A double-layer dispatching method and system considering electrolytic aluminum load participating in grid operation

Abstract

The present application relates to a kind of double-layer scheduling method and system considering electrolytic aluminium load participating in grid operation, belong to electric power engineering technical field.The method includes: constructing high-load electrolytic aluminium upper model, including: high-load electrolytic aluminium in self-provided coal-fired power plant model, considering new energy consumption high-load electrolytic aluminium load model and considering demand response technology high-load electrolytic aluminium load model;Constitute the lower model of grid day-ahead market clearing;According to high-load electrolytic aluminium upper model, the lower model of grid day-ahead market clearing is constituted double-layer model, then double-layer model is converted into mixed integer linear model, then solve, according to the scheduling result of solving result.This method can effectively reduce the output of coal-fired power plant, improve photovoltaic power generation utilization level, provide a good choice for promoting local renewable energy consumption and improving energy efficiency.

Description

Technical Field

[0001] This invention belongs to the field of power engineering technology, specifically relating to a two-layer dispatching method and system that takes into account the participation of electrolytic aluminum load in power grid operation. Background Technology

[0002] Based on the electrolytic aluminum load model, this paper studies a two-level optimal scheduling method for high-energy-consuming electrolytic aluminum loads under the power market environment.

[0003] Currently, there are some studies on two-layer optimal scheduling methods for energy systems that integrate emerging technologies. For example, in a multi-energy market environment, a two-layer collaborative optimization model for thermal power units in the electricity and heat market under a real-time heat pricing mechanism is proposed. The upper layer model is the optimal scheduling model for thermal power units, and the lower layer is the clearing model for the electricity and heat market. Finally, the two-layer model is transformed into a mixed-integer linear programming model for solution using KKT conditions.

[0004] The integration of electrolytic aluminum loads has, on the one hand, increased the total regional load and enhanced the power grid's hydropower absorption capacity; on the other hand, due to its rapid growth rate, large load volume, and centralized power supply characteristics, the large-scale integration of electrolytic aluminum loads has also significantly impacted the safety and stability of the power grid. Electrolytic aluminum industrial loads are classified as Class I loads, requiring high power quality and reliability. Voltage fluctuations on the grid side, the occurrence of the anode effect, and changes in the number of electrolytic cells can all cause current fluctuations. Without proper current stabilization control, this will not only affect the quality of aluminum production but also increase electricity consumption per ton of aluminum. Therefore, overcoming the shortcomings of existing technologies is a pressing issue in the field of power engineering technology. Summary of the Invention

[0005] The purpose of this invention is to address the shortcomings of existing technologies and to provide a two-layer scheduling method and system that takes into account the participation of electrolytic aluminum loads in grid operation, specifically addressing the scheduling optimization problem of electrolytic aluminum loads in grid operation.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0007] A two-layer dispatching method that takes into account the participation of electrolytic aluminum load in power grid operation includes the following steps:

[0008] S1. Construct an upper-level model for high-energy-consuming electrolytic aluminum; the upper-level model for high-energy-consuming electrolytic aluminum includes: a self-owned coal-fired power plant model for high-energy-consuming electrolytic aluminum, a high-energy-consuming electrolytic aluminum load model that takes into account the consumption of new energy sources, and a high-energy-consuming electrolytic aluminum load model that takes into account demand response technology.

[0009] The self-owned coal-fired power plant model in high-energy-consuming electrolytic aluminum uses the power generation cost of the self-owned power plant as the objective function and the output power constraint and ramp-up power constraint of the self-owned power plant as the constraint conditions.

[0010] The high-energy-consuming electrolytic aluminum load model that takes into account the consumption of new energy sources uses the distributed generation price as the objective function and the photovoltaic output constraint as the constraint condition.

[0011] The high-energy-consuming electrolytic aluminum load model that takes demand response technology into account takes the compensation cost of demand response as the objective function and non-transferable load constraints, load transfer-in constraints, and load transfer-out constraints as constraints.

[0012] S2. Construct a lower-level model for day-ahead market clearing in the power grid;

[0013] The aforementioned lower-level model for day-ahead market clearing in the power grid takes social welfare maximization as its objective function and uses the power balance equation, photovoltaic output constraints, coal-fired power unit operation constraints, coal-fired power unit ramping constraints, active load constraints, and line transmission capacity constraints as constraints.

[0014] S3. Construct a two-layer model based on the upper-layer model of high-energy-consuming electrolytic aluminum and the lower-layer model of the day-ahead market clearing of the power grid. The objective function of the two-layer model is to minimize the cost of coal-fired power generation and photovoltaic power generation. Then, the two-layer model is transformed into a mixed integer linear model and solved. The scheduling is carried out based on the solution results.

[0015] Furthermore, preferably, the construction of the high-energy-carrying electrolytic aluminum upper layer model in step S1 is as follows:

[0016] S11. Construct a model of a self-owned coal-fired power plant in high-energy-consuming electrolytic aluminum, as shown in equations (1)-(4):

[0017] Target cost of power generation from self-owned coal-fired power plant C coal for:

[0018]

[0019] Among them, P t coal Let c be the output power of a self-owned coal-fired power plant at time t. coal The unit power generation cost of a self-owned coal-fired power plant;

[0020] Self-owned coal-fired power plants must meet the following operational constraints:

[0021] P min ≤P t coal ≤P max (2)

[0022]

[0023] Among them, P max P min and P rampThese are the maximum output power, minimum output power, and maximum ramp power of a self-owned coal-fired power plant, respectively.

[0024] Carbon emission costs are calculated as follows:

[0025]

[0026] In the formula, c carbon and e coal These represent the unit carbon price and the carbon emission factor of coal, respectively.

[0027] S12. Construct a high-energy-consuming electrolytic aluminum load model that takes into account the consumption of new energy sources, as shown in equations (5)-(9):

[0028] First, it is necessary to ensure that the output of distributed photovoltaic power is within a certain confidence level η:

[0029]

[0030] in, To predict photovoltaic power output; The actual output of photovoltaic power; pr(·) represents the probability that the actual output of photovoltaic power is less than the predicted output of photovoltaic power.

[0031] Assuming the actual output of photovoltaic power Follows Gaussian distribution but:

[0032]

[0033] in Represents the cumulative distribution function of photovoltaic power. For variance;

[0034] Thus, the standard form is obtained:

[0035]

[0036] Let N(0,1) be the cumulative distribution function of the standard normal distribution.

[0037] At confidence level η, the photovoltaic output constraints are as follows:

[0038]

[0039] in, Let N(0,1) be the inverse cumulative distribution function of the standard normal distribution.

[0040] The price of distributed generation is:

[0041]

[0042] Where lcoe is the cost per kilowatt-hour of photovoltaic power generation;

[0043] S13. Construct a high-energy-consuming electrolytic aluminum load model that incorporates demand response technology, as shown in equations (10)-(16):

[0044] Assume that the high-energy-consuming electrolytic aluminum load includes non-transferable load and transferable load; the non-transferable load must be satisfied within a scheduling cycle, while the transferable load is transferred in and out within a scheduling cycle based on actual demand, as follows:

[0045] P t e =P t e，fix +P t e，in -P t e，out (10)

[0046] 0≤P t e，in ≤αP t e (11)

[0047] 0≤P t e，out ≤αP t e (12)

[0048]

[0049] Wherein, formula (10) represents the electrolytic aluminum load P at each time t. t e Due to non-transferable load P t e，fix , load transfer P t e，in and the load P t e，out The composition, formulas (11)-(12) respectively constrain the maximum inbound load and the maximum outbound load at each moment, where α is the maximum transferable load coefficient, formula (13) guarantees that the total inbound load within a scheduling cycle is equal to the total outbound load, where T is a scheduling cycle; the compensation cost for demand response is expressed as follows:

[0050]

[0051] In the formula, c pe Indicates the unit compensation cost for demand response;

[0052] Since the electrolytic aluminum load can purchase electricity from the electricity market, the power balance of the electrolytic aluminum load itself is:

[0053] P t coa l+P t PV +P t EA =P t e (15)

[0054] In the formula, P t EA This represents the amount of electricity purchased from the electricity market by the electrolytic aluminum load at time t.

[0055] Total day-ahead operating cost C for electrolytic aluminum:

[0056]

[0057] in, For the electricity purchase cost from the electricity market for electrolytic aluminum loads, λ t This represents the marginal electricity price at time t.

[0058] Furthermore, preferably, σ t，fore Set as 5%.

[0059] Furthermore, preferably, the lower-level model for the day-ahead market clearing of the power grid in step S2 is as follows:

[0060] The electricity market is cleared by an independent power system operator with the goal of maximizing social welfare, i.e. minimizing the day-ahead operating costs of the power system, as shown in equation (17).

[0061]

[0062]

[0063]

[0064]

[0065]

[0066]

[0067]

[0068] In the formula, the subscript t represents the scheduling time, the subscript m represents photovoltaic power generation, the subscript n represents the coal-fired unit, the subscript l represents the electrical load, and the subscript i represents the bus; Ω RES Ω represents a photovoltaic power generation collection. CG Ω represents a collection of coal-fired power plants. U Π represents the set of electrical loads; This represents the output electrical power of photovoltaic m at time t. This represents the output electrical power of coal-fired unit n at time t. This represents the electrical load demand of user l at time t; This represents the unit power generation cost of coal-fired unit n. This represents the unit electricity cost for user l; ε(i) indicates whether there is an electrolytic aluminum load connected to the i-th busbar, with 1 for connected electrolytic aluminum load and 0 for unconnected load; b ij θ represents the line susceptance connecting busbars i and j. it θ represents the voltage phase angle of the i-th bus at time t; jt This represents the voltage phase angle of the j-th bus at time t; Δt represents the unit power generation cost of photovoltaic m, and Δt represents the dispatching time period;

[0069] Equation (17) minimizes the day-ahead operating cost of the power system; Equation (18) is the power balance equation; Equation (19) is the photovoltaic output constraint, where This represents the unit output electrical power of photovoltaic m at time t. The maximum configuration capacity of photovoltaic m is given; Equation (20) represents the operating constraints of the coal-fired power unit, where... and Let P be the maximum and minimum output power of the nth unit, respectively; Equation (21) is the ramp-up constraint for the coal-fired power unit, where P n ramp Let be the maximum ramping power of the nth unit; Equation (22) represents the active power load constraint, where Let P be the maximum electrical load of user l; Equation (23) is the line transmission capacity constraint, where P ij，max This indicates the transmission capacity of line ij.

[0070] Furthermore, preferably, the specific method of step S3 is as follows:

[0071] A two-layer model is constructed based on the upper-layer model of high-energy-consuming electrolytic aluminum and the lower-layer model of day-ahead market clearing in the power grid. The two-layer model is as follows:

[0072]

[0073] The constraints are as follows:

[0074] Equations (2), (3), (8)-(15), (18)-(23) (25)

[0075]

[0076]

[0077]

[0078]

[0079]

[0080]

[0081]

[0082]

[0083]

[0084]

[0085]

[0086]

[0087]

[0088]

[0089] Where, λ t λ represents the marginal electricity price at time t; mt The marginal electricity price of photovoltaic power generation m at time t, λ nt Represents the boundary electricity price of coal-fired unit n at time t, λ lt λ represents the marginal electricity price of the electrical load l at time t; it , λ jt Let i and j represent the marginal electricity prices of bus i and j at time t, respectively; (25) represents the constraints between the upper and lower level models; (25)-(29) represent the Lagrangian functions of the original problem with respect to the variables of the original problem. and θ it The gradient is 0; (30)-(39) are complementary constraints on the inequality equations of the lower-level original problem; α mt β mt , χ nt δ nt φ nt , γ lt κ lt , and ω ij,t The variables are the dual variables of the left and right sides of the inequalities in equations (19)-(23), respectively; then the bi-level model is transformed into a mixed integer linear model, as follows:

[0090] The Big M method is used to process the two-layer model, as follows:

[0091]

[0092]

[0093]

[0094]

[0095]

[0096]

[0097]

[0098]

[0099]

[0100]

[0101] In the formula, and All are auxiliary 0-1 integer variables; M is a specific value;

[0102] Consider the existence of a nonlinear term λ in the objective function (24) t P t EA Using strong duality theory, the objective function of the lower-level model is expressed as follows:

[0103]

[0104] Where, λ t P t EA It can be represented as:

[0105]

[0106] Therefore, the objective function (24) is expressed as

[0107]

[0108] The linear constraints are as follows:

[0109] (25)-(29) and (40)-(49) (53)

[0110] The above models are typical mixed-integer linear programming problems, such as equations (25)-(29), (40)-(49) and (52), which are then solved and scheduled according to the solution results.

[0111] Furthermore, preferably, the solution is obtained using the CPLEX solver or the GUROBI solver.

[0112] This invention also provides a two-layer dispatching system that takes into account the participation of electrolytic aluminum load in grid operation, comprising:

[0113] The first processing module is used to construct the upper-level model of high-energy-consuming electrolytic aluminum; the upper-level model of high-energy-consuming electrolytic aluminum includes: a self-owned coal-fired power plant model in high-energy-consuming electrolytic aluminum, a high-energy-consuming electrolytic aluminum load model that takes into account the consumption of new energy sources, and a high-energy-consuming electrolytic aluminum load model that takes into account demand response technology.

[0114] The second processing module is used to construct the lower-level model of the day-ahead market clearing of the power grid; the lower-level model of the day-ahead market clearing of the power grid takes the maximization of social welfare as the objective function and uses the power balance equation constraint, photovoltaic power output constraint, coal-fired power unit operation constraint, coal-fired power unit ramping constraint, active load constraint and line transmission capacity constraint as the constraint conditions.

[0115] The third processing module is used to construct a two-layer model based on the upper-layer model of high-energy-consuming electrolytic aluminum and the lower-layer model obtained from the day-ahead market clearing of the power grid. The objective function of the two-layer model is to minimize the cost of coal-fired power generation and photovoltaic power generation. Then, the two-layer model is transformed into a mixed-integer linear model and then solved.

[0116] The scheduling and control module is used to perform scheduling and control based on the solution results.

[0117] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor, when executing the program, implements the steps of the above-mentioned two-layer scheduling method that takes into account the participation of electrolytic aluminum load in grid operation.

[0118] The present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, when the computer program is executed by a processor, it implements the steps of the above-mentioned two-layer scheduling method that takes into account the participation of electrolytic aluminum load in grid operation.

[0119] In this invention, α is the maximum transferable load factor, and its value ranges from 0 to 1.

[0120] Predicted photovoltaic output values ​​can be obtained through PV, developed by the U.S. National Renewable Energy Agency.

[0121] This invention constructs a two-layer optimized scheduling method for the power system based on upper and lower layer models, which considers electrolytic aluminum load and new energy consumption. M is a specific value that is as small as possible while satisfying constraints.

[0122] Compared with the prior art, the beneficial effects of this invention are as follows:

[0123] Based on energy internet technology, this invention establishes a linear functional relationship equation between energy conservation and emission reduction targets and energy efficiency and economic constraints, thereby achieving the goal of energy conservation and emission reduction in electrolytic aluminum production.

[0124] The method of this invention can effectively reduce the output of coal-fired power plants and improve the utilization level of photovoltaic power generation, providing a good option for promoting local renewable energy consumption and improving energy efficiency.

[0125] This invention provides a method for controlling high-energy-consuming loads on a large scale. As the second most commonly used metal after steel, China's electrolytic aluminum production reached 35.75 million tons in 2019, with Shandong accounting for 21%, Xinjiang 15.5%, Inner Mongolia 13%, Guangxi, Qinghai, Yunnan, and Gansu each around 7%, and Henan around 5%. Producing aluminum at this scale requires trillions of kilowatt-hours of electricity. Electricity costs account for over 30% of aluminum production costs. Henan's weighted average electricity price of 0.406 yuan / kWh is more than double that of Xinjiang. Therefore, combined with China's increasingly stringent emission reduction policies, high-energy-consuming electrolytic aluminum enterprises, especially in regions with higher weighted average electricity prices, have a strong incentive to participate in grid ancillary services. This invention can directly benefit the profitability of the electrolytic aluminum sector. The current pre-tax profit per ton of aluminum has reached 3,000 yuan, the highest level in nearly 10 years. The domestic production capacity of the entire electrolytic aluminum sector is nearing its ceiling. Attached Figure Description

[0126] Figure 1 This is a diagram of a two-layer model structure;

[0127] Figure 2 It is an IEEE-39 transmission network node;

[0128] Figure 3 Contribute to the unit's photovoltaic power generation;

[0129] Figure 4 For power balance in scenario S1;

[0130] Figure 5 For power balance in scenario S2;

[0131] Figure 6 For power balance in S3 scenario;

[0132] Figure 7 For power balance in scenario S4;

[0133] Figure 8 The changes in day-ahead operating costs and carbon emissions with carbon prices;

[0134] Figure 9 The day-ahead operating costs and carbon emissions vary with the price per kilowatt-hour.

[0135] Figure 10 This is a schematic diagram of the structure of the two-layer dispatching system of the present invention, which takes into account the participation of electrolytic aluminum load in power grid operation;

[0136] Figure 11 This is a schematic diagram of the electronic device structure of the present invention. Detailed Implementation

[0137] The present invention will now be described in further detail with reference to the embodiments.

[0138] Those skilled in the art will understand that the following embodiments are for illustrative purposes only and should not be construed as limiting the scope of the invention. Where specific techniques or conditions are not specified in the embodiments, they are performed in accordance with the techniques or conditions described in the literature in the field or according to the product instructions. Materials or equipment whose manufacturers are not specified are all conventional products that can be obtained by purchase.

[0139] like Figure 1 As shown, the present invention provides a two-layer dispatching method that takes into account the participation of electrolytic aluminum load in power grid operation, comprising the following steps:

[0140] S1. Construct an upper-level model for high-energy-consuming electrolytic aluminum; the upper-level model for high-energy-consuming electrolytic aluminum includes: a self-owned coal-fired power plant model for high-energy-consuming electrolytic aluminum, a high-energy-consuming electrolytic aluminum load model that takes into account the consumption of new energy sources, and a high-energy-consuming electrolytic aluminum load model that takes into account demand response technology.

[0141] The self-owned coal-fired power plant model in high-energy-consuming electrolytic aluminum uses the power generation cost of the self-owned power plant as the objective function and the output power constraint and ramp-up power constraint of the self-owned power plant as the constraint conditions.

