Low-carbon scheduling method considering comprehensive demand response of electricity and carbon of electrolytic aluminum enterprise

Abstract

The invention provides a low-carbon scheduling method considering an electrolytic aluminum enterprise electricity-carbon comprehensive demand response, and the method comprises the steps: building a day-ahead prediction typical scene set with probability features based on the historical data of day-ahead prediction errors, and building a novel power system double-layer low-carbon scheduling model containing an electrolytic aluminum enterprise participating electricity-carbon comprehensive demand response project. Wherein the lower-layer model is an electrolytic aluminum enterprise optimization scheduling model, and the upper-layer model is a novel power system low-carbon scheduling model. The nonlinear links of the model are linearized through related optimization theories, and then the mixed integer double-layer quadratic programming model obtained after conversion is iteratively solved to obtain a novel power system day-ahead scheduling scheme and dynamic carbon emission factors in each time period, so that the power system day-ahead scheduling scheme is optimized while economical efficiency and safety of a power grid and a user are considered. And the low-carbon properties of power system scheduling and enterprise production operation are improved.

Description

Technical Field

[0001] The present invention relates to a low-carbon dispatching method that takes into account the comprehensive electricity-carbon demand response of an electrolytic aluminum enterprise, and belongs to the technical field of power dispatching. Background Art

[0002] Due to the operating characteristics of real-time power balance in the power system, increasing the proportion of renewable energy generation on the power generation side does not simply mean increasing the proportion of wind power and photovoltaic installed capacity, but relies more on the power system to adopt reasonable and effective optimization and scheduling methods.

[0003] However, the dynamic carbon emission factors used in existing research are typically calculated solely using the region's baseload curve and the day-ahead forecast curve for renewable energy output, without considering the impact of user demand response behavior. When large industrial loads such as electrolytic aluminum use this factor to optimize their electricity consumption, this inevitably leads to changes in the region's load curve, which in turn affects the unit mix and dynamic carbon emission factor values ​​for each time period. Therefore, the carbon emissions estimated by the dynamic carbon emission factors published by power companies for electrolytic aluminum enterprises may differ significantly from their actual carbon emissions after the end of the day's operations. Clearly, there is an information asymmetry between power companies and electrolytic aluminum enterprises regarding dynamic carbon emission factors. This unfairness may increase enterprise concerns and even undermine their enthusiasm for participating in demand response to reduce carbon emissions. Summary of the Invention

[0004] The purpose of this invention is to provide a low-carbon scheduling method that takes into account the comprehensive electricity and carbon demand response of electrolytic aluminum enterprises. Through iterative calculation of the feedback mechanism, the convergence value of the dynamic carbon emission factor is obtained, thereby improving the accuracy of the electrolytic aluminum enterprises' estimation of their own carbon emissions, thereby enhancing their willingness to participate in the comprehensive electricity and carbon demand response plan.

[0005] In order to achieve the above objectives / solve the above technical problems, the present invention is implemented by adopting the following technical solutions.

[0006] In one aspect, the present invention provides a low-carbon scheduling method that considers the comprehensive electricity and carbon demand response of an electrolytic aluminum enterprise, comprising the following steps:

[0007] Obtain a typical scenario set for day-ahead forecasts of new energy and load power;

[0008] Taking the maximum expected value of the net profit of the electrolytic aluminum enterprise under the typical scenario set of new energy and load power day-ahead forecast as the first optimization scheduling objective function, combined with the operating constraints of the electrolytic aluminum enterprise, an optimization scheduling model for the electrolytic aluminum enterprise is constructed;

[0009] Taking the minimum cost of power system operation under the typical scenario set of new energy and load power forecast as the second optimization dispatch objective function, combined with the power system operation constraints, the power system side optimization dispatch model is constructed.

[0010] A two-layer low-carbon dispatch model is constructed with the electrolytic aluminum enterprise optimization dispatch model as the lower model and the power system optimization dispatch model as the upper model. The nonlinear terms in the two-layer low-carbon dispatch model are linearly optimized to obtain a two-layer mixed integer quadratic programming model.

[0011] By iteratively solving a two-level mixed-integer quadratic programming model, we obtained a coordinated dispatch plan for the power system and the aluminum smelter that responds to the integrated electricity and carbon demand, as well as the latest dynamic carbon emission factor value for the current period released by the power system.

[0012] Among them, the solution method of the two-level mixed integer quadratic programming model is as follows:

[0013] Based on the iterative calculation between the upper model transferring the dynamic carbon emission factor, spare capacity demand and flexible adjustment power demand to the lower model and the lower model transferring its own load curve to the upper model, the latest coordinated scheduling plan and dynamic carbon emission factor are solved.

[0014] Furthermore, the iterative calculation between transferring the dynamic carbon emission factor, spare capacity requirement, and flexible power adjustment requirement from the upper model to the lower model and transferring the load curve from the lower model to the upper model to solve the latest coordinated scheduling solution and dynamic carbon emission factor specifically includes:

[0015] The upper model transmits dynamic carbon emission factors, reserve capacity requirements, and flexible power requirements to the lower model;

[0016] The lower-level model transmits the dynamic carbon emission factor, spare capacity demand, and flexible power demand information based on the obtained model, optimizes the load curve of the electrolytic aluminum enterprise, and returns it to the upper-level model to participate in the power balance constraint, so that the upper-level model optimizes the new unit combination plan and updates the dynamic carbon emission factor, spare capacity demand, and flexible power demand;

[0017] Until the maximum deviation of the electrolytic aluminum enterprise load curve obtained from two consecutive iterations is less than the set threshold, the latest upper-level model unit combination plan, dynamic carbon emission factor, spare capacity demand and flexible adjustment power demand are obtained.

[0018] Furthermore, the method for obtaining a typical scenario set for new energy and load power day-ahead prediction specifically includes:

[0019] The principal component analysis method is used to obtain the boundary scenarios in the historical dataset of new energy and load power day-ahead forecast errors, including:

[0020] The principal component analysis method is used to obtain the characteristic vector of the day-ahead forecast error of new energy and load power, which is expressed as follows:

[0021] ;

[0022] Among them, U is the historical dataset matrix of the day-ahead forecast error of new energy and load power, is the zero-mean matrix of U, is the average value of all samples in the matrix U, N is the total number of samples in the historical dataset of the forecast error of new energy and load power, S is the covariance matrix of the historical dataset of the forecast error of new energy and load power, q h is the hth eigenvector of S, λ h For q h The corresponding eigenvalue, Λ is the eigenvalue λ h The diagonal matrix formed, Q is the vector feature q h The matrix formed, is the operator that converts a vector into a diagonal matrix, is the transpose, N RES is the total number of renewable energy power generation stations in the power system, N bus is the total number of load nodes in the power system, and T is the total number of dispatch periods;

[0023] Will Each sample in is converted to each feature vector q h In the coordinate system formed, the expression is:

[0024] ;

[0025] in, for The oth sample in , for In each eigenvector q h The mapping points in the coordinate system are: for In the eigenvector q h Projection value in direction;

[0026] After all samples are projected, each eigenvector q is determined h The coordinates of the two vertices in the direction are expressed as:

[0027] ;

[0028] in, is the minimum vertex coordinate in the direction of the hth eigenvector in the eigenvector coordinate system, is the maximum vertex coordinate in the direction of the hth eigenvector in the eigenvector coordinate system, e h is the unit column vector in the direction of the hth eigenvector, for The Nth sample in In the eigenvector q h Projection value in direction;

[0029] The scaling factor η is introduced to expand the above vertex coordinates so that the high-dimensional polyhedron surrounded by the expanded vertices can contain all the points corresponding to the historical data of the source-load day-ahead forecast error in space. The expression is:

[0030] ;

[0031] ;

[0032] Among them, ||β o ||1 is β o 1-norm, η is the scaling factor of the expanded vertex, β o is the minimum ratio that the o-th sample needs to expand in each eigenvector direction; is the oth sample in the (N RES +N bus ) The minimum ratio of expansion required to move toward the maximum vertex in the direction of T eigenvectors;

[0033] Get the vertex coordinates of the expanded high-dimensional polyhedron in the original coordinate system. The expression is:

[0034] ;

[0035] in, is the minimum vertex coordinate in the direction of the hth eigenvector of the high-dimensional polyhedron in the original coordinate system; is the maximum vertex coordinate in the direction of the hth eigenvector of the high-dimensional polyhedron in the original coordinate system;

[0036] Adjust the unreasonable scenes in the high-dimensional polyhedron vertex correspondence scene. The expression is:

[0037] ;

[0038] in, represent or The e-th component of is the e-th component of the N-th sample in the matrix U;

[0039] The adjusted high-dimensional polyhedron vertex set is recorded as the boundary scene set u vtx , including 2(N RES +N bus )T vertices, the expression is:

[0040] ;

[0041] in: is the boundary scene set u vtx The j-th vertex in ;

[0042] Statistics on the attribution of all samples in the historical dataset of new energy and load power day-ahead forecast errors to each vertex, including:

[0043] Calculate u vtx The Euclidean distance between each vertex in and any sample point in the historical data of the new energy and load power day-ahead forecast error is expressed as:

[0044] ;

[0045] Among them, d o,j is the historical data sample u of the oth renewable energy and load power day-ahead forecast error o and the jth vertex The Euclidean distance between

[0046] will u o Belong to the vertex closest to it, and count the number of samples belonging to each vertex and save it in array n array , the expression is:

[0047] ;

[0048] in, For array n array Used to store the The number of sample points, array n array is the 1×2(N b +N w )T-dimensional array,

[0049] The array n array All components are initialized to 0. Whenever a sample point is assigned to the jth vertex, n array The jth element of is added by one, and the above operation is performed on each sample point, and the final result is n array That is, it reflects the attribution of the sample points of the historical dataset of the day-ahead forecast error to each vertex;

[0050] The k-medoids algorithm is used to obtain the cluster center scenario in the historical data set of the day-ahead forecast error. The expression is:

[0051] ;

[0052] in, is the λth cluster center, is the corresponding probability, n cluis the number of cluster centers;

[0053] The cluster center scene u clu With the boundary scene u vtx By combining the above, we can obtain a set of representative day-ahead forecast error scenarios for renewable energy and load power, as well as the probability of each representative day-ahead forecast error scenario, namely:

[0054] ;

[0055] Among them, u typ is the scenario set of new energy and load power day-ahead forecast errors, n sce for u typ The total number of scenes included, is the sth new energy and load power day-ahead forecast error scenario,

[0056] The expressions for the components involved are:

[0057] ;

[0058] Among them, N RES and N bus are the number of new energy power stations and load nodes respectively, for The day-ahead power forecast error of the r-th renewable energy power station in the middle period t is: for The day-ahead power error of the load at the bth node in period t, where T is the total number of scheduling periods;

[0059] Determine the initial probability of each new energy and load power day-ahead forecast error scenario, expressed as:

[0060] ;

[0061] Among them, p s for The initial probability of

[0062] Obtain the basic forecast scenario, and build a corresponding typical scenario set for day-ahead forecast based on the day-ahead forecast error scenario of new energy and load power combined with the day-ahead forecast basic scenario.

[0063] By combining boundary scenarios and cluster center scenarios, the accuracy of the typical scenario set for day-ahead prediction of new energy and load power is more in line with the actual situation, which will help the subsequent optimization scheduling model of this application to obtain a more reasonable scheduling plan.

[0064] Furthermore, the operating constraints of the electrolytic aluminum enterprise include:

[0065] The temperature-power coupling constraint is expressed as:

[0066] ;

[0067] ;

[0068] in, is the electrolyte temperature of the i-th electrolytic aluminum production line in the basic scenario, is the lower limit of the electrolyte temperature of the i-th electrolytic aluminum production line, is the upper limit of the electrolyte temperature of the i-th aluminum electrolytic production line, c is the specific heat capacity of the molten electrolyte, m is the mass of the molten electrolyte, is the rated power of the i-th electrolytic aluminum production line, is the power of the i-th electrolytic aluminum production line in period t under the basic scenario, The duration of each scheduling period;

[0069] The power upper and lower limit constraints are expressed as:

[0070] ;

[0071] in, is the minimum power of the i-th electrolytic aluminum production line in period t, is the maximum power of the i-th electrolytic aluminum production line during period t;

[0072] The spare capacity constraint is expressed as:

[0073] ;

[0074] ;

[0075] ;

[0076] ;

[0077] ;

[0078] in, The maximum spare capacity provided for the i-th electrolytic aluminum production line in period t, is the reserve capacity provided by the i-th electrolytic aluminum production line in period t, t′ is the continuous duration of the delivered reserve capacity,

[0079] The electrolyte heat gain and loss caused by the continuous delivery of spare capacity from the beginning of time period t to time period t+t′; The electrolyte heat gain and loss caused by the continuous delivery of spare capacity from the start of time t to time t+t′;

[0080] Flexible adjustment of power constraints, the expression is:

[0081] The flexible power adjustment provided by the electrolytic aluminum production line cannot exceed the reserved spare capacity and cannot be in both the upward and downward adjustment states at the same time. To this end, a 0-1 auxiliary variable is introduced to construct the following constraints:

[0082] ;

[0083] ;

[0084] ;

[0085] in, is a 0-1 variable indicating that the i-th electrolytic aluminum production line is in an upward adjustment state in the s-th typical scenario period t; is a 0-1 variable indicating that the i-th electrolytic aluminum production line is in a downward adjustment state in the s-th typical scenario period t;

[0086] is the electrolyte temperature of the i-th electrolytic aluminum production line in the s-th typical scenario, period t, and M is a preset auxiliary parameter, which can be 1×10 8 ; Provide flexible power adjustment for the i-th electrolytic aluminum production line in the s-th typical scenario and time period t; The flexible power adjustment provided for the i-th electrolytic aluminum production line in the s-th typical scenario and time period t;

[0087] The daily output constraint is expressed as:

[0088] ;

[0089] ;

[0090] Among them, Q Al,i,s,t is the aluminum output of the ith electrolytic aluminum production line in period t under the sth typical scenario, P Al,i,s,t is the power of the i-th electrolytic aluminum production line in the s-th typical scenario period t, is the minimum power that the electrolytic aluminum production line i can maintain, f(P Al,i,s,t )for The yield calculation function when is the electrochemical equivalent of aluminum, η i is the electrolysis efficiency of the i-th electrolytic aluminum production line, n Al,i is the number of electrolytic cells included in the i-th electrolytic aluminum production line, R i is the equivalent resistance of the i-th electrolytic aluminum production line, E iis the sum of the back electromotive force, anode overvoltage and cathode overvoltage of the i-th electrolytic aluminum production line, is the rated daily output of aluminum, α min is the ratio of the lower limit of daily output to the rated daily output, is the number of electrolytic aluminum production lines.

[0091] Furthermore, the first optimization scheduling objective function is expressed as:

[0092] ;

[0093] in, Subsidies for providing spare capacity to electrolytic aluminum enterprises during period t, is the aluminum product revenue in period t under the sth typical scenario, Subsidies obtained by providing flexible capacity adjustment to electrolytic aluminum enterprises in period t under the sth typical scenario, is the operation and maintenance cost of the electrolytic aluminum production line of the electrolytic aluminum enterprise in period t under the sth typical scenario, is the carbon emission cost of the electrolytic aluminum enterprise in period t under the sth typical scenario, is the number of typical scenarios;

[0094] ;

[0095] Among them, N Al is the number of electrolytic aluminum production lines; is the unit subsidy price for electrolytic aluminum production lines to participate in demand response and provide spare capacity. The maximum spare capacity provided for the i-th electrolytic aluminum production line in period t, The minimum spare capacity provided for the i-th electrolytic aluminum production line in period t;

[0096] ;

[0097] Among them, λ Al is the unit net profit of aluminum products, Q Al,i,s,t is the aluminum output of the i-th electrolytic aluminum production line in period t under the s-th typical scenario;

[0098] ;

[0099] in, Provide a unit subsidy price that allows flexible adjustment of power levels for electrolytic aluminum enterprises participating in demand response during period t. Provide flexible power adjustment for the i-th electrolytic aluminum production line in the s-th typical scenario and time period t; The flexible power adjustment provided for the i-th electrolytic aluminum production line in the s-th typical scenario and time period t;

[0100] ;

[0101] Among them, c om is the cost coefficient of electrolytic aluminum load operation and maintenance, P Al,i,s,t is the power of the i-th electrolytic aluminum production line in the s-th typical scenario during the t-th period;

[0102] ;

[0103] Among them, c carb is the unit carbon emission cost coefficient; EF Al The carbon emission intensity of the internal process of the electrolytic aluminum production line corresponding to the unit electricity consumption; is the dynamic carbon emission factor value in period t.

[0104] By combining the operating constraints of electrolytic aluminum enterprises and the first optimization scheduling objective function, an optimization scheduling model for electrolytic aluminum enterprises was constructed, which accurately considered the impact of the temperature constraints of the electrolytic aluminum production line on its provision of backup capacity to the power system. It helps to explore the ability of electrolytic aluminum enterprises to participate in the demand response of the power system while ensuring production safety, while improving the low-carbon attributes of the enterprise's operation.

