CN108469797B - Neural network and evolutionary computation based ore grinding process modeling method - Google Patents

Neural network and evolutionary computation based ore grinding process modeling method Download PDