality constraints is constructed into a vertex set.

4. The method for calculating the optimal benchmark carbon price according to claim 1, characterized in that: The method of solving the interior point of the parameter space according to the benchmark carbon price that meets the preset requirements is as follows: Select an arbitrary first benchmark carbon price to solve the lower model and obtain the corresponding optimal output of the first generator set; substitute the first benchmark carbon price and the optimal output of the first generator set into all inequality constraints as the first inequality constraints, and determine whether all first inequality constraints meet the preset conditions; if so, take the first benchmark carbon price as the interior point; If not, the first inequality constraint is updated by the second benchmark carbon price and the optimal output of the second generator set until the updated first inequality constraints meet the preset conditions; the second benchmark carbon price and the optimal output of the second generator set are obtained by solving the obstacle function in a preset search area; the obstacle function is constructed according to different obstacle factors, the first benchmark carbon price and the optimal output of the first generator set.

5. The method for calculating the optimal benchmark carbon price according to claim 4, characterized in that: Substituting the first benchmark carbon price and the first generator set optimal output into all inequality constraints, and judging whether all first inequality constraints meet preset conditions, is specifically: Convert all inequality constraints to standard form: g j (x)≥0,1≤j≤t; In the formula, g j (x) represents the standard form of the inequality constraint; x represents the variable associated with the inequality constraint; t is the number of all inequality constraints; j is the number of the jth inequality constraint; The first benchmark carbon price and the optimal output of the first generator set are substituted into all inequality constraints, and the subscripts of the standard form of the inequality constraints whose values ​​are not greater than zero are constructed into a first subscript set; if the first subscript set is an empty set, all first inequality constraints meet the preset conditions.

6. The method for calculating the optimal benchmark carbon price according to claim 4, characterized in that: The first inequality constraint is updated by the second benchmark carbon price and the second generator set optimal output until the updated first inequality constraint satisfies the preset conditions, specifically: Constructing a barrier function according to the barrier factor, the first benchmark carbon price and the optimal output of the first generator set; solving the barrier function in a preset search area to obtain the second benchmark carbon price and the optimal output of the second generator set; Substitute the second benchmark carbon price and the optimal output of the second generator set into all inequality constraints to update the first inequality constraint, and determine whether all updated first inequality constraints meet preset conditions; if not, update the barrier factor and re-solve the second benchmark carbon price and the optimal output of the second generator set.

7. The method for calculating the optimal benchmark carbon price according to claim 2, characterized in that: The first decision mapping function of the lower model is constructed by taking the benchmark carbon price of the lower model as the input variable and the output of the corresponding generator set as the output variable, which is specifically: Among them, H i,t is the first decision mapping function, is the benchmark carbon price, P i,t is the output of traditional generator set i in period t.

8. The method for calculating the optimal benchmark carbon price according to claim 7, characterized in that: The step of constructing a maximum error function according to the first decision mapping function, solving the maximum error function according to the second constraint and the system operation cost approximate value constraint, and generating a maximum error function value is specifically as follows: Obtain the benchmark carbon price in the simplex space to construct an interpolation matrix, and construct an optimal decision matrix according to the output of the power generation unit corresponding to the benchmark carbon price; use linear interpolation to establish an approximate expression of the first decision mapping function: Among them, M is the interpolation matrix, X i,t is the optimal decision matrix; is the first decision mapping function, is the approximate expression of the first decision mapping function; G q is a simplex space; q is a simplex space G q Serial number; Using the approximate expression of the first decision mapping function as the first decision mapping function; Construct the maximum error function: Among them, CP is the maximum error function, Represents the mapping function based on the first decision The calculated approximate value of the system operation cost; T represents an economic dispatch cycle; N c Indicates the number of generator sets in the system; a i b represents the quadratic coefficient of the operating cost of traditional generator set i; i represents the linear coefficient of the operating cost of the traditional generator set i; c i The constant term representing the operating cost of traditional generator set i; μ i represents the carbon emission coefficient per unit of electricity generated by generator set i, and ε represents the free carbon emission coefficient per unit of electricity generated; represents the carbon trading price of power generation unit i; The approximate value constraint of the system operation cost is: According to the second constraint and the approximate value constraint of the system operating cost, the maximum error function is solved to obtain the maximum error value and the optimal solution of the benchmark carbon price.

9. A method for calculating the optimal benchmark carbon price according to claim 8, characterized in that: The first decision mapping function whose maximum error function value is less than the first threshold is used as the second decision mapping function, and if the second decision mapping function satisfies the preset space set condition, it also includes: If the solution of the third objective function is not less than the first threshold, then according to the optimal solution of the benchmark carbon price, s interior points are selected from the simplex space, and corresponding new first simplex spaces are respectively constructed; according to the new first simplex space, the interpolation matrix and the optimal decision matrix are updated; according to the updated interpolation matrix and the optimal decision matrix, the first decision mapping function is updated; s is a positive integer; If all simplex spaces of the second decision mapping function do not satisfy the preset space set condition, then s+1 interior points are reselected to form a new second simplex space, and the first decision mapping function is reconstructed; The preset space set condition is: Among them, G1-G q are the 1st, 2nd, ..., qth simplex spaces respectively.

10. The method for calculating the optimal benchmark carbon price according to claim 1, characterized in that: The single-layer optimization model is specifically: The third objective function of the single-layer optimization model is to minimize the benchmark carbon price: min The third constraint of the single-layer optimization model includes: a first constraint and a second decision mapping function; The first constraint is: E S ≤(1-λ redu )E S0 ; Among them, E S represents the actual total carbon emissions of the generator set considering carbon trading; redu Represents the carbon reduction coefficient; E S0 represents the actual total carbon emissions of the power generation unit without carbon trading; T represents an economic dispatch cycle; N c Indicates the number of generator sets in the system; μ i P represents the carbon emission coefficient of unit power generation of generator set i; i,t represents the output of traditional generator set i in period t; in, is the second decision mapping function on the nth simplex space, G n is the nth simplex space; n=1,...,k,...q; k and q are both positive integers, is the benchmark carbon price.

Images