[0105] Furthermore, the second optimization scheduling objective function is expressed as:

[0106] ;

[0107] in, is the total operating cost of the fossil energy unit in period t under the basic scenario, Subsidies for providing spare capacity to electrolytic aluminum enterprises during period t, is the unit price of electric energy exchange between the power system and other external power systems during period t, The power exchanged between the power system and other external power systems during period t, with the purchase of power from other external power systems being considered positive; is the regulation cost of the fossil energy unit in period t under the sth typical scenario, Subsidies obtained by providing flexible capacity adjustment to electrolytic aluminum enterprises in period t under the sth typical scenario, is the reliability cost of power system operation in period t under the sth typical scenario;

[0108] ;

[0109] Among them, a g For the g-th fossil energy unit and The corresponding power generation cost coefficient, b g For the g-th fossil energy unit and The corresponding power generation cost coefficient, c g is the fixed power generation cost coefficient of the g-th fossil energy unit when it is in operation, 、 are the costs of starting and stopping the g-th fossil energy unit, 、 are the indicator variables for the on / off state switching of the g-th fossil energy unit during period t, The unit subsidy price for providing spare capacity for fossil energy units, 、 The upper and lower reserve capacities provided for the g-th fossil energy unit during period t, is the power generation capacity of the g-th fossil energy unit in period t, is the number of fossil energy units;

[0110] ;

[0111] in, The unit cost of providing flexible power adjustment for the units in time period t; 、 are the upper and lower flexible adjustment powers provided by unit g in the sth typical scenario period t respectively;

[0112] ;

[0113] Among them, N RES is the total number of renewable energy power generation stations in the power system, N bus is the total number of load nodes in the power system, c cur 、c ld These are the unit penalty prices for curtailed electricity and load shedding electricity from new energy power stations; is the power curtailment of the rth renewable energy power station in period t under the sth typical scenario, is the load shedding power of the bth node in period t under the sth typical scenario;

[0114] Power system operation constraints, including:

[0115] The power balance constraint is expressed as:

[0116] ;

[0117] ;

[0118] in, is the predicted power of the rth new energy power station in time period t, is the day-ahead predicted power of the b-th node load in period t; is the power exchanged between the power system and the external power system during period t, Provide flexible power adjustment for the i-th electrolytic aluminum production line in the s-th typical scenario and time period t; The flexible power adjustment provided for the i-th electrolytic aluminum production line in the s-th typical scenario and time period t;

[0119] The line transmission capacity constraint is expressed as:

[0120] ;

[0121] ;

[0122] Among them, k lb is the power transfer factor of node b with respect to line l, f lmax is the maximum transmission power of line l, 、 、 They represent the connection of unit g, new energy power station r, and electrolytic aluminum production line i to the power system node b to access the power system;

[0123] The unit output and climbing constraints are expressed as:

[0124] ;

[0125] Among them, P gmin 、P gmax are the lower and upper limits of the output of unit g respectively; I g,t is the start / stop state variable of the g-th unit in period t, and its value is 1 or 0, indicating the start / stop state respectively;

[0126] ;

[0127] Among them, UR g , DR g are the up and down climbing rates of unit g respectively;

[0128] The minimum start and stop time constraint of the unit is expressed as:

[0129] ;

[0130] in, 、 are the durations of continuous startup and shutdown of unit g during period t-1 respectively; 、 are the minimum startup and shutdown durations of unit g respectively;

[0131] Spare capacity constraints, specifically:

[0132] The upper and lower limit constraints of the spare capacity provided by the unit are expressed as:

[0133] ;

[0134] in, is the maximum power generation capacity of the g-th fossil energy unit, is the minimum power generation capacity of the g-th fossil energy unit;

[0135] Flexible adjustment of power constraints, the expression is:

[0136] ;

[0137] The constraints on the curtailed power and load shedding power of new energy are expressed as follows:

[0138] .

[0139] By combining the power system's operational constraints with the second optimal dispatch objective function, a new power system low-carbon dispatch model was constructed. Together with the aforementioned optimal dispatch model for aluminum smelters, this model forms a new two-tiered low-carbon dispatch model for the power system, including aluminum smelters' participation in a comprehensive electricity-carbon demand response program. This model fully taps the potential of aluminum smelters to participate in power system demand response programs, while balancing the operational safety and economics of both the power system and aluminum smelters. Furthermore, it utilizes dynamic carbon emission factors to guide aluminum smelters to shift some of their electricity demand to periods of low power system carbon emission intensity, without compromising their economic efficiency. This enhances the low-carbon nature of both the power system and aluminum smelters.

[0140] Furthermore, the nonlinear terms in the two-layer low-carbon scheduling model are linearly optimized, including:

[0141] The maximum power that can be achieved when the electrolytic aluminum production line is adjusted only once for 1 hour within 24 hours is used as the P in the piecewise linearization process. Al,i,s,t The upper limit of the power limit constraint is Δt=1, then:

[0142] ;

[0143] The interval Divide into n segments, and use linear interpolation function to replace each segment The daily output constraint can be transformed into the following equivalent form:

[0144] ;

[0145] Where: n is the number of evenly spaced segments, x0,⋯, x k , x k+1 ,⋯, x n For interval The endpoints of the evenly divided segments;

[0146] x k and x k+1 are the left and right endpoints of the k+1th evenly spaced segment, k=0,...,n-1, x0= , x n = ;

[0147] is the maximum operating power that the electrolytic aluminum production line i can withstand, z1,⋯,z n To show that P Al,i,s,t In the range The 0-1 auxiliary variable for the 1st to nth average partition segment, z0 is used to indicate P Al,i,s,t < Auxiliary variables, M is the preset auxiliary parameter, for The yield calculation function value at time .

[0148] Through the above-mentioned linearization process, the new two-level low-carbon dispatch model of the power system containing electrolytic aluminum enterprises participating in the integrated electricity-carbon demand response project, which cannot be directly solved, is converted into a two-level mixed integer quadratic programming model that can be solved by a commercial optimization solver, thereby improving the efficiency of model solving.

[0149] Furthermore, the calculation method of the dynamic carbon emission factor is expressed as follows:

[0150] ;

[0151] in, is the dynamic carbon emission factor value in period t, is the power generation capacity of the g-th fossil energy unit in period t, is the number of fossil energy units, EF g is the carbon emission intensity of the g-th unit, is the power exchanged between the power system and other external power systems during period t, is the dynamic carbon emission factor corresponding to the amount of electricity purchased from the external power system during period t.

[0152] In a second aspect, the present invention provides a low-carbon dispatching device that takes into account the comprehensive electricity and carbon demand response of electrolytic aluminum enterprises, comprising:

[0153] The typical scenario set module for day-ahead prediction is used to obtain the typical scenario set for day-ahead prediction of new energy and load power;

[0154] The electrolytic aluminum enterprise-side optimization scheduling module is used to construct an electrolytic aluminum enterprise-side optimization scheduling model based on the maximum expected value of the electrolytic aluminum enterprise's net profit under a typical scenario set of new energy and load power day-ahead forecasts as the first optimization scheduling objective function and combined with the electrolytic aluminum enterprise's operating constraints;

[0155] The power system optimization and dispatching module is used to construct a power system optimization and dispatching model based on the minimum power system operation cost under the typical scenario set of new energy and load power forecasting as the second optimization and dispatching objective function, combined with the power system operation constraints.

[0156] A two-layer module is used to construct a two-layer low-carbon scheduling model with the electrolytic aluminum enterprise-side optimization scheduling model as the lower model and the power system-side optimization scheduling model as the upper model. The nonlinear terms in the two-layer low-carbon scheduling model are linearly optimized to obtain a two-layer mixed integer quadratic programming model.

[0157] Dynamic carbon emission factor solution module: This module is used to iteratively solve a two-level mixed-integer quadratic programming model to obtain a coordinated dispatch plan for the power system and the electrolytic aluminum enterprise that responds to the integrated electricity and carbon demand, as well as the latest dynamic carbon emission factor value for the current period released by the power system.

[0158] Among them, the solution method of the two-level mixed integer quadratic programming model is as follows:

[0159] Based on the iterative calculation between the upper model transferring the dynamic carbon emission factor, spare capacity demand and flexible adjustment power demand to the lower model and the lower model transferring its own load curve to the upper model, the latest coordinated scheduling plan and dynamic carbon emission factor are solved.

[0160] In a third aspect, the present invention provides a low-carbon dispatching system that takes into account the comprehensive electricity and carbon demand response of electrolytic aluminum enterprises, including:

[0161] Memory, used to store computer programs / instructions;

[0162] A processor is used to execute the computer program / instructions to implement the steps of the above-mentioned low-carbon scheduling method that takes into account the comprehensive electricity and carbon demand response of electrolytic aluminum enterprises.

