A method for finding the global optimal cost-effectiveness ratio when the explosive composition is determined

Abstract

The present invention discloses a method for finding the global optimal cost-effectiveness ratio when determining explosive components. The method relates to the field of explosive component ratios and includes the following steps: S1: establishing a calculation model; S2: applying inequality constraints; and S3: finding the optimal value using a differential evolution algorithm based on constraint satisfaction. The method for finding the global optimal cost-effectiveness ratio is proposed in open-pit blasting projects. When determining explosive components, the method can find the global optimal cost-effectiveness ratio within the reasonableness of the component ratios within a maximum cost limit and based on a given expected oxygen balance condition.

Description

Technical Field

[0001] The present invention relates to the field of explosive component ratio, and in particular to a method for seeking a global optimal cost-performance ratio when explosive components are determined. Background Art

[0002] In mining projects, the quality of blasting effects directly affects the subsequent processes and the economic benefits of the mine, and is related to production safety. How to correctly select explosives with excellent performance, reasonable proportions and appropriate prices to meet different blasting requirements is of great significance.

[0003] Wang Xuguang believes that when the combustible agent in the explosive is completely oxidized, the energy released by the explosive is the greatest and the harmful gases generated are the least. Therefore, oxygen balance must be given priority when designing the explosive formula.

[0004] Regarding the optimization of explosive performance, for example, the study of emulsion explosive structure model and optimization technology. Beijing General Research Institute of Mining and Metallurgy Appraisal Data. 1997, emulsion explosive formulation optimization based on hybrid artificial bee colony algorithm [J]. Journal of Applied Functional Analysis, 2015, 17(4): 423-430, and formulation design and optimization of multi-component mixed explosives [J]. Explosive Materials. 1992(05). In these three papers, the researchers only considered the maximization of detonation heat when considering explosive performance.

[0005] The paper "Principles and Mathematical Models for Emulsion Explosive Formulation Optimization" [J]. Nonferrous Mining and Metallurgy, 1999(06): 1-4, points out that explosive volume plays a significant role in the explosive's power contribution. Therefore, in the design of emulsion explosives, the product of explosive heat and explosive volume, along with the total raw material cost, was used as the optimization objective. A mathematical model was also established, taking into account other constraints. Due to the complexity of this model, the authors did not provide an optimization method in this paper.

[0006] Research on the optimization of the proportions of multi-material mixed explosives is not common. Mathematical modeling that takes into account the maximum cost-effectiveness, oxygen balance conditions, and the rational range of each component as constraints is mostly only found in emulsion explosives, and few simple and fast solutions are provided.

[0007] Commonly used explosives in open-pit blasting projects fall into three categories: ammonium nitrate, heavy ammonium nitrate, and emulsion explosives. These explosives are composed of two or more substances mixed through a specific process. The evaluation of explosive performance (explosion heat and volume) is often based on empirical calculations, in addition to laboratory measurements. Furthermore, the design of emulsion explosive formulations has traditionally relied on a combination of empirical analogies, repeated calculations, and performance testing to achieve a slightly negative oxygen balance. However, these calculations and tests are often limited by budget and experience, making satisfactory results difficult to achieve. This paper proposes a method for determining explosive composition based on a given desired oxygen balance, within a maximum cost constraint and a reasonable ratio of components, to find the optimal global cost-effectiveness ratio.

[0008] Therefore, when inventing an explosive composition, it is necessary to seek a method with the best global cost-effectiveness ratio to solve the above problems. Summary of the Invention

[0009] The purpose of the present invention is to provide a method for seeking the global optimal cost-performance ratio when determining the explosive composition, so as to solve the problems raised in the above background technology.

[0010] To achieve the above object, the present invention provides the following technical solution: a method for seeking the global optimal cost-performance ratio when determining the explosive composition, comprising the following steps:

[0011] S1: Establish a calculation model. When the composition of the explosive (mixture of multiple substances) is determined, the Brinkley-Wilson rule chemical reaction equation is determined based on the expected value of the oxygen balance of the designed explosive, so as to establish the corresponding RWS (relative weight strength) and corresponding cost calculation model;

[0012] S2: Inequality constraints, which take the cost performance of explosives as the optimization objective, the mass fraction of the components as the variable, and the sum of the mass fractions of the components as 100% and the expected value of the oxygen balance of the designed explosive as the equality constraints. The maximum cost limit and the reasonable range of the mass fractions of the components are the inequality constraints.