[0142] The high-energy-consuming electrolytic aluminum load model that takes into account the consumption of new energy sources uses the distributed generation price as the objective function and the photovoltaic output constraint as the constraint condition.

[0143] The high-energy-consuming electrolytic aluminum load model that takes demand response technology into account takes the compensation cost of demand response as the objective function and non-transferable load constraints, load transfer-in constraints, and load transfer-out constraints as constraints.

[0144] S2. Construct a lower-level model for day-ahead market clearing in the power grid;

[0145] The aforementioned lower-level model for day-ahead market clearing in the power grid takes social welfare maximization as its objective function and uses the power balance equation, photovoltaic output constraints, coal-fired power unit operation constraints, coal-fired power unit ramping constraints, active load constraints, and line transmission capacity constraints as constraints.

[0146] S3. Construct a two-layer model based on the upper-layer model of high-energy-consuming electrolytic aluminum and the lower-layer model of the day-ahead market clearing of the power grid. The objective function of the two-layer model is to minimize the cost of coal-fired power generation and photovoltaic power generation. Then, the two-layer model is transformed into a mixed integer linear model and solved. The scheduling is carried out based on the solution results.

[0147] S11, Model of self-contained coal-fired power plant in high-energy-consuming electrolytic aluminum, as shown in equations (1)-(4):

[0148] Target cost of power generation from self-owned coal-fired power plant C coal for:

[0149]

[0150] Among them, P t coal Let c be the output power of a self-owned coal-fired power plant at time t. coal The unit power generation cost of a self-owned coal-fired power plant;

[0151] Self-owned coal-fired power plants must meet the following operational constraints:

[0152] P min ≤P t coal ≤P max (2)

[0153]

[0154] Among them, P max P min and P ramp These are the maximum output power, minimum output power, and maximum ramp power of a self-owned coal-fired power plant, respectively.

[0155] Carbon emission costs are calculated as follows:

[0156]

[0157] In the formula, c carbon and e coal These represent the unit carbon price and the carbon emission factor of coal, respectively.

[0158] S12, High-energy-consuming electrolytic aluminum considering the consumption of new energy sources

[0159] Utilizing new energy sources in high-energy-consuming electrolytic aluminum production can effectively improve the clean energy consumption of the system. Taking distributed photovoltaic (PV) systems as an example, for the safe and stable operation of the system, it is first necessary to ensure that the output of distributed PV systems is within a certain confidence level η.

[0160]

[0161] in, To predict photovoltaic (PV) output, we assume the actual PV output. Follows Gaussian distribution The above formula can be rewritten as:

[0162]

[0163] in Let represent the cumulative distribution function of photovoltaic power, which can be rewritten in the following standard form:

[0164]

[0165] Let N(0,1) be the cumulative distribution function of the standard normal distribution.

[0166] At confidence level η, the photovoltaic output constraints are as follows:

[0167]

[0168] in Let be the inverse cumulative distribution function of the standard normal distribution N(0,1). The predicted value of photovoltaic power output can be obtained from the PV model developed by the U.S. National Renewable Energy Laboratory, where σ... t，fore Set as standard predicted value 5%.

[0169] The price of distributed generation is:

[0170]

[0171] Where lcoe is the cost per kilowatt-hour of photovoltaic power generation.

[0172] S13, High-energy-consuming electrolytic aluminum considering demand response technology

[0173] As a significant industrial load, electrolytic aluminum can serve as an important demand-side resource, enabling peak shaving and valley filling of electricity load, and improving the reliability and economy of the power system. It is assumed that the high-energy-consuming electrolytic aluminum load includes both non-transferable and transferable loads. Non-transferable loads must be satisfied within a scheduling cycle, while transferable loads can be transferred in and out based on actual demand within a scheduling cycle, as detailed below:

[0174] P t e =P t e，fix +P t e，in -P t e，out (10)

[0175] 0≤P t e，in≤αP t e (11)

[0176] 0≤P t e，out ≤αP t e (12)

[0177]

[0178] In the formula (10), the electrolytic aluminum load P at each time t is represented. t e Due to non-transferable load P t e，fix , load transfer P t e ，in and the load P t e，out The composition, formulas (11)-(12) respectively constrain the maximum inbound load and the maximum outbound load at each moment, where α is the maximum transferable load coefficient, formula (13) guarantees that the total inbound load within a scheduling cycle is equal to the total outbound load, where T is a scheduling cycle; the compensation cost for demand response is expressed as follows:

[0179]

[0180] In the formula, c pe This represents the unit compensation cost for the demand response.

[0181] Furthermore, the electrolytic aluminum load can purchase electricity from the electricity market, therefore the power balance of the electrolytic aluminum load itself is:

[0182] P t coal +P t PV +P t EA =P t e (15)

[0183] In the formula, P t EA This represents the amount of electricity purchased from the electricity market by the electrolytic aluminum load at time t.

[0184] The total day-ahead operating cost C of the electrolytic aluminum load is as follows, where the third item is the cost of purchasing electricity from the electricity market for the electrolytic aluminum load, λ. t This represents the marginal electricity price at time t.

[0185]

[0186] in, For the electricity purchase cost from the electricity market for electrolytic aluminum loads, λ t This represents the marginal electricity price at time t.