[0163] Compared with the existing technology, the beneficial effects achieved by the present invention are as follows: This application proposes a low-carbon scheduling method that takes into account the comprehensive electricity-carbon demand response of electrolytic aluminum enterprises, constructs a two-layer low-carbon scheduling model with the electrolytic aluminum enterprise side optimization scheduling model as the lower model and the power system side optimization scheduling model as the upper model, and performs linear optimization on the nonlinear terms in the two-layer low-carbon scheduling model to construct a two-layer mixed integer quadratic programming model; in the process of solving the two-layer mixed integer quadratic programming model, the feedback mechanism is used to iteratively calculate the convergence value of the dynamic carbon emission factor, thereby improving the accuracy of users' estimates of their own carbon emissions, thereby enhancing their willingness to participate in the comprehensive electricity-carbon demand response plan.

[0164] The proposed method not only mitigates peak-valley fluctuations in the power grid and optimizes power resource allocation, but also addresses the cost-cutting and efficiency-enhancing needs of businesses. Furthermore, this strategy promotes the efficient use of renewable energy and the development of green energy, providing theoretical and methodological support for achieving sustainable and healthy social development. BRIEF DESCRIPTION OF THE DRAWINGS

[0165] Figure 1 Shown is a schematic diagram of the present invention. DETAILED DESCRIPTION

[0166] It should be noted that:

[0167] The technical solution of the present invention is described in detail below through the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present invention and the specific features in the embodiments are detailed descriptions of the technical solution of the present invention, rather than limitations on the technical solution of the present invention. In the absence of conflict, the embodiments of the present invention and the technical features in the embodiments can be combined with each other.

[0168] The term "and / or" simply describes a relationship between related objects, indicating that three possible relationships exist. For example, "A and / or B" can mean: A exists alone, A and B exist simultaneously, or B exists alone. Additionally, the character " / " generally indicates an "or" relationship between the related objects.

[0169] Example 1

[0170] like Figure 1 An embodiment shown in FIG. 1 provides a low-carbon scheduling method that considers the comprehensive electricity and carbon demand response of an electrolytic aluminum enterprise, including the following steps:

[0171] Step 1: Obtain a typical scenario set for new energy and load power day-ahead forecasts, including:

[0172] Step 1.1: Obtain boundary scenarios from the historical dataset of new energy and load power day-ahead forecast errors using principal component analysis, including:

[0173] The principal component analysis method is used to obtain the characteristic vector of the day-ahead forecast error of new energy and load power, which is expressed as follows:

[0174] ;

[0175] Among them, U is the historical dataset matrix of the day-ahead forecast error of new energy and load power, is the zero-mean matrix of U, is the average value of all samples in the matrix U, N is the total number of samples in the historical dataset of the forecast error of new energy and load power, S is the covariance matrix of the historical dataset of the forecast error of new energy and load power, q h is the hth eigenvector of S, λ h For q h The corresponding eigenvalue, Λ is the eigenvalue λ h The diagonal matrix formed, Q is the vector feature q h The matrix formed, is the operator that converts a vector into a diagonal matrix, is the transpose, N RES is the total number of renewable energy power generation stations in the power system, N bus is the total number of load nodes in the power system, and T is the total number of dispatch periods;

[0176] Will Each sample in is converted to each feature vector q h In the coordinate system formed, the expression is:

[0177] ;

[0178] in, for The oth sample in , for In each eigenvector q h The mapping points in the coordinate system are: for In the eigenvector q h Projection value in direction;

[0179] After all samples are projected, each eigenvector q is determined h The coordinates of the two vertices in the direction are expressed as:

[0180] ;

[0181] in, is the minimum vertex coordinate in the direction of the hth eigenvector in the eigenvector coordinate system, is the maximum vertex coordinate in the direction of the hth eigenvector in the eigenvector coordinate system, e h is the unit column vector in the direction of the hth eigenvector, for The Nth sample in In the eigenvector q h Projection value in direction;

[0182] The scaling factor η is introduced to expand the above vertex coordinates so that the high-dimensional polyhedron surrounded by the expanded vertices can contain all the points corresponding to the historical data of the source-load day-ahead forecast error in space. The expression is:

[0183] ;

[0184] ;

[0185] Among them, ||β o ||1 is β o 1-norm, η is the scaling factor of the expanded vertex, β o is the minimum ratio that the o-th sample needs to expand in each eigenvector direction; is the oth sample in the (N RES +N bus ) The minimum ratio of expansion required to move toward the maximum vertex in the direction of T eigenvectors;

[0186] Get the vertex coordinates of the expanded high-dimensional polyhedron in the original coordinate system. The expression is:

[0187] ;

[0188] in, is the minimum vertex coordinate in the direction of the hth eigenvector of the high-dimensional polyhedron in the original coordinate system; is the maximum vertex coordinate in the direction of the hth eigenvector of the high-dimensional polyhedron in the original coordinate system;

[0189] Adjust the unreasonable scenes in the high-dimensional polyhedron vertex correspondence scene. The expression is:

[0190] ;

[0191] in, represent or The e-th component of is the e-th component of the N-th sample in the matrix U;

[0192] The adjusted high-dimensional polyhedron vertex set is recorded as the boundary scene set u vtx , including 2(NRES +N bus )T vertices, the expression is:

[0193] ;

[0194] in: is the boundary scene set u vtx The j-th vertex in ;

[0195] Statistics on the attribution of all samples in the historical dataset of new energy and load power day-ahead forecast errors to each vertex, including:

[0196] Calculate u vtx The Euclidean distance between each vertex in and any sample point in the historical data of the new energy and load power day-ahead forecast error is expressed as:

[0197] ;

[0198] Among them, d o,j is the historical data sample u of the oth renewable energy and load power day-ahead forecast error o and the jth vertex The Euclidean distance between

[0199] will u o Belong to the vertex closest to it, and count the number of samples belonging to each vertex and save it in array n array , the expression is:

[0200] ;

[0201] in, For array n array Used to store the The number of sample points in the array n array is the 1×2(N b +N w )T-dimensional array,

[0202] The array n array All components are initialized to 0. Whenever a sample point is assigned to the jth vertex, n array The jth element of is added by one, and the above operation is performed on each sample point, and the final result is n array That is, it reflects the attribution of the sample points of the historical dataset of the day-ahead forecast error to each vertex;

[0203] Step 1.2: Use the k-medoids algorithm to obtain the cluster center scenario in the historical data set of day-ahead forecast errors. The expression is:

[0204] ;

[0205] in, is the λth cluster center, is the corresponding probability, n clu is the number of cluster centers;

[0206] The cluster center scene u clu With the boundary scene u vtx By combining the above, we can obtain a set of representative day-ahead forecast error scenarios for renewable energy and load power, as well as the probability of each representative day-ahead forecast error scenario, namely:

[0207] ;

[0208] Among them, u typ is the scenario set of new energy and load power day-ahead forecast errors, n sce for u typ The total number of scenes included, is the sth new energy and load power day-ahead forecast error scenario,

[0209] The expressions for the components involved are:

[0210] ;

[0211] Among them, N RES and N bus are the number of new energy power stations and load nodes respectively, for The day-ahead power forecast error of the r-th renewable energy power station in the middle period t is: for The day-ahead power error of the load at the bth node in period t, where T is the total number of scheduling periods;

[0212] Determine the initial probability of each new energy and load power day-ahead forecast error scenario, expressed as:

[0213] ;

[0214] Among them, p s for The initial probability of

[0215] Step 1.3: Obtain the basic day-ahead forecast scenarios for power system load and renewable energy power, and construct a corresponding day-ahead forecast typical scenario set based on the day-ahead forecast error scenarios for renewable energy and load power combined with the basic day-ahead forecast scenarios.