[0013] S3: A method for seeking optimal values ​​based on the differential evolution algorithm with constraint satisfaction.

[0014] Preferably, when an explosive component is determined, a method for seeking a global optimal cost-performance ratio is expressed as follows:

[0015]

[0016]

[0017]

[0018] cost≤p*

[0019] lower i ≤x i ≤upper i (1)

[0020] x i ——mass fraction of the i-th component, %;

[0021] B i ——oxygen balance value of the i-th component, g / g;

[0022] OB*——expected oxygen balance value of explosive, g / g;

[0023] p*——maximum fee limit, yuan / kg;

[0024] lower i ——Minimum limit of mass fraction of the i-th component, %;

[0025] upper i ——The maximum limit of mass fraction of the i-th component, %.

[0026] Preferably, the method for seeking the optimal value based on the constraint satisfaction differential evolution algorithm described in S3 further includes the following steps:

[0027] Step 1: Initialize the parameters: population size is pop_num, maximum number of iterations is iter_max, strategy control parameter P_cost, expected number of individuals satisfying the constraint is T_mun, strategy control adjustment coefficient α, number of variables n, and generate pop_num individuals that weakly violate the constraint as a population;

[0028] Step 2: Start iteration: In each iteration, generate a random number and compare it with P_cost, and then decide whether to proceed to step 3 or step 4;

[0029] Step 3: Unconstrained single-objective optimization: Use the conventional differential evolution algorithm to perform crossover mutation on the population, use a greedy strategy in selection, merge the mutant population and the contemporary population, sort according to the function objective value, select the first pop_num individuals as the next generation population, and go to step 5;

[0030] Step 4: Constraint satisfaction optimization: Use the conventional differential evolution algorithm to perform crossover mutation on the population. Use the greedy strategy in selection to merge the mutant population and the contemporary population. Calculate the condition value of each constraint and convert it into constraint violation degree according to formula (2). Then use the non-dominated sorting to sort the constraint violation degree. Select the first pop_num individuals with higher dominance level as the next generation population and go to step 5.

[0031] gi (X)<=b i

[0032]

[0033] c i =0 else (2)

[0034] Step 5: After sorting according to the constraint dominance, record the number of undominated solutions (i.e., feasible solutions). If there are too few feasible solutions, use formula (3) to update P_cost. If there are too many feasible solutions, use formula (4) to update P_cost and then sort according to the objective function value and record the current optimal individual.

[0035] P_cost=(1-α)×P_cost (3)

[0036] P_cost=1-(1-P_cost)×(1-α) (4)

[0037] Step 6: Determine the termination condition of the iteration. If it is not met, go to step 2 and continue iterating; if it is met, output the optimal individual and end.

[0038] Preferably, in the optimal value seeking method based on the constraint satisfaction differential evolution algorithm, a random number is generated and compared with P_cost in each iteration, and step three is entered when performing unconstrained single-objective optimization.

[0039] Preferably, in the optimal value seeking method based on the constraint satisfaction differential evolution algorithm, a random number is generated and compared with P_cost in each iteration, and step 4 is entered when the constraint satisfaction optimization is performed.

[0040] Preferably, when determining the explosive composition described in S1, the method for seeking the global optimal cost-performance ratio further includes the following steps:

[0041] A1: Select the ingredients of the explosive and obtain the chemical formula, formation enthalpy (kJ / mol), unit price (yuan / kg) and oxygen balance value (g / g) of each ingredient from the database;

[0042] A2: Establish a mathematical model for calculating RWS by calculating oxygen balance, using the Brinkley-Wilson rule chemical equation, and using the Gass theorem to calculate explosion heat and explosion volume.

[0043]

[0044] Q vi ——calculated detonation heat of explosive, kJ / kg;

[0045] V i ——Calculated explosive capacity of explosive, L / kg;

[0046] Q ANFO ——explosion heat of porous granular ammonium nitrate explosive, kJ / kg;

[0047] V ANFO ——Explosion volume of porous granular ammonium nitrate explosive, L / kg.

[0048] A3: The calculation of the cost is as follows:

[0049]

[0050] x i ——Mass fraction (percentage) of the i-th component, %;

[0051] P i ——Unit price of the ith ingredient, yuan / kg.

