Steel enterprise oxygen system optimization scheduling model building and solving method

Abstract

The invention provides an iron and steel enterprise oxygen system optimization scheduling model construction and solving method, relates to the technical field of industrial information, establishes an iron and steel enterprise oxygen system optimization scheduling model, and considers oxygen demand constraint, unit production process constraint, minimum startup and shutdown time constraint and climbing constraint. The model decides the production operation state and the oxygen yield of each unit in each time period so as to reduce the total operation cost of an oxygen system. According to the characteristics of the model, a Lagrange relaxation algorithm is designed for solving. According to the method, material balance constraint is relaxed into a target function through a Lagrange multiplier, and an original problem is decomposed into an air separation unit sub-problem, a gasifier unit sub-problem and a storage system sub-problem. In order to accelerate the sub-problem solving speed, an optimal property acceleration strategy is provided, a subgradient algorithm is designed, and a multiplier updating strategy is provided. Oxygen diffusion can be remarkably reduced, the production and operation cost of an oxygen system is reduced, and safe and stable operation of the oxygen system is guaranteed.

Description

Technical Field

[0001] The invention relates to the field of industrial information technology, and in particular to a construction and solution method for an optimal scheduling model of an oxygen system in a steel enterprise. Background Art

[0002] Oxygen is an essential energy medium in the steel production process, and is mainly used in blast furnace ironmaking, converter steelmaking and other production processes. In order to ensure the smooth progress of the steel production process, steel companies generally configure oxygen systems to ensure the oxygen demand for ironmaking, steelmaking and other processes. The preparation of oxygen consumes a lot of electricity, and the power consumption of the oxygen system accounts for about 20% of the total power consumption of the whole plant, which belongs to the first category of power load.

[0003] The oxygen emission rate of my country's steel enterprises is generally high, and most enterprises have an oxygen emission rate of 2% to 8%. The main reason for the high oxygen emission rate is the imbalance between the intermittent oxygen use of the converter and the continuous production of the air separation unit. When the blowing rhythm is accelerated, the supply is less than the demand, and the oxygen supply is insufficient; while when the converter fails, is under maintenance, and the molten iron supply is insufficient, the supply exceeds the demand, and oxygen needs to be released. At present, the relevant research on the optimization and scheduling of oxygen systems in steel enterprises at home and abroad and patents is as follows:

[0004] Jia Tianyun et al. (An oxygen optimization scheduling system and method for steel enterprises, application number: 201510997648.5) used a series of equipment working condition modules to solve the problem of fine oxygen scheduling under complex working condition combinations, and provided an oxygen scheduling system that is easy to interact with, thereby reducing oxygen costs, reducing the frequency of oxygen production equipment adjustment, and improving production stability. However, its scheduling strategy is limited to the equipment level, lacks comprehensive consideration of the constraints of the entire system, and it is difficult to obtain the global optimal scheduling solution;

[0005] Zhang Ziyang et al. (A method for optimizing the allocation of oxygen production systems in steel enterprises, application number 201811181626.1) established a two-stage model and obtained the global optimal solution or the bound of the optimal solution of the allocation problem by solving the model. The allocation plan is planned through the optimal solution to improve the allocation efficiency. At the same time, within the prescribed range of the oxygen pipeline pressure, the oxygen utilization rate is improved and energy is saved. In the constraint processing of the oxygen pipeline network, this method does not fully consider the matching between the operating specifications and the actual process, which easily causes the optimization results to be inconsistent with the actual needs, affecting the scheduling effect.

[0006] Chen Guiling et al. (Research on Energy Saving Optimization of Air Separation Oxygen Production in Steel Enterprises [D]. Shandong Jianzhu University, 2015.) established a mathematical model with the goal of minimizing the oxygen emission rate and maximizing the economic benefits of the oxygen system, and solved the model in combination with an improved multi-objective particle swarm algorithm. Numerical experiments have shown the feasibility of this optimization method. However, the model lacks comprehensive consideration of equipment operation safety and energy utilization efficiency, and it is difficult to meet the comprehensive needs under complex working conditions;

[0007] Zhang Qianqian et al. (Research on Two-stage Balanced Scheduling of Oxygen System in Steel Enterprises [D]. Dalian University of Technology, 2016.) considered both oxygen and nitrogen products of oxygen generators at the same time, and established and solved the mathematical model with the goal of minimizing the oxygen and nitrogen emission rate. In the numerical experiment part, the superiority of the scheduling scheme was demonstrated under four typical oxygen and nitrogen supply and demand imbalances. This study was limited to the static optimization scheduling of two products, and did not take into account the dynamic supply and demand balance of the oxygen system, making it difficult to cope with emergencies caused by changes in demand during actual production.

[0008] The above studies mostly focus on the optimization of specific scenarios or goals, lacking a global consideration of the interactions between oxygen preparation, storage, transportation and use. In addition, equipment operating limitations and process specifications have not been fully incorporated into the scheduling model, affecting the feasibility of the scheduling plan and the feasibility of the scheduling decision. Summary of the invention

[0009] In view of the shortcomings of the prior art, the purpose of the present invention is to propose a steel enterprise oxygen system optimization scheduling model construction and solution method, including:

[0010] Step 1: Establish an optimal scheduling model for the oxygen system of a steel enterprise, and consider actual production constraints. The optimal scheduling model includes actual production constraints and an objective function for optimizing the oxygen system. The actual production constraints include upper and lower limits of oxygen production for the air separation unit and the gasifier unit, ramp constraints for the air separation unit and the gasifier unit, on / off constraints for the air separation unit, storage system-related constraints, and material balance constraints for the oxygen system.

[0011] Step 2: Use the Lagrange multiplier vector {λ t},{μ t}, relax the oxygen system material balance constraints into the objective function, generate a Lagrangian relaxation problem, and record it as (LR), construct a Lagrangian dual problem, and record it as (LD), the Lagrangian dual problem is the maximum value of the Lagrangian relaxation problem; for a given multiplier vector {λ t},{μ t}, decompose the Lagrangian relaxation problem (LR) into air separation unit subproblem, gasifier unit subproblem and storage system subproblem;

[0012] Step 3: Calculate the solutions of the air separation unit subproblem, the gasifier unit subproblem and the storage system subproblem, and then obtain a scheduling schedule, which includes the start and stop times of the air separation unit and the gasifier unit, the gaseous oxygen production and liquid oxygen production of the air separation unit, and the gaseous oxygen production of the gasifier unit;

[0013] Step 3.1: Set the initial number of iterations, λt , μ t The initial value and maximum number of iterations are used to initialize the dual function value of the Lagrangian dual problem, and the initial number of iterations is used as the current number of iterations. t The initial value of λ t The current value of μ t The initial value of μ t Current value: the dual function value of the initial Lagrangian dual problem is used as the current dual function value of the Lagrangian dual problem;

[0014] Step 3.2: Solve the air separation unit subproblem, gasifier unit subproblem and storage system subproblem;

[0015] Step 3.3: The solutions of the air separation unit subproblem, the gasifier unit subproblem and the storage system subproblem are used as the solutions of the Lagrangian relaxation problem, and then the value of the Lagrangian relaxation problem is calculated. Among the values ​​of the relaxation problem at the current iteration number and the current dual function value, the one with the largest value is obtained as the new current dual function value, and the new current dual function value is used as the lower bound of the objective function;

[0016] Step 3.4: When the scheduling time is set, the solution of the Lagrangian relaxation problem is used as the initial solution, and the heuristic algorithm is used to transform the initial solution into a feasible solution of the optimization scheduling model to obtain the start-stop sequence of the air separation unit and the gasifier unit;

[0017] Step 3.5: According to the start-stop sequence, use GUROBI to build a model and solve it to obtain a scheduling schedule under the current number of iterations;

[0018] Step 3.6: Calculate the value of the objective function based on the feasible solution and the objective function, and use the value of the objective function as the upper bound of the objective function;

[0019] Step 3.7: Calculate the duality gap GAP of the current iteration number k, and determine whether the duality gap GAP is less than the preset threshold. If the duality gap GAP is less than the set value, output the scheduling plan under the current iteration number; if the duality gap GAP is not less than the set value, execute step 3.8;

[0020] Step 3.8: Determine whether the current number of iterations is less than or equal to the maximum number of iterations. If the current number of iterations is greater than the maximum number of iterations, output the scheduling plan under the current number of iterations; if the current number of iterations is less than or equal to the maximum number of iterations, add one to the current number of iterations and execute step 3.9;

[0021] Step 3.9: Use the subgradient algorithm to update the Lagrange multiplier and return to step 3.2.