[0216] Step 2: Obtain the operating parameters of the electrolytic aluminum enterprise, including: electrolytic aluminum production line parameter information and the daily output requirements of the electrolytic aluminum enterprise. On this basis, the maximum expected value of the net profit of the electrolytic aluminum enterprise under the typical scenario set of new energy and load power forecast is used as the first optimization scheduling objective function. Combined with the operating constraints of the electrolytic aluminum enterprise, an optimization scheduling model for the electrolytic aluminum enterprise is constructed, which specifically includes:

[0217] Step 2.1: Obtain the operating constraints of the electrolytic aluminum enterprise, including:

[0218] The temperature-power coupling constraint is expressed as:

[0219] ;

[0220] ;

[0221] in, is the electrolyte temperature of the i-th electrolytic aluminum production line in the basic scenario, is the lower limit of the electrolyte temperature of the i-th electrolytic aluminum production line, is the upper limit of the electrolyte temperature of the i-th aluminum electrolytic production line, c is the specific heat capacity of the molten electrolyte, m is the mass of the molten electrolyte, is the rated power of the i-th electrolytic aluminum production line, is the power of the i-th electrolytic aluminum production line in period t under the basic scenario, The duration of each scheduling period;

[0222] The power upper and lower limit constraints are expressed as:

[0223] ;

[0224] in, is the minimum power of the i-th electrolytic aluminum production line in period t, is the maximum power of the i-th electrolytic aluminum production line during period t;

[0225] The spare capacity constraint is expressed as:

[0226] ;

[0227] ;

[0228] ;

[0229] ;

[0230] ;

[0231] in, The maximum spare capacity provided for the i-th electrolytic aluminum production line in period t, is the reserve capacity provided by the i-th electrolytic aluminum production line in period t, t′ is the continuous duration of the delivered reserve capacity,

[0232] The electrolyte heat gain and loss caused by the continuous delivery of spare capacity from the beginning of time period t to time period t+t′; The electrolyte heat gain and loss caused by the continuous delivery of spare capacity from the start of time t to time t+t′;

[0233] Flexible adjustment of power constraints, the expression is:

[0234] The flexible power adjustment provided by the electrolytic aluminum production line cannot exceed the reserved spare capacity and cannot be in both the upward and downward adjustment states at the same time. To this end, a 0-1 auxiliary variable is introduced to construct the following constraints:

[0235] ;

[0236] ;

[0237] ;

[0238] in, is a 0-1 variable indicating that the i-th electrolytic aluminum production line is in an upward adjustment state in the s-th typical scenario period t; is a 0-1 variable indicating that the i-th electrolytic aluminum production line is in a downward adjustment state in the s-th typical scenario period t;

[0239] is the electrolyte temperature of the i-th electrolytic aluminum production line in the s-th typical scenario, period t, and M is a preset auxiliary parameter, which can be 1×10 8 ; Provide flexible power adjustment for the i-th electrolytic aluminum production line in the s-th typical scenario and time period t; The flexible power adjustment provided for the i-th electrolytic aluminum production line in the s-th typical scenario and time period t;

[0240] The daily output constraint is expressed as:

[0241] ;

[0242] ;

[0243] Among them, Q Al,i,s,t is the aluminum output of the ith electrolytic aluminum production line in period t under the sth typical scenario, P Al,i,s,t is the power of the i-th electrolytic aluminum production line in the s-th typical scenario period t, is the minimum power that the electrolytic aluminum production line i can maintain, f(P Al,i,s,t )for The yield calculation function when is the electrochemical equivalent of aluminum, η i is the electrolysis efficiency of the i-th electrolytic aluminum production line, n Al,i is the number of electrolytic cells included in the i-th electrolytic aluminum production line, R i is the equivalent resistance of the i-th electrolytic aluminum production line, E i is the sum of the back electromotive force, anode overvoltage and cathode overvoltage of the i-th electrolytic aluminum production line, is the rated daily output of aluminum, α min is the ratio of the lower limit of daily output to the rated daily output, is the number of electrolytic aluminum production lines.

[0244] Step 2.2: Obtain the first optimization scheduling objective function, which is expressed as:

[0245] ;

[0246] in, Subsidies for providing spare capacity to electrolytic aluminum enterprises during period t, is the aluminum product revenue in period t under the sth typical scenario, Subsidies obtained by providing flexible capacity adjustment to electrolytic aluminum enterprises in period t under the sth typical scenario, is the operation and maintenance cost of the electrolytic aluminum production line of the electrolytic aluminum enterprise in period t under the sth typical scenario, is the carbon emission cost of the electrolytic aluminum enterprise in period t under the sth typical scenario, is the number of typical scenarios;

[0247] ;

[0248] Among them, N Al is the number of electrolytic aluminum production lines; is the unit subsidy price for electrolytic aluminum production lines to participate in demand response and provide spare capacity. The maximum spare capacity provided for the i-th electrolytic aluminum production line in period t, The minimum spare capacity provided for the i-th electrolytic aluminum production line in period t;

[0249] ;

[0250] Among them, λ Al is the unit net profit of aluminum products, Q Al,i,s,t is the aluminum output of the i-th electrolytic aluminum production line in period t under the s-th typical scenario;

[0251] ;

[0252] in, Provide a unit subsidy price that allows flexible adjustment of power levels for electrolytic aluminum enterprises participating in demand response during period t. Provide flexible power adjustment for the i-th electrolytic aluminum production line in the s-th typical scenario and time period t; The flexible power adjustment provided for the i-th electrolytic aluminum production line in the s-th typical scenario and time period t;

[0253] ;

[0254] Among them, c om is the cost coefficient of electrolytic aluminum load operation and maintenance, P Al,i,s,t is the power of the i-th electrolytic aluminum production line in the s-th typical scenario during the t-th period;

[0255] ;

[0256] Among them, c carb is the unit carbon emission cost coefficient; EF Al The carbon emission intensity of the internal process of the electrolytic aluminum production line corresponding to the unit electricity consumption; is the dynamic carbon emission factor value in period t.

[0257] Step 3: Obtain power system operating parameters, including parameter information of generator sets and transmission lines. Based on these parameters, the minimum power system operating cost under a typical scenario set of renewable energy and load power forecasts is used as the second optimization scheduling objective function. Combined with the power system operation constraints, an optimization scheduling model for the power system is constructed. Specifically, the following steps are involved:

[0258] Step 3.1: Obtain the second optimization scheduling objective function, which is expressed as:

[0259] ;

[0260] in, is the total operating cost of the fossil energy unit in period t under the basic scenario, Subsidies for providing spare capacity to electrolytic aluminum enterprises during period t, is the unit price of electric energy exchange between the power system and other external power systems during period t, The power exchanged between the power system and other external power systems during period t, with the purchase of power from other external power systems being considered positive; is the regulation cost of the fossil energy unit in period t under the sth typical scenario, Subsidies obtained by providing flexible capacity adjustment to electrolytic aluminum enterprises in period t under the sth typical scenario, is the reliability cost of power system operation in period t under the sth typical scenario;

[0261] ;

[0262] Among them, a g For the g-th fossil energy unit and The corresponding power generation cost coefficient, b g For the g-th fossil energy unit and The corresponding power generation cost coefficient, c g is the fixed power generation cost coefficient of the g-th fossil energy unit when it is in operation, 、 are the costs of starting and stopping the g-th fossil energy unit, 、 are the indicator variables for the on / off state switching of the g-th fossil energy unit during period t, The unit subsidy price for providing spare capacity for fossil energy units, 、 The upper and lower reserve capacities provided for the g-th fossil energy unit during period t, is the power generation capacity of the g-th fossil energy unit in period t, is the number of fossil energy units;

[0263] ;

[0264] in, The unit cost of providing flexible power adjustment for the units in time period t; 、 are the upper and lower flexible adjustment powers provided by unit g in the sth typical scenario period t respectively;

[0265] ;

[0266] Among them, N RES is the total number of renewable energy power generation stations in the power system, N bus is the total number of load nodes in the power system, c cur 、c ld These are the unit penalty prices for curtailed electricity and load shedding electricity from new energy power stations; is the power curtailment of the rth renewable energy power station in period t under the sth typical scenario, is the load shedding power of the bth node in period t under the sth typical scenario;

[0267] Step 3.2: Obtain power system operating constraints, including:

[0268] The power balance constraint is expressed as:

[0269] ;

[0270] ;

[0271] in, is the predicted power of the rth new energy power station in time period t, is the day-ahead predicted power of the b-th node load in period t; is the power exchanged between the power system and the external power system during period t, Provide flexible power adjustment for the i-th electrolytic aluminum production line in the s-th typical scenario and time period t; The flexible power adjustment provided for the i-th electrolytic aluminum production line in the s-th typical scenario and time period t;

[0272] The line transmission capacity constraint is expressed as:

[0273] ;

[0274] ;

[0275] Among them, k lb is the power transfer factor of node b with respect to line l, f lmax is the maximum transmission power of line l, 、 、 They represent the connection of unit g, new energy power station r, and electrolytic aluminum production line i to the power system node b to access the power system;

[0276] The unit output and climbing constraints are expressed as:

[0277] ;

[0278] Among them, P gmin 、P gmax are the lower and upper limits of the output of unit g respectively; I g,t is the start / stop state variable of the g-th unit in period t, and its value is 1 or 0, indicating the start / stop state respectively;

[0279] ;

[0280] Among them, UR g , DR g are the up and down climbing rates of unit g respectively;

[0281] The minimum start and stop time constraint of the unit is expressed as:

[0282] ;

[0283] in, 、 are the durations of continuous startup and shutdown of unit g during period t-1 respectively; 、 are the minimum startup and shutdown durations of unit g respectively;

[0284] Spare capacity constraints, specifically:

[0285] The upper and lower limit constraints of the spare capacity provided by the unit are expressed as:

[0286] ;

[0287] in, is the maximum power generation capacity of the g-th fossil energy unit, is the minimum power generation capacity of the g-th fossil energy unit;

[0288] Flexible adjustment of power constraints, the expression is:

[0289] ;

[0290] The constraints on the curtailed power and load shedding power of new energy are expressed as follows:

[0291] .

[0292] Step 4: Construct a two-layer low-carbon dispatch model with the electrolytic aluminum enterprise optimization dispatch model as the lower model and the power system optimization dispatch model as the upper model. Perform linear optimization on the nonlinear terms in the two-layer low-carbon dispatch model to obtain a two-layer mixed integer quadratic programming model, which specifically includes:

[0293] The maximum power that can be achieved when the electrolytic aluminum production line is adjusted only once for 1 hour within 24 hours is used as the P in the piecewise linearization process. Al,i,s,t The upper limit of the power limit constraint is Δt=1, then:

[0294] ;

[0295] The interval Divide into n segments, and use linear interpolation function to replace each segment The daily output constraint can be transformed into the following equivalent form:

[0296] ;

[0297] Where: n is the number of evenly spaced segments, x0,⋯, x k , x k+1 ,⋯, x n For interval The endpoints of the evenly divided segments;

[0298] x k and x k+1 are the left and right endpoints of the k+1th evenly spaced segment, k=0,...,n-1, x0= , x n = ;

[0299] is the maximum operating power that the electrolytic aluminum production line i can withstand, z1,⋯,z n To show that P Al,i,s,t In the range The 0-1 auxiliary variable for the 1st to nth average partition segment, z0 is used to indicate P Al,i,s,t < Auxiliary variables, M is the preset auxiliary parameter, for The yield calculation function value at time .

[0300] Step 5: Iteratively solve the two-level mixed-integer quadratic programming model to obtain a coordinated dispatch plan for the power system and the electrolytic aluminum enterprises that responds to the integrated electricity and carbon demand, as well as the latest dynamic carbon emission factor value for the current period released by the power system. Specifically, the solution includes:

[0301] Step 5.1: Obtain the dynamic carbon emission factor value for the current period and initialize the power system's reserve capacity and flexible power demand for the electrolytic aluminum enterprise in the current period;

[0302] Step 5.2: Input the dynamic carbon emission factor value, spare capacity, and flexible power demand into the electrolytic aluminum enterprise-side optimization scheduling model to obtain the load curve of the electrolytic aluminum enterprise after participating in the comprehensive electricity-carbon demand response;

[0303] Step 5.3: Input the load curve of the electrolytic aluminum enterprise into the power system optimization scheduling model, and optimize and solve to obtain the output plan of all units, the updated power system's spare capacity and flexible power demand for the electrolytic aluminum enterprise, and the improved dynamic carbon emission factor value;

[0304] Step 5.4: Repeat the loop of steps 5.2 to 5.3 using the improved dynamic carbon emission factor value and the updated power system's reserve capacity and flexible adjustment power demand for the electrolytic aluminum enterprise until the maximum deviation of the electrolytic aluminum enterprise load curve obtained from two consecutive iterations is less than the set threshold. The improved dynamic carbon emission factor value is output as the final published carbon signal, and the current reserve and flexible adjustment power demand of the electrolytic aluminum enterprise is obtained at the same time.

[0305] The calculation method of the dynamic carbon emission factor is as follows:

[0306] ;

[0307] in, is the dynamic carbon emission factor value in period t, is the power generation capacity of the g-th fossil energy unit in period t, is the number of fossil energy units, EF g is the carbon emission intensity of the g-th unit, is the power exchanged between the power system and other external power systems during period t, is the dynamic carbon emission factor corresponding to the amount of electricity purchased from the external power system during period t.

[0308] Example 2

[0309] This embodiment provides a low-carbon scheduling device that takes into account the comprehensive electricity and carbon demand response of electrolytic aluminum enterprises, including:

[0310] The typical scenario set module for day-ahead prediction is used to obtain the typical scenario set for day-ahead prediction of new energy and load power;

[0311] The electrolytic aluminum enterprise-side optimization scheduling module is used to construct an electrolytic aluminum enterprise-side optimization scheduling model based on the maximum expected value of the electrolytic aluminum enterprise's net profit under a typical scenario set of new energy and load power day-ahead forecasts as the first optimization scheduling objective function and combined with the electrolytic aluminum enterprise's operating constraints;

[0312] The power system optimization and dispatching module is used to construct a power system optimization and dispatching model based on the minimum power system operation cost under the typical scenario set of new energy and load power forecasting as the second optimization and dispatching objective function, combined with the power system operation constraints.

[0313] A two-layer module is used to construct a two-layer low-carbon scheduling model with the electrolytic aluminum enterprise-side optimization scheduling model as the lower model and the power system-side optimization scheduling model as the upper model. The nonlinear terms in the two-layer low-carbon scheduling model are linearly optimized to obtain a two-layer mixed integer quadratic programming model.

[0314] Dynamic carbon emission factor solution module: This module is used to iteratively solve a two-level mixed-integer quadratic programming model to obtain a coordinated dispatch plan for the power system and the electrolytic aluminum enterprise that responds to the integrated electricity and carbon demand, as well as the latest dynamic carbon emission factor value for the current period released by the power system.

[0315] Among them, the solution method of the two-level mixed integer quadratic programming model is as follows:

[0316] Based on the iterative calculation between the upper model transferring the dynamic carbon emission factor, spare capacity demand and flexible adjustment power demand to the lower model and the lower model transferring its own load curve to the upper model, the latest coordinated scheduling plan and dynamic carbon emission factor are solved.

[0317] Example 3

[0318] This embodiment provides a low-carbon scheduling system that considers the comprehensive electricity and carbon demand response of electrolytic aluminum enterprises, including:

[0319] Memory, used to store computer programs / instructions;

[0320] A processor is used to execute the computer program / instructions to implement the steps of the above-mentioned low-carbon scheduling method that takes into account the comprehensive electricity and carbon demand response of electrolytic aluminum enterprises.

[0321] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0322] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0323] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0324] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 The steps for the function specified in one or more boxes.

[0325] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the purpose of the present invention and the claims, which are all protected by the present invention.

Claims

1. A low-carbon scheduling method considering the comprehensive electricity and carbon demand response of electrolytic aluminum enterprises, characterized by: The following steps are involved: Obtain a typical scenario set for day-ahead forecasts of new energy and load power; Taking the maximum expected value of the net profit of the electrolytic aluminum enterprise under the typical scenario set of new energy and load power day-ahead forecast as the first optimization scheduling objective function, combined with the operating constraints of the electrolytic aluminum enterprise, an optimization scheduling model for the electrolytic aluminum enterprise is constructed; Taking the minimum cost of power system operation under the typical scenario set of new energy and load power forecast as the second optimization dispatch objective function, combined with the power system operation constraints, the power system side optimization dispatch model is constructed. A two-layer low-carbon dispatch model is constructed with the electrolytic aluminum enterprise optimization dispatch model as the lower model and the power system optimization dispatch model as the upper model. The nonlinear terms in the two-layer low-carbon dispatch model are linearly optimized to obtain a two-layer mixed integer quadratic programming model. By iteratively solving a two-level mixed-integer quadratic programming model, we obtained a coordinated dispatch plan for the power system and the aluminum smelter that responds to the integrated electricity and carbon demand, as well as the latest dynamic carbon emission factor value for the current period released by the power system. Among them, the solution method of the two-level mixed integer quadratic programming model is as follows: Based on the iterative calculation between the upper model transferring the dynamic carbon emission factor, spare capacity demand and flexible adjustment power demand to the lower model and the lower model transferring its own load curve to the upper model, the latest coordinated scheduling plan and dynamic carbon emission factor are solved.

2. The low-carbon scheduling method considering the comprehensive electricity and carbon demand response of electrolytic aluminum enterprises according to claim 1 is characterized in that: The iterative calculation between transferring the dynamic carbon emission factor, spare capacity requirement and flexible power adjustment requirement from the upper model to the lower model and transferring the load curve from the lower model to the upper model to solve the latest coordinated scheduling solution and dynamic carbon emission factor specifically includes: The upper model transmits dynamic carbon emission factors, reserve capacity requirements, and flexible power requirements to the lower model; The lower-level model transmits the dynamic carbon emission factor, spare capacity demand, and flexible power demand information based on the obtained model, optimizes the load curve of the electrolytic aluminum enterprise, and returns it to the upper-level model to participate in the power balance constraint, so that the upper-level model optimizes the new unit combination plan and updates the dynamic carbon emission factor, spare capacity demand, and flexible power demand; Until the maximum deviation of the electrolytic aluminum enterprise load curve obtained from two consecutive iterations is less than the set threshold, the latest upper-level model unit combination plan, dynamic carbon emission factor, spare capacity demand and flexible adjustment power demand are obtained.

3. The low-carbon scheduling method considering the comprehensive electricity and carbon demand response of electrolytic aluminum enterprises according to claim 1 is characterized in that: The method for obtaining a typical scenario set for new energy and load power day-ahead prediction specifically includes: The principal component analysis method is used to obtain the boundary scenarios in the historical dataset of new energy and load power day-ahead forecast errors, including: The principal component analysis method is used to obtain the characteristic vector of the day-ahead forecast error of new energy and load power, which is expressed as follows: ； Among them, U is the historical dataset matrix of the day-ahead forecast error of new energy and load power, is the zero-mean matrix of U, is the average value of all samples in the matrix U, N is the total number of samples in the historical dataset of the forecast error of new energy and load power, S is the covariance matrix of the historical dataset of the forecast error of new energy and load power, q h is the hth eigenvector of S, λ h For q h The corresponding eigenvalue, Λ is the eigenvalue λ h The diagonal matrix formed, Q is the vector feature q h The matrix formed, is the operator that converts a vector into a diagonal matrix, is the transpose, N RES is the total number of renewable energy power generation stations in the power system, N bus is the total number of load nodes in the power system, and T is the total number of dispatch periods; Will Each sample in is converted to each feature vector q h In the coordinate system formed, the expression is: ； in, for The oth sample in , for In each eigenvector q h The mapping points in the coordinate system are: for In the eigenvector q h Projection value in direction; After all samples are projected, each eigenvector q is determined h The coordinates of the two vertices in the direction are expressed as: ； in, is the minimum vertex coordinate in the direction of the hth eigenvector in the eigenvector coordinate system, is the maximum vertex coordinate in the direction of the hth eigenvector in the eigenvector coordinate system, e h is the unit column vector in the direction of the hth eigenvector, for The Nth sample in In the eigenvector q h Projection value in direction; The scaling factor η is introduced to expand the above vertex coordinates so that the high-dimensional polyhedron surrounded by the expanded vertices can contain all the points corresponding to the historical data of the source-load day-ahead forecast error in space. The expression is: ； ； Among them, ||β o ||1 is β o 1-norm, η is the scaling factor of the expanded vertex, β o is the minimum ratio that the o-th sample needs to expand in each eigenvector direction; is the oth sample in the (N RES +N bus ) The minimum ratio of expansion required to move toward the maximum vertex in the direction of T eigenvectors; Get the vertex coordinates of the expanded high-dimensional polyhedron in the original coordinate system. The expression is: ； in, is the minimum vertex coordinate in the direction of the hth eigenvector of the high-dimensional polyhedron in the original coordinate system; is the maximum vertex coordinate in the direction of the hth eigenvector of the high-dimensional polyhedron in the original coordinate system; Adjust the unreasonable scenes in the high-dimensional polyhedron vertex correspondence scene. The expression is: ； in, represent or The e-th component of is the e-th component of the N-th sample in the matrix U; The adjusted high-dimensional polyhedron vertex set is recorded as the boundary scene set u vtx , including 2(N RES +N bus )T vertices, the expression is: ； in: is the boundary scene set u vtx The j-th vertex in ; Statistics on the attribution of all samples in the historical dataset of new energy and load power day-ahead forecast errors to each vertex, including: Calculate u vtx The Euclidean distance between each vertex in and any sample point in the historical data of the new energy and load power day-ahead forecast error is expressed as: ； Among them, d o,j is the historical data sample u of the oth renewable energy and load power day-ahead forecast error o and the jth vertex The Euclidean distance between will u o Belong to the vertex closest to it, and count the number of samples belonging to each vertex and save it in array n array , the expression is: ； in, For array n array Used to store the The number of sample points in the array n array is the 1×2(N b +N w )T-dimensional array, The array n array All components are initialized to 0. Whenever a sample point is assigned to the jth vertex, n array The jth element of is added by one, and the above operation is performed on each sample point, and the final result is n array That is, it reflects the attribution of the sample points of the historical dataset of the day-ahead forecast error to each vertex; The k-medoids algorithm is used to obtain the cluster center scenario in the historical data set of the day-ahead forecast error. The expression is: ； in, is the λth cluster center, is the corresponding probability, n clu is the number of cluster centers; The cluster center scene u clu With the boundary scene u vtx By combining the two, we can obtain a set of representative day-ahead forecast error scenarios for renewable energy and load power, as well as the probability of each representative day-ahead forecast error scenario, namely: ； Among them, u typ is the scenario set of new energy and load power day-ahead forecast errors, n sce for u typ The total number of scenes included, is the sth new energy and load power day-ahead forecast error scenario, The expressions for the components involved are: ； Among them, N RES and N bus are the number of new energy power stations and load nodes respectively, for The day-ahead power forecast error of the r-th renewable energy power station in the middle period t is: for The day-ahead power error of the load at the bth node in period t, where T is the total number of scheduling periods; Determine the initial probability of each new energy and load power day-ahead forecast error scenario, expressed as: ； Among them, p s for The initial probability of Obtain the basic scenario for day-ahead prediction, and construct a corresponding typical scenario set for day-ahead prediction based on the day-ahead prediction error scenarios of new energy and load power combined with the basic scenario for day-ahead prediction.

4. The low-carbon scheduling method considering the comprehensive electricity-carbon demand response of electrolytic aluminum enterprises according to claim 1 is characterized in that: The operating constraints of the electrolytic aluminum enterprises include: The temperature-power coupling constraint is expressed as: ； ； in, is the electrolyte temperature of the i-th electrolytic aluminum production line in the basic scenario, is the lower limit of the electrolyte temperature of the i-th electrolytic aluminum production line, is the upper limit of the electrolyte temperature of the i-th aluminum electrolytic production line, c is the specific heat capacity of the molten electrolyte, m is the mass of the molten electrolyte, is the rated power of the i-th electrolytic aluminum production line, is the power of the i-th electrolytic aluminum production line in period t under the basic scenario, The duration of each scheduling period; The power upper and lower limit constraints are expressed as: ； in, is the minimum power of the i-th electrolytic aluminum production line in period t, is the maximum power of the i-th electrolytic aluminum production line during period t; The spare capacity constraint is expressed as: ； ； ； ； ； in, The maximum spare capacity provided for the i-th electrolytic aluminum production line in period t, is the reserve capacity provided by the i-th electrolytic aluminum production line in period t, t′ is the continuous duration of the delivered reserve capacity, The electrolyte heat gain and loss caused by the continuous delivery of spare capacity from the beginning of time period t to time period t+t′; The electrolyte heat gain and loss caused by the continuous delivery of spare capacity from the start of time t to time t+t′; Flexible adjustment of power constraints, the expression is: ； ； ； in, is a 0-1 variable indicating that the i-th electrolytic aluminum production line is in an upward adjustment state in the s-th typical scenario period t; is a 0-1 variable indicating that the i-th electrolytic aluminum production line is in a downward adjustment state in the s-th typical scenario period t; is the electrolyte temperature of the i-th electrolytic aluminum production line in the s-th typical scenario and time period t, and M is the preset auxiliary parameter; Provide flexible power adjustment for the i-th electrolytic aluminum production line in the s-th typical scenario and time period t; The flexible power adjustment provided for the i-th electrolytic aluminum production line in the s-th typical scenario and time period t; The daily output constraint is expressed as: ； ； Among them, Q Al,i,s,t is the aluminum output of the ith electrolytic aluminum production line in period t under the sth typical scenario, P Al,i,s,t is the power of the i-th electrolytic aluminum production line in the s-th typical scenario period t, is the minimum power that the electrolytic aluminum production line i can maintain, f(P Al,i,s,t )for The yield calculation function when is the electrochemical equivalent of aluminum, η i is the electrolysis efficiency of the i-th electrolytic aluminum production line, n Al,i is the number of electrolytic cells included in the i-th electrolytic aluminum production line, R i is the equivalent resistance of the i-th electrolytic aluminum production line, E i is the sum of the back electromotive force, anode overvoltage and cathode overvoltage of the i-th electrolytic aluminum production line, is the rated daily output of aluminum, α min is the ratio of the lower limit of daily output to the rated daily output, is the number of electrolytic aluminum production lines.

