CN114242179A - Method and system for predicting yield of refined alloy element, electronic device, and medium - Google Patents

Method and system for predicting yield of refined alloy element, electronic device, and medium Download PDF

Info

Publication number
CN114242179A
CN114242179A CN202111500819.0A CN202111500819A CN114242179A CN 114242179 A CN114242179 A CN 114242179A CN 202111500819 A CN202111500819 A CN 202111500819A CN 114242179 A CN114242179 A CN 114242179A
Authority
CN
China
Prior art keywords
alloying
data
predicted
current
heat
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202111500819.0A
Other languages
Chinese (zh)
Inventor
严莹子
张璟涵
汤槟
张晓辉
蒲大志
向山林
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
CISDI Chongqing Information Technology Co Ltd
Original Assignee
CISDI Chongqing Information Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by CISDI Chongqing Information Technology Co Ltd filed Critical CISDI Chongqing Information Technology Co Ltd
Priority to CN202111500819.0A priority Critical patent/CN114242179A/en
Publication of CN114242179A publication Critical patent/CN114242179A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
    • G16C20/00—Chemoinformatics, i.e. ICT specially adapted for the handling of physicochemical or structural data of chemical particles, elements, compounds or mixtures
    • G16C20/30—Prediction of properties of chemical compounds, compositions or mixtures
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/04—Manufacturing
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
    • G16C20/00—Chemoinformatics, i.e. ICT specially adapted for the handling of physicochemical or structural data of chemical particles, elements, compounds or mixtures
    • G16C20/70—Machine learning, data mining or chemometrics
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/30—Computing systems specially adapted for manufacturing

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Theoretical Computer Science (AREA)
  • Human Resources & Organizations (AREA)
  • Economics (AREA)
  • Strategic Management (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Marketing (AREA)
  • Chemical & Material Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Tourism & Hospitality (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Computing Systems (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Crystallography & Structural Chemistry (AREA)
  • Game Theory and Decision Science (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • Evolutionary Computation (AREA)
  • Medical Informatics (AREA)
  • Software Systems (AREA)
  • Quality & Reliability (AREA)
  • Operations Research (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Development Economics (AREA)
  • Primary Health Care (AREA)
  • Manufacturing & Machinery (AREA)
  • Investigating And Analyzing Materials By Characteristic Methods (AREA)

Abstract

The invention relates to the technical field of refined alloy, and discloses a method, a system, electronic equipment and a medium for predicting the yield of refined alloy elements, wherein the method obtains the current data of a to-be-predicted heat, obtains a plurality of historical data, determines similar data corresponding to the to-be-predicted heat from the historical data according to the current pre-alloying characteristics, determines the predicted alloying process characteristics corresponding to the to-be-predicted heat according to the similar data, inputs the current pre-alloying characteristics and the predicted alloying process characteristics into a prediction model to obtain the yield of the predicted elements of the refined alloy of the to-be-predicted heat, obtains the predicted alloying process of the to-be-predicted heat through the historical data and the current data, and further obtains the yield of the predicted elements through inputting the current data and the predicted alloying process characteristics into the successfully trained prediction model to obtain the yield of the predicted elements, so that the input data of the prediction model has more reliability, the accuracy of the yield of the predicted elements of the alloy refined by the furnace to be predicted is improved.

