Molten iron pretreatment effect big data intelligent online evaluation and optimization method

Abstract

The invention belongs to the technical field of ferrous metallurgy, and particularly relates to a big data intelligent online evaluation and optimization method for a molten iron pretreatment effect, which can intelligently retrieve corresponding data from big data systems such as production and quality management of iron and steel enterprises after molten iron pretreatment is finished. Molten iron pretreatment effect evaluation of carbon emission, metallurgical efficiency improvement, cost control, steel quality control and the like of a current pretreatment furnace is completed immediately, evaluation is completed rapidly, relevant optimization can be conducted immediately, the defects that in the past, manual evaluation of the molten iron pretreatment production effect is not in time and not comprehensive, and carbon dioxide emission evaluation is lacked are overcome, and the production efficiency is improved. The method plays an important role in molten iron pretreatment low-carbon control, efficiency improvement and steel quality control improvement in China.

Description

Technical Field

[0001] The present invention relates to the technical field of iron and steel metallurgy, and particularly to a big data intelligent online evaluation and optimization method for the effect of hot metal pretreatment. Background Technique

[0002] Hot metal pretreatment is an important process for the production of high-quality steel. Its energy conservation, consumption reduction, efficiency improvement, and cost control are particularly important for enhancing the new quality productivity of modern iron and steel production in China and achieving low-carbon and efficient iron and steel smelting. However, for a long time, due to the lack of a big data system for iron and steel production and the absence of carbon emission management, the evaluation of the carbon emission intensity level of hot metal pretreatment production has not been carried out. The evaluation of the production cost control, pretreatment efficiency, and operation level of hot metal pretreatment cannot be carried out online and immediately, and it cannot timely feedback and reflect the effects and levels of the current heat pretreatment process in many important aspects such as carbon emission control, efficiency, and cost control. It cannot be evaluated, analyzed, and optimized in time, which is not conducive to the optimization of the hot metal pretreatment process and operation level, nor to the improvement of production management. Summary of the Invention

[0003] To solve the problems existing in the prior art, the main object of the present invention is to propose a big data intelligent online evaluation and optimization method for the effect of hot metal pretreatment.

[0004] According to one aspect of the present invention, the following technical solution is provided:

[0005] A big data intelligent online evaluation and optimization method for the effect of hot metal pretreatment. After the hot metal pretreatment is completed, the big data of the hot metal pretreatment process of this heat is used to determine the metallurgical effect factor of this heat. According to the metallurgical effect factor of the heat, the effect of hot metal pretreatment is evaluated. At the same time, according to the evaluation results, an artificial intelligence algorithm is used to analyze the subjective and objective factors affecting the effect of hot metal pretreatment and determine the optimization direction. In the present invention, the effects of hot metal pretreatment include carbon dioxide emission intensity, cost, efficiency, temperature control, and end-point sulfur control, etc.

[0006] As a preferred scheme of the big data intelligent online evaluation and optimization method for the effect of hot metal pretreatment described in the present invention, wherein: the metallurgical effect factor R is used to evaluate the comprehensive metallurgical effect of the corresponding heat. The metallurgical effect factor R of this heat ranks in the top 10% within the recent evaluation period of the corresponding type of hot metal pretreatment, and the evaluation result is excellent; when it ranks in the top 20%, the evaluation result is determined to be good; when it is lower than the top 20%, improvement is required.

[0007] As a preferred scheme of the big data intelligent online evaluation and optimization method for the effect of hot metal pretreatment described in the present invention, wherein: the metallurgical effect factor R is:

[0008] R = a1×CE + a2×EF + a3×ΔT + a4×S + a5×Cost + a6×Slag

[0009] In the formula: R is the metallurgical effect factor;

[0010] a1 is the carbon emission intensity weighting factor; CE is the carbon emission intensity level;

[0011] a2 is the efficiency weighting factor; EF is the efficiency level;

[0012] a3 is the temperature drop control weighting factor during the pretreatment process; ΔT is the temperature drop control level during the pretreatment process;

[0013] a4 is the sulfur control weighting factor at the end of the pretreatment; S is the sulfur control level at the end of the pretreatment;

[0014] a5 is the cost control weighting factor during the pretreatment process; Cost is the cost control level during the pretreatment process;

[0015] a6 is the slag removal control weighting factor after desulfurization; Slag is the slag removal control level during the pretreatment process.

[0016] The beneficial effects of the present invention are as follows:

[0017] The present invention proposes a big data intelligent online evaluation and optimization method for the effect of hot metal pretreatment. After the hot metal pretreatment is completed, corresponding data can be intelligently retrieved from big data systems such as steel enterprise production and quality management. Immediately complete the evaluation of the hot metal pretreatment effect such as carbon emission, improvement of metallurgical efficiency, cost control, and steel quality control for the current pretreatment heat. If the carbon emission intensity level, efficiency level, process temperature drop control level, end sulfur control level, cost control level, and slag removal control level of this heat are lower than 0.7, it indicates that the control in this aspect is poor and corresponding improvements are needed. The evaluation is completed quickly and relevant optimizations can be carried out immediately, overcoming the shortcomings of untimely and incomplete manual evaluation of the production effect of hot metal pretreatment in the past and the lack of carbon dioxide emission evaluation, and playing an important role in the low-carbon control, efficiency improvement, and steel quality control improvement of hot metal pretreatment in China. Specific embodiments

[0018] The technical solutions in the embodiments will be described clearly and completely below. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0019] The present invention provides a big data intelligent online evaluation and optimization method for the effect of hot metal pretreatment. By using the metallurgical big data systems being established by various steel enterprises, a method for evaluating the carbon dioxide emission intensity, pretreatment effect and operation level of hot metal pretreatment is established. This method immediately applies the big data during the hot metal pretreatment process to online weighted intelligent evaluation of the pretreatment effect, operation level and carbon dioxide emission intensity of this heat of hot metal. The evaluation model comprehensively considers the carbon dioxide emission intensity, cost, pretreatment efficiency, temperature control and end-point sulfur control of this heat of hot metal pretreatment, etc. Through the weighted coefficient, it intelligently evaluates the comprehensive metallurgical effect and operation level of this heat of hot metal pretreatment, and adjusts the focus direction of the intelligent optimization of the hot metal pretreatment effect evaluation through the weighted coefficient. At the same time, by applying artificial intelligence algorithms, it analyzes the subjective and objective factors affecting the hot metal pretreatment effect and determines the optimization direction.

[0020] According to one aspect of the present invention, the present invention provides the following technical solution:

[0021] A big data intelligent online evaluation and optimization method for the effect of hot metal pretreatment. After the hot metal pretreatment is completed, the big data during the hot metal pretreatment process of this heat is used to determine the metallurgical effect factors of this heat, and the pretreatment effect of hot metal is comprehensively evaluated according to the metallurgical effect factors of the heat. At the same time, according to the evaluation results, by applying artificial intelligence algorithms, it analyzes the subjective and objective factors affecting the hot metal pretreatment effect and determines the optimization direction. In the present invention, the hot metal pretreatment effect includes carbon dioxide emission intensity, cost, efficiency, temperature control and end-point sulfur control, etc.

[0022] Preferably, the metallurgical effect factor R is used to evaluate the comprehensive metallurgical effect of the corresponding heat. If the metallurgical effect factor R of this heat ranks in the top 10% within the recent evaluation period (week, month, quarter, year) of the corresponding category of hot metal pretreatment, the evaluation result is excellent; if it ranks in the top 20%, the evaluation result is determined to be good; if it is lower than the top 20%, improvement is required.

[0023] Preferably, the metallurgical effect factor R is:

[0024] R = a1×CE + a2×EF + a3×ΔT + a4×S + a5×Cost + a6×Slag

[0025] In the formula: R is the metallurgical effect factor;

[0026] a1 is the carbon emission intensity weighting factor, and its value ranges from 0 to 0.9; CE is the carbon emission intensity level;

[0027] a2 is the efficiency weighting factor, and its value ranges from 0.1 to 0.9; EF is the efficiency level;

[0028] a3 is the temperature drop control weighting factor during the pretreatment process, and its value range is 0 to 0.6; ΔT is the temperature drop control level during the pretreatment process;

[0029] a4 is the weighted factor for sulfur control at the end of pretreatment, with a value range of 0 - 0.8; S is the sulfur control level at the end of pretreatment;

[0030] a5 is the weighted factor for cost control during pretreatment, with a value range of 0 - 0.8; Cost is the cost control level during pretreatment;

[0031] a6 is the weighted factor for slag removal control after desulfurization, with a value range of 0.2 - 0.8; Slag is the slag removal control level during pretreatment.