[0022] Optionally, step 1 specifically includes:

[0023] Step 1.1: According to the operation status of the air separation unit and gasifier unit in the oxygen system of the steel enterprise, establish the upper and lower limits of oxygen production of the air separation unit and gasifier unit, including:

[0024] (1) Upper and lower limits of gaseous oxygen production of air separation units:

[0025]

[0026] Where i represents the air separation unit, I represents the set of air separation units; t represents the scheduling time period, T represents the set of discrete time points in the scheduling outlook period; u i,t Indicates the start and stop status of air separation unit i during period t, 0 means shutdown, 1 means startup; pg i,t represents the gaseous oxygen production of air separation unit i during period t; represents the lower limit of the gaseous oxygen production of air separation unit i, represents the upper limit of the gaseous oxygen production of air separation unit i;

[0027] (2) Upper and lower limits of liquid oxygen production of air separation units:

[0028]

[0029] Among them, pl i,t represents the liquid oxygen production of air separation unit i during period t; , represents the lower limit of liquid oxygen production of air separation unit i, represents the upper limit of liquid oxygen production of air separation unit i;

[0030] (3) Upper and lower limits of gaseous oxygen production of gasifier units:

[0031]

[0032] Wherein, j represents the gasifier unit, J represents the set of gasifier units; u j,t Indicates the start and stop status of the vaporizer unit j during period t, 0 means shutdown, 1 means startup, pg j,t represents the gaseous oxygen production of gasifier unit j during period t; , represents the lower limit of gaseous oxygen production of gasifier unit j; represents the upper limit of the gaseous oxygen production of gasifier unit j;

[0033] Step 1.2: According to the oxygen production change rate of the air separation unit and the gasifier unit during operation, establish the ramp constraints of the air separation unit and the gasifier unit, including:

[0034] (1) Gaseous oxygen production ramp-up constraints of air separation units:

[0035]

[0036] in, It represents the upper limit of the oxygen production rate of change when air separation unit i is running. Indicates the upper limit of the oxygen production rate change when air separation unit i is started. Indicates the upper limit of the oxygen production rate when air separation unit i is shut down, pg i,t-1 represents the gaseous oxygen production of air separation unit i during period t-1, u i,t-1 Indicates the start and stop status of air separation unit i during period t-1;

[0037] (2) Liquid oxygen production ramp-up constraints of air separation units:

[0038]

[0039] in, It represents the upper limit of the rate of change of liquid oxygen production when air separation unit i is in operation; It represents the upper limit of the rate of change of liquid oxygen production when air separation unit i is started; Indicates the upper limit of the liquid oxygen production rate when air separation unit i is shut down, pl i,t-1 represents the liquid oxygen production of air separation unit i during period t-1;

[0040] (3) Gasifier unit gaseous oxygen production ramp constraints:

[0041]

[0042] in, It represents the upper limit of the rate of change of gaseous oxygen production when gasifier unit j is running; It represents the upper limit of the rate of change of gaseous oxygen production when gasifier unit j is started; pg represents the upper limit of the rate of change of gaseous oxygen production when gasifier unit j is shut down; j,t-1 represents the gaseous oxygen production of gasifier unit j during period t-1, u j,t-1 Indicates the start and stop status of gasifier unit j during period t-1;

[0043] Step 1.3: According to the on / off clearance of the air separation unit during actual operation, the on / off constraints of the air separation unit are established as follows:

[0044]

[0045] in, It represents the continuous operation time of air separation unit i before time period t, T represents the continuous shutdown time of air separation unit i before time period t; i on Indicates the minimum startup time of air separation unit i, T i off Indicates the minimum shutdown time of air separation unit i;

[0046] Step 1.4: Establish storage system related constraints based on the storage characteristics of liquid oxygen and gaseous oxygen;

[0047] (1) Establish constraints related to gaseous oxygen storage;

[0048]

[0049] Among them, sg t represents the storage amount of gaseous oxygen in period t, sg min Indicates the lower limit of gaseous oxygen inventory, sg max Indicates the upper limit of gaseous oxygen inventory;

[0050] (2) Establish constraints related to liquid oxygen storage;

[0051]

[0052] Among them, sl t represents the storage volume of liquid oxygen during period t, sl min Indicates the lower limit of liquid oxygen inventory, sl max Indicates the upper limit of liquid oxygen inventory;

[0053] Step 1.5: Establish oxygen system material balance constraints for liquid oxygen and gaseous oxygen;

[0054] (1) Establishing gaseous oxygen material balance constraints:

[0055]

[0056] in, represents the demand for gaseous oxygen during period t, sg t-1 Indicates the storage amount of gaseous oxygen in period t-1;

[0057] (2) Establish liquid oxygen material balance constraints:

[0058]

[0059] in, Indicates the conversion factor of gaseous oxygen into liquid oxygen, sl t-1 represents the storage amount of liquid oxygen during the period t-1, represents the demand for liquid oxygen during period t;

[0060] Step 1.6: Optimize the objective function of the oxygen system and establish a minimization objective function, wherein the objective function is composed of the operation and start-up and shutdown costs of the air separation unit, the operation and start-up and shutdown costs of the gasifier unit, and the inventory cost of the oxygen system;

[0061] (1) Operation and start-up and shutdown costs of air separation units:

[0062] C i,t =(α0 i pg i,t +α1 i pl i,t )·u i,t +S i ·u i,t (1-u i,t-1 ); (12)

[0063] Among them, C i,t represents the cost of air separation unit i in period t, S i represents the startup cost of air separation unit i, α0 i ,α1 i represents the operating cost coefficient of air separation unit i;

[0064] (2) Gasifier unit operation and start-up and shutdown costs:

[0065] C j,t =β j ·pg j,t ·u j,t +S j ·u j,t (1-u j,t-1 ); (13)

[0066] Among them, C j,t represents the cost of gasifier unit j in period t, S j represents the startup cost of gasifier unit j, β j represents the operating cost coefficient of gasifier unit j;

[0067] (3) Establishing oxygen system inventory costs;

[0068]

[0069] Among them, C t represents the inventory cost of the oxygen system during period t, including the inventory cost of gaseous oxygen and the inventory cost of liquid oxygen, γ g represents the inventory cost coefficient of gaseous oxygen, γ l represents the inventory cost factor for liquid oxygen;

[0070] Based on this, the objective function can be summarized as follows:

[0071]

[0072] Based on this, the upper and lower limit constraints of oxygen production of the air separation unit and the gasifier unit, the ramping constraints of the air separation unit and the gasifier unit, the on-off constraints of the air separation unit, the storage system related constraints, the material balance constraints of the oxygen system and the objective function of optimizing the oxygen system were obtained.