[0052] Preferably, when determining the explosive composition described in S2, a method for seeking a globally optimal cost-performance ratio is employed. Issues that need to be considered in the design of the objective function of S2 include: the ratio of explosive performance to explosive cost, the constraint that the sum of the mass fractions of each component is 100%, the constraint on the expected oxygen balance, the maximum cost constraint, and the maximum and minimum constraints on the mass fractions of each component:

[0053] B1: The function F with the cost-effectiveness of explosives as the optimization objective can be expressed as:

[0054]

[0055] B2: Components and constraints:

[0056]

[0057] B3: Specifying the oxygen balance value as a constraint. A better oxygen balance design is conducive to the energy utilization of the explosion system, produces less harmful gases, and provides a better explosion effect. The constraint is expressed as:

[0058]

[0059] OB*—as specified by the designer or calculated using initial input (initial group allocation ratio), g / g;

[0060] B4: Fee Limitation

[0061] cost≤p* (10)

[0062] p*——maximum limit of explosive cost, yuan / kg;

[0063] B5: Constraints on the rationality of the allocation ratio of each group

[0064] loweri ≤x i ≤upper i (11)

[0065] lower i ——Minimum limit of mass fraction of the i-th component, %;

[0066] upper i ——The maximum limit of mass fraction of the i-th component, %.

[0067] Preferably, the algorithm of S3 is the invention of this invention, which is applied to such multi-constraint, nonlinear objective function to solve the optimal input engineering problem, which is fast and reliable, and is specifically implemented as follows:

[0068] C1: Initialization parameters: population size is pop_num, maximum number of iterations is iter_max, strategy control parameter P_cost, expected number of individuals satisfying the constraint is T_mun, strategy control adjustment coefficient α, and the number of independent variables n is used to generate pop_num individuals that weakly violate the constraint as a population;

[0069] C2: Start iteration: In each iteration, a random number is generated and compared with P_cost, and the selection is made;

[0070] C3: Unconstrained single-objective optimization: Use the conventional differential evolution algorithm to perform crossover mutation on the population, use a greedy strategy in selection, merge the mutant population and the contemporary population, sort according to the function objective value, select the first pop_num individuals as the next generation population, and go to C5;

[0071] C4: Constraint satisfaction optimization: Use the conventional differential evolution algorithm to perform crossover mutation on the population. Use the greedy strategy in selection to merge the mutant population and the contemporary population. Calculate the condition value of each constraint and convert it into constraint violation according to formula (12). Then use the non-dominated sorting to sort the constraint violation. Select the first pop_num individuals with higher dominance level as the next generation population and go to C5.

[0072] g i (X)<=b i

[0073]

[0074] c i =0 else (12)

[0075] C5: After sorting according to the constraint dominance, record the number of undominated solutions (i.e., feasible solutions). If there are too few feasible solutions, use formula (13) to update P_cost. If there are too many feasible solutions, use formula (14) to update P_cost. Then, sort according to the objective function value and record the current optimal individual.

[0076] P_cost=(1-α)×P_cost (13)

[0077] P_cost=1-(1-P_cost)×(1-α) (14)

[0078] C6: Determine the iteration termination condition. If it is not met, go to C2 and continue iterating. If it is met, output the optimal individual and end.

[0079] Preferably, in each iteration of C2, a random number is generated and compared with P_cost, and a judgment is made to perform unconstrained single-objective optimization and enter C3.

[0080] Preferably, when the explosive composition is determined, a method for seeking the global optimal cost-performance ratio is used. In each iteration of C2, a random number is generated and compared with P_cost, and a judgment is made to select and perform constraint satisfaction optimization, and then enter C4.

[0081] The technical effects and advantages of the present invention are as follows:

[0082] The present invention provides a method for seeking a globally optimal cost-effective ratio when determining explosive components. In open-pit blasting projects, the present invention provides a method for seeking a globally optimal cost-effective ratio when determining explosive components based on given expected oxygen balance conditions, within a maximum cost limit, and within a reasonable range of the ratios of the components. BRIEF DESCRIPTION OF THE DRAWINGS

[0083] Figure 1 This is a schematic diagram of step one and step two of the present invention.

[0084] Figure 2 This is a schematic diagram of step three of the present invention.

[0085] Figure 3 This is a comparison diagram of the results before and after optimization of the present invention.

[0086] Figure 4 Enumerate feasible solution point diagrams for the present invention. DETAILED DESCRIPTION

[0087] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0088] The present invention provides Figure 1-4 When the explosive composition is determined, a method for seeking the global optimal cost-performance ratio includes the following steps:

[0089] S1: Establish a calculation model. When the composition of the explosive (mixture of multiple substances) is determined, the Brinkley-Wilson rule chemical reaction equation is determined based on the expected value of the oxygen balance of the designed explosive, so as to establish the corresponding RWS (relative weight strength) and corresponding cost calculation model;

[0090] S2: Inequality constraints, which take the cost performance of explosives as the optimization objective, the mass fraction of the components as the variable, and the sum of the mass fractions of the components as 100% and the expected value of the oxygen balance of the designed explosive as the equality constraints. The maximum cost limit and the reasonable range of the mass fractions of the components are the inequality constraints.

[0091] S3: A method for seeking optimal values ​​based on the differential evolution algorithm with constraint satisfaction.

[0092] A method for finding the global optimal cost-effectiveness ratio when the explosive composition is determined is expressed as:

[0093]

[0094]

[0095]

[0096] cost≤p*

[0097] lower i ≤x i ≤upper i (1)

[0098] x i ——mass fraction of the i-th component, %;

[0099] B i ——oxygen balance value of the i-th component, g / g;

[0100] OB*——expected oxygen balance value of explosive, g / g;

[0101] p*——maximum fee limit, yuan / kg;

[0102] lower i ——Minimum limit of mass fraction of the i-th component, %;

[0103] upper i ——The maximum limit of mass fraction of the i-th component, %.

[0104] The optimal value seeking method of S3 based on the constraint satisfaction differential evolution algorithm also includes the following steps:

[0105] Step 1: Initialize the parameters: population size is pop_num, maximum number of iterations is iter_max, strategy control parameter P_cost, expected number of individuals satisfying the constraint is T_mun, strategy control adjustment coefficient α, number of variables n, and generate pop_num individuals that weakly violate the constraint as a population;

[0106] Step 2: Start iteration: In each iteration, generate a random number and compare it with P_cost, and then decide whether to proceed to step 3 or step 4;

[0107] Step 3: Unconstrained single-objective optimization: Use the conventional differential evolution algorithm to perform crossover mutation on the population, use a greedy strategy in selection, merge the mutant population and the contemporary population, sort according to the function objective value, select the first pop_num individuals as the next generation population, and go to step 5;

[0108] Step 4: Constraint satisfaction optimization: Use the conventional differential evolution algorithm to perform crossover mutation on the population. Use the greedy strategy in selection to merge the mutant population and the contemporary population. Calculate the condition value of each constraint and convert it into constraint violation degree according to formula (2). Then use the non-dominated sorting to sort the constraint violation degree. Select the first pop_num individuals with higher dominance level as the next generation population and go to step 5.

[0109] g i (X)<=b i

[0110]

[0111] c i =0 else (2)

[0112] Step 5: After sorting according to the constraint dominance, record the number of undominated solutions (i.e., feasible solutions). If there are too few feasible solutions, use formula (3) to update P_cost. If there are too many feasible solutions, use formula (4) to update P_cost and then sort according to the objective function value and record the current optimal individual.

[0113] P_cost=(1-α)×P_cost (3)

[0114] P_cost=1-(1-P_cost)×(1-α) (4)

[0115] Step 6: Determine the termination condition of the iteration. If it is not met, go to step 2 and continue iterating; if it is met, output the optimal individual and end.

[0116] The optimal value seeking method based on the constraint satisfaction differential evolution algorithm generates a random number and compares it with P_cost in each iteration, and then proceeds to step three when performing unconstrained single-objective optimization.

[0117] The optimal value seeking method of the differential evolution algorithm based on constraint satisfaction, in each iteration, a random number is generated and compared with P_cost, and step four is entered when the constraint satisfaction optimization is performed.

[0118] When an explosive component of S1 is determined, the method for seeking the global optimal cost-performance ratio further includes the following steps:

[0119] A1: Select the ingredients of the explosive and obtain the chemical formula, formation enthalpy (kJ / mol), unit price (yuan / kg) and oxygen balance value (g / g) of each ingredient from the database;

[0120] A2: Establish a mathematical model for calculating RWS by calculating oxygen balance, using the Brinkley-Wilson rule chemical equation, and using the Gass theorem to calculate explosion heat and explosion volume.