[0187] In step S2, the lower-level model is obtained through the day-ahead market clearing of the power grid, as detailed below:

[0188] The electricity market is cleared by independent electricity system operators with the goal of maximizing social welfare, as specifically characterized by the following:

[0189]

[0190]

[0191]

[0192]

[0193]

[0194]

[0195]

[0196] In the formula, the subscript t represents the scheduling time, the subscript m represents photovoltaic power generation, the subscript n represents the coal-fired unit, the subscript l represents the electrical load, and the subscript i represents the bus; Ω RES Ω represents a photovoltaic power generation collection. CG Ω represents a collection of coal-fired power plants. U Π represents the set of electrical loads; This represents the output electrical power of photovoltaic m at time t. This represents the output electrical power of coal-fired unit n at time t. This represents the electrical load demand of user l at time t; This represents the unit power generation cost of coal-fired unit n. This represents the unit electricity cost for user l; ε(i) indicates whether there is an electrolytic aluminum load connected to the i-th busbar, with 1 for connected electrolytic aluminum load and 0 for unconnected load; b ij θ represents the line susceptance connecting busbars i and j. it θ represents the voltage phase angle of the i-th bus at time t; jt This represents the voltage phase angle of the j-th bus at time t; Δt represents the unit power generation cost of photovoltaic m, and Δt represents the dispatching time period;

[0197] Equation (17) minimizes the day-ahead operating cost of the power system; Equation (18) is the power balance equation; Equation (19) is the photovoltaic output constraint, where This represents the unit output electrical power of photovoltaic m at time t. The maximum configuration capacity of photovoltaic m is given; Equation (20) represents the operating constraints of the coal-fired power unit, where... and Let P be the maximum and minimum output power of the nth unit, respectively; Equation (21) is the ramp-up constraint for the coal-fired power unit, where P n ramp Let be the maximum ramping power of the nth unit; Equation (22) represents the active power load constraint, where Let P be the maximum electrical load of user l; Equation (23) is the line transmission capacity constraint, where P ij,max This indicates the transmission capacity of line ij.

[0198] Based on the upper and lower layer models in steps S1 and S2, a compact form of the two-layer optimization model is proposed, as follows:

[0199]

[0200] The constraints are as follows:

[0201] (2), (3), (8)-(15), (18)-(23) (25)

[0202]

[0203]

[0204]

[0205]

[0206]

[0207]

[0208]

[0209]

[0210]

[0211]

[0212]

[0213]

[0214]

[0215]

[0216] Where, λ tλ represents the marginal electricity price at time t; mt The marginal electricity price of photovoltaic power generation m at time t, λ nt Represents the boundary electricity price of coal-fired unit n at time t, λ lt λ represents the marginal electricity price of the electrical load l at time t; it , λ jt Let i and j represent the marginal electricity prices of bus i and j at time t, respectively; (25) represents the constraints between the upper and lower level models; (25)-(29) represent the Lagrangian functions of the original problem with respect to the variables of the original problem. and θ it The gradient is 0; (30)-(39) are complementary constraints on the inequality equations of the lower-level original problem; α mt β mt , χ nt δ nt φ nt , γ lt κ lt , and ω ij，t The variables are the dual variables of the left and right sides of the inequalities in equations (19)-(23), respectively; then the bi-level model is transformed into a mixed integer linear model, as follows:

[0217] Considering that complementary relaxation constraints cannot be solved directly, the Big M method is further used for further processing, as follows:

[0218]

[0219]

[0220]

[0221]

[0222]

[0223]

[0224]

[0225]

[0226]

[0227]

[0228] In the formula, and All are auxiliary 0-1 integer variables. Furthermore, consider the presence of a nonlinear term λ in the objective function (24).t P t EA Furthermore, strong duality theory can be applied, and the lower-level objective function can be expressed as follows:

[0229]

[0230] Where, λ t P t EA It can be represented as:

[0231]

[0232] Therefore, the objective function (24) can be expressed as follows:

[0233]

[0234] The linear constraints are as follows:

[0235] (25)-(29) and (40)-(49) (53)

[0236] The above model is a typical mixed-integer linear programming problem, which can be solved directly using commercial solvers.

[0237] The solver is either the CPLEX solver or the GUROBI solver.

[0238] like Figure 10 The two-layer dispatching system shown includes consideration of electrolytic aluminum load participation in grid operation, comprising:

[0239] The first processing module 101 is used to construct an upper-level model of high-energy-consuming electrolytic aluminum; the upper-level model of high-energy-consuming electrolytic aluminum includes: a self-owned coal-fired power plant model in high-energy-consuming electrolytic aluminum, a high-energy-consuming electrolytic aluminum load model that takes into account the consumption of new energy sources, and a high-energy-consuming electrolytic aluminum load model that takes into account demand response technology.

[0240] The second processing module 102 is used to construct a lower-level model for the day-ahead market clearing of the power grid. The lower-level model for the day-ahead market clearing of the power grid takes social welfare maximization as the objective function and uses the power balance equation constraint, photovoltaic power output constraint, coal-fired power unit operation constraint, coal-fired power unit ramping constraint, active load constraint and line transmission capacity constraint as constraint conditions.

[0241] The third processing module 103 is used to construct a two-layer model based on the upper-layer model of high-energy-consuming electrolytic aluminum and the lower-layer model of the day-ahead market clearing of the power grid. The two-layer model takes the minimum cost of coal-fired power generation and photovoltaic power generation as the objective function. Then, the two-layer model is transformed into a mixed integer linear model and then solved.

[0242] The scheduling control module 104 is used to perform scheduling control based on the solution results.

[0243] The system provided in this embodiment of the invention is used to execute the above-described method embodiments. For specific processes and details, please refer to the above embodiments, which will not be repeated here.

[0244] Figure 11 This is a schematic diagram of the electronic device structure provided in an embodiment of the present invention, with reference to... Figure 11 The electronic device may include a processor 201, a communications interface 202, a memory 203, and a communications bus 204, wherein the processor 201, the communications interface 202, and the memory 203 communicate with each other through the communications bus 204. Processor 201 can call logic instructions in memory 203 to execute the following methods: Constructing an upper-level model for high-energy-consuming electrolytic aluminum; the upper-level model includes: a model of self-owned coal-fired power plants in high-energy-consuming electrolytic aluminum, a load model of high-energy-consuming electrolytic aluminum considering renewable energy consumption, and a load model of high-energy-consuming electrolytic aluminum considering demand response technology; Constructing a lower-level model for day-ahead market clearing; the lower-level model for day-ahead market clearing uses social welfare maximization as its objective function, and uses power balance equation constraints, photovoltaic output constraints, coal-fired power unit operation constraints, coal-fired power unit ramping constraints, active load constraints, and line transmission capacity constraints as constraints; Constructing a two-layer model based on the upper-level model for high-energy-consuming electrolytic aluminum and the lower-level model for day-ahead market clearing, the two-layer model using the minimum cost of coal-fired and photovoltaic power generation as its objective function; Then, converting the two-layer model into a mixed-integer linear model and solving it; Performing scheduling control based on the solution results.

