Ingredient optimization method and system

By constructing a ingredients optimization model and using genetic algorithms to solve it, the problem of insufficient stability and accuracy of ingredients solutions in traditional ingredients methods is solved, and real-time optimization and stability improvement of ingredients solutions in metallurgy and chemical industries is achieved.

CN114266526BActive Publication Date: 2024-11-29CISDI ENGINEERING CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202210068004.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-01-20
Publication Date
2024-11-29
Estimated Expiration
2042-01-20

AI Technical Summary

Technical Problem

Traditional ingredients methods rely on experience and are difficult to ensure the accuracy and stability of ingredients solutions, especially in metallurgy and chemical industries that require high stability.

Method used

By obtaining raw material data, calculating the ingredients, price and inventory stability indicators of raw materials, building a ingredient optimization model, combining genetic algorithms to solve it, and finally obtaining the optimal ingredient solution.

Benefits of technology

It realizes real-time optimization of the ingredients solution during the production process, ensures the stability of product quality, and improves the accuracy and stability of the ingredients solution.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114266526B_ABST
    Figure CN114266526B_ABST
Patent Text Reader

Abstract

The present invention proposes a batching optimization method and system, comprising: obtaining raw material data, obtaining corresponding raw material stability indicators according to the raw material data, wherein the raw material stability indicators include: component stability indicators, price stability indicators and inventory stability indicators; constructing an objective function and conditional constraints of a batching optimization model according to the raw material stability indicators, wherein the conditional constraints include: ratio constraints, component constraints and particle size constraints; inputting the batching cost and the raw material stability indicators into the batching optimization model to obtain an optimal batching plan; the present invention can realize online optimization of the batching plan and effectively guide the smooth development of production activities.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of intelligent metallurgy, and in particular to a batching optimization method and system. Background Art

[0002] In process manufacturing industries such as metallurgy and chemical industry, the batching process of various raw materials is involved. The raw material composition, particle size and other attribute indicators in the batching process have a great influence on the product performance. In the traditional batching method, the main considerations are the product ingredient requirement constraints, the material matching experience constraints, and the batching cost goals. On this basis, linear programming, traversal solution, and bionic algorithms are used to optimize the batching scheme.

[0003] However, in actual production, we need to consider not only the above requirements, but also the continuity and stability of industrial production, especially the continuous process manufacturing represented by the metallurgical and chemical industries, which have high requirements for the stability of product quality. The stability of the batching scheme is an important indicator for evaluating the batching scheme, which includes the requirements for the stability of raw material ingredients, storage stability and price stability. The traditional method that relies on experience is difficult to guarantee the accuracy and stability of the acquisition plan, and it is difficult to apply it to production activities with high requirements for stability. Summary of the invention

[0004] In view of the above problems existing in the prior art, the present invention proposes a batching optimization method and system, which mainly solves the problem that the existing method relies on experience to generate a batching scheme with insufficient stability and accuracy.

[0005] In order to achieve the above purpose and other purposes, the technical solution adopted by the present invention is as follows.

[0006] A batching optimization method, comprising:

[0007] Acquire raw material data, and acquire corresponding raw material stability indicators according to the raw material data, wherein the raw material stability indicators include: ingredient stability indicators, price stability indicators, and inventory stability indicators;

[0008] Constructing the objective function and conditional constraints of the batching optimization model according to the raw material stability index, wherein the conditional constraints include: ratio constraints, component constraints and particle size constraints;

[0009] The batching cost and the raw material stability index are input into the batching optimization model to obtain the optimal batching solution.

[0010] Optionally, the calculation method of the component stability index includes:

[0011]

[0012]

[0013] Among them, q ki represents the feeding amount of material K in feeding time period i, σ ki(a1) Represents the standard deviation of component a1 of material K, σ ki(a2) Represents the standard deviation of component a2 of material K, σ ki(an) Indicates the standard deviation of the component an of material K; C kj is the ratio of the kth material in the batching scheme j.

[0014] Optionally, the price stability index is calculated by:

[0015] Obtain price data of the corresponding material in a historical time period T, obtain price distribution of the price data, and divide the price distribution into multiple continuous price intervals;

[0016] According to the frequency of occurrence of the price data in each price range, the probability of the price data falling into each price range is obtained;

[0017] The price stability index is calculated based on the probability that the price data falls into each price range.

[0018] Optionally, the price stability indicator is expressed as:

[0019]

[0020] Among them, P k is the probability that the current price of material k belongs to the price range; C kj is the percentage of material k in batching scheme j; δ pj is the price stability index of ingredient solution j.

[0021] 10. The ingredient optimization method according to claim 1, characterized in that the calculation method of the inventory stability index includes:

[0022]

[0023]

[0024]

[0025] Among them, Y is the average daily planned output; D is the planned material cycle; W k is the current inventory of material k; C kj is the percentage of material k in batching scheme j; μ is the yield coefficient in the production process.

[0026] 11. The batching optimization method according to claim 5, characterized in that the value range of the yield coefficient is 0.6-0.9, or it is calculated by the proportion of the residual amount after deducting moisture and burnt loss in the mixed material of batching scheme j, and the calculation method is:

[0027] μ1=(1-Loi j )*(1-H2O j )

[0028] Among them, H2O j Excluding the water content, Loi j Indicates the burn loss ratio.

[0029] Optionally, constructing an objective function and conditional constraints of a batching optimization model according to the raw material stability index includes:

[0030] The objective function is expressed as:

[0031]

[0032] The ratio constraint is expressed as:

[0033] ∑X k =1

[0034] XL k ≤X k ≤XH k

[0035] The composition constraint is expressed as:

[0036]

[0037] The granularity constraint is expressed as:

[0038]

[0039] Among them, M i P is the consumption of tons of raw materials in the batching scheme; k is the current price of the kth raw material, in yuan / ton; X k is the kth raw material ratio; XL i and XH i are the lower and upper limits of the i-th raw material, in %; C k,j is the percentage of component j in raw material k, Loi is the loss on burning, H2O is the percentage of water; CL j and CH j are the lower and upper limits of component j, respectively, in %; G k,d is the percentage of particle size index d of raw material k; GL d and GH dThey are respectively the lower limit and upper limit of the mixed powder particle size index d.

[0040] Optionally, inputting the batching cost and the raw material stability index into the batching optimization model to obtain the optimal batching solution includes:

[0041] The optimal solution of the batching optimization model is obtained by a genetic algorithm as the optimal batching scheme, which includes the following steps:

[0042] Random initialization generates a population size of size = Q parent individuals, where each parent individual corresponds to a batching scheme;

[0043] Calculate the fitness value of each individual;

[0044] Perform crossover and mutation operations on individuals according to the fitness values ​​to generate new Qx individuals, with the crossover probability set to p1 and the mutation probability set to p2;

[0045] Select the top Q-Qx individuals in the parent generation in terms of fitness value as the parent generation elites, and form the next generation population with the newly added individuals;

[0046] Let the number of iterations k = k + 1;

[0047] Repeat the above steps until the number of iterations k = the total number of iterations K;

[0048] The individuals of the last evolved offspring population are sorted and the individuals with the highest fitness are selected as the optimal batching solution.

[0049] A batching optimization system, comprising:

[0050] An index acquisition module, used to acquire raw material data, and acquire corresponding raw material stability indexes according to the raw material data, wherein the raw material stability indexes include: ingredient stability index, price stability index and inventory stability index;

[0051] A model building module, used to build an objective function and condition constraints of a batching optimization model according to the raw material stability index, wherein the condition constraints include: a ratio constraint, a component constraint, and a particle size constraint;

[0052] The solution optimization module is used to input the ingredient cost and the raw material stability index into the ingredient optimization model to obtain the optimal ingredient solution.

[0053] As described above, the method and system for optimizing ingredients of the present invention have the following beneficial effects.

[0054] Through online ingredient detection, the real-time composition of each raw material in the actual batching is measured, and the stability index of each raw material component is calculated. Combined with the batching cost and the inventory and price stability index of the raw materials, a multi-objective evaluation system is established for batching. The batching plan can be optimized in real time during the production process to ensure the smooth progress of production activities. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 Schematic diagram of the process of optimizing ingredients in one embodiment of the present invention.

[0056] Figure 2 The figure is a schematic diagram of the process of genetic algorithm optimization in one embodiment of the present invention. DETAILED DESCRIPTION

[0057] The following describes the embodiments of the present invention by specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict.

[0058] It should be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present invention, and thus the drawings only show components related to the present invention rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed arbitrarily, and the component layout may also be more complicated.

[0059] See also Figure 1 The present invention provides a method for optimizing ingredients, comprising the following steps:

[0060] Step S01, obtaining raw material data, and obtaining corresponding raw material stability indicators according to the raw material data, wherein the raw material stability indicators include: component stability indicators, price stability indicators and inventory stability indicators;

[0061] Step S02, constructing an objective function and condition constraints of a batching optimization model according to the raw material stability index, wherein the condition constraints include: a batching ratio constraint, a component constraint, and a particle size constraint;

[0062] Step S03, inputting the batching cost and the raw material stability index into the batching optimization model to obtain the optimal batching solution.

[0063] Specifically, taking the optimization of ingredients in the sintering process of steel metallurgy as an example, the raw materials for metallurgical production usually include TFe, Al and Si.

[0064] 1. Data Collection

[0065] (1) A weight detector and an online composition detector are installed on the batching belt to collect and store the composition and weight data of different mineral powder raw materials supplied in real time.

[0066] (2) The external MES system collects and stores the actual execution information of different raw material supply in the batching process, mainly including the name of the raw material, the start time of the supply, and the end time of the supply;

[0067] (3) Collect and store price and inventory information data of raw materials used in the production process;

[0068] 2. Calculation of raw material composition stability

[0069] (1) Calculate the feeding quantity q of material k in each feeding time period i ki , standard deviation σ of TFe ki(TFe) , standard deviation σ of Al ki(Al) , the standard deviation σ of Si ki(si) .

[0070] (2) Within the set time T (T = 90 days, assuming that there are R feeding time periods within the time T), a total of 10 materials are involved in batching, and the component stability index of material k is calculated based on its various components.

[0071]

[0072]

[0073] In the formula, C kj is the ratio of the kth material in scheme j, %.

[0074] 3. Calculation of raw material price stability

[0075] Material price stability index δ P , based on the price data within the historical time T, the historical price distribution of material k is divided into 20 consecutive intervals;

[0076]

[0077] Max k is the historical maximum price of material k;

[0078] Min k is the historical minimum price of material k;

[0079] n is the price range step, set n = 5;

[0080] Calculate the probability P of each interval based on the frequency of price data in 20 intervalski .

[0081]

[0082] m ki is the frequency of the historical price of material k being within interval i;

[0083] M k The total frequency of price records for material k.

[0084]

[0085] P k is the probability that the current price of material k belongs to the historical price range, %;

[0086] C kj is the percentage of material k in batching scheme j, %;

[0087] δ pj is the price stability index of ingredient solution j.

[0088] 4. Calculation of raw material inventory stability

[0089] Inventory stability index δ S It is to evaluate the degree to which the raw materials involved in the batching scheme can meet the planned material cycle. The inventory stability index is used to evaluate the quality of the batching scheme. Therefore, the raw material inventory stability corresponding to a certain scheme j is δ Sj .

[0090]

[0091]

[0092]

[0093] Y is the average daily planned output, tons;

[0094] D is the planned material cycle, days;

[0095] W k is the current inventory of material k, tons;

[0096] C kj is the percentage of material k in batching scheme j, %;

[0097] μ=0.86.

[0098] 5. Establish a batching calculation model

[0099] Objective function:

[0100]

[0101] Ratio constraints:

[0102] ∑X k =1

[0103] XL k ≤X k ≤XH k

[0104] Ingredient constraints:

[0105]

[0106] Granularity constraints:

[0107]

[0108] Among them, M i P is the iron consumption per ton of the kth raw material in the batching scheme; k is the current price of the kth raw material, in yuan / ton; X k is the kth raw material ratio; XL i and XH i are the lower and upper limits of the i-th raw material, in %; C k,j is the percentage of component j in raw material k, loi is the loss on burning, H2O is the water content, the unit is %; CL j and CH j are the lower and upper limits of component j, respectively, in %; G k,d is the percentage of particle size index d of raw material k, unit %; GL d and GH d They are respectively the lower limit and upper limit of the mixed powder particle size index d, in %.

[0109] Table 1 Examples of sinter composition constraint values

[0110] Material Name TFe <![CDATA[SiO 2 ]]> CaO MgO <![CDATA[Al 2 THE 3 ]]> P S <![CDATA[CL j ]]> 53.00 4 6 1 0 0 0 <![CDATA[CH j ]]> 62.50 5.3 7.5 2 2.7 0.01 0.02

[0111] 6. Establish a genetic algorithm to solve

[0112] See also Figure 2 , the constraint programming model in the above step (5) is solved by using a genetic algorithm to find the optimal solution of the model. The steps are as follows:

[0113] (1) Randomly initialize the population size to 1000 parent individuals (dosage plan);

[0114] (2) Calculate the fitness value of each individual;

[0115] (3) Perform crossover and mutation operations on individuals according to their fitness values ​​to generate new Qx individuals, with a crossover probability of p1 = 0.3 and a mutation probability of p2 = 0.1;

[0116] (4) Select the top 1000-Qx individuals in the parent generation in terms of fitness value as the parent generation elites, and form the next generation population with the newly added individuals;

[0117] (5) The number of iterations k = k + 1;

[0118] (6) Repeat the above 2 to 5 until the number of iterations k = the total number of iterations K;

[0119] (7) Sort the individuals in the last evolved offspring population and select the one with the highest fitness as the optimal batching solution.

[0120] It is worth noting that the databases of the present invention include but are not limited to ORACLE, DB2, SQL Server, Sybase, Informix, MySQL, VF and Access. The above-mentioned material weight measuring equipment includes but is not limited to weighing instruments of different sensor types such as photoelectric, hydraulic, electromagnetic, capacitive, magnetic pole change, vibration, gyroscopic, and resistance strain. The above-mentioned material component detection equipment includes but is not limited to X-ray fluorescence spectrometer. The processing of constraints in the above-mentioned genetic algorithm includes but is not limited to the feasible matrix method, penalty function and other processing methods.

[0121] This embodiment also provides a batch optimization system for executing the batch optimization method described in the above method embodiment. Since the technical principle of the system embodiment is similar to that of the above method embodiment, the same technical details will not be repeated.

[0122] In one embodiment, an ingredient optimization system includes: an index acquisition module, which is used to acquire raw material data and acquire corresponding raw material stability indicators based on the raw material data, wherein the raw material stability indicators include: ingredient stability indicators, price stability indicators and inventory stability indicators; a model construction module, which is used to construct an objective function and conditional constraints of an ingredient optimization model based on the raw material stability indicators, wherein the conditional constraints include: ratio constraints, ingredient constraints and particle size constraints; a solution optimization module, which is used to input ingredient costs and the raw material stability indicators into the ingredient optimization model to obtain an optimal ingredient solution.

[0123] The above embodiments are merely illustrative of the principles and effects of the present invention, and are not intended to limit the present invention. Anyone familiar with the art may modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by a person of ordinary skill in the art without departing from the spirit and technical concept disclosed by the present invention shall still be covered by the claims of the present invention.

Claims

1. A batching optimization method, characterized in that: include: Obtain raw material data, and obtain corresponding raw material stability indicators according to the raw material data, wherein the raw material stability indicators include: component stability indicators, price stability indicators, and inventory stability indicators; the calculation method of the component stability indicators includes: Among them, q ki represents the feeding amount of material K in feeding time period i, σ ki(a1) Represents the standard deviation of component a1 of material K, σ ki(a2) Represents the standard deviation of component a2 of material K, σ ki(an) Indicates the standard deviation of the component an of material K; C kj is the ratio of the kth material in the batching scheme j; The price stability index is expressed as: Among them, P k is the probability that the current price of material k belongs to the price range; C kj is the percentage of material k in batching scheme j; δ pj is the price stability index of ingredient solution j; The calculation method of the inventory stability index includes: Among them, Y is the average daily planned output; D is the planned material cycle; W k is the current inventory of material k; C kj is the percentage of material k in batching scheme j; μ is the yield coefficient in the production process; Constructing the objective function and conditional constraints of the batching optimization model according to the raw material stability index, wherein the conditional constraints include: ratio constraints, component constraints and particle size constraints; The batching cost and the raw material stability index are input into the batching optimization model to obtain the optimal batching solution.

2. The method for optimizing ingredients according to claim 1, characterized in that: The calculation method of the price stability index includes: Obtain price data of the corresponding material in a historical time period T, obtain price distribution of the price data, and divide the price distribution into multiple continuous price intervals; According to the frequency of occurrence of the price data in each price range, the probability of the price data falling into each price range is obtained; The price stability index is calculated based on the probability that the price data falls into each price range.

3. The ingredient optimization method according to claim 1, characterized in that: The value range of the yield coefficient is 0.6-0.9, or it can be calculated by the proportion of the residual amount after deducting the moisture and the burning loss in the mixed material of the batching scheme j, and the calculation method is: μ=(1-Loi j )*(1-H2O j ) Among them, H2O j Excluding the water content, Loi j Indicates the burn loss ratio.

4. The method for optimizing ingredients according to claim 1, characterized in that: The objective function and condition constraints of the batching optimization model are constructed according to the raw material stability index, including: The objective function is expressed as: The ratio constraint is expressed as: ∑X k =1 XL k ≤X k ≤XH k The composition constraint is expressed as: The granularity constraint is expressed as: Among them, M i P is the consumption of tons of raw materials in the batching scheme; k is the current price of the kth raw material, in yuan / ton; X k is the kth raw material ratio; XL k and XH k are the lower and upper limits of the kth raw material, in %; C k,j is the percentage content of component j in raw material k; CL j and CH j are the lower and upper limits of component j, respectively, in %; G k,d is the percentage of particle size index d of raw material k; GL d and GH d They are respectively the lower limit and upper limit of the mixed powder particle size index d.

5. The method for optimizing ingredients according to claim 1, characterized in that: Inputting the batching cost and the raw material stability index into the batching optimization model to obtain the optimal batching solution, including: The optimal solution of the batching optimization model is obtained by a genetic algorithm as the optimal batching scheme, which includes the following steps: Random initialization generates a population size of size = Q parent individuals, where each parent individual corresponds to a batching scheme; Calculate the fitness value of each individual; Perform crossover and mutation operations on individuals according to the fitness values ​​to generate new Qx individuals, with the crossover probability set to p1 and the mutation probability set to p2; Select the top Q-Qx individuals in the parent generation in terms of fitness value as the parent generation elites, and form the next generation population with the newly added individuals; Let the number of iterations k = k + 1; Repeat the above steps until the number of iterations k = the total number of iterations K; The individuals of the last evolved offspring population are sorted and the individuals with the highest fitness are selected as the optimal batching solution.

6. A batching optimization system, characterized in that: include: The index acquisition module is used to acquire raw material data, and acquire the corresponding raw material stability index according to the raw material data, wherein the raw material stability index includes: component stability index, price stability index and inventory stability index; the calculation method of the component stability index includes: Among them, q ki represents the feeding amount of material K in feeding time period i, σ ki(a1) Represents the standard deviation of component a1 of material K, σ ki(a2) Represents the standard deviation of component a2 of material K, σ ki(an) Indicates the standard deviation of the component an of material K; C kj is the ratio of the kth material in the batching scheme j; The price stability index is expressed as: Among them, P k is the probability that the current price of material k belongs to the price range; C kj is the percentage of material k in batching scheme j; δ pj is the price stability index of ingredient solution j; The calculation method of the inventory stability index includes: Among them, Y is the average daily planned output; D is the planned material cycle; W k is the current inventory of material k; C kj is the percentage of material k in batching scheme j; μ is the yield coefficient in the production process; A model building module, used to build an objective function and condition constraints of a batching optimization model according to the raw material stability index, wherein the condition constraints include: a ratio constraint, a component constraint, and a particle size constraint; The solution optimization module is used to input the ingredient cost and the raw material stability index into the ingredient optimization model to obtain the optimal ingredient solution.