[0073] Optionally, the Lagrangian relaxation problem is specifically expressed by the following formula:

[0074] (LR)L(λ,μ)=minZ LR ; (16)

[0075] Among them, Z LR It can be expressed by the following formula:

[0076]

[0077] Among them, the parameters in formula (17) satisfy the constraints of formula (1)-formula (9);

[0078] The Lagrangian dual problem is expressed by the following formula:

[0079] (LD)maxL(λ,μ); (18)

[0080] Based on this, we obtain the Lagrangian relaxation problem, namely formula (17), and the Lagrangian dual problem, namely formula (18).

[0081] Optionally, the air distribution unit subproblem, the gasifier unit subproblem and the storage system subproblem in step 2 are expressed by the following formula:

[0082] (1) Air separation unit sub-problem

[0083]

[0084] Wherein, the parameters in formula (19) satisfy the constraints of formula (1), formula (2), formula (4), formula (5) and formula (7);

[0085] (2) Gasifier unit sub-problem

[0086]

[0087] Among them, the parameters in formula (20) satisfy the constraints of formula (3) and formula (6);

[0088] (3) Storage system sub-issue

[0089]

[0090] Among them, the parameters in formula (21) satisfy the constraints of formula (8) and formula (9).

[0091] Optionally, step 3.2 specifically includes:

[0092] Step 3.2.1: Solve equation (19) for the air separation unit subproblem, where α0 i ,α1 i and S i The value of λ is known. t is t The current value of μ t μ t Current value, solve the minimum value of formula (19), and get its corresponding pg i,t The value of pl i,t The value of u i,t The value of u i,t-1 The value of pg i,t The value of pl i,t The value of u i,t The value of u i,t-1 The value of is taken as the solution to the air separation unit subproblem, that is, the solution to formula (19);

[0093] Step 3.2.2: Solve equation (20) for the gasifier unit subproblem, where β j , and S j The value of λ is known. t is t The current value of μ t μ t Current value, solve the minimum value of formula (20), and get its corresponding pg j,t The value of u j,t The value of u j,t-1 The value of pg j,t The value of u j,t The value of u j,t-1 The value of is taken as the solution to the gasifier unit subproblem, i.e., the solution to formula (20);

[0094] Step 3.2.3: Solve equation (21) for the storage system subproblem, where λ t is t The current value of μ t μ t Current value, solve the minimum value of formula (21), and get its corresponding C t The value of sg t The value of sg t-1 The value of sg t The value and sl t-1The value of C t The value of sg t The value of sg t-1 The value of sg t The value and sl t-1 The numerical value of is taken as the solution to the storage system subproblem, that is, the solution to formula (21).

[0095] Optionally, step 3.4 specifically includes:

[0096] Step 3.4.1: For each time period, the solution of the Lagrangian relaxation problem is used as the initial solution, and the initial solution is substituted into the formula on the left side of the inequality sign of formula (10) to obtain the total output of gaseous oxygen, and determine the magnitude of the total output of gaseous oxygen and the demand for gaseous oxygen. Specifically, when the total output of gaseous oxygen is greater than the demand for gaseous oxygen, the gasifier units are sorted from high to low according to the operating cost to obtain the first sequence, and the gasifier units that are shut down in sequence according to the sequence numbers from small to large in the first sequence are sorted from high to low according to the operating cost to satisfy the formula ( 7) sort the air separation units with the minimum startup time constraint in the first sequence to obtain a second sequence, and at the same time, shut down the air separation units in the second sequence in sequence from small to large, and store the gaseous oxygen that exceeds the demand for gaseous oxygen until it is released; when the total output of gaseous oxygen is less than the demand for gaseous oxygen, start the gasifier units in sequence in sequence from large to small in the first sequence, and at the same time, start the air separation units in the second sequence in sequence in sequence from large to small, until the total output of oxygen is greater than or equal to the demand for gaseous oxygen;

[0097] Step 3.4.2: For each time period, the solution of the Lagrangian relaxation problem is used as the initial solution, and the initial solution is substituted into the formula on the left side of the inequality sign of formula (11) to obtain the total output of liquid oxygen, and determine the total output of liquid oxygen and the demand for liquid oxygen. Specifically, when the total output of liquid oxygen is greater than the demand for liquid oxygen, the liquid oxygen in excess of the demand for liquid oxygen is stored; when the total output of liquid oxygen is less than the demand for liquid oxygen, the output of gaseous oxygen of the operating air separation unit is increased, and then the output of liquid oxygen is increased until the output of liquid oxygen is equal to or less than the demand for liquid oxygen. The total oxygen output is greater than the demand for liquid oxygen, or the gaseous oxygen output of all operating air separation units reaches the upper limit. When the gaseous oxygen output of all operating air separation units reaches the upper limit and the total output of liquid oxygen is less than the demand for liquid oxygen, the air separation units that meet the minimum shutdown time constraint in formula (7) are sorted from low to high according to the operating cost to obtain a third sequence. The air separation units are turned on according to the sequence numbers in the third sequence from small to large, and the gaseous oxygen output of the turned on air separation units is increased at the same time until the liquid oxygen output is greater than or equal to the demand for liquid oxygen.

[0098] Step 3.4.3: Based on the adjustments to the start and stop of the air separation unit and the gasifier unit in steps 3.4.1 and 3.4.2, determine the start and stop sequence of the air separation unit and the gasifier unit. Based on the parameters involved in steps 3.4.1 and 3.4.2, update the initial solution to obtain a feasible solution.

[0099] Optionally, the duality gap GAP of the current iteration number k is calculated in step 3.7, which is specifically implemented by the following formula:

[0100]

[0101] Among them, L * is the minimum objective function value up to the current iteration number k, L k is the current dual function value for the current iteration number k.

[0102] Optionally, in step 3.9, a subgradient algorithm is used to update the Lagrange multiplier, which is specifically implemented by the following formula:

[0103] μ k+1 =μ k +t k g k (twenty three)

[0104]

[0105] Among them, μ k+1 is the Lagrange multiplier at iteration number k+1, μ k is the Lagrange multiplier at the iteration number k, t k is the step size of the current iteration number k, g k is the subgradient of the current iteration number k, and γ is the step size coefficient.

[0106] The beneficial effects of adopting the above technical solution are:

[0107] The present invention provides a method for constructing and solving an optimization scheduling model for an oxygen system in a steel enterprise. The method is highly practical, reduces actual production problems to unit combination problems, takes into account a variety of production constraints involved in actual applications, and can effectively obtain deployment results. The application of this method can effectively improve the utilization rate of the oxygen system in a steel enterprise, reduce the emission of oxygen, and reduce the total cost of the production and operation of the oxygen system. At the same time, it can meet the requirements of user needs, safety and stability of pipeline pressure, and economic benefits, and ensure the economic and stable operation of the oxygen system while saving energy and reducing emissions. BRIEF DESCRIPTION OF THE DRAWINGS

[0108] Figure 1A schematic diagram of a flow chart of a method for constructing and solving an optimal scheduling model for an oxygen system of a steel enterprise in an embodiment of the present invention;

[0109] Figure 2 The figure is a schematic diagram of the process of obtaining the scheduling plan table in the embodiment of the present invention. DETAILED DESCRIPTION

[0110] The specific implementation of the present invention is further described in detail below in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0111] In view of the problems existing in the prior art, the present invention provides a steel enterprise oxygen system optimization scheduling model construction and solution method, so as to achieve the stability, reliability and accuracy of the steel enterprise oxygen system, and consider the various constraints in the actual operation of the steel enterprise, to provide support for energy management personnel to formulate oxygen scheduling plans, and to minimize the cost of operation while meeting the oxygen demand of downstream users, which can help enterprises save energy and reduce consumption and reduce emissions. Figure 1 , which may include the following steps:

[0112] Step 1: Establish an optimal scheduling model for the oxygen system of a steel enterprise, and consider actual production constraints. The optimal scheduling model includes actual production constraints and an objective function for optimizing the oxygen system. The actual production constraints include upper and lower limits of oxygen production for the air separation unit and the gasifier unit, ramp constraints for the air separation unit and the gasifier unit, on / off constraints for the air separation unit, storage system-related constraints, and material balance constraints for the oxygen system.