[0245] Furthermore, the logical instructions in the aforementioned memory 203 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0246] On the other hand, embodiments of the present invention also provide a non-transitory computer-readable storage medium storing a computer program thereon. When executed by a processor, this computer program implements the two-layer scheduling method for considering the participation of electrolytic aluminum load in grid operation provided in the above embodiments. For example, it includes: constructing an upper-layer model for high-energy-consuming electrolytic aluminum; the upper-layer model for high-energy-consuming electrolytic aluminum includes: a model of self-owned coal-fired power plants in high-energy-consuming electrolytic aluminum, a high-energy-consuming electrolytic aluminum load model considering renewable energy consumption, and a high-energy-consuming electrolytic aluminum load model considering demand response technology; and constructing a lower-layer model for day-ahead market clearing in the grid. The model consists of a lower-level model for day-ahead market clearing of the power grid, with social welfare maximization as the objective function and constraints including power balance equation, photovoltaic output, coal-fired power unit operation, coal-fired power unit ramping, active load, and line transmission capacity. A two-layer model is constructed based on the upper-level model for high-energy-consuming electrolytic aluminum and the lower-level model for day-ahead market clearing of the power grid. This two-layer model has the objective function of minimizing the costs of coal-fired and photovoltaic power generation. The two-layer model is then transformed into a mixed-integer linear model and solved. Scheduling control is then implemented based on the solution results.

[0247] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0248] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0249] Application Examples

[0250] Taking the IEEE-39 transmission network node computational example (e.g.) Figure 2As shown in the figure, the electrolytic aluminum load is connected to node 22 to verify the effectiveness of the method proposed in this invention. The relevant system parameters are as follows: the maximum output power of the self-owned coal-fired power plant is 100MW, and the ramp-up capability is 50MW; the photovoltaic configuration capacity is 150MW, and the unit photovoltaic output per typical day is as follows: Figure 3 As shown, the maximum demand response power is 0.4 times the predicted electrical load at the current time. In the basic scenario, the unit generation cost of a self-owned coal-fired power plant is ¥0.12 / kWh, the unit cost of photovoltaic power generation is ¥0.26 / kWh, the unit cost of demand response is ¥0.02 / kWh, and the unit carbon emission cost is ¥150 / ton. The following four scenarios are set: 1) Electrolytic aluminum with a self-owned coal-fired power plant interacting with the grid; 2) Based on 1), considering the carbon emission cost of the coal-fired power plant; 3) Based on 2), further considering the electrolytic aluminum load including renewable energy generation; 4) Based on 3), further considering the impact of demand response.

[0251] (1) Economic comparison

[0252] Table 1 presents a comparison of economic efficiency under four scenarios. The table shows that in scenario S1, the cost of generating electricity from a self-supplied coal-fired power plant for electrolytic aluminum loads is 266.41 × 10⁻⁶. 3 The cost of purchasing electricity from the electricity market is ¥10.19 × 10 3 In this scenario, the carbon emissions from a self-owned coal-fired power plant are 2.33 tons. In scenario S2, after considering carbon emission costs, the power generation cost of a self-owned coal-fired power plant is reduced to 114.40 × 10⁻⁶ tons. 3 The cost of purchasing electricity from the electricity market increased to 337.73 × 10⁻⁶ yuan. 3 The carbon emission cost in this scenario is ¥150.14 × 10 3 The higher carbon price led to a reduction in carbon emissions from the electrolytic aluminum load to 1.0 ton. In scenario S3, by considering the integration of photovoltaics into the electrolytic aluminum load, the cost of generating electricity from the self-owned coal-fired power plant was further reduced, while the cost of purchasing electricity from the electricity market increased slightly compared to S2; in addition, carbon emission costs decreased compared to S2, resulting in a reduction of carbon emissions of 0.81 tons. In scenario S4, after further considering demand response, the cost of generating electricity from the self-owned coal-fired power plant, carbon emission costs, and carbon emissions decreased further, while the cost of purchasing electricity from the electricity market continued to increase. Furthermore, considering demand response increased the cost of photovoltaic power generation, indicating that demand response technology can significantly reduce the output of self-owned coal-fired power generation while increasing photovoltaic output.

[0253] Table 1. Economic Comparison in Four Scenarios

[0254]

[0255] (2) Analysis of typical daily operation scenarios

[0256] Figures 4 to 6The power balance of the electrolytic aluminum load under S1 and S2 is given. From Figure 4 It can be seen that the electrolytic aluminum load, including its own coal-fired power plant, prioritizes power supply from its own coal-fired power plant, only considering purchasing a small amount of electricity from the electricity market during the peak electricity demand period from 17:00 to 22:00. Based on S1, after considering the carbon emission costs of the self-owned coal-fired power plant, the amount of electricity supplied by the electrolytic aluminum load from its own coal-fired power plant is significantly reduced, and it mainly operates between 4:00-11:00 and 15:00-20:00, while purchasing electricity from the upstream grid during other times. Since the maximum output of photovoltaic power is concentrated at noon, and wind power output is mainly concentrated between 20:00 and 4:00 the next morning, the electrolytic aluminum load mainly purchases electricity from the electricity market during these periods.

[0257] Figure 6 and Figure 7 The power balance of the electrolytic aluminum load under S3 and S4 conditions is given respectively. From Figure 6 It can be seen that after considering the allocation of new energy sources to the electrolytic aluminum load, the electrolytic aluminum load uses its own new energy sources for power supply during the periods of 6:00-10:00 and 16:00-17:00. It should be noted that the new energy power generation cost of the electrolytic aluminum load in this invention adopts the levelized cost of power generation over the entire life cycle of the equipment, so its cost is relatively high, resulting in the electrolytic aluminum load not maximizing the power generation of new energy sources. Figure 7 The power balance of the electrolytic aluminum load after further application of demand response technology is presented. This is achieved through... Figure 6 The comparison revealed that the electrolytic aluminum load primarily shifts its electricity demand to other times during the 0:00-4:00 and 14:00-15:00 periods. This leads to a further reduction in the power generation from the electrolytic aluminum load's own coal-fired power plants. Photovoltaic output further increases, and electricity purchases from the electricity market also increase slightly. Therefore, further consideration of demand response technologies can improve the utilization level of photovoltaic power generation and reduce the carbon emissions from the electrolytic aluminum load.

[0258] (3) Sensitivity analysis of key parameters

[0259] To study the impact of carbon prices and the cost per kilowatt-hour of photovoltaic power generation on the optimal operation of electrolytic aluminum load under market conditions, Figure 8 The graph shows the day-ahead operating costs and carbon emissions for electrolytic aluminum load as the carbon price changes from 0 to 300 RMB / ton. As can be seen from the graph, the electrolytic aluminum load gradually increases with the increase in carbon price. When the carbon price is greater than 200 RMB / ton, the day-ahead operating costs remain unchanged. This is mainly because the higher carbon price causes the electrolytic aluminum load to no longer rely on its own coal-fired power plant for power generation, but instead uses renewable energy generation and purchases electricity from the electricity market. The carbon emissions data shows that when the carbon price is below 100 RMB / ton, the carbon emissions remain unchanged. This is attributed to the fact that the lower carbon price makes the power generation cost of its own coal-fired power plant price competitive, and the electrolytic aluminum load still prioritizes the use of its own coal-fired power plant for power generation. Figure 9 The report presents the day-ahead operating costs and carbon emissions of electrolytic aluminum loads as the photovoltaic (PV) electricity price changes from ¥0.2 to ¥0.3 / kWh. It can be seen that both day-ahead operating costs and carbon emissions increase with the increase in PV electricity price, but their trends differ significantly. Day-ahead operating costs increase rapidly when the PV electricity price changes from ¥0.2 to ¥0.275 / kWh, and remain relatively stable above ¥0.275 / kWh; while carbon emissions increase slightly when the PV electricity price changes from ¥0.2 to ¥0.275 / kWh, and increase dramatically above ¥0.275 / kWh.

[0260] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of this invention is defined by the appended claims and their equivalents.

Claims

1. A two-layer dispatching method considering the participation of electrolytic aluminum load in power grid operation, characterized in that, Includes the following steps: S1. Construct an upper-level model for high-energy-consuming electrolytic aluminum; the upper-level model for high-energy-consuming electrolytic aluminum includes: a self-owned coal-fired power plant model for high-energy-consuming electrolytic aluminum, a high-energy-consuming electrolytic aluminum load model that takes into account the consumption of new energy sources, and a high-energy-consuming electrolytic aluminum load model that takes into account demand response technology. The self-owned coal-fired power plant model in high-energy-consuming electrolytic aluminum uses the power generation cost of the self-owned power plant as the objective function and the output power constraint and ramp-up power constraint of the self-owned power plant as the constraint conditions. The high-energy-consuming electrolytic aluminum load model that takes into account the consumption of new energy sources uses the distributed generation price as the objective function and the photovoltaic output constraint as the constraint condition. The high-energy-consuming electrolytic aluminum load model that takes demand response technology into account takes the compensation cost of demand response as the objective function and non-transferable load constraints, load transfer-in constraints, and load transfer-out constraints as constraints. S2. Construct a lower-level model for day-ahead market clearing in the power grid; The aforementioned lower-level model for day-ahead market clearing in the power grid takes social welfare maximization as its objective function and uses the power balance equation, photovoltaic output constraints, coal-fired power unit operation constraints, coal-fired power unit ramping constraints, active load constraints, and line transmission capacity constraints as constraints. S3. Construct a two-layer model based on the upper-layer model of high-energy-consuming electrolytic aluminum and the lower-layer model of the day-ahead market clearing of the power grid. The objective function of the two-layer model is to minimize the cost of coal-fired power generation and photovoltaic power generation. Then, the two-layer model is transformed into a mixed integer linear model and solved. The scheduling is carried out based on the solution results. The construction of the upper-layer model for high-energy-carrying electrolytic aluminum in step S1 is as follows: S11. Construct a model of a self-owned coal-fired power plant in high-energy-consuming electrolytic aluminum, as shown in equations (1)-(4): Target cost of power generation from self-owned coal-fired power plants C coal for: （1） in, For self-owned coal-fired power plants t Output power at any time c coal The unit power generation cost of a self-owned coal-fired power plant; Self-owned coal-fired power plants must meet the following operational constraints: （2） （3） in, , and These are the maximum output power, minimum output power, and maximum ramp power of a self-owned coal-fired power plant, respectively. Carbon emission costs are calculated as follows: （4） In the formula, and These represent the unit carbon price and the carbon emission factor of coal, respectively. S12. Construct a high-energy-consuming electrolytic aluminum load model that takes into account the consumption of new energy sources, as shown in equations (5)-(9): First, it is necessary to ensure that the output of distributed photovoltaic power is at a certain confidence level. η Within: （5） in, To predict photovoltaic power output; To contribute to the actual power of photovoltaics; pr (·) indicates the probability that the actual photovoltaic output is less than the predicted photovoltaic output; Assuming the actual output of photovoltaic power Follows Gaussian distribution ,but: （6） in Represents the cumulative distribution function of photovoltaic power. For variance; Thus, the standard form is obtained: （7） (·) represents the standard normal distribution The cumulative distribution function; At confidence level η The photovoltaic output constraints are as follows: （8） in, It follows a standard normal distribution. The inverse cumulative distribution function; The price of distributed generation is: （9） in, lcoe The cost per kilowatt-hour of photovoltaic power generation; S13. Construct a high-energy-consuming electrolytic aluminum load model that incorporates demand response technology, as shown in equations (10)-(16): Assume that the high-energy-consuming electrolytic aluminum load includes non-transferable load and transferable load; the non-transferable load must be satisfied within a scheduling cycle, while the transferable load is transferred in and out within a scheduling cycle based on actual demand, as follows: （10） （11） （12） （13） Wherein, formula (10) represents the electrolytic aluminum load at each time t. From non-transferable loads , load transfer and transferred load The formulas (11) and (12) constrain the maximum inbound load and maximum outbound load at each time point, respectively, where α is the maximum transferable load coefficient. Formula (13) ensures that the total inbound load within a scheduling cycle equals the total outbound load. T For one scheduling cycle, the compensation cost for demand response is expressed as follows: （14） In the formula, c pe Indicates the unit compensation cost for demand response; Since the electrolytic aluminum load can purchase electricity from the electricity market, the power balance of the electrolytic aluminum load itself is: （15） In the formula, Indicates the load of electrolytic aluminum at t Electricity purchased from the electricity market at all times; Total operating cost per day of electrolytic aluminum load C : （16） in, The cost of purchasing electricity from the electricity market for the electrolytic aluminum load. Indicates in t Marginal electricity price at any given moment; The underlying model for the day-ahead market clearing of the power grid in step S2 is as follows: The electricity market is cleared by an independent power system operator with the goal of maximizing social welfare, i.e. minimizing the day-ahead operating costs of the power system, as shown in equation (17). （17） （18） （19） （20） （21） （22） （23） In the formula, the subscript t Indicates the scheduling time, subscript m Indicates photovoltaic power generation, subscript n Indicates coal-fired power unit, subscript l Indicates electrical load, subscript i Indicates the busbar; Indicates a photovoltaic power generation collection, Indicates a coal-fired power generation complex. Represents the set of electrical loads. Represents the set of busbars; Indicates photovoltaic m exist t Output power at any time Indicates coal-fired power unit n exist t Output power at any time Indicates user l exist t The electrical load demand at any given time; Indicates coal-fired power unit n The unit cost of electricity generation Indicates user l The unit cost of electricity; Indicates the first i Whether an electrolytic aluminum load is connected to the busbar; if an electrolytic aluminum load is connected, take 1; otherwise, take 0. b ij Indicates connecting busbar i and j Line susceptance, θ it Indicates the first i busbars in t Voltage phase angle at any given moment; Indicates that the j-th busbar is in t Voltage phase angle at any given moment; This represents the unit power generation cost of photovoltaic (m). Indicates the scheduling time period; Equation (17) minimizes the day-ahead operating cost of the power system; Equation (18) is the power balance equation; Equation (19) is the photovoltaic output constraint, where Indicates photovoltaic m exist t The unit output electrical power at any given time, For photovoltaic m Maximum configured capacity; Equation (20) represents the operating constraints of the coal-fired power unit, where and The first n The maximum and minimum output power of the unit; Equation (21) is the ramp-up constraint of the coal-fired power unit, where For the first n The maximum ramping power of the generator set; Equation (22) is the active power load constraint, where For users l The maximum electrical load; Equation (23) is the line transmission capacity constraint, where Indicates the line ij The power transmission capacity.