[0121]

[0122] Q vi ——calculated detonation heat of explosive, kJ / kg;

[0123] V i ——Calculated explosive capacity of explosive, L / kg;

[0124] Q ANFO ——explosion heat of porous granular ammonium nitrate explosive, kJ / kg;

[0125] V ANFO ——Explosion volume of porous granular ammonium nitrate explosive, L / kg.

[0126] A3: The calculation of the cost is as follows:

[0127]

[0128] x i ——Mass fraction (percentage) of the i-th component, %;

[0129] P i ——Unit price of the ith ingredient, yuan / kg.

[0130] The design of the objective function of S2 needs to consider the following issues: the ratio of explosive performance to explosive cost, the constraint that the sum of the mass fractions of each component is 100%, the constraint on the desired oxygen balance, the maximum cost limit, and the maximum and minimum limits on the mass fractions of each component:

[0131] B1: The function F with the cost-effectiveness of explosives as the optimization objective can be expressed as:

[0132]

[0133] B2: Components and constraints:

[0134]

[0135] B3: Specifying the oxygen balance value as a constraint. A better oxygen balance design is conducive to the energy utilization of the explosion system, produces less harmful gases, and provides a better explosion effect. The constraint is expressed as:

[0136]

[0137] OB*—as specified by the designer or calculated using initial input (initial group allocation ratio), g / g;

[0138] B4: Fee Limitation

[0139] cost≤p* (10)

[0140] p*——maximum limit of explosive cost, yuan / kg;

[0141] B5: Constraints on the rationality of the allocation ratio of each group

[0142] lower i ≤x i ≤upper i (11)

[0143] lower i ——Minimum limit of mass fraction of the i-th component, %;

[0144] upper i ——The maximum limit of mass fraction of the i-th component, %.

[0145] The S3 algorithm is an invention of this time and is applied to engineering problems with multiple constraints and nonlinear objective functions to solve optimal inputs. It is fast and reliable. The specific implementation is as follows:

[0146] C1: Initialization parameters: population size is pop_num, maximum number of iterations is iter_max, strategy control parameter P_cost, expected number of individuals satisfying the constraint is T_mun, strategy control adjustment coefficient α, and the number of independent variables n is used to generate pop_num individuals that weakly violate the constraint as a population;

[0147] C2: Start iteration: In each iteration, a random number is generated and compared with P_cost, and the selection is made;

[0148] C3: Unconstrained single-objective optimization: Use the conventional differential evolution algorithm to perform crossover mutation on the population, use a greedy strategy in selection, merge the mutant population and the contemporary population, sort according to the function objective value, select the first pop_num individuals as the next generation population, and go to C5;

[0149] C4: Constraint satisfaction optimization: Use the conventional differential evolution algorithm to perform crossover mutation on the population. Use the greedy strategy in selection to merge the mutant population and the contemporary population. Calculate the condition value of each constraint and convert it into constraint violation according to formula (12). Then use the non-dominated sorting to sort the constraint violation. Select the first pop_num individuals with higher dominance level as the next generation population and go to C5.

[0150] g i (X)<=b i

[0151]

[0152] c i =0 else (12)

[0153] C5: After sorting according to the constraint dominance, record the number of undominated solutions (i.e., feasible solutions). If there are too few feasible solutions, use formula (13) to update P_cost. If there are too many feasible solutions, use formula (14) to update P_cost. Then, sort according to the objective function value and record the current optimal individual.

[0154] P_cost=(1-α)×P_cost (13)

[0155] P_cost=1-(1-P_cost)×(1-α) (14)

[0156] C6: Determine the iteration termination condition. If it is not met, go to C2 and continue iterating. If it is met, output the optimal individual and end.

[0157] In each iteration of C2, a random number is generated and compared with P_cost, and a judgment is made to perform unconstrained single-objective optimization and enter C3.

[0158] In each iteration of C2, a random number is generated and compared with P_cost, and a judgment is made to perform constraint satisfaction optimization and enter C4.

[0159] The present invention is demonstrated below by taking into account an emulsion explosive with a certain composition (mixed by 7 substances in a specific process).