[0113] Step 1.1: According to the operation status of the air separation unit and gasifier unit in the oxygen system of the steel enterprise, establish the upper and lower limits of oxygen production of the air separation unit and gasifier unit, including:

[0114] (1) Upper and lower limits of gaseous oxygen production of air separation units:

[0115]

[0116] Where i represents the air separation unit, I represents the set of air separation units; t represents the scheduling time period, T represents the set of discrete time points in the scheduling outlook period; u i,t Indicates the start and stop status of air separation unit i during period t, 0 means shutdown, 1 means startup; pg i,t represents the gaseous oxygen production of air separation unit i during period t; represents the lower limit of the gaseous oxygen production of air separation unit i, represents the upper limit of the gaseous oxygen production of air separation unit i;

[0117] (2) Upper and lower limits of liquid oxygen production of air separation units:

[0118]

[0119] Among them, pl i,t represents the liquid oxygen production of air separation unit i during period t; , represents the lower limit of liquid oxygen production of air separation unit i, represents the upper limit of liquid oxygen production of air separation unit i;

[0120] (3) Upper and lower limits of gaseous oxygen production of gasifier units:

[0121]

[0122] Wherein, j represents the gasifier unit, J represents the set of gasifier units; u j,t Indicates the start and stop status of the vaporizer unit j during period t, 0 means shutdown, 1 means startup, pg j,t represents the gaseous oxygen production of gasifier unit j during period t; , represents the lower limit of gaseous oxygen production of gasifier unit j; represents the upper limit of the gaseous oxygen production of gasifier unit j;

[0123] Step 1.2: According to the oxygen production change rate of the air separation unit and the gasifier unit during operation, establish the ramp constraints of the air separation unit and the gasifier unit, including:

[0124] (1) Gaseous oxygen production ramp-up constraints of air separation units:

[0125]

[0126] in, It represents the upper limit of the oxygen production rate of change when air separation unit i is running. Indicates the upper limit of the oxygen production rate change when air separation unit i is started. Indicates the upper limit of the oxygen production rate when air separation unit i is shut down, pg i,t-1 represents the gaseous oxygen production of air separation unit i during period t-1, u i,t-1 Indicates the start and stop status of air separation unit i during period t-1;

[0127] (2) Liquid oxygen production ramp-up constraints of air separation units:

[0128]

[0129] in, It represents the upper limit of the rate of change of liquid oxygen production when air separation unit i is in operation; It represents the upper limit of the rate of change of liquid oxygen production when air separation unit i is started; Indicates the upper limit of the liquid oxygen production rate when air separation unit i is shut down, pl i,t-1 represents the liquid oxygen production of air separation unit i during period t-1;

[0130] (3) Gasifier unit gaseous oxygen production ramp constraints:

[0131]

[0132] in, It represents the upper limit of the rate of change of gaseous oxygen production when gasifier unit j is running; It represents the upper limit of the rate of change of gaseous oxygen production when gasifier unit j is started; pg represents the upper limit of the rate of change of gaseous oxygen production when gasifier unit j is shut down; j,t-1 represents the gaseous oxygen production of gasifier unit j during period t-1, u j,t-1 Indicates the start and stop status of gasifier unit j during period t-1;

[0133] Step 1.3: According to the on / off clearance of the air separation unit during actual operation, the on / off constraints of the air separation unit are established as follows:

[0134]

[0135] in, It represents the continuous operation time of air separation unit i before time period t, T represents the continuous shutdown time of air separation unit i before time period t; i on Indicates the minimum startup time of air separation unit i, T i off Indicates the minimum shutdown time of air separation unit i;

[0136] Step 1.4: Establish storage system related constraints based on the storage characteristics of liquid oxygen and gaseous oxygen;

[0137] (1) Establish constraints related to gaseous oxygen storage;

[0138]

[0139] Among them, sg t represents the storage amount of gaseous oxygen in period t, sg min Indicates the lower limit of gaseous oxygen inventory, sg max Indicates the upper limit of gaseous oxygen inventory;

[0140] (2) Establish constraints related to liquid oxygen storage;

[0141]

[0142] Among them, sl t represents the storage volume of liquid oxygen during period t, sl min Indicates the lower limit of liquid oxygen inventory, sl max Indicates the upper limit of liquid oxygen inventory;

[0143] Step 1.5: Establish oxygen system material balance constraints for liquid oxygen and gaseous oxygen;

[0144] (1) Establishing gaseous oxygen material balance constraints:

[0145]

[0146] in, represents the demand for gaseous oxygen during period t, sg t-1 Indicates the storage amount of gaseous oxygen in period t-1;

[0147] (2) Establish liquid oxygen material balance constraints:

[0148]

[0149] in, Indicates the conversion factor of gaseous oxygen into liquid oxygen, sl t-1 represents the storage amount of liquid oxygen during the period t-1, represents the demand for liquid oxygen during period t;

[0150] Step 1.6: Optimize the objective function of the oxygen system and establish a minimization objective function, wherein the objective function is composed of the operation and start-up and shutdown costs of the air separation unit, the operation and start-up and shutdown costs of the gasifier unit, and the inventory cost of the oxygen system;

[0151] (1) Operation and start-up and shutdown costs of air separation units:

[0152] C i,t =(α0 i pg i,t +α1 i pl i,t )·u i,t +S i ·u i,t (1-u i,t-1 ); (12)

[0153] Among them, C i,t represents the cost of air separation unit i in period t, S i represents the startup cost of air separation unit i, α0 i ,α1 i represents the operating cost coefficient of air separation unit i;

[0154] (2) Gasifier unit operation and start-up and shutdown costs:

[0155] C j,t =β j ·pg j,t ·u j,t +S j ·u j,t (1-u j,t-1 ); (13)

[0156] Among them, C j,t represents the cost of gasifier unit j in period t, S j represents the startup cost of gasifier unit j, β j represents the operating cost coefficient of gasifier unit j;

[0157] (3) Establishing oxygen system inventory costs;

[0158]

[0159] Among them, C t represents the inventory cost of the oxygen system during period t, including the inventory cost of gaseous oxygen and the inventory cost of liquid oxygen, γ g represents the inventory cost coefficient of gaseous oxygen, γ l represents the inventory cost factor for liquid oxygen;

[0160] Based on this, the objective function can be summarized as follows:

[0161]

[0162] Based on this, the upper and lower limit constraints of oxygen production of the air separation unit and the gasifier unit, the ramping constraints of the air separation unit and the gasifier unit, the on-off constraints of the air separation unit, the storage system related constraints, the material balance constraints of the oxygen system and the objective function of optimizing the oxygen system were obtained.

[0163] Wherein, when step 1 is implemented by software, since formula (1)-formula (15) belong to nonlinear programming models, they cannot be directly solved by optimization software, so they need to be converted into linear programming models;

[0164] The ramp-up constraint of the gaseous oxygen production of the air separation unit in formula (4) is linearized as follows:

[0165]

[0166] The liquid oxygen production ramp constraint of the air separation unit in formula (5) is linearized as follows:

[0167]

[0168] The air separation unit on / off constraints in formula (7) are linearized as follows:

[0169]

[0170] The ramp constraint of gaseous oxygen production of the gasifier unit in formula (6) is linearized as follows:

[0171]

[0172] According to the decomposable structural characteristics of the established oxygen scheduling mathematical model, the coupling constraints in the model are relaxed into the objective function using Lagrange multipliers. This way, the original problem can be decomposed into three easily solvable sub-problems of a single unit: air separation unit, gasifier unit, and storage system. This is specifically achieved through step 2.

[0173] Step 2: Use the Lagrange multiplier vector {λ t},{μ t}, relax the oxygen system material balance constraints into the objective function, generate a Lagrangian relaxation problem, and record it as (LR), construct a Lagrangian dual problem, and record it as (LD), the Lagrangian dual problem is the maximum value of the Lagrangian relaxation problem; for a given multiplier vector {λ t},{μ t}, decompose the Lagrangian relaxation problem (LR) into air separation unit subproblem, gasifier unit subproblem and storage system subproblem;

[0174] Among them, the Lagrangian relaxation problem is specifically expressed by the following formula:

[0175] (LR)L(λ,μ)=minZ LR ; (16)

[0176] Among them, Z LR It can be expressed by the following formula:

[0177]

[0178] Among them, the parameters in formula (17) satisfy the constraints of formula (1)-formula (9);

[0179] The Lagrangian dual problem is expressed by the following formula:

[0180] (LD)maxL(λ,μ); (18)

[0181] Based on this, we obtain the Lagrangian relaxation problem, namely formula (17), and the Lagrangian dual problem, namely formula (18).

[0182] Based on this, for a given multiplier vector {λ t},{μ t}, decompose the Lagrangian relaxation problem (LR) into the air separation unit subproblem, the gasifier unit subproblem and the storage system subproblem, where the air separation unit subproblem, the gasifier unit subproblem and the storage system subproblem are expressed by the following formula:

[0183] (1) Air separation unit sub-problem

[0184]

[0185] Wherein, the parameters in formula (19) satisfy the constraints of formula (1), formula (2), formula (4), formula (5) and formula (7);

[0186] (2) Gasifier unit sub-problem

[0187]

[0188] Among them, the parameters in formula (20) satisfy the constraints of formula (3) and formula (6);

[0189] (3) Storage system sub-issue

[0190]

[0191] Among them, the parameters in formula (21) satisfy the constraints of formula (8) and formula (9).

[0192] Step 3: Calculate the air separation unit subproblem, the gasifier unit subproblem and the storage system subproblem, and then obtain a scheduling schedule, which includes the start and stop times of the air separation unit and the gasifier unit, the gaseous oxygen production and liquid oxygen production of the air separation unit, and the gaseous oxygen production of the gasifier unit; Combined Figure 2 , specifically including the following steps:

[0193] Step 3.1: Set the initial number of iterations, λ t , μ t The initial value and maximum number of iterations are used to initialize the dual function value of the Lagrangian dual problem, and the initial number of iterations is used as the current number of iterations. t The initial value of λ t The current value of μ t The initial value of μ t Current value: the dual function value of the initial Lagrangian dual problem is used as the current dual function value of the Lagrangian dual problem;

[0194] Step 3.2: Solve the air separation unit subproblem, the gasifier unit subproblem and the storage system subproblem; the air separation unit subproblem and the gasifier unit subproblem are solved using dynamic programming, and the storage system subproblem is solved using GUROBI.

[0195] Step 3.2.1: Solve equation (19) for the air separation unit subproblem, where α0 i ,α1 i and S i The value of λ is known. t is t The current value of μ t μ t Current value, solve the minimum value of formula (19), and get its corresponding pg i,t The value of pl i,t The value of u i,t The value of u i,t-1 The value of pg i,t The value of pl i,t The value of u i,t The value of u i,t-1 The value of is taken as the solution to the air separation unit subproblem, that is, the solution to formula (19);

[0196] Step 3.2.2: Solve equation (20) for the gasifier unit subproblem, where β j , and S j The value of λ is known. t is t The current value of μ t μ t Current value, solve the minimum value of formula (20), and get its corresponding pg j,t The value of u j,t The value of u j,t-1 The value of pg j,t The value of u j,t The value of u j,t-1 The value of is taken as the solution to the gasifier unit subproblem, i.e., the solution to formula (20);

[0197] Step 3.2.3: Solve equation (21) for the storage system subproblem, where λ t is t The current value of μ t μ t Current value, solve the minimum value of formula (21), and get its corresponding C t The value of sg t The value of sg t-1 The value of sg t The value and sl t-1 The value of C t The value of sg t The value of sg t-1 The value of sgt The value and sl t-1 The numerical value of is taken as the solution to the storage system subproblem, that is, the solution to formula (21).

[0198] In order to speed up the solution of the sub-problem, the present invention proposes an acceleration strategy, which is proposed based on the optimality condition. When the optimality condition is met, the optimal solution of the sub-problem can be directly obtained. The optimality condition and the optimal solution under the optimality condition are introduced as follows:

[0199] 1) Air separation unit sub-problem:

[0200] Optimal properties: When α0 i -λ t >0 and α1 i -μ t >0, if the unit continues to run less than the minimum start-up time when it drops to the minimum power, it will continue to run at low power. When the continuous running time is equal to the minimum start-up time, it will be shut down directly; if the unit continues to run greater than the minimum start-up time when it drops to the minimum power, it will be shut down directly.

[0201] 2) Gasifier unit sub-problem:

[0202] Optimal properties: When When the load is reduced, the speed is reduced at the fastest load reduction rate and then shut down.

[0203] Step 3.3: The solutions of the air separation unit subproblem, the gasifier unit subproblem and the storage system subproblem are used as the solutions of the Lagrangian relaxation problem, and then the value of the Lagrangian relaxation problem is calculated. Among the values ​​of the relaxation problem at the current iteration number and the current dual function value, the one with the largest value is obtained as the new current dual function value, and the new current dual function value is used as the lower bound of the objective function;

[0204] Since the oxygen supply and demand balance constraints and liquid oxygen supply and demand balance constraints are relaxed in the relaxation problem, the solution of the relaxation problem is often infeasible for the original problem. Therefore, this chapter designs a two-stage heuristic algorithm to convert the solution of the relaxation problem into a feasible solution of the original problem. The construction method is as follows:

[0205] In the first stage, the minimum on / off time constraint and ramp constraint of the air separation unit and the ramp constraint of the gasifier are considered to determine the start / stop sequence of the air separation equipment and the gasification equipment.

[0206] In the second stage, based on the start-stop sequence given in the first stage, the oxygen and liquid oxygen production of the air separation unit, the oxygen production of the vaporizer, and the storage capacity of oxygen and liquid oxygen in each period are calculated.