Description

Method and system for predicting yield of refined alloy element, electronic device, and medium
Technical Field
The invention relates to the technical field of refined alloy, in particular to a method and a system for predicting yield of refined alloy elements, electronic equipment and a medium.
Background
With the progress of modern science and technology and the development of industry, the quality of high-efficiency continuous casting technology and steel is continuously improved, the requirements on the characteristics of molten steel such as temperature, element components, gas content and the like are strict, and common steelmaking primary furnaces such as an electric arc furnace, a converter and the like are difficult to meet the industrial requirements. Therefore, a part of the steel-making tasks is transferred to the outside of the ordinary steel-making primary Furnace to be completed By an external refining method such as an LF (Ladle Furnace), a CAS (Ladle Argon blowing) method, a VD (Vacuum degassing) Furnace, a VOD (Vacuum And Stir Oxygen Injection) Furnace, And the like, so as to improve the production efficiency And shorten the smelting time.
At present, the external refining method needs to predict the element yield of the alloy refined by the heat, further calculates the alloy batching scheme, and adjusts the alloy content in molten steel in the alloying process so as to provide a more scientific alloy feeding table, improve the utilization rate of alloy batching, effectively improve the phenomena of alloy waste, product control fluctuation and the like, and further achieve the purposes of reducing the refining cost, improving the refining efficiency and stabilizing the refining quality.
However, since the refined alloy is a highly complex nonlinear process, and the alloying process is summarized, not only can complex 'gas-solid', 'solid-solid' and 'solid-liquid' reactions occur, but also can be accompanied by phenomena such as high temperature, multiphase coexistence, chemical reaction, energy transfer and the like, and alloy elements are lost in various complex modes, so that the accuracy of the predicted element yield of the alloy refined by the furnace to be predicted in the related technology is low, and the industrial requirement cannot be met.
Disclosure of Invention
The following presents a simplified summary in order to provide a basic understanding of some aspects of the disclosed embodiments. This summary is not an extensive overview nor is intended to identify key/critical elements or to delineate the scope of such embodiments but rather as a prelude to the more detailed description that is presented later.
In view of the above-described shortcomings of the prior art, the present invention discloses a refined alloy element yield prediction method, system, electronic device, and medium to improve the accuracy of furnace predicted element yields.
The invention provides a method for predicting yield of refined alloy elements, which comprises the following steps: acquiring current data of a heat to be predicted and acquiring a plurality of historical data, wherein the current data comprises current pre-alloying characteristics; determining similar data corresponding to the to-be-predicted heat from each historical data according to the current pre-alloying characteristics, and determining predicted alloying process characteristics corresponding to the to-be-predicted heat according to the similar data; obtaining a model training result of a prediction model, wherein the prediction model is obtained by training a plurality of sample data corresponding to the to-be-predicted heat, and the sample data comprises a sample characteristic before alloying, a sample characteristic in an alloying process and an element yield label; and if the model training result comprises successful training, inputting the current pre-alloying characteristics and the predicted alloying process characteristics into the prediction model to obtain the predicted element yield of the alloy refined by the to-be-predicted heat.
Optionally, determining similar data corresponding to the to-be-predicted heat from each historical data according to the current pre-alloying feature includes: the historical data comprises historical pre-alloying features and historical alloying process features, and the current pre-alloying features and the historical pre-alloying features each comprise a plurality of pre-alloying sub-features; acquiring a characteristic weighted value corresponding to each pre-alloying sub-characteristic, and determining the similarity between the current data and each historical data according to the current pre-alloying characteristic, the historical pre-alloying characteristic and each characteristic weighted value; and determining similar data from the historical data according to the similarity.
Optionally, the similarity between the current data and the historical data is determined by:
Figure BDA0003401561660000021
wherein Similarity is the Similarity between the current data and the historical data, ScorelA characteristic weighted value, x, corresponding to the first pre-alloying sub-characteristic of the current pre-alloying characteristiclA characteristic value, x, of the l pre-alloying sub-characteristic of the historical pre-alloying characteristicl' is a feature value of the l pre-alloying sub-feature of the current pre-alloying feature, and N is the number of the pre-alloying sub-features.
Optionally, determining a model training result of the predictive model by: dividing the plurality of sample data into a plurality of training data and a plurality of verification data; training a preset neural network model through each training data to obtain a prediction model; determining an error value for the prediction model from each of the validation data; if the error value is smaller than a preset error threshold value, determining the model training result as successful training; and if the error value is greater than or equal to the preset error threshold value, determining the model training result as training failure.
Optionally, the sample data is obtained by the following method, including: determining a plurality of alternative data corresponding to current heat information from each historical data, wherein the current heat information comprises at least one of manufacturer information, steel type information, alloying time information and a time interval corresponding to the heat to be predicted, and the alternative data comprises sample characteristics before alloying and sample characteristics in an alloying process; if the number of the alternative data is larger than or equal to a training number threshold, acquiring an element yield corresponding to each alternative data, and determining the element yield corresponding to the alternative data as an element yield label of the alternative data to obtain sample data; and if the quantity of the screening data is less than the threshold value of the training quantity, continuously acquiring and recording new heat production data added to the historical data to obtain new sample data.
Optionally, after obtaining the model training result of the prediction model, the method further includes at least one of: if the model training result comprises training failure, acquiring element yield corresponding to the similar data, and determining the predicted element yield according to the element yield corresponding to each similar data; before inputting the current pre-alloying feature and the predicted alloying process feature into the prediction model, obtaining the update time of the prediction model, and presetting a first preset condition, wherein the first preset condition is that the model training result comprises successful training and the update time is within a preset time interval, and if the first preset condition is met, the current pre-alloying feature and the predicted alloying process feature are input into the prediction model; and before inputting the current pre-alloying feature and the predicted alloying process feature into the prediction model, obtaining the update time of the prediction model, and presetting a first preset condition, wherein the first preset condition is that the model training result comprises successful training and the update time is within a preset time interval, and if the model training result does not meet the first preset condition, obtaining update data to perform update training on the prediction model, wherein the update data is obtained according to historical data after the update time.
Optionally, the element yield is determined by:
Figure BDA0003401561660000031
wherein n is the number of alloying times,
Figure BDA0003401561660000032
the yield of the element, delta w, of the n-th alloying of the j elementjIs the difference between the mass of the j element in the 1 st molten steel and the mass of the j element in the n-th molten steel, G is the initial weight of the molten steel, G ismAdding weight, g, for the m-th alloyiThe added mass of the i-type alloy,
Figure BDA0003401561660000033
is the content of the j element in the i-type alloy.
The invention provides a refined alloy element yield prediction system, which comprises: the device comprises a first obtaining module, a second obtaining module and a third obtaining module, wherein the first obtaining module is used for obtaining current data of a to-be-predicted heat and obtaining a plurality of historical data, and the current data comprises current pre-alloying characteristics; the determining module is used for determining similar data corresponding to the to-be-predicted heat from each historical data according to the current pre-alloying characteristics and determining predicted alloying process characteristics corresponding to the to-be-predicted heat according to the similar data; the second obtaining module is used for obtaining a model training result of a prediction model, wherein the prediction model is obtained by training a plurality of sample data corresponding to the to-be-predicted heat, and the sample data comprises a sample characteristic before alloying, a sample characteristic in an alloying process and an element yield label; and the prediction module is used for inputting the current pre-alloying characteristics and the predicted alloying process characteristics into the prediction model to obtain the yield of the predicted elements of the alloy refined by the to-be-predicted heat if the model training result comprises successful training.
The present invention provides an electronic device, including: a processor and a memory; the memory is used for storing computer programs, and the processor is used for executing the computer programs stored by the memory so as to make the electronic equipment execute the method.
The present invention provides a computer-readable storage medium having stored thereon a computer program for: which when executed by a processor implements the method described above
The invention has the beneficial effects that: the method comprises the steps of obtaining current data of a to-be-predicted heat and obtaining a plurality of historical data, determining similar data corresponding to the to-be-predicted heat from the historical data according to current pre-alloying characteristics of the current data, determining predicted alloying process characteristics corresponding to the to-be-predicted heat according to the similar data, and inputting the current pre-alloying characteristics and the predicted alloying process characteristics into a prediction model if a model training result comprises successful training to obtain the yield of predicted elements of an alloy refined by the to-be-predicted heat. Therefore, the predicted alloying process of the heat to be predicted is obtained through the historical data and the current data, and the predicted element yield is obtained through inputting the current data and the characteristics of the predicted alloying process into the successfully trained prediction model, so that the input data of the prediction model has higher reliability, and the accuracy of the predicted element yield of the alloy refined by the heat to be predicted is improved.
Drawings
FIG. 1 is a schematic flow chart of a method for predicting the yield of a refined alloying element in an embodiment of the present invention;