5. The low-carbon scheduling method considering the comprehensive electricity and carbon demand response of electrolytic aluminum enterprises according to claim 1 is characterized in that: The first optimization scheduling objective function is expressed as: ； in, Subsidies for providing spare capacity to electrolytic aluminum enterprises during period t, is the aluminum product revenue in period t under the sth typical scenario, Subsidies obtained by providing flexible capacity adjustment to electrolytic aluminum enterprises in period t under the sth typical scenario, is the operation and maintenance cost of the electrolytic aluminum production line of the electrolytic aluminum enterprise in period t under the sth typical scenario, is the carbon emission cost of the electrolytic aluminum enterprise in period t under the sth typical scenario, is the number of typical scenarios; ； Among them, N Al is the number of electrolytic aluminum production lines; is the unit subsidy price for electrolytic aluminum production lines to participate in demand response and provide spare capacity. The maximum spare capacity provided for the i-th electrolytic aluminum production line in period t, The minimum spare capacity provided for the i-th electrolytic aluminum production line in period t; ； Among them, λ Al is the unit net profit of aluminum products, Q Al,i,s,t is the aluminum output of the i-th electrolytic aluminum production line in period t under the s-th typical scenario; ； in, Provide a unit subsidy price that allows flexible adjustment of power levels for electrolytic aluminum enterprises participating in demand response during period t. Provide flexible power adjustment for the i-th electrolytic aluminum production line in the s-th typical scenario and time period t; The flexible power adjustment provided for the i-th electrolytic aluminum production line in the s-th typical scenario and time period t; ； Among them, c om is the cost coefficient of electrolytic aluminum load operation and maintenance, P Al,i,s,t is the power of the i-th electrolytic aluminum production line in the s-th typical scenario during the t-th period; ； Among them, c carb is the unit carbon emission cost coefficient; EF Al The carbon emission intensity of the internal process of the electrolytic aluminum production line corresponding to the unit electricity consumption; is the dynamic carbon emission factor value in period t.

6. The low-carbon scheduling method considering the comprehensive electricity and carbon demand response of electrolytic aluminum enterprises according to claim 1 is characterized in that: The second optimization scheduling objective function is expressed as: ； in, is the total operating cost of the fossil energy unit in period t under the basic scenario, Subsidies for providing spare capacity to electrolytic aluminum enterprises during period t, is the unit price of electric energy exchange between the power system and other external power systems during period t, The power exchanged between the power system and other external power systems during period t, with the purchase of power from other external power systems being considered positive; is the regulation cost of the fossil energy unit in period t under the sth typical scenario, Subsidies obtained by providing flexible capacity adjustment to electrolytic aluminum enterprises in period t under the sth typical scenario, is the power system operation reliability cost in period t under the sth typical scenario; ； Among them, a g For the g-th fossil energy unit and The corresponding power generation cost coefficient, b g For the g-th fossil energy unit and The corresponding power generation cost coefficient, c g is the fixed power generation cost coefficient of the g-th fossil energy unit when it is in operation, 、 are the costs of starting and stopping the g-th fossil energy unit, 、 are the indicator variables for the on / off state switching of the g-th fossil energy unit during period t, The unit subsidy price for providing spare capacity for fossil energy units, 、 The upper and lower reserve capacities provided for the g-th fossil energy unit during period t, is the power generation capacity of the g-th fossil energy unit in period t, is the number of fossil energy units; ； in, The unit cost of providing flexible power adjustment for the units in time period t; 、 are the upper and lower flexible adjustment powers provided by unit g in the sth typical scenario period t respectively; ； Among them, N RES is the total number of renewable energy power generation stations in the power system, N bus is the total number of load nodes in the power system, c cur 、c ld These are the unit penalty prices for curtailed electricity and load shedding electricity from new energy power stations; is the power curtailment of the rth renewable energy power station in period t under the sth typical scenario, is the load shedding power of the bth node in period t under the sth typical scenario; Power system operation constraints, including: The power balance constraint is expressed as: ； ； in, is the predicted power of the rth new energy power station in time period t, is the day-ahead predicted power of the b-th node load in period t; is the power exchanged between the power system and the external power system during period t, Provide flexible power adjustment for the i-th electrolytic aluminum production line in the s-th typical scenario and time period t; The flexible power adjustment provided for the i-th electrolytic aluminum production line in the s-th typical scenario and time period t; The line transmission capacity constraint is expressed as: ； ； Among them, k lb is the power transfer factor of node b with respect to line l, f lmax is the maximum transmission power of line l, 、 、 They represent the connection of unit g, new energy power station r, and electrolytic aluminum production line i to the power system node b to access the power system; The unit output and climbing constraints are expressed as: ； Among them, P gmin 、P gmax are the lower and upper limits of the output of unit g respectively; I g,t is the start and stop state variable of the g-th unit in period t; ； Among them, UR g , DR g are the up and down climbing rates of unit g respectively; The minimum start and stop time constraint of the unit is expressed as: ； in, 、 are the durations of continuous startup and shutdown of unit g during period t-1 respectively; 、 are the minimum startup and shutdown durations of unit g respectively; Spare capacity constraints, specifically: The upper and lower limit constraints of the spare capacity provided by the unit are expressed as: ； in, is the maximum power generation capacity of the g-th fossil energy unit, is the minimum power generation capacity of the g-th fossil energy unit; Flexible adjustment of power constraints, the expression is: ； The constraints on the curtailed power and load shedding power of new energy are expressed as follows: 。 7. The low-carbon scheduling method considering the comprehensive electricity and carbon demand response of electrolytic aluminum enterprises according to claim 5 is characterized in that: Linear optimization of nonlinear terms in the two-layer low-carbon dispatch model is performed, including: Let Δt=1 in the power upper and lower limit constraints, then: ； The interval Divide into n segments, and use linear interpolation function to replace each segment The daily output constraint can be transformed into the following equivalent form: ； Where: n is the number of evenly spaced segments, x0,..., x k , x k+1 ,..., x n For interval The endpoints of the evenly divided segments; x k and x k+1 are the left and right endpoints of the k+1th evenly spaced segment, k=0,...,n-1, x0= , x n = ; is the maximum operating power that the electrolytic aluminum production line i can withstand, z1,...,z n To show that P Al,i,s,t In the range The 0-1 auxiliary variable for the 1st to nth average partition segment, z0 is used to indicate P Al,i,s,t < Auxiliary variables, M is the preset auxiliary parameter, for The yield calculation function value at time .

8. The low-carbon scheduling method considering the comprehensive electricity-carbon demand response of electrolytic aluminum enterprises according to claim 1 is characterized in that: The calculation method of the dynamic carbon emission factor is expressed as follows: ； in, is the dynamic carbon emission factor value in period t, is the power generation capacity of the g-th fossil energy unit in period t, is the number of fossil energy units, EF g is the carbon emission intensity of the g-th unit, is the power exchanged between the power system and other external power systems during period t, is the dynamic carbon emission factor corresponding to the amount of electricity purchased from the external power system during period t.

9. A low-carbon dispatching device that takes into account the comprehensive electricity and carbon demand response of electrolytic aluminum enterprises, characterized by: include: The typical scenario set module for day-ahead prediction is used to obtain the typical scenario set for day-ahead prediction of new energy and load power; The electrolytic aluminum enterprise-side optimization scheduling module is used to construct an electrolytic aluminum enterprise-side optimization scheduling model based on the maximum expected value of the electrolytic aluminum enterprise's net profit under a typical scenario set of new energy and load power day-ahead forecasts as the first optimization scheduling objective function and combined with the electrolytic aluminum enterprise's operating constraints; The power system optimization and dispatching module is used to construct a power system optimization and dispatching model based on the minimum power system operation cost under the typical scenario set of new energy and load power forecasting as the second optimization and dispatching objective function, combined with the power system operation constraints. A two-layer module is used to construct a two-layer low-carbon scheduling model with the electrolytic aluminum enterprise-side optimization scheduling model as the lower model and the power system-side optimization scheduling model as the upper model. The nonlinear terms in the two-layer low-carbon scheduling model are linearly optimized to obtain a two-layer mixed integer quadratic programming model. Dynamic carbon emission factor solution module: This module is used to iteratively solve a two-level mixed-integer quadratic programming model to obtain a coordinated dispatch plan for the power system and the electrolytic aluminum enterprise that responds to the integrated electricity and carbon demand, as well as the latest dynamic carbon emission factor value for the current period released by the power system. Among them, the solution method of the two-level mixed integer quadratic programming model is as follows: Based on the iterative calculation between the upper model transferring the dynamic carbon emission factor, spare capacity demand and flexible adjustment power demand to the lower model and the lower model transferring its own load curve to the upper model, the latest coordinated scheduling plan and dynamic carbon emission factor are solved.

10. A low-carbon dispatching system that considers the comprehensive electricity and carbon demand response of electrolytic aluminum enterprises, characterized by: include: Memory, used to store computer programs / instructions; A processor for executing the computer program / instructions to implement the steps of the low-carbon scheduling method according to any one of claims 1-8 that takes into account the comprehensive electricity-carbon demand response of the electrolytic aluminum enterprise.

Images