[0207] Among them, in the first stage, when judging the start and stop status of each device, the start and stop status of each device is determined by oxygen demand first. Only when liquid oxygen is insufficient will the air separation unit be turned on to prevent the dead loop caused by cycle adjustment. The heuristic algorithm repair process is as shown in step 3.4;

[0208] Step 3.4: When the scheduling time is set, the solution of the Lagrangian relaxation problem is used as the initial solution, and the heuristic algorithm is used to transform the initial solution into a feasible solution of the optimization scheduling model to obtain the start-stop sequence of the air separation unit and the gasifier unit;

[0209] Step 3.4.1: For each time period, the solution of the Lagrangian relaxation problem is used as the initial solution, and the initial solution is substituted into the formula on the left side of the inequality sign of formula (10) to obtain the total output of gaseous oxygen, and determine the magnitude of the total output of gaseous oxygen and the demand for gaseous oxygen. Specifically, when the total output of gaseous oxygen is greater than the demand for gaseous oxygen, the gasifier units are sorted from high to low according to the operating cost to obtain the first sequence, and the gasifier units that are shut down in sequence according to the sequence numbers from small to large in the first sequence are sorted from high to low according to the operating cost to satisfy the formula ( 7) sort the air separation units with the minimum startup time constraint in the first sequence to obtain a second sequence, and at the same time, shut down the air separation units in the second sequence in sequence from small to large, and store the gaseous oxygen that exceeds the demand for gaseous oxygen until it is released; when the total output of gaseous oxygen is less than the demand for gaseous oxygen, start the gasifier units in sequence in sequence from large to small in the first sequence, and at the same time, start the air separation units in the second sequence in sequence in sequence from large to small, until the total output of oxygen is greater than or equal to the demand for gaseous oxygen;

[0210] Step 3.4.2: For each time period, the solution of the Lagrangian relaxation problem is used as the initial solution, and the initial solution is substituted into the formula on the left side of the inequality sign of formula (11) to obtain the total output of liquid oxygen, and determine the total output of liquid oxygen and the demand for liquid oxygen. Specifically, when the total output of liquid oxygen is greater than the demand for liquid oxygen, the liquid oxygen in excess of the demand for liquid oxygen is stored; when the total output of liquid oxygen is less than the demand for liquid oxygen, the output of gaseous oxygen of the operating air separation unit is increased, and then the output of liquid oxygen is increased until the output of liquid oxygen is equal to or less than the demand for liquid oxygen. The total oxygen output is greater than the demand for liquid oxygen, or the gaseous oxygen output of all operating air separation units reaches the upper limit. When the gaseous oxygen output of all operating air separation units reaches the upper limit and the total output of liquid oxygen is less than the demand for liquid oxygen, the air separation units that meet the minimum shutdown time constraint in formula (7) are sorted from low to high according to the operating cost to obtain a third sequence. The air separation units are turned on according to the sequence numbers in the third sequence from small to large, and the gaseous oxygen output of the turned on air separation units is increased at the same time until the liquid oxygen output is greater than or equal to the demand for liquid oxygen.

[0211] It should be noted that after reading the data, the air separation units and gasifier units are sorted in ascending order according to the operating costs to ensure that the heuristic algorithm gives priority to starting the equipment with lower operating costs when oxygen or liquid oxygen is insufficient, and gives priority to shutting down the equipment with higher operating costs when shutting down the equipment.

[0212] Step 3.4.3: Based on the adjustments to the start and stop of the air separation unit and the gasifier unit in steps 3.4.1 and 3.4.2, determine the start and stop sequence of the air separation unit and the gasifier unit. Based on the parameters involved in steps 3.4.1 and 3.4.2, update the initial solution to obtain a feasible solution.

[0213] Step 3.5: According to the start-stop sequence, use GUROBI to build a model and solve it to obtain a scheduling schedule under the current number of iterations;

[0214] Step 3.6: Calculate the value of the objective function based on the feasible solution and the objective function, and use the value of the objective function as the upper bound of the objective function;

[0215] Step 3.7: Calculate the duality gap GAP of the current iteration number k, and determine whether the duality gap GAP is less than the preset threshold. If the duality gap GAP is less than the set value, output the scheduling plan under the current iteration number; if the duality gap GAP is not less than the set value, execute step 3.8;

[0216] The calculation of the duality gap GAP of the current iteration number k is specifically implemented by the following formula:

[0217]

[0218] Among them, L * is the minimum objective function value up to the current iteration number k, L k is the current dual function value for the current iteration number k.

[0219] Step 3.8: Determine whether the current number of iterations is less than or equal to the maximum number of iterations. If the current number of iterations is greater than the maximum number of iterations, output the scheduling plan under the current number of iterations; if the current number of iterations is less than or equal to the maximum number of iterations, add one to the current number of iterations and execute step 3.9;

[0220] Step 3.9: Use the subgradient algorithm to update the Lagrange multiplier and return to step 3.2.

[0221] Among them, the subgradient algorithm is used to update the Lagrange multiplier, which is specifically implemented by the following formula:

[0222] μ k+1 =μ k +t k g k (twenty three)

[0223]

[0224] Among them, μ k+1 is the Lagrange multiplier at iteration number k+1, μ k is the Lagrange multiplier at the iteration number k, t k is the step size of the current iteration number k, g k is the subgradient of the current iteration number k, and γ is the step size coefficient.

[0225] The following is an example to further describe the specific implementation of the present invention in detail. A steel enterprise has four air separation units, two gasifier units, and one storage system. The oxygen required by the workshop is supplied by the pipeline network of the whole plant. The volume of the 2# air separation unit is 19985 cubic meters. The volume of the 3# air separation unit is 18281 cubic meters. The volume of the 4# air separation unit is 18,301 cubic meters. The volume of the 1# air separation unit is 18156 cubic meters. The volume of the 2# air separation unit is 26,185 cubic meters. The capacity of the 3# air separation unit is 26,200 cubic meters. The volume of the 4# air separation unit is 27631 cubic meters. The gasifier unit has a capacity of 26,030 cubic meters. is 1100 cubic meters, The scheduling cycle T = 24 is taken as an example, the basic scheduling data and parameters of the steel enterprise are sorted out, and the solution algorithm is used for iterative calculation. The experimental results are shown in Table 1, which mainly include the information of the start and stop time, storage volume and release volume in a period.

[0226] Table 1 24h air separation plant scheduling results

[0227]

[0228]

[0229] According to the above constraints, the corresponding oxygen distribution scheme can be obtained by solving the algorithm, and compared with the results obtained by the solver. The results are shown in Table 2, where ACT represents the average calculation time of each scale example, and GAP represents the duality gap.

[0230] Table 2 Comparison results between LR algorithm and solver

[0231]

[0232] After the entire distribution target is completed, from the perspective of oxygen utilization, while meeting the needs of oxygen and liquid oxygen users, the production and operation costs of the oxygen system can be effectively reduced. The basis of economic dispatch of oxygen systems in steel enterprises is the oxygen optimization distribution model. Under the concept of advocating efficient energy use, different dispatch plans can be formulated for different users, effectively reducing the operating costs of the oxygen system and reducing oxygen emissions while ensuring safe and stable operation.

[0233] The above description is only a preferred embodiment of the present disclosure and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the embodiments of the present disclosure is not limited to the technical solutions formed by a specific combination of the above-mentioned technical features, but should also cover other technical solutions formed by any combination of the above-mentioned technical features or their equivalent features without departing from the above-mentioned inventive concept. For example, the above-mentioned features are replaced with the technical features with similar functions disclosed in the embodiments of the present disclosure (but not limited to) to form a technical solution.