2. The two-layer dispatching method for considering the participation of electrolytic aluminum load in grid operation according to claim 1, characterized in that: Set as 5%.

3. The two-layer dispatching method for considering the participation of electrolytic aluminum load in grid operation according to claim 1, characterized in that: The specific method for step S3 is as follows: A two-layer model is constructed based on the upper-layer model of high-energy-consuming electrolytic aluminum and the lower-layer model of day-ahead market clearing in the power grid. The two-layer model is as follows: （24） The constraints are as follows: Equations (2), (3), (8)-(15), (18)-(23) (25) （26） （27） （28） （29） （30） （31） （32） （33） （34） （35） （36） （37） （38） （39） in, This represents the marginal electricity price at time t; Indicates photovoltaic power generation m The marginal electricity price at time t, Indicates coal-fired power unit n The boundary electricity price at time t, Indicates electrical load l The marginal electricity price at time t; , Let i and j represent the marginal electricity prices of bus i and j at time t, respectively; (25) represents the constraints between the upper and lower level models; (25)-(29) represent the Lagrangian functions of the original problem with respect to the variables of the original problem. , , and The gradient is 0; (30)-(39) are complementary constraints of the inequality equations of the lower-level original problem; , , , , , , , , and The variables are the dual variables of the left and right sides of the inequalities in equations (19)-(23), respectively; then the bi-level model is transformed into a mixed integer linear model, as follows: The Big M method is used to process the two-layer model, as follows: （40） （41） （42） （43） （44） （45） （46） （47） （48） （49） In the formula, , , , , , , , , and All are auxiliary 0-1 integer variables; M For a specific value; Consider the presence of a nonlinear term in the objective function (24) Using strong duality theory, the objective function of the lower-level model is expressed as follows: (50) in, It can be represented as: (51) Therefore, the objective function (24) is expressed as (52) The linear constraints are as follows: (25)-(29) and (40)-(49) (53) Equations (25)-(29), (40)-(49) and (52) are typical mixed integer linear programming problems. They are then solved, and scheduling is carried out based on the solution results.

4. The two-layer dispatching method for considering the participation of electrolytic aluminum load in grid operation according to claim 3, characterized in that: The solution is obtained using the CPLEX solver or the GUROBI solver.

5. A two-layer dispatching system considering the participation of electrolytic aluminum load in grid operation, executing the two-layer dispatching method considering the participation of electrolytic aluminum load in grid operation as described in any one of claims 1 to 4, characterized in that, include: The first processing module is used to construct the upper-layer model of high-energy-consuming electrolytic aluminum. The aforementioned upper-level model for high-energy-consuming electrolytic aluminum includes: a model of self-owned coal-fired power plants in high-energy-consuming electrolytic aluminum, a load model of high-energy-consuming electrolytic aluminum that takes into account the consumption of new energy sources, and a load model of high-energy-consuming electrolytic aluminum that takes into account demand response technology; The second processing module is used to construct the lower-level model of the day-ahead market clearing of the power grid; the lower-level model of the day-ahead market clearing of the power grid takes the maximization of social welfare as the objective function and uses the power balance equation constraint, photovoltaic power output constraint, coal-fired power unit operation constraint, coal-fired power unit ramping constraint, active load constraint and line transmission capacity constraint as the constraint conditions. The third processing module is used to construct a two-layer model based on the upper-layer model of high-energy-consuming electrolytic aluminum and the lower-layer model of the day-ahead market clearing of the power grid. The objective function of the two-layer model is to minimize the cost of coal-fired power generation and photovoltaic power generation. Then, the two-layer model is transformed into a mixed integer linear model and then solved. The scheduling and control module is used to perform scheduling and control based on the solution results.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the two-layer scheduling method for taking into account the participation of electrolytic aluminum load in grid operation as described in any one of claims 1 to 4.

7. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the steps of the two-level dispatching method as described in any one of claims 1 to 4, which takes into account the participation of electrolytic aluminum load in grid operation.

Images