[0032] Preferably, the calculation methods for the carbon emission intensity level CE, efficiency level EF, temperature drop control level ΔT during pretreatment, sulfur control level S at the end of pretreatment, cost control level Cost during pretreatment, and slag removal control level Slag during pretreatment are as follows:

[0033] CE = 2 - Ce / Ce min

[0034] In the formula: Ce is the carbon dioxide emission intensity during hot metal pretreatment, tons of carbon dioxide per ton of iron; Ce min is the minimum carbon dioxide emission intensity within a recent evaluation period (week, month, quarter, year), tons of carbon dioxide per ton of iron; and Ce = {ΔW × Ke + ∑SM i × K Si + ∑G i × K Gi} / W HM

[0035] In the formula: ΔW is the power consumption during hot metal pretreatment, kW·h; Ke is the carbon dioxide emission factor for power consumption in hot metal pretreatment of this enterprise, tons of carbon dioxide per kW·h, taking the value of the carbon dioxide emission factor for electricity consumption of this enterprise verified by the national authoritative department; SM i is the mass of desulfurizer i added during pretreatment, tons; K Si is the carbon emission factor of desulfurizer i, tons of carbon dioxide per ton; G i is the volume of gas i consumed during pretreatment, m 3 ; K Gi is the carbon emission factor of gas i, tons of carbon dioxide per m 3 ; W HM is the mass of pretreated hot metal, tons.

[0036] EF = Ef / Ef max

[0037] In the formula: Ef is the pretreatment efficiency; Ef max is the highest pretreatment efficiency within a recent evaluation period (week, month, quarter, year), and Ef = {[%S] 0-[%S] e} / t

[0038] Where: [%S] 0 , [%S] e are the initial and final sulfur contents of hot metal pretreatment, wt%; t is the pretreatment time.

[0039] ΔT = 2 - {T 0 - T e} / {T 0 - T e} min

[0040] Where: T e , T 0 are the final and initial temperatures of hot metal pretreatment, K; {T 0 - T e} min is the minimum temperature drop during the pretreatment process in the recent evaluation period (week, month, quarter, year).

[0041] S = 2 - (0.002 + [%S] e ) / (0.002 + [%S] min )

[0042] Where: [%S] e is the sulfur content of molten iron at the end of pretreatment, wt%; [%S] min is the minimum sulfur content of molten iron at the end of pretreatment in the recent evaluation period (week, month, quarter, year), wt%;

[0043] Cost = 2 - Co / Co Min

[0044] Where: Co is the cost of hot metal pretreatment for this heat, including the costs of desulfurizer, electricity, gas consumed during pretreatment and the cost of iron loss, yuan; Co Min is the minimum value of the pretreatment cost for this steel grade or similar steel grades in the recent evaluation period (week, month, quarter, year), yuan;

[0045] And Co = {ΔW × Pr e + ∑SM i × Pr Si + ∑G i × Pr Gi + M L × Pr L} / W HM

[0046] Where: Pr e is the unit price of electricity, yuan / kW·h; Pr Si is the unit price of desulfurizer i, yuan / ton; PrGi is the unit price of gas i, yuan / m 3 ; M L is the iron loss in the pretreatment process, tons; Pr L is the unit price of hot metal, yuan / ton.

[0047] If there is no slag skimming after hot metal pretreatment, Slag = 0;

[0048] If slag skimming is carried out after hot metal pretreatment, Slag = 2 - (1 + Qslag) / (1 + Qslag min );

[0049] In the formula: Qslag is the remaining slag amount after slag skimming, tons; Qslag min is the lowest remaining slag amount after slag skimming within a recent evaluation period (week, month, quarter, year), tons.

[0050] Preferably, the subjective and objective factors include: initial hot metal composition, temperature, stirring paddle speed, type and dosage of desulfurizer, carrier gas flow rate, slag skimming operation, stirring paddle insertion depth, treatment time.

[0051] Preferably, on the basis of determining the optimization direction, an optimization model is established to optimize the hot metal pretreatment effect. The optimization model establishes a single-objective mathematical programming model or a multi-objective mathematical programming model from one or more of the five aspects of carbon emission, desulfurization efficiency, temperature drop control, cost control, and end-point sulfur control, and uses artificial intelligence algorithms such as NSGAⅡ, genetic algorithm, non-dominated sorting genetic algorithm, multi-objective ant colony optimization algorithm, etc. to solve the model to obtain the operation parameter combination that meets the objectives.

[0052] Preferably, the data used in the evaluation model comes from the production and quality big data system of the steel enterprise. After the hot metal pretreatment is completed, the corresponding data can be retrieved from the big data system for the evaluation of this heat. The weighted factors in the evaluation model can be adjusted accordingly according to the different emphases of the current evaluation.

[0053] Preferably, CE, EF, ΔT, S, Cost, and Slag can also be evaluated separately to evaluate the smelting level and control effect of the hot metal pretreatment of this heat in the corresponding aspects, and optimize the hot metal pretreatment and the previous hot metal production process according to the corresponding data.

[0054] Preferably, for the evaluation of the metallurgical effect of the first 10 heats of hot metal pretreatment adopted by this method, since Ce min , Ef max , {T 0 -T e} min , [%S] min and Co Min , Qslag minThe numerical value is missing. The corresponding initial value can be set first according to the hot metal pretreatment process and operation level of the corresponding category for evaluation.

[0055] Preferably, according to the calculation model and optimization algorithm of the metallurgical effect factor R, the corresponding program is compiled and placed in the hot metal pretreatment operation system, and data exchange is carried out with the corresponding production and quality management big data system of the steel enterprise. After the hot metal pretreatment is completed, various data are intelligently retrieved by the system for automatic evaluation and optimization, and the corresponding evaluation data and indicators are stored in the system; technicians, operators and managers can retrieve the corresponding data indicators for evaluation and carry out process and operation optimization.

[0056] Preferably, when evaluating and optimizing according to the metallurgical effect factor R, the hot metal pretreatment needs to be subdivided into corresponding categories according to the differences in the requirements of the hot metal pretreatment target. According to the requirements for sulfur control at the pretreatment end point, it is subdivided into low-sulfur (end point sulfur ≥ 0.003 wt%) hot metal treatment and ultra-low-sulfur (end point sulfur < 0.003 wt%) hot metal treatment, and the corresponding categories of Ce min 、Ef max 、{T 0 -T e} min 、[%S] min and Co Min are used for intelligent classification evaluation of data.

[0057] Preferably, when calculating the carbon emission intensity during the hot metal pretreatment process, the value range of the carbon dioxide emission factor of each substance is calculated according to the values in Table 1.

[0058] Table 1

[0059]

[0060] The technical solution of the present invention will be further described below in conjunction with specific embodiments.

[0061] Example 1

[0062] A certain steel plant uses the KR method to desulfurize hot metal. The main components of the hot metal before pretreatment are (Si: 0.67 wt%, P: 0.131 wt%, S: 0.024 wt%). After pretreatment, the sulfur content is required to be less than 0.0030 wt%. After pretreatment, the main components of the hot metal are (Si: 0.66 wt%, P: 0.128 wt%, S: 0.0028 wt%). The consumption of raw and auxiliary materials for this heat is as follows: lime: 0.531 t; electricity: 0.4 MW·h; argon: 17 m 3 . The weight of the hot metal is 195 t. This heat is subjected to slag skimming treatment, and the remaining slag amount after slag skimming is 1 t.

[0063] The calculation method of the carbon dioxide emission intensity Ce for this heat is as follows:

[0064] Ce = {ΔW·Ke + ∑SM i ·K Si + ∑G i ·K Gi} / W HM = (ΔW·Ke + SM 石灰 ×K 石灰 + G 氩气 ×K 氩

[0065] 气 ) / W HM = (0.4×0.504 + 0.531×0.95 + 17 / 1000×0.103) / 195 = 0.0036。

[0066] The calculation method for the carbon emission intensity level CE of this heat is as follows:

[0067] CE = 2 - Ce / Ce min = 2 - 0.0036 / 0.003 = 0.8

[0068] The calculation method for the pretreatment efficiency of this heat is as follows:

[0069] Ef = {[%S] 0 - [%S] e} / t = (0.024 - 0.0028) / 34 = 0.00062

[0070] The calculation method for the pretreatment efficiency level of this heat is as follows:

[0071] EF = Ef / Ef max = 0.00062 / 0.0008 = 0.775

[0072] The calculation method for the temperature drop control level ΔT of this heat is as follows:

[0073] ΔT = 2 - {T 0 - T e} / {T 0 - T e} min = 2 - (1683 - 1600) / 54 = 0.463

[0074] The calculation method for the sulfur control level S at the end of pretreatment of this heat is as follows:

[0075] S = 2 - (0.002 + [%S] e ) / (0.002 + [%S] min ) = 2 - (0.002 + 0.0028) / (0.002 + 0.002)

[0076] = 0.80

[0077] The calculation method of the hot metal pretreatment cost Co for this heat is as follows:

[0078] Co = {ΔW × Pr e + ∑SM i × Pr Si + ∑G i × Pr Gi + M L × Pr L} / W HM = {ΔW × Pr e + M 石灰 × Pr 石灰

[0079] + G 氩气 × Pr 氩气 + M L × Pr L} / W HM = (400 × 0.7 + 531 × 0.54 + 17 × 2.5 + 0.39 × 2457) / 195 = 8.04

[0080] The calculation method of the cost control level Cost for this heat is as follows:

[0081] Cost = 2 - Co / Co Min = 2 - 8.04 / 6 = 0.66

[0082] The calculation method of the slag removal control level Slag for this heat is as follows:

[0083] Slag = 2 - (1 + Qslag) / (1 + Qslag min ) = 2 - (1 + 1) / (1 + 0.8) = 0.89

[0084] The calculation method of the evaluation effect for this heat is as follows:

[0085] R = a1 × CE + a2 × EF + a3 × ΔT + a4 × S + a5 × Cost + a6 × Slag = 0.6 × 0.8 + 0.3 × 0.775 + 0.4 × 0.46 + 0.6 × 0.8 + 0.5 × 0.66 + 0.6 × 0.89 = 2.16

[0086] The comprehensive evaluation factor R of the pretreatment effect and operation level of this heat is 2.16, ranking in the top 32% among the ultra-low sulfur pretreatment R rankings in the recent 3 months, indicating that the metallurgical effect of this heat is not good and the operation level is slightly poor. There is still much room for improvement in the hot metal pretreatment operation of this heat in terms of carbon emission control, desulfurization efficiency, temperature drop control, end-point sulfur control, and cost control, and improvement is needed; especially the temperature drop control level is 0.463 and the cost control level is 0.66, with a large improvement space.

[0087] The metallurgical effect optimization method for this heat is a multi-objective mathematical model with constraints. The model takes the lowest carbon emission and the highest desulfurization efficiency as the dual objectives, combines with the classic multi-objective optimization algorithm NSGAⅡ algorithm for solution calculation, and generates a set of Pareto optimal solution sets to provide decision support for the actual operation parameters in production, truly achieving the purpose of optimizing the smelting effect of hot metal pretreatment. The optimized results are as follows: the lime addition amount is 0.7t, an increase of 0.17t compared with the original process; the treatment time is 30 minutes, a decrease of 4 minutes compared with the original process; the KR stirring speed is 140 revolutions per minute, an increase of 20 revolutions per minute compared with the original process; the remaining slag amount after slag skimming is 0.5t, a decrease of 0.5t compared with the original process.

[0088] Example 2

[0089] A certain steel plant uses the rotary injection method to desulfurize semi-steel. The main components of the hot metal before pretreatment are (C: 3.17wt%, S: 0.1021wt%). After pretreatment of this heat, the hot metal composition requirements are 0.0030wt% - 0.0035%, and the main components of the hot metal after pretreatment are (C: 3.06wt%, S: 0.0033wt%). The consumption of raw and auxiliary materials for this heat are as follows: lime: 1.659t; magnesium powder: 0.4t; electricity: 0.48MW·h; argon: 30m 3 . The hot metal weight is 233t. This heat is subjected to slag skimming treatment, and the remaining slag amount after slag skimming is 1.5t.

[0090] The calculation method of the carbon dioxide emission intensity Ce for this heat is:

[0091] Ce = {ΔW×Ke + ∑SM i ×K Si +∑G i ×K Gi} / W HM =(ΔW×Ke + SM 石灰 ×K 石灰 +SM 镁 ×K 镁

[0092] +G 氩气 ×K 氩气 ) / W HM

[0093] =(0.48×0.504 + 1.659×0.95 + 0.4×33.3 + 30 / 1000×0.103) / 233 = 0.065。

[0094] The calculation method for the carbon emission intensity level CE of this heat is as follows:

[0095] CE = 2 - Ce / Ce min = 2 - 0.065 / 0.06 = 0.92

[0096] The calculation method for the pretreatment efficiency of this heat is as follows:

[0097] Ef = {[%S] 0 - [%S] e} / t = (0.1021 - 0.0033) / 18 = 0.0055

[0098] The calculation method for the pretreatment efficiency level of this heat is as follows:

[0099] EF = Ef / Ef max = 0.0055 / 0.008 = 0.69

[0100] The calculation method for the temperature drop control level ΔT of this heat is as follows:

[0101] ΔT = 2 - {T 0 - T e} / {T 0 - T e} min = 2 - (1656 - 1586) / 40 = 0.225

[0102] The calculation method for the sulfur control level S at the end of pretreatment of this heat is as follows:

[0103] S = 2 - (0.002 + [%S] e ) / (0.002 + [%S] min ) = 2 - 0.0053 / 0.004 = 0.68

[0104] The calculation method for the hot metal pretreatment cost Co of this heat is as follows:

[0105] Co = {ΔW × Pr e + ∑SM i × Pr Si + ∑G i × Pr Gi + M L × Pr L} / W HM = {ΔW × Pr e + M 石灰×Pr 石灰

[0106] +M 镁 ×Pr 镁+ G 氩气 ×Pr 氩气 +M L ×Pr L} / W HM =(480×0.7 + 1659×0.54 + 400×13.2 + 30×2.5 + 0.45×2457) / 233 = 10.35

[0107] The calculation method for the cost control level Cost of this heat is as follows:

[0108] Cost = 2 - Co / Co Min = 2 - 10.35 / 8 = 0.71

[0109] The calculation method for the slag removal control level Slag of this heat is as follows:

[0110] Slag = 2 - (1 + Qslag) / (1 + Qslag min ) = 2 - (1 + 1.5) / (1 + 1) = 0.75

[0111] The calculation method for the evaluation effect of this heat is as follows:

[0112] R = a1×CE + a2×EF + a3×ΔT + a4×S + a5×Cost + a6×Slag = 0.5×0.73 + 0.3×0.69 + 0.4×0.225 + 0.5×0.68 + 0.6×0.71 + 0.4×0.5 = 1.628

[0113] The evaluation level of this heat is 1.628.

[0114] The comprehensive evaluation factor R for the pretreatment effect and operation level of this heat is 1.628, ranking in the top 29% among the low-sulfur hot metal pretreatment in the recent 3 months, indicating that the metallurgical effect and operation level of this heat are not good. There is still much room for improvement in the carbon emission control, desulfurization efficiency, temperature drop control, end-point sulfur control, and cost control of the hot metal pretreatment operation of this heat, and improvement is needed.

[0115] The metallurgical effect optimization method for this heat has a multi-objective mathematical model with constraints. The model takes the lowest carbon emission, the highest desulfurization efficiency, and the lowest cost as the three objectives, combines with the multi-objective optimization classic algorithm NSGAⅡ algorithm for solution calculation, and generates a set of Pareto optimal solutions to provide decision support for the actual operation parameters in production, truly achieving the purpose of optimizing the smelting effect of hot metal pretreatment. The optimized results are as follows: the lime addition amount is 1.5t, which is 0.159t less than the original process; the magnesium powder addition amount is 0.3t, which is 0.1t less than the original process; the treatment time is 17 minutes, which is 1 minute less than the original process; the stirring speed is 160 revolutions per minute, which is 15 revolutions per minute more than the original process; the slag amount after slag skimming is 1.2t, which is 0.3t less than the original process.

[0116] The above are only the preferred embodiments of the present invention, and do not limit the patent scope of the present invention. Any equivalent structural transformation made using the content of the specification of the present invention under the inventive concept of the present invention, or direct / indirect application in other related technical fields, is included in the patent protection scope of the present invention.

Claims

1. A big data intelligent online evaluation and optimization method for molten iron pretreatment effect, characterized in that: After the molten iron pretreatment is completed, the big data of the molten iron pretreatment process of this furnace is used to determine the metallurgical effect factor of this furnace, and the molten iron pretreatment effect is evaluated based on the furnace metallurgical effect factor. At the same time, based on the evaluation results, the artificial intelligence algorithm is used to analyze the subjective and objective factors affecting the molten iron pretreatment effect and determine the optimization direction.

2. The method for intelligent online evaluation and optimization of molten iron pretreatment effect based on big data according to claim 1, characterized in that: The metallurgical effect factor R is: R=a1×CE+a2×EF+a3×ΔT+a4×S+a5×Cost+a6×Slag Where: R is the metallurgical effect factor; a1 is the carbon emission intensity weighting factor; CE is the carbon emission intensity level; a2 is the efficiency weighting factor; EF is the efficiency level; a3 is the weighting factor of temperature drop control in the pretreatment process; ΔT is the temperature drop control level in the pretreatment process; a4 is the sulfur control weighting factor at the pretreatment endpoint; S is the sulfur control level at the pretreatment endpoint; a5 is the weighting factor of pretreatment process cost control; Cost is the pretreatment process cost control level; a6 is the weighted factor for slag removal control after desulfurization is completed; Slag is the slag removal control level during the pretreatment process.

3. The method for intelligent online evaluation and optimization of molten iron pretreatment effect based on big data according to claim 1, characterized in that: The calculation method of carbon emission intensity level CE is: CE=2-Ce / Ce min Where: Ce is the carbon dioxide emission intensity of the hot metal pretreatment process, tons of carbon dioxide / ton of iron; Ce min is the minimum carbon dioxide emission intensity in the last evaluation period, tons of carbon dioxide / ton of iron; and Ce={ΔW×Ke+∑SM i ×K Si +∑G i ×K Gi } / W HM Where: ΔW is the power consumption of hot metal pretreatment process, kW·h; Ke is the carbon dioxide emission factor of hot metal pretreatment power consumption, tons of carbon dioxide / kW·h; SM i The mass of desulfurizer i added in the pretreatment process, tons; K Si is the carbon emission factor of desulfurizer i, tons of carbon dioxide / ton; G i is the volume of gas i consumed in the pretreatment process, m 3 ; K Gi is the carbon emission factor of gas i, tons of carbon dioxide / m 3 ; W HM For pretreatment of molten iron mass, tons.

4. The method for intelligent online evaluation and optimization of molten iron pretreatment effect based on big data according to claim 1, characterized in that: The efficiency level EF is calculated as: IF=If / If max Where: Ef is the pretreatment efficiency; Ef max is the highest pretreatment efficiency in the last evaluation cycle, and Ef = {[%S]0-[%S] e } / t Where: [%S]0, [%S] e are the initial and final sulfur contents of molten iron pretreatment, wt%; t is the pretreatment time.

5. The method for intelligent online evaluation and optimization of molten iron pretreatment effect based on big data according to claim 1, characterized in that: The calculation method of the temperature drop control level ΔT in the pretreatment process is: ΔT=2-{T0-T e } / {T0-T e } min Where: T e T0 and T1 are the end point and initial temperature of hot metal pretreatment, respectively, K; {T0-T e } min It is the minimum temperature drop of the pretreatment process in the last evaluation cycle.

6. The method for intelligent online evaluation and optimization of molten iron pretreatment effect based on big data according to claim 1, characterized in that: The calculation method of the pretreatment endpoint sulfur control level S is: S=2-(0.002+[%S] e ) / (0.002+[%S] min ) Where: [%S] e is the sulfur content of the molten iron at the end of pretreatment, wt%; [%S] min It is the minimum sulfur content at the end point of molten iron pretreatment in the last evaluation cycle, wt%.

7. The method for intelligent online evaluation and optimization of molten iron pretreatment effect based on big data according to claim 1, characterized in that: The calculation method of the cost control level Cost of the pretreatment process is: Cost=2-What / What Min Where: Co is the pretreatment cost of the hot metal of this batch, including the cost of desulfurizer, electricity, gas and iron loss consumed in the pretreatment process, RMB; Co Min The lowest value of the pretreatment cost of the steel grade produced in the recent evaluation cycle, RMB; and Co = {ΔW × Pr e + ∑SM i × Pr Si + ∑G i × Pr Gi + M L × Pr L} / W HM Where: Pr e is the unit price of electricity, yuan / kW·h; Pr Si is the unit price of desulfurizer i, yuan / ton; Pr Gi is the unit price of gas i, yuan / m 3 ; M L is the iron loss in the pretreatment process, tons; Pr L is the unit price of molten iron, Yuan / ton.

8. The method for intelligent online evaluation and optimization of molten iron pretreatment effect based on big data according to claim 1, characterized in that: The method for determining the slag control level Slag in the pretreatment process is: If there is no slag removal after molten iron pretreatment, Slag = 0; If slag is removed after molten iron pretreatment, Slag = 2-(1+Qslag) / (1+Qslag min ); Where: Qslag is the amount of slag remaining after slagging, tons; Qslag min It is the minimum amount of residual slag after slagging in the last evaluation cycle, tons.

Images