Claims

1. A steel enterprise oxygen system optimization scheduling model construction and solution method, characterized in that: include: Step 1: Establish an optimal scheduling model for the oxygen system of a steel enterprise, and consider actual production constraints. The optimal scheduling model includes actual production constraints and an objective function for optimizing the oxygen system. The actual production constraints include upper and lower limits of oxygen production for the air separation unit and the gasifier unit, ramp constraints for the air separation unit and the gasifier unit, on / off constraints for the air separation unit, storage system-related constraints, and material balance constraints for the oxygen system. Step 2: Use the Lagrange multiplier vector {λ t },{μ t }, relax the oxygen system material balance constraints into the objective function, generate a Lagrangian relaxation problem, and record it as (LR), construct a Lagrangian dual problem, and record it as (LD), the Lagrangian dual problem is the maximum value of the Lagrangian relaxation problem; for a given multiplier vector {λ t },{μ t }, decompose the Lagrangian relaxation problem (LR) into air separation unit subproblem, gasifier unit subproblem and storage system subproblem; Step 3: Calculate the solutions of the air separation unit subproblem, the gasifier unit subproblem and the storage system subproblem, and then obtain a scheduling schedule, which includes the start and stop times of the air separation unit and the gasifier unit, the gaseous oxygen production and liquid oxygen production of the air separation unit, and the gaseous oxygen production of the gasifier unit; Step 3.1: Set the initial number of iterations, λ t , μ t The initial value and maximum number of iterations are used to initialize the dual function value of the Lagrangian dual problem, and the initial number of iterations is used as the current number of iterations. t The initial value of λ t The current value of μ t The initial value of μ t Current value: the dual function value of the initial Lagrangian dual problem is used as the current dual function value of the Lagrangian dual problem; Step 3.2: Solve the air separation unit subproblem, the gasifier unit subproblem and the storage system subproblem; Step 3.3: The solutions of the air separation unit subproblem, the gasifier unit subproblem and the storage system subproblem are used as the solutions of the Lagrangian relaxation problem, and then the value of the Lagrangian relaxation problem is calculated. Among the values ​​of the relaxation problem at the current iteration number and the current dual function value, the one with the largest value is obtained as the new current dual function value, and the new current dual function value is used as the lower bound of the objective function; Step 3.4: When the scheduling time is set, the solution of the Lagrangian relaxation problem is used as the initial solution, and the heuristic algorithm is used to transform the initial solution into a feasible solution of the optimization scheduling model to obtain the start-stop sequence of the air separation unit and the gasifier unit; Step 3.5: According to the start-stop sequence, use GUROBI to build a model and solve it to obtain a scheduling schedule under the current number of iterations; Step 3.6: Calculate the value of the objective function based on the feasible solution and the objective function, and use the value of the objective function as the upper bound of the objective function; Step 3.7: Calculate the duality gap GAP of the current iteration number k, determine whether the duality gap GAP is less than a preset threshold, and if the duality gap GAP is less than the set value, output the scheduling plan under the current iteration number; When the dual gap GAP is not less than the set value, execute step 3.8; Step 3.8: Determine whether the current number of iterations is less than or equal to the maximum number of iterations. If the current number of iterations is greater than the maximum number of iterations, output the scheduling plan under the current number of iterations; When the current number of iterations is less than or equal to the maximum number of iterations, the current number of iterations is increased by one, and step 3.9 is executed; Step 3.9: Use the subgradient algorithm to update the Lagrange multiplier and return to step 3.

2.

2. The method for constructing and solving the optimization scheduling model of the oxygen system of a steel enterprise according to claim 1 is characterized in that: Step 1 specifically includes: Step 1.1: According to the operation status of the air separation unit and gasifier unit in the oxygen system of the steel enterprise, establish the upper and lower limits of oxygen production of the air separation unit and gasifier unit, including: (1) Upper and lower limits of gaseous oxygen production of air separation units: Where i represents the air separation unit, I represents the set of air separation units; t represents the scheduling time period, T represents the set of discrete time points in the scheduling outlook period; u i,t Indicates the start and stop status of air separation unit i during period t, 0 means shutdown, 1 means startup; pg i,t represents the gaseous oxygen production of air separation unit i during period t; represents the lower limit of the gaseous oxygen production of air separation unit i, represents the upper limit of the gaseous oxygen production of air separation unit i; (2) Upper and lower limits of liquid oxygen production of air separation units: Among them, pl i,t represents the liquid oxygen production of air separation unit i during period t; , represents the lower limit of liquid oxygen production of air separation unit i, represents the upper limit of liquid oxygen production of air separation unit i; (3) Upper and lower limits of gaseous oxygen production of gasifier units: Wherein, j represents the gasifier unit, J represents the set of gasifier units; u j,t Indicates the start and stop status of the vaporizer unit j during period t, 0 means shutdown, 1 means startup, pg j,t represents the gaseous oxygen production of gasifier unit j during period t; represents the lower limit of the gaseous oxygen production of gasifier unit j; represents the upper limit of the gaseous oxygen production of gasifier unit j; Step 1.2: According to the oxygen production change rate of the air separation unit and the gasifier unit during operation, establish the ramp constraints of the air separation unit and the gasifier unit, including: (1) Gaseous oxygen production ramp-up constraints of air separation units: in, It represents the upper limit of the oxygen production rate of change when air separation unit i is running. Indicates the upper limit of the oxygen production rate change when air separation unit i is started. Indicates the upper limit of the oxygen production rate when air separation unit i is shut down, pg i,t-1 represents the gaseous oxygen production of air separation unit i during period t-1, u i,t-1 Indicates the start and stop status of air separation unit i during period t-1; (2) Liquid oxygen production ramp-up constraints of air separation units: in, It represents the upper limit of the rate of change of liquid oxygen production when air separation unit i is in operation; It represents the upper limit of the rate of change of liquid oxygen production when air separation unit i is started; Indicates the upper limit of the liquid oxygen production rate when air separation unit i is shut down, pl i,t-1 represents the liquid oxygen production of air separation unit i during period t-1; (3) Gasifier unit gaseous oxygen production ramp constraints: in, It represents the upper limit of the rate of change of gaseous oxygen production when gasifier unit j is in operation; It represents the upper limit of the rate of change of gaseous oxygen production when gasifier unit j is started; pg represents the upper limit of the rate of change of gaseous oxygen production when gasifier unit j is shut down; j,t-1 represents the gaseous oxygen production of gasifier unit j during period t-1, u j,t-1 Indicates the start and stop status of gasifier unit j during period t-1; Step 1.3: According to the on / off clearance of the air separation unit during actual operation, the on / off constraints of the air separation unit are established as follows: in, It represents the continuous operation time of air separation unit i before time period t, T represents the continuous shutdown time of air separation unit i before time period t; i on Indicates the minimum startup time of air separation unit i, T i off Indicates the minimum shutdown time of air separation unit i; Step 1.4: Establish storage system related constraints based on the storage characteristics of liquid oxygen and gaseous oxygen; (1) Establish constraints related to gaseous oxygen storage; Among them, sg t Indicates the storage amount of gaseous oxygen during period t, sg min Indicates the lower limit of gaseous oxygen inventory, sg max Indicates the upper limit of gaseous oxygen inventory; (2) Establish constraints related to liquid oxygen storage; Among them, sl t represents the storage volume of liquid oxygen during period t, sl min Indicates the lower limit of liquid oxygen inventory, sl max Indicates the upper limit of liquid oxygen inventory; Step 1.5: Establish oxygen system material balance constraints for liquid oxygen and gaseous oxygen; (1) Establishing gaseous oxygen material balance constraints: Among them, D t g represents the demand for gaseous oxygen during period t, sg t-1 Indicates the storage amount of gaseous oxygen during period t-1; (2) Establish liquid oxygen material balance constraints: in, Indicates the conversion factor of gaseous oxygen into liquid oxygen, sl t-1 represents the storage amount of liquid oxygen during the period t-1, represents the demand for liquid oxygen during period t; Step 1.6: Optimize the objective function of the oxygen system and establish a minimization objective function, wherein the objective function is composed of the operation and start-up and shutdown costs of the air separation unit, the operation and start-up and shutdown costs of the gasifier unit, and the inventory cost of the oxygen system; (1) Operation and start-up and shutdown costs of air separation units: C i,t =(α0 i pg i,t +α1 i pl i,t )·u i,t +S i ·the i,t (1-th i,t-1 );(12) Among them, C i,t represents the cost of air separation unit i in period t, S i represents the startup cost of air separation unit i, α0 i ,α1 i represents the operating cost coefficient of air separation unit i; (2) Gasifier unit operation and start-up and shutdown costs: C j,t =β j pg j,t u j,t +S j u j,t (1-u j,t-1 );(13) Among them, C j,t represents the cost of gasifier unit j in period t, S j represents the startup cost of gasifier unit j, β j represents the operating cost coefficient of gasifier unit j; (3) Establishing oxygen system inventory costs; Among them, C t represents the inventory cost of the oxygen system during period t, including the inventory cost of gaseous oxygen and the inventory cost of liquid oxygen, γ g represents the inventory cost coefficient of gaseous oxygen, γ l represents the inventory cost factor for liquid oxygen; Based on this, the objective function can be summarized as follows: Based on this, the upper and lower limit constraints of oxygen production of the air separation unit and the gasifier unit, the ramping constraints of the air separation unit and the gasifier unit, the on-off constraints of the air separation unit, the storage system related constraints, the material balance constraints of the oxygen system and the objective function of optimizing the oxygen system were obtained.

3. The method for constructing and solving the optimization scheduling model of the oxygen system of a steel enterprise according to claim 2 is characterized in that: The Lagrangian relaxation problem is specifically expressed by the following formula: (LR)L(λ,μ)=minZ LR (16) Among them, Z LR It can be expressed by the following formula: Among them, the parameters in formula (17) satisfy the constraints of formula (1)-formula (9); The Lagrangian dual problem is expressed by the following formula: (LD)maxL(λ,μ); (18) Based on this, we obtain the Lagrangian relaxation problem, namely formula (17), and the Lagrangian dual problem, namely formula (18).

4. The method for constructing and solving the optimization scheduling model of oxygen system in a steel enterprise according to claim 3 is characterized in that: In step 2, the air distribution unit subproblem, the gasifier unit subproblem and the storage system subproblem are expressed by the following formula: (1) Air separation unit sub-problem Wherein, the parameters in formula (19) satisfy the constraints of formula (1), formula (2), formula (4), formula (5) and formula (7); (2) Gasifier unit sub-problem Among them, the parameters in formula (20) satisfy the constraints of formula (3) and formula (6); (3) Storage system sub-issue Among them, the parameters in formula (21) satisfy the constraints of formula (8) and formula (9).

5. The method for constructing and solving the optimization scheduling model of oxygen system in a steel enterprise according to claim 4 is characterized in that: Step 3.2 specifically includes: Step 3.2.1: Solve equation (19) for the air separation unit subproblem, where α0 i ,α1 i and S i The value of λ is known. t is t The current value of μ t μ t Current value, solve the minimum value of formula (19), and get its corresponding pg i,t The value of pl i,t The value of u i,t The value of u i,t-1 The value of pg i,t The value of pl i,t The value of u i,t The value of u i,t-1 The value of is taken as the solution to the air separation unit subproblem, that is, the solution to formula (19); Step 3.2.2: Solve equation (20) for the gasifier unit subproblem, where β j , and S j The value of λ is known. t is t The current value of μ t μ t Current value, solve the minimum value of formula (20), and get its corresponding pg j,t The value of u j,t The value of u j,t-1 The value of pg j,t The value of u j,t The value of u j,t-1 The value of is taken as the solution to the gasifier unit subproblem, i.e., the solution to formula (20); Step 3.2.3: Solve equation (21) for the storage system subproblem, where λ t is t The current value of μ t μ t Current value, solve the minimum value of formula (21), and get its corresponding C t The value of sg t The value of sg t-1 The value of sg t The value and sl t-1 The value of C t The value of sg t The value of sg t-1 The value of sg t The value and sl t-1 The numerical value of is taken as the solution to the storage system subproblem, that is, the solution to formula (21).

6. The method for constructing and solving the optimization scheduling model of oxygen system in a steel enterprise according to claim 5 is characterized in that: Step 3.4 specifically includes: Step 3.4.1: For each time period, the solution of the Lagrangian relaxation problem is used as the initial solution, and the initial solution is substituted into the formula on the left side of the inequality sign of formula (10) to obtain the total output of gaseous oxygen, and determine the magnitude of the total output of gaseous oxygen and the demand for gaseous oxygen. Specifically, when the total output of gaseous oxygen is greater than the demand for gaseous oxygen, the gasifier units are sorted from high to low according to the operating cost to obtain the first sequence, and the gasifier units that are shut down in sequence according to the sequence numbers from small to large in the first sequence are sorted from high to low according to the operating cost to satisfy the formula ( 7) sort the air separation units with the minimum startup time constraint in the first sequence to obtain a second sequence, and at the same time, shut down the air separation units in the second sequence in sequence from small to large, and store the gaseous oxygen that exceeds the demand for gaseous oxygen until it is released; when the total output of gaseous oxygen is less than the demand for gaseous oxygen, start the gasifier units in sequence in sequence from large to small in the first sequence, and at the same time, start the air separation units in the second sequence in sequence in sequence from large to small, until the total output of oxygen is greater than or equal to the demand for gaseous oxygen; Step 3.4.2: For each time period, the solution of the Lagrangian relaxation problem is used as the initial solution, and the initial solution is substituted into the formula on the left side of the inequality sign of formula (11) to obtain the total output of liquid oxygen, and determine the total output of liquid oxygen and the demand for liquid oxygen. Specifically, when the total output of liquid oxygen is greater than the demand for liquid oxygen, the liquid oxygen in excess of the demand for liquid oxygen is stored; when the total output of liquid oxygen is less than the demand for liquid oxygen, the output of gaseous oxygen of the operating air separation unit is increased, and then the output of liquid oxygen is increased until the output of liquid oxygen is equal to or less than the demand for liquid oxygen. The total oxygen output is greater than the demand for liquid oxygen, or the gaseous oxygen output of all operating air separation units reaches the upper limit. When the gaseous oxygen output of all operating air separation units reaches the upper limit and the total output of liquid oxygen is less than the demand for liquid oxygen, the air separation units that meet the minimum shutdown time constraint in formula (7) are sorted from low to high according to the operating cost to obtain a third sequence. The air separation units are turned on according to the sequence numbers in the third sequence from small to large, and the gaseous oxygen output of the turned on air separation units is increased at the same time until the liquid oxygen output is greater than or equal to the demand for liquid oxygen. Step 3.4.3: Based on the adjustments to the start and stop of the air separation unit and the gasifier unit in steps 3.4.1 and 3.4.2, determine the start and stop sequence of the air separation unit and the gasifier unit. Based on the parameters involved in steps 3.4.1 and 3.4.2, update the initial solution to obtain a feasible solution.

7. The method for constructing and solving the optimization scheduling model of oxygen system in a steel enterprise according to claim 1 is characterized in that: In step 3.7, the duality gap GAP of the current iteration number k is calculated, which is specifically implemented by the following formula: Among them, L * is the minimum objective function value up to the current iteration number k, L k is the current dual function value for the current iteration number k.

8. The method for constructing and solving the optimization scheduling model of oxygen system in a steel enterprise according to claim 1 is characterized in that: In step 3.9, the subgradient algorithm is used to update the Lagrange multiplier, which is specifically implemented by the following formula: m k+1 =μ k +t k g k (23) Among them, μ k+1 is the Lagrange multiplier at iteration number k+1, μ k is the Lagrange multiplier at the iteration number k, t k is the step size of the current iteration number k, g k is the subgradient of the current iteration number k, and γ is the step size coefficient.

Images