[0160] Complete step one of the technical solution: query and calculation of relevant data.

[0161] The design takes the ratio of emulsion explosive performance to cost as the optimization target, the mass percentage and the slightly negative oxygen balance (calculated by initial value input) as the equality constraints, and the maximum cost limit and the maximum and minimum limits of each component are determined by experience. Its mathematical expression is:

[0162]

[0163]

[0164]

[0165] cost≤1.7005

[0166] 40≤x1≤80

[0167] 3≤x2≤10

[0168] 8≤x3≤20

[0169] 2≤x4≤5

[0170] 1≤x5≤3

[0171] 2≤x6≤5

[0172] 0.01≤x7≤0.10 (15)

[0173] Based on the global optimal solution search method in step 3, a set of optimal solutions is obtained. The comparison results before and after optimization are as follows: Figure 3 , where the feasible solution point diagram is based on enumeration as Figure 4 .

Claims

1. A method for finding the global optimal cost-performance ratio when determining explosive components, characterized by: The following steps are involved: S1: Establish a calculation model. When the explosive composition is determined, the Brinkley-Wilson rule chemical reaction equation is determined based on the expected value of the oxygen balance of the designed explosive, so as to establish the corresponding RWS and corresponding cost calculation model; S2: Inequality constraints, which takes the performance-price ratio of explosives as the optimization objective and the mass fraction of components as the variable. The sum of the mass fractions of each component is 100% and the expected value of the oxygen balance of the designed explosive is considered as the equality constraint. The maximum cost limit and the reasonable range of the mass fraction of each component are used as the inequality constraints. S3: A method for finding the optimal value based on the differential evolution algorithm with constraint satisfaction; The following steps are also included: A1: Select the ingredients of the explosive and obtain the chemical formula, formation enthalpy, unit price and oxygen balance value of each ingredient from the database; A2: Establish a mathematical model for calculating RWS by calculating oxygen balance, using the Brinkley-Wilson rule chemical equation, and using the Gass theorem to calculate explosion heat and explosion volume. Q vi ——calculated detonation heat of explosive, kJ / kg; V i ——Calculated explosive capacity of explosive, L / kg; Q ANFO ——explosion heat of porous granular ammonium nitrate explosive, kJ / kg; V ANFO ——Explosion volume of porous granular ammonium nitrate explosive, L / kg, A3: The calculation of the cost is as follows: x i ——mass fraction of the i-th component, %; P i ——Unit price of the i-th component, yuan / kg; The design of the objective function of S2 needs to consider the following issues: the ratio of explosive performance to explosive cost, the constraint that the sum of the mass fractions of each component is 100%, the constraint on the desired oxygen balance, the maximum cost constraint, and the maximum and minimum constraints on the mass fractions of each component: B1: The function F with the cost-effectiveness of explosives as the optimization objective can be expressed as: B2: Components and constraints: B3: Specifying the oxygen balance value as a constraint. A better oxygen balance design is conducive to the energy utilization of the explosion system, produces less harmful gases, and provides a better explosion effect. The constraint is expressed as: OB * ——As specified by the designer or calculated using the initial input initial group ratio, g / g; B4: Fee Limitation cost≤p * (10) p * ——The maximum limit of explosive cost, yuan / kg; B5: Constraints on the rationality of the allocation ratio of each group lower i ≤x i ≤upper i (11) lower i ——Minimum limit of mass fraction of the i-th component, %; upper i ——The maximum limit of mass fraction of the i-th component, %.

2. The method for seeking the global optimal cost-performance ratio when determining explosive components according to claim 1, characterized in that: It is expressed as: cost≤p * lower i ≤x i ≤upper i (1) x i ——mass fraction of the i-th component, %; B i ——oxygen balance value of the i-th component, g / g; OB * —expected oxygen balance value of explosive, g / g; p * ——Maximum fee limit, RMB / kg; lower i ——Minimum limit of mass fraction of the i-th component, %; upper i ——The maximum limit of mass fraction of the i-th component, %.

3. The method for seeking the global optimal cost-performance ratio when determining explosive components according to claim 1, characterized in that: The following steps are also included: Step 1: Initialize the parameters: population size is pop_num, maximum number of iterations is iter_max, strategy control parameter P_cost, expected number of individuals satisfying the constraint is T_mun, strategy control adjustment coefficient α, number of variables n, and generate pop_num individuals that weakly violate the constraint as a population; Step 2: Start iteration: In each iteration, generate a random number and compare it with P_cost, and then decide whether to proceed to step 3 or step 4; Step 3: Unconstrained single-objective optimization: Use the conventional differential evolution algorithm to perform crossover mutation on the population, use a greedy strategy in selection, merge the mutant population and the contemporary population, sort according to the function objective value, select the first pop_num individuals as the next generation population, and go to step 5; Step 4: Constraint satisfaction optimization: Use the conventional differential evolution algorithm to perform crossover mutation on the population. Use the greedy strategy in selection to merge the mutant population and the contemporary population. Calculate the condition value of each constraint and convert it into constraint violation degree according to formula (2). Then use the non-dominated sorting to sort the constraint violation degree. Select the first pop_num individuals with higher dominance level as the next generation population and go to step 5. g i (X)<=b i if g i (X)>b i ci=0else(2) Step 5: After sorting according to the constraint dominance, record the number of non-dominated solutions. If there are too few feasible solutions, use formula (3) to update P_cost. If there are too many feasible solutions, use formula (4) to update P_cost and then sort according to the objective function value and record the current optimal individual. P_cost=(1-α)×P_cost(3) P_cost=1-(1-P_cost)×(1-α)(4) Step 6: Determine the iteration termination condition. If it is not met, go to step 2 and continue iterating. If it is met, output the optimal individual and end.

4. The method for seeking the global optimal cost-performance ratio when determining explosive components according to claim 3, characterized in that: In each iteration, a random number is generated and compared with P_cost, and step three is entered for unconstrained single-objective optimization.

5. The method for seeking the global optimal cost-performance ratio when determining explosive components according to claim 3, characterized in that: In each iteration, a random number is generated and compared with P_cost, and step 4 is entered when the constraint satisfaction optimization is performed.

6. The method for seeking the global optimal cost-performance ratio when determining explosive components according to claim 1, characterized in that: The S3 algorithm is an invention of this invention and is applied to engineering problems with multiple constraints and nonlinear objective functions to solve optimal inputs. It is fast and reliable and is specifically implemented as follows: C1: Initialization parameters: population size is pop_num, maximum number of iterations is iter_max, strategy control parameter P_cost, expected number of individuals satisfying the constraint is T_mun, strategy control adjustment coefficient α, and the number of independent variables n is used to generate pop_num individuals that weakly violate the constraint as a population; C2: Start iteration: In each iteration, a random number is generated and compared with P_cost, and the selection is made; C3: Unconstrained single-objective optimization: Use the conventional differential evolution algorithm to perform crossover mutation on the population, use a greedy strategy in selection, merge the mutant population and the contemporary population, sort according to the function objective value, select the first pop_num individuals as the next generation population, and go to C5; C4: Constraint satisfaction optimization: Use the conventional differential evolution algorithm to perform crossover mutation on the population. Use the greedy strategy in selection to merge the mutant population and the contemporary population. Calculate the condition value of each constraint and convert it into constraint violation according to formula (12). Then use the non-dominated sorting to sort the constraint violation. Select the first pop_num individuals with higher dominance level as the next generation population and go to C5. g i (X)<=b i if g i (X)>b i ci=0 else (12) C5: After sorting according to the constraint dominance, record the number of non-dominated solutions. If there are too few feasible solutions, use formula (13) to update P_cost. If there are too many feasible solutions, use formula (14) to update P_cost. Then, sort according to the objective function value and record the current optimal individual. P_cost=(1-α)×P_cost (13) P_cost=1-(1-P_cost)×(1-α) (14) C6: Determine the iteration termination condition. If it is not met, go to C2 and continue iterating. If it is met, output the optimal individual and end.

7. The method for seeking the global optimal cost-performance ratio when determining explosive components according to claim 6, characterized in that: In each iteration of C2, a random number is generated and compared with P_cost, and a judgment is made to perform unconstrained single-objective optimization and enter C3.

8. The method for seeking the global optimal cost-performance ratio when determining explosive components according to claim 6, characterized in that: In each iteration of C2, a random number is generated and compared with P_cost, and a judgment is made to perform constraint satisfaction optimization and enter C4.

Images