FIG. 2 is a schematic flow chart of another method for predicting the yield of a refined alloying element in an embodiment of the present invention;
FIG. 3 is a flow chart illustrating a method for training a predictive model according to an embodiment of the invention;
FIG. 4 is a schematic flow chart of another method for predicting the yield of refined alloying elements in an embodiment of the present invention;
FIG. 5 is a schematic diagram showing the construction of a system for predicting the yield of a refined alloying element in an embodiment of the present invention;
fig. 6 is a schematic structural diagram of an electronic device in an embodiment of the present invention.
Detailed Description
The embodiments of the present invention are described below with reference to specific embodiments, and other advantages and effects of the present invention will be easily understood by those skilled in the art from the disclosure of the present specification. The invention is capable of other and different embodiments and of being practiced or of being carried out in various ways, and its several details are capable of modification in various respects, all without departing from the spirit and scope of the present invention. It should be noted that, in the following embodiments and examples, subsamples may be combined with each other without conflict.
It should be noted that the drawings provided in the following embodiments are only for illustrating the basic idea of the present invention, and the components related to the present invention are only shown in the drawings rather than drawn according to the number, shape and size of the components in actual implementation, and the type, quantity and proportion of the components in actual implementation may be changed freely, and the layout of the components may be more complicated.
In the following description, numerous details are set forth to provide a more thorough explanation of embodiments of the present invention, however, it will be apparent to one skilled in the art that embodiments of the present invention may be practiced without these specific details, and in other embodiments, well-known structures and devices are shown in block diagram form, rather than in detail, in order to avoid obscuring embodiments of the present invention.
The terms "first," "second," and the like in the description and in the claims, and the above-described drawings of embodiments of the present disclosure, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It should be understood that the data so used may be interchanged under appropriate circumstances such that embodiments of the present disclosure described herein may be made. Furthermore, the terms "comprising" and "having," as well as any variations thereof, are intended to cover non-exclusive inclusions.
The term "plurality" means two or more unless otherwise specified.
In the embodiment of the present disclosure, the character "/" indicates that the preceding and following objects are in an or relationship. For example, A/B represents: a or B.
The term "and/or" is an associative relationship that describes objects, meaning that three relationships may exist. For example, a and/or B, represents: a or B, or A and B.
Referring to fig. 1, an embodiment of the present disclosure provides a method for predicting yield of refined alloy elements, including:
step S101, obtaining current data of a heat to be predicted and obtaining a plurality of historical data;
wherein the current data includes current pre-alloying features;
step S102, determining similar data corresponding to the to-be-predicted heat from various historical data according to the current pre-alloying characteristics, and determining predicted alloying process characteristics corresponding to the to-be-predicted heat according to the similar data;
step S103, obtaining the model training result of the prediction model,
the prediction model is obtained by training a plurality of sample data corresponding to the to-be-predicted heat, wherein the sample data comprises sample characteristics before alloying, sample characteristics in the alloying process and element yield labels;
and step S104, if the model training result includes successful training, inputting the current pre-alloying characteristics and the predicted alloying process characteristics into the prediction model to obtain the predicted element yield of the alloy refined by the to-be-predicted heat.
By adopting the method for predicting the yield of the refined alloy elements, provided by the embodiment of the disclosure, the current data of the to-be-predicted heat is obtained, a plurality of historical data are obtained, the similar data corresponding to the to-be-predicted heat is determined from the historical data according to the current pre-alloying characteristics of the current data, the predicted alloying process characteristics corresponding to the to-be-predicted heat are determined according to the similar data, and if the model training result comprises successful training, the current pre-alloying characteristics and the predicted alloying process characteristics are input into the prediction model, so that the predicted element yield of the refined alloy of the to-be-predicted heat is obtained. Therefore, the predicted alloying process of the heat to be predicted is obtained through the historical data and the current data, and the predicted element yield is obtained through inputting the current data and the characteristics of the predicted alloying process into the successfully trained prediction model, so that the input data of the prediction model has higher reliability, and the accuracy of the predicted element yield of the alloy refined by the heat to be predicted is improved. Meanwhile, the alloy content in the molten steel is adjusted in the alloying process by calculating and determining the alloy batching scheme through predicting the element yield, so that a more scientific alloy feeding table is provided, the utilization rate of alloy batching is improved, the phenomena of alloy waste, quality control fluctuation and the like are effectively improved, and the aims of reducing the refining cost, improving the refining efficiency and stabilizing the refining quality are fulfilled.
Optionally, historical data and current data of the heat to be predicted are collected by the industrial production system.
Optionally, the molten steel elements in the molten steel include at least one of carbon, silicon, phosphorus, sulfur, manganese, iron, calcium, and the like.
Alternatively, the alloy species to which the molten steel is added includes at least one of carbon powder, ferrosilicon powder, fluorite, lime, and the like.
Optionally, the historical data includes historical pre-alloying features and historical alloying process features.
Optionally, the current pre-alloying feature, the historical pre-alloying feature, and the pre-alloying sample feature each include a plurality of pre-alloying sub-features, and the pre-alloying sub-features include a first feature field and a first feature value, wherein the first feature field includes a plurality of the initial weight of molten steel, the initial temperature of molten steel, the initial content of each molten steel element, the target content of each molten steel element, and the like.
Optionally, there is a same first feature field between the current pre-alloying feature, the historical pre-alloying feature, and the pre-alloying sample feature.
Optionally, the predicted alloying process characteristic, the historical alloying process characteristic, and the alloying process sample characteristic all include a plurality of alloying process sub-characteristics, and the alloying process sub-characteristics include a second characteristic field and a second characteristic value, wherein the second characteristic field includes a plurality of alloying time, mass fraction change rate before and after alloying of each molten steel element, addition quality of each alloy type, argon blowing time after adding the alloy, current fluctuation after adding the alloy, and the like.
Optionally, there is a second feature field that is the same between the predicted alloying process feature, the historical alloying process feature, the alloying process sample feature.
Optionally, the historical data comprises a plurality of production sub-features, the production sub-features comprising an alloying pre-sub-feature and an alloying process sub-feature, wherein the production sub-features are 2000-.
Optionally, the alloying precursor sub-feature and the alloying process sub-feature are determined by: determining the importance of each production sub-feature through a preset importance rule; determining a plurality of important sub-features from each production sub-feature according to the importance; classifying the important sub-features to obtain a plurality of alloying front sub-features and a plurality of alloying process sub-features, wherein the preset importance rule comprises at least one of a shapley value, an integrated Tree model, a GDBT (Gradient Boosting Decision Tree) model, an XGboost (Extreme Gradient Boosting) model, a Tree (forest) model, a random forest model and the like, the number of the alloying front sub-features is 10-20, and the number of the alloying process sub-features is 10-20.
Optionally, determining similar data corresponding to the to-be-predicted heat from each historical data according to the current pre-alloying features, including: the historical data comprises historical pre-alloying characteristics and historical alloying process characteristics, and the current pre-alloying characteristics and the historical pre-alloying characteristics both comprise a plurality of pre-alloying sub-characteristics; acquiring a characteristic weighted value corresponding to each pre-alloying sub-characteristic, and determining the similarity between the current data and each historical data according to the current pre-alloying characteristic, the historical pre-alloying characteristic and each characteristic weighted value; and determining similar data from the historical data according to the similarity. Therefore, because the influence degrees of the sub-characteristics before alloying on the yield of the predicted elements are different, the similarity between the current data and the historical data is enabled to be more reliable by obtaining the characteristic weighted value of the sub-characteristics before alloying, then determining the similarity between the current data and the historical data according to the current pre-alloying characteristic, the historical pre-alloying characteristic and the characteristic weighted value, further determining the similar data according to the similarity, determining the predicted alloying process characteristic corresponding to the heat to be predicted according to the similar data, and then inputting the current data and the predicted alloying process characteristic into the successfully trained prediction model to obtain the yield of the predicted elements, so that the input data of the prediction model is enabled to be more reliable, and the accuracy of the yield of the predicted elements of the refined alloy of the heat to be predicted is improved.
Optionally, the characteristic weighting value corresponding to each alloying precursor characteristic is determined by: inputting the current pre-alloying features into a preset classification learning model, and calculating to obtain a weighted value array, wherein the weighted value array comprises feature weighted values corresponding to the pre-alloying sub-features, and the preset classification learning model comprises at least one of a GDBT (generalized differential bit rate) model, an XGBoost model, a lightBoost model and a random forest model.
Optionally, the similarity between the current data and the historical data is determined by:
Figure BDA0003401561660000071
wherein, Similarity is the Similarity between the current data and the historical data, ScorelA characteristic weighted value, x, corresponding to the first pre-alloying sub-characteristic of the current pre-alloying characteristiclCharacteristic value, x, of the first pre-alloying sub-characteristic which is a pre-historical pre-alloying characteristicl' is a feature value of the l pre-alloying sub-feature of the current pre-alloying feature, and N is the number of pre-alloying sub-features. Therefore, because the influence degrees of the sub-characteristics before alloying on the yield of the predicted elements are different, the similarity between the current data and the historical data is enabled to be more reliable by obtaining the characteristic weighted value of the sub-characteristics before alloying, then determining the similarity between the current data and the historical data according to the current pre-alloying characteristic, the historical pre-alloying characteristic and the characteristic weighted value, further determining the similar data according to the similarity, determining the predicted alloying process characteristic corresponding to the heat to be predicted according to the similar data, and then inputting the current data and the predicted alloying process characteristic into the successfully trained prediction model to obtain the yield of the predicted elements, so that the input data of the prediction model is enabled to be more reliable, and the accuracy of the yield of the predicted elements of the refined alloy of the heat to be predicted is improved.
Optionally, determining similar data from the historical data according to the similarity includes: presetting a similarity threshold; and if the similarity between the current data and the historical data is greater than the similarity threshold, determining the historical data as similar data. In some embodiments, similar data is determined from historical data based on nearest neighbor methods and respective similarities, wherein the nearest neighbor methods include one or more of decision trees, rough sets, neural networks, ensemble learning, and the like.
Optionally, determining a predicted alloying process characteristic corresponding to the heat to be predicted according to the similar data, including: removing similar data containing abnormal values to obtain a plurality of pieces of removed abnormal data, wherein the abnormal values are second characteristic values within the abnormal characteristic value interval; and determining the average value of the second characteristic values of the second characteristic fields of the rejection data, and taking the determined average value of the second characteristic values as the second characteristic values of the second characteristic fields of the predicted alloying process characteristic to generate the predicted alloying process characteristic.
Optionally, the model training result of the prediction model is determined by: dividing a plurality of sample data into a plurality of training data and a plurality of verification data; training a preset neural network model through each training data to obtain a prediction model; determining an error value of the prediction model according to each verification data; if the error value is smaller than the preset error threshold value, determining the model training result as successful training; and if the error value is greater than or equal to the preset error threshold value, determining the model training result as the training failure. Therefore, the prediction model is obtained through the training data, the error value of the prediction model is determined through the verification data, the model training result is determined according to the error value, and the reliability of the prediction model is improved.
In some embodiments, the predetermined neural network model is an XGBoost model.
In some embodiments, the predetermined error threshold is 3-5%.
Optionally, if the error value is greater than or equal to the preset error threshold, modifying the model parameter of the prediction model until the error value of the modified prediction model is less than the preset error threshold, where the model parameter includes a model structure, a number of model nodes, a model learning rate, and the like.
Optionally, the error value of the prediction model is determined by:
Figure BDA0003401561660000081
wherein, RMSEjIs an error value of the prediction model,
Figure BDA0003401561660000082
the yield of the prediction element of the j element output by the prediction model,
Figure BDA0003401561660000083
to predict the yield of elements for the j element of the validation data, M is the number of validation data.
Optionally, the sample data is obtained by the following method, including: determining a plurality of alternative data corresponding to current heat information from each historical data, wherein the current heat information comprises at least one of manufacturer information, steel type information, alloying time information and a time interval corresponding to a heat to be predicted, and the alternative data comprises sample characteristics before alloying and sample characteristics in an alloying process; if the number of the alternative data is larger than or equal to the training number threshold, acquiring the element yield corresponding to each alternative data, and determining the element yield corresponding to the alternative data as an element yield label of the alternative data to obtain sample data; and if the quantity of the screening data is less than the threshold value of the training quantity, continuously collecting and recording new heat production data added to the historical data to obtain new sample data. Therefore, the current heat information comprises manufacturer information, steel type information, alloying time information, time interval and other environment information corresponding to the heat to be predicted, the alternative data is determined through the current heat information, and the sample data is obtained according to the alternative data, so that the sample data is more consistent with the alloying environment of the heat to be predicted, the trained prediction model is more reliable, and the accuracy of the yield of the prediction elements of the alloy refined by the heat to be predicted is improved.
Optionally, the historical data includes a plurality of historical feature values, and determining a plurality of candidate data corresponding to the current heat information from each historical data includes: determining a screening condition according to the current heat information, and determining a plurality of screened data from various historical data according to the screening condition, wherein the screening condition comprises that the manufacturer information of the historical data is the same as that of the current data, the steel grade information of the historical data is the same as that of the current data, the recording time of the historical data is within a preset screening time period and the like, and the preset screening time period comprises 1-3 months; deleting screened data with missing historical characteristic values or illegal historical characteristic values from the screened data to obtain a plurality of initialized data, wherein if the historical characteristic values are within a preset illegal threshold interval, the historical characteristic values are determined to be illegal; and carrying out normalization processing on each initialization data to obtain a plurality of alternative data.
Optionally, the initialization data is normalized by the following method:
Figure BDA0003401561660000091
wherein a is the normalized characteristic value of the alternative data, amaxTo normalize the maximum value, aminTo normalize the minimum, b is the historical characteristic of the initialization data, bmaxFor the maximum value among the historical characteristic values of the respective initialization data, bminIs the minimum value of the historical characteristic values of the initialization data, amax1 and amin＝-1。
Optionally, after obtaining the model training result of the prediction model, the method further includes at least one of: if the model training result comprises training failure, acquiring element yield corresponding to the similar data, and determining predicted element yield according to the element yield corresponding to each similar data; before inputting the current pre-alloying characteristics and the predicted alloying process characteristics into the prediction model, obtaining the updating time of the prediction model, and presetting a first preset condition, wherein the first preset condition is that the model training result comprises successful training and the updating time is within a preset time interval, and if the first preset condition is met, inputting the current pre-alloying characteristics and the predicted alloying process characteristics into the prediction model; before inputting the current pre-alloying characteristics and the predicted alloying process characteristics into the prediction model, obtaining the updating time of the prediction model, and presetting a first preset condition, wherein the first preset condition is that the model training result comprises successful training and the updating time is within a preset time interval, if the first preset condition is not met, obtaining updating data to update and train the prediction model, and the updating data is obtained according to historical data after the updating time. Therefore, if the training result of the model is training failure, the yield of the prediction elements is determined through the yield of the elements of the similar data, and compared with the method that the yield of the prediction elements is obtained only through the prediction model, the application range and the reliability are improved. Meanwhile, if the updating time of the prediction model is not within the preset time interval, the updating data after the updating time is obtained to update and train the prediction model, so that the prediction model is always the latest model, the accuracy and the reliability of the prediction model are improved, and the accuracy of the yield of the prediction elements of the alloy refined by the to-be-predicted heat is improved.
In some embodiments, the predetermined time interval is 15 days to 90 days.
Optionally, the element yield is determined by:
Figure BDA0003401561660000092
wherein n is the number of alloying times,
Figure BDA0003401561660000093
the yield of the element, delta w, of the n-th alloying of the j elementjIs the difference between the mass of the j element in the 1 st molten steel and the mass of the j element in the n-th molten steel, G is the initial weight of the molten steel, G ismAdding weight, g, for the m-th alloyiThe added mass of the i-type alloy,
Figure BDA0003401561660000094
is the content of the j element in the i-type alloy.
In some embodiments, a plurality of historical data is collected from a stream of historical heat production information; screening out historical data of manufacturer A, steel type B27036 and the last half year from the historical data, and deleting missing data from the screened historical data to obtain 1146 alternative data; extracting historical pre-alloying characteristics and historical alloying process characteristics in the alternative data, wherein the historical pre-alloying characteristics comprise 12 first characteristic fields of initial weight of molten steel, initial temperature of molten steel, initial content of carbon element of molten steel, initial content of silicon element of molten steel, initial content of phosphorus element of molten steel, initial content of sulfur element of molten steel, initial content of manganese element of molten steel, target content of carbon element of molten steel, target content of silicon element of molten steel, target content of phosphorus element of molten steel, target content of sulfur element of molten steel and target content of manganese element of molten steel, and the historical alloying process characteristics comprise alloying time, mass fraction change rate of sulfur element before and after alloying, mass fraction change rate of carbon element before and after alloying, mass fraction change rate of manganese element before and after alloying, mass fraction change rate of silicon element before and after alloying, and the mass fraction change rate of manganese element after alloying, The adding amount of carbon powder, the adding amount of silicon iron powder, the adding amount of fluorite, the adding amount of lime, the argon blowing time after adding alloy and the current fluctuation after adding alloy are 11 second characteristic fields; normalizing each extracted alternative data and determining the element yield of each alternative data to obtain 1146 pieces of sample data; randomly selecting 854 training data and 146 verification data from 1146 sample data, training a preset neural network model through each training data to respectively obtain prediction models of manganese elements and silicon elements, and determining error values of the prediction models according to each verification data; modifying the model parameters of the prediction model according to the error value until the error value of the modified prediction model is smaller than a preset error threshold value; acquiring current data of the heat to be predicted, determining similar data corresponding to the heat to be predicted from various historical data according to the current pre-alloying characteristics, and determining predicted alloying process characteristics corresponding to the heat to be predicted according to the similar data; respectively inputting the current pre-alloying characteristics and the predicted alloying process characteristics into prediction models of manganese elements and silicon elements to obtain the predicted element yield of the manganese elements and the silicon elements of the alloy refined by the furnace to be predicted.
Referring to fig. 2, an embodiment of the present disclosure provides a method for predicting yield of refined alloy elements, including:
step S201, obtaining current data of a heat to be predicted, and collecting a plurality of historical data from a production information flow of a historical heat;
wherein the current data includes current pre-alloying features;
step S202, determining a plurality of alternative data corresponding to the current heat information from each historical data;
the current heat information comprises at least one of manufacturer information, steel type information, alloying time information and time intervals corresponding to the heat to be predicted;
step S203, judging whether the alternative data is greater than or equal to a training number threshold value; if yes, go to step S204; if not, executing step S201;
step S204, obtaining the element yield corresponding to each alternative data, determining the element yield corresponding to the alternative data as an element yield label of the alternative data to obtain sample data, and executing step S205;
wherein the sample data comprises a plurality of training data and a plurality of validation data;
step S205, training a preset neural network model through sample data to obtain a prediction model;
step S206, determining an error value of the prediction model to obtain a model training result of the prediction model;
step S207, determining similar data corresponding to the to-be-predicted heat from various historical data according to the current pre-alloying characteristics, and determining predicted alloying process characteristics corresponding to the to-be-predicted heat according to the similar data;
step S208, judging whether the model training result is successful; if yes, go to step S209; if not, go to step S210;
and S209, inputting the current pre-alloying characteristics and the predicted alloying process characteristics into a prediction model to obtain the predicted element yield of the alloy refined by the furnace to be predicted.
Step S210, obtaining the element yield corresponding to the similar data, and determining the predicted element yield according to the element yield corresponding to each similar data.
By adopting the method for predicting the yield of the refined alloy elements, provided by the embodiment of the disclosure, the current data of the to-be-predicted heat is obtained, a plurality of historical data are obtained, the similar data corresponding to the to-be-predicted heat is determined from the historical data according to the current pre-alloying characteristics of the current data, the predicted alloying process characteristics corresponding to the to-be-predicted heat are determined according to the similar data, and if the model training result comprises successful training, the current pre-alloying characteristics and the predicted alloying process characteristics are input into the prediction model, so that the predicted element yield of the refined alloy of the to-be-predicted heat is obtained. Therefore, the predicted alloying process of the heat to be predicted is obtained through the historical data and the current data, and the predicted element yield is obtained through inputting the current data and the characteristics of the predicted alloying process into the successfully trained prediction model, so that the input data of the prediction model has higher reliability, and the accuracy of the predicted element yield of the alloy refined by the heat to be predicted is improved. Meanwhile, the alloy content in the molten steel is adjusted in the alloying process by calculating and determining the alloy batching scheme through predicting the element yield, so that a more scientific alloy feeding table is provided, the utilization rate of alloy batching is improved, the phenomena of alloy waste, quality control fluctuation and the like are effectively improved, and the aims of reducing the refining cost, improving the refining efficiency and stabilizing the refining quality are fulfilled.
Referring to fig. 3, an embodiment of the present disclosure provides a method for training a prediction model, including:
step S301, collecting a plurality of historical data from the historical heat production information flow;
step S302, determining a plurality of alternative data from each historical data according to a preset screening rule;
the preset screening rule comprises at least one of manufacturer information, steel grade information, alloying time information and a time interval;
step S303, judging whether the alternative data is greater than or equal to a training number threshold value; if yes, go to step S304; if not, go to step S307;
step S304, extracting the pre-alloying history characteristics and the alloying history characteristics in the history data;
s305, determining the element yield of each molten steel element corresponding to each historical data according to the pre-alloying characteristics and the alloying process characteristics to obtain a sample data set corresponding to each molten steel element;
s306, acquiring an element prediction model corresponding to each molten steel element according to the sample data set corresponding to each molten steel element;
the element prediction model comprises a prediction model corresponding to the heat to be predicted.
And step S307, continuously collecting and recording new heat production data added to historical data to acquire new sample data.
By adopting the training method of the prediction model provided by the embodiment of the disclosure, the alternative data is obtained according to the preset screening rule, the sample data is obtained, the element prediction model of the prediction model corresponding to the heat to be predicted is obtained according to the sample data, the current data of the heat to be predicted is obtained, a plurality of historical data are obtained, the similar data corresponding to the heat to be predicted is determined from the historical data according to the current pre-alloying characteristics of the current data, the predicted alloying process characteristics corresponding to the heat to be predicted are determined according to the similar data, and if the model training result comprises the training success, the current pre-alloying characteristics and the predicted alloying process characteristics are input into the prediction model, so that the predicted element yield of the alloy refined by the heat to be predicted is obtained. Therefore, the predicted alloying process of the heat to be predicted is obtained through the historical data and the current data, and the predicted element yield is obtained through inputting the current data and the characteristics of the predicted alloying process into the successfully trained prediction model, so that the input data of the prediction model has higher reliability, and the accuracy of the predicted element yield of the alloy refined by the heat to be predicted is improved. Meanwhile, the alloy content in the molten steel is adjusted in the alloying process by calculating and determining the alloy batching scheme through predicting the element yield, so that a more scientific alloy feeding table is provided, the utilization rate of alloy batching is improved, the phenomena of alloy waste, quality control fluctuation and the like are effectively improved, and the aims of reducing the refining cost, improving the refining efficiency and stabilizing the refining quality are fulfilled.
Referring to fig. 4, an embodiment of the present disclosure provides a method for predicting yield of refined alloy elements, including:
step S401, obtaining current data of a heat to be predicted;
wherein the current data includes current pre-alloying features;
step S402, determining similar data corresponding to the heat to be predicted from various historical data according to the current pre-alloying characteristics;
step S403, obtaining a prediction model corresponding to the heat to be predicted, and judging whether the model training result of the prediction model is successful; if yes, go to step S404; if not, go to step S408;
s404, determining predicted alloying process characteristics corresponding to the heat to be predicted according to the similar data;
step S405, obtaining the updating time of the prediction model, and judging whether the updating time is within a preset time interval; if yes, go to step S406; if not, executing step S407;
and step S406, inputting the current pre-alloying characteristics and the predicted alloying process characteristics into a prediction model to obtain the predicted element yield of the alloy refined by the furnace to be predicted.
Step S407, acquiring updating data to perform updating training on the prediction model;
wherein the update data is obtained from historical data after the update time.
In step S408, sample data is obtained to train the prediction model.
By adopting the method for predicting the yield of the refined alloy elements, provided by the embodiment of the disclosure, the current data of the to-be-predicted heat is obtained, a plurality of historical data are obtained, the similar data corresponding to the to-be-predicted heat is determined from the historical data according to the current pre-alloying characteristics of the current data, the predicted alloying process characteristics corresponding to the to-be-predicted heat are determined according to the similar data, and if the model training result comprises successful training, the current pre-alloying characteristics and the predicted alloying process characteristics are input into the prediction model, so that the predicted element yield of the refined alloy of the to-be-predicted heat is obtained. Therefore, the predicted alloying process of the heat to be predicted is obtained through the historical data and the current data, and the predicted element yield is obtained through inputting the current data and the characteristics of the predicted alloying process into the successfully trained prediction model, so that the input data of the prediction model has higher reliability, and the accuracy of the predicted element yield of the alloy refined by the heat to be predicted is improved. Meanwhile, the alloy content in the molten steel is adjusted in the alloying process by calculating and determining the alloy batching scheme through predicting the element yield, so that a more scientific alloy feeding table is provided, the utilization rate of alloy batching is improved, the phenomena of alloy waste, quality control fluctuation and the like are effectively improved, and the aims of reducing the refining cost, improving the refining efficiency and stabilizing the refining quality are fulfilled.
Referring to fig. 5, an embodiment of the present disclosure provides a refined alloying element yield prediction system, which includes a first obtaining module 501, a determining module 502, a second obtaining module 503, and a prediction module 504. The first obtaining module 501 is configured to obtain current data of a heat to be predicted, and obtain a plurality of historical data, where the current data includes current pre-alloying features. The determining module 502 is configured to determine similar data corresponding to the to-be-predicted heat from each historical data according to the current pre-alloying feature, and determine a predicted alloying process feature corresponding to the to-be-predicted heat according to the similar data. The second obtaining module 503 is configured to obtain a model training result of the prediction model, where the prediction model is obtained by training a plurality of sample data corresponding to the to-be-predicted heat, and the sample data includes a sample characteristic before alloying, a sample characteristic in an alloying process, and an element yield label. The prediction module 504 is configured to, if the model training result includes successful training, input the current pre-alloying feature and the predicted alloying process feature into the prediction model to obtain the predicted element yield of the alloy refined by the to-be-predicted heat.
By adopting the refined alloy element yield prediction system provided by the embodiment of the disclosure, the current data of the to-be-predicted heat is obtained, a plurality of historical data are obtained, the similar data corresponding to the to-be-predicted heat is determined from the historical data according to the current pre-alloying characteristics of the current data, the predicted alloying process characteristics corresponding to the to-be-predicted heat are determined according to the similar data, and if the model training result comprises successful training, the current pre-alloying characteristics and the predicted alloying process characteristics are input into the prediction model, so that the predicted element yield of the alloy refined by the to-be-predicted heat is obtained. Therefore, the predicted alloying process of the heat to be predicted is obtained through the historical data and the current data, and the predicted element yield is obtained through inputting the current data and the characteristics of the predicted alloying process into the successfully trained prediction model, so that the input data of the prediction model has higher reliability, and the accuracy of the predicted element yield of the alloy refined by the heat to be predicted is improved. Meanwhile, the alloy content in the molten steel is adjusted in the alloying process by calculating and determining the alloy batching scheme through predicting the element yield, so that a more scientific alloy feeding table is provided, the utilization rate of alloy batching is improved, the phenomena of alloy waste, quality control fluctuation and the like are effectively improved, and the aims of reducing the refining cost, improving the refining efficiency and stabilizing the refining quality are fulfilled.
As shown in fig. 6, the present embodiment discloses an electronic device, including: a processor (processor)600 and a memory (memory) 601; the memory is used for storing computer programs, and the processor is used for executing the computer programs stored in the memory so as to enable the terminal to execute the method in the embodiment. Optionally, the electronic device may further include a Communication Interface 602 and a bus 603. The processor 600, the communication interface 602, and the memory 601 may communicate with each other via a bus 603. The communication interface 602 may be used for information transfer. The processor 600 may call logic instructions in the memory 601 to perform the methods in the embodiments described above.
Optionally, the electronic device comprises a computer, a tablet, a smartphone, a server, or the like.
In addition, the logic instructions in the memory 601 may be implemented in the form of software functional units and stored in a computer readable storage medium when the logic instructions are sold or used as independent products.
The memory 601 is a computer-readable storage medium, and can be used for storing software programs, computer-executable programs, such as program instructions/modules corresponding to the methods in the embodiments of the present disclosure. The processor 600 executes the functional application and data processing by executing the program instructions/modules stored in the memory 601, i.e. implements the method in the above-described embodiments.
The memory 601 may include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function; the storage data area may store data created according to the use of the terminal device, and the like. In addition, the memory 601 may include a high speed random access memory, and may also include a non-volatile memory.
By adopting the electronic equipment provided by the embodiment of the disclosure, the current data of the heat to be predicted is obtained, a plurality of historical data are obtained, the similar data corresponding to the heat to be predicted is determined from the historical data according to the current pre-alloying characteristics of the current data, the predicted alloying process characteristics corresponding to the heat to be predicted are determined according to the similar data, and if the model training result comprises successful training, the current pre-alloying characteristics and the predicted alloying process characteristics are input into the prediction model, so that the predicted element yield of the alloy refined by the heat to be predicted is obtained. Therefore, the predicted alloying process of the heat to be predicted is obtained through the historical data and the current data, and the predicted element yield is obtained through inputting the current data and the characteristics of the predicted alloying process into the successfully trained prediction model, so that the input data of the prediction model has higher reliability, and the accuracy of the predicted element yield of the alloy refined by the heat to be predicted is improved. Meanwhile, the alloy content in the molten steel is adjusted in the alloying process by calculating and determining the alloy batching scheme through predicting the element yield, so that a more scientific alloy feeding table is provided, the utilization rate of alloy batching is improved, the phenomena of alloy waste, quality control fluctuation and the like are effectively improved, and the aims of reducing the refining cost, improving the refining efficiency and stabilizing the refining quality are fulfilled.
The present embodiment also discloses a computer-readable storage medium on which a computer program is stored, which when executed by a processor implements any of the methods in the present embodiments.
The computer-readable storage medium in the present embodiment can be understood by those skilled in the art as follows: all or part of the steps for implementing the above method embodiments may be performed by hardware associated with a computer program. The aforementioned computer program may be stored in a computer readable storage medium. When executed, the program performs steps comprising the method embodiments described above; and the aforementioned storage medium includes: various media that can store program codes, such as ROM, RAM, magnetic or optical disks.
The electronic device disclosed in this embodiment includes a processor, a memory, a transceiver, and a communication interface, where the memory and the communication interface are connected to the processor and the transceiver and perform mutual communication, the memory is used to store a computer program, the communication interface is used to perform communication, and the processor and the transceiver are used to run the computer program, so that the electronic device performs the steps of the above method.
In this embodiment, the Memory may include a Random Access Memory (RAM), and may also include a non-volatile Memory (non-volatile Memory), such as at least one disk Memory.
The Processor may be a general-purpose Processor, and includes a Central Processing Unit (CPU), a Network Processor (NP), and the like; the Integrated Circuit may also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, a discrete Gate or transistor logic device, or a discrete hardware component.
The above description and drawings sufficiently illustrate embodiments of the disclosure to enable those skilled in the art to practice them. Other embodiments may incorporate structural, logical, electrical, process, and other changes. The examples merely typify possible variations. Individual components and functions are optional unless explicitly required, and the sequence of operations may vary. Portions and subsamples of some embodiments may be included in or substituted for portions and subsamples of other embodiments. Furthermore, the words used in the specification are words of description only and are not intended to limit the claims. As used in the description of the embodiments and the claims, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. Similarly, the term "and/or" as used in this application is meant to encompass any and all possible combinations of one or more of the associated listed. Furthermore, the terms "comprises," "comprising," and variations thereof, when used in this application, specify the presence of stated sub-samples, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other sub-samples, integers, steps, operations, elements, components, and/or groups thereof. Without further limitation, an element defined by the phrase "comprising an …" does not exclude the presence of other like elements in a process, method or apparatus that comprises the element. In this document, each embodiment may be described with emphasis on differences from other embodiments, and the same and similar parts between the respective embodiments may be referred to each other. For methods, products, etc. of the embodiment disclosures, reference may be made to the description of the method section for relevance if it corresponds to the method section of the embodiment disclosure.
Those of skill in the art would appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software may depend upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the disclosed embodiments. It can be clearly understood by the skilled person that, for convenience and brevity of description, the specific working processes of the system, the apparatus and the unit described above may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
In the embodiments disclosed herein, the disclosed methods, products (including but not limited to devices, apparatuses, etc.) may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the units may be only one logical division, and there may be other divisions in actual implementation, for example, a plurality of units or components may be combined or may be integrated into another system, or some subsamples may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form. The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to implement the present embodiment. In addition, functional units in the embodiments of the present disclosure may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit.
The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). In some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. In the description corresponding to the flowcharts and block diagrams in the figures, operations or steps corresponding to different blocks may also occur in different orders than disclosed in the description, and sometimes there is no specific order between the different operations or steps. For example, two sequential operations or steps may in fact be executed substantially concurrently, or they may sometimes be executed in the reverse order, depending upon the functionality involved. Each block of the block diagrams and/or flowchart illustrations, and combinations of blocks in the block diagrams and/or flowchart illustrations, can be implemented by special purpose hardware-based systems that perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

Claims (10)

1. A method for predicting the yield of a refined alloy element, comprising:
acquiring current data of a heat to be predicted and acquiring a plurality of historical data, wherein the current data comprises current pre-alloying characteristics;
determining similar data corresponding to the to-be-predicted heat from each historical data according to the current pre-alloying characteristics, and determining predicted alloying process characteristics corresponding to the to-be-predicted heat according to the similar data;
obtaining a model training result of a prediction model, wherein the prediction model is obtained by training a plurality of sample data corresponding to the to-be-predicted heat, and the sample data comprises a sample characteristic before alloying, a sample characteristic in an alloying process and an element yield label;
and if the model training result comprises successful training, inputting the current pre-alloying characteristics and the predicted alloying process characteristics into the prediction model to obtain the predicted element yield of the alloy refined by the to-be-predicted heat.
2. The method of claim 1, wherein determining similar data corresponding to the heat to be predicted from each of the historical data according to the current pre-alloying features comprises:
the historical data comprises historical pre-alloying features and historical alloying process features, and the current pre-alloying features and the historical pre-alloying features each comprise a plurality of pre-alloying sub-features;
acquiring a characteristic weighted value corresponding to each pre-alloying sub-characteristic, and determining the similarity between the current data and each historical data according to the current pre-alloying characteristic, the historical pre-alloying characteristic and each characteristic weighted value;
and determining similar data from the historical data according to the similarity.
3. The method of claim 2, wherein the similarity between the current data and the historical data is determined by:
Figure FDA0003401561650000011
wherein Similarity is the Similarity between the current data and the historical data, ScorelA characteristic weighted value, x, corresponding to the first pre-alloying sub-characteristic of the current pre-alloying characteristiclA characteristic value, x, of the l pre-alloying sub-characteristic of the historical pre-alloying characteristicl' is a feature value of the l pre-alloying sub-feature of the current pre-alloying feature, and N is the number of the pre-alloying sub-features.
4. A method according to any one of claims 1 to 3, wherein the model training result of the predictive model is determined by:
dividing the plurality of sample data into a plurality of training data and a plurality of verification data;
training a preset neural network model through each training data to obtain a prediction model;
determining an error value for the prediction model from each of the validation data;
if the error value is smaller than a preset error threshold value, determining the model training result as successful training;
and if the error value is greater than or equal to the preset error threshold value, determining the model training result as training failure.
5. The method of claim 1, wherein obtaining sample data comprises:
determining a plurality of alternative data corresponding to current heat information from each historical data, wherein the current heat information comprises at least one of manufacturer information, steel type information, alloying time information and a time interval corresponding to the heat to be predicted, and the alternative data comprises sample characteristics before alloying and sample characteristics in an alloying process;
if the number of the alternative data is larger than or equal to a training number threshold, acquiring an element yield corresponding to each alternative data, and determining the element yield corresponding to the alternative data as an element yield label of the alternative data to obtain sample data;
and if the quantity of the screening data is less than the threshold value of the training quantity, continuously acquiring and recording new heat production data added to the historical data to obtain new sample data.
6. The method of claim 1, wherein after obtaining the model training results for the predictive model, the method further comprises at least one of:
if the model training result comprises training failure, acquiring element yield corresponding to the similar data, and determining the predicted element yield according to the element yield corresponding to each similar data;
before inputting the current pre-alloying feature and the predicted alloying process feature into the prediction model, obtaining the update time of the prediction model, and presetting a first preset condition, wherein the first preset condition is that the model training result comprises successful training and the update time is within a preset time interval, and if the first preset condition is met, the current pre-alloying feature and the predicted alloying process feature are input into the prediction model;
and before inputting the current pre-alloying feature and the predicted alloying process feature into the prediction model, obtaining the update time of the prediction model, and presetting a first preset condition, wherein the first preset condition is that the model training result comprises successful training and the update time is within a preset time interval, and if the model training result does not meet the first preset condition, obtaining update data to perform update training on the prediction model, wherein the update data is obtained according to historical data after the update time.
7. The method according to claim 5 or 6, characterized in that the element yield is determined by:
Figure FDA0003401561650000031
wherein n is the number of alloying times,
Figure FDA0003401561650000032
the yield of the element, delta w, of the n-th alloying of the j elementjIs the difference between the mass of the j element in the 1 st molten steel and the mass of the j element in the n-th molten steel, G is the initial weight of the molten steel, G ismAdding weight, g, for the m-th alloyiThe added mass of the i-type alloy,
Figure FDA0003401561650000033
is the content of the j element in the i-type alloy.
8. A refined alloying element yield prediction system comprising:
the device comprises a first obtaining module, a second obtaining module and a third obtaining module, wherein the first obtaining module is used for obtaining current data of a to-be-predicted heat and obtaining a plurality of historical data, and the current data comprises current pre-alloying characteristics;
the determining module is used for determining similar data corresponding to the to-be-predicted heat from each historical data according to the current pre-alloying characteristics and determining predicted alloying process characteristics corresponding to the to-be-predicted heat according to the similar data;
the second obtaining module is used for obtaining a model training result of a prediction model, wherein the prediction model is obtained by training a plurality of sample data corresponding to the to-be-predicted heat, and the sample data comprises a sample characteristic before alloying, a sample characteristic in an alloying process and an element yield label;
and the prediction module is used for inputting the current pre-alloying characteristics and the predicted alloying process characteristics into the prediction model to obtain the yield of the predicted elements of the alloy refined by the to-be-predicted heat if the model training result comprises successful training.
9. An electronic device, comprising: a processor and a memory;
the memory is for storing a computer program, and the processor is for executing the computer program stored by the memory to cause the electronic device to perform the method of any of claims 1-7.
10. A computer-readable storage medium having stored thereon a computer program, characterized in that: the computer program, when executed by a processor, implements the method of any one of claims 1 to 7.
CN202111500819.0A 2021-12-09 2021-12-09 Method and system for predicting yield of refined alloy element, electronic device, and medium Pending CN114242179A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111500819.0A CN114242179A (en) 2021-12-09 2021-12-09 Method and system for predicting yield of refined alloy element, electronic device, and medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111500819.0A CN114242179A (en) 2021-12-09 2021-12-09 Method and system for predicting yield of refined alloy element, electronic device, and medium

Publications (1)

Publication Number Publication Date
CN114242179A true CN114242179A (en) 2022-03-25

Family

ID=80754384

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111500819.0A Pending CN114242179A (en) 2021-12-09 2021-12-09 Method and system for predicting yield of refined alloy element, electronic device, and medium

Country Status (1)

Country Link
CN (1) CN114242179A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116469481A (en) * 2023-06-19 2023-07-21 苏州方兴信息技术有限公司 LF refined molten steel composition forecasting method based on XGBoost algorithm

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH07150329A (en) * 1993-11-29 1995-06-13 Kawasaki Steel Corp Method for controlling alloying of galvannealed steel sheet
CN111192144A (en) * 2020-01-03 2020-05-22 湖南工商大学 Financial data prediction method, device, equipment and storage medium
US20200257933A1 (en) * 2019-02-05 2020-08-13 Imagars Llc Machine Learning to Accelerate Alloy Design
CN112036081A (en) * 2020-08-26 2020-12-04 北京科技大学 Method for determining addition amount of silicon-manganese alloy in converter tapping based on yield prediction
CN112215442A (en) * 2020-11-27 2021-01-12 中国电力科学研究院有限公司 Method, system, device and medium for predicting short-term load of power system
CN113617851A (en) * 2021-06-23 2021-11-09 武汉钢铁有限公司 Online feedback control method and device for short-process production line and electronic equipment

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH07150329A (en) * 1993-11-29 1995-06-13 Kawasaki Steel Corp Method for controlling alloying of galvannealed steel sheet
US20200257933A1 (en) * 2019-02-05 2020-08-13 Imagars Llc Machine Learning to Accelerate Alloy Design
CN111192144A (en) * 2020-01-03 2020-05-22 湖南工商大学 Financial data prediction method, device, equipment and storage medium
CN112036081A (en) * 2020-08-26 2020-12-04 北京科技大学 Method for determining addition amount of silicon-manganese alloy in converter tapping based on yield prediction
CN112215442A (en) * 2020-11-27 2021-01-12 中国电力科学研究院有限公司 Method, system, device and medium for predicting short-term load of power system
CN113617851A (en) * 2021-06-23 2021-11-09 武汉钢铁有限公司 Online feedback control method and device for short-process production line and electronic equipment

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
GHAMARIAN, I等: "Developing a phenomenological equation to predict yield strength from composition and microstructure in β processed Ti-6Al-4V", MATERIALS SCIENCE AND ENGINEERING A-STRUCTURAL MATERIALS PROPERTIES MICROSTRUCTURE AND PROCESSING, vol. 660, 13 April 2016 (2016-04-13), pages 172 - 180, XP029470257, DOI: 10.1016/j.msea.2016.02.052 *
沈守娟;: "合金收得率影响因素与预测模型", 电子质量, no. 12, 20 December 2019 (2019-12-20), pages 1 - 7 *
谭敦强等: "Al-Fe系合金中的相及相转变", 材料导报, vol. 17, no. 05, 15 May 2003 (2003-05-15), pages 18 - 27 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116469481A (en) * 2023-06-19 2023-07-21 苏州方兴信息技术有限公司 LF refined molten steel composition forecasting method based on XGBoost algorithm
CN116469481B (en) * 2023-06-19 2023-08-29 苏州方兴信息技术有限公司 LF refined molten steel composition forecasting method based on XGBoost algorithm

Similar Documents

Publication Publication Date Title
Sallet et al. EuGene: an automated integrative gene finder for eukaryotes and prokaryotes
Long et al. Dynamic scheduling in steelmaking-continuous casting production for continuous caster breakdown
CN114242179A (en) Method and system for predicting yield of refined alloy element, electronic device, and medium
JP2023106037A (en) Driving assist system, driving assist method, and program
Navarra et al. Quantitative methods to support data acquisition modernization within copper smelters
EP4083872A1 (en) Intention feature value extraction device, learning device, method, and program
CN111627210B (en) Traffic flow prediction method, device, equipment and medium
CN113705074A (en) Chemical accident risk prediction method and device
CN116341750A (en) Steel inclusion level prediction method, device, terminal and storage medium
Cateni et al. The importance of variable selection for neural networks-based classification in an industrial context
Liu et al. XGBoost-based model for predicting hydrogen content in electroslag remelting
Jarosz et al. Metamodeling and optimization of a blister copper two-stage production process
CN115543638A (en) Uncertainty-based edge calculation data collection and analysis method, system and equipment
Liu et al. Optimization of a molten iron scheduling problem with uncertain processing time using variable neighborhood search algorithm
Murua et al. A slag prediction model in an electric arc furnace process for special steel production
CN111832815A (en) Scientific research hotspot prediction method and system
CN113617851A (en) Online feedback control method and device for short-process production line and electronic equipment
Touil et al. A hybrid metaheuristic method to optimize the order of the sequences in continuous-casting
Niyayesh et al. Predicting Steel Grade Based on Electric Arc Furnace End Point Parameters
CN113987008B (en) Blast furnace slag peel falling judgment method, terminal equipment and storage medium
CN117313529A (en) Converter steelmaking oxygen blowing amount prediction method, electronic equipment and storage medium
Zhang et al. Improved back propagation neural network method for predicting sulfur content in hot metal
Majumder et al. Use of artificial neural network to predict the yield of sinter plant as a function of production parameters
Zhao et al. Prediction of desulfurization efficiency and costs during Kanbara reactor hot metal treatment using machine learning
CN114496110B (en) Method and system for predicting concentration of cobalt ions at outlet of zinc hydrometallurgy purification cobalt removal reactor

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
CB02 Change of applicant information

Address after: 401329 No. 5-6, building 2, No. 66, Nongke Avenue, Baishiyi Town, Jiulongpo District, Chongqing

Applicant after: MCC CCID information technology (Chongqing) Co.,Ltd.

Address before: 401329 No. 5-6, building 2, No. 66, Nongke Avenue, Baishiyi Town, Jiulongpo District, Chongqing

Applicant before: CISDI CHONGQING INFORMATION TECHNOLOGY Co.,Ltd.

CB02 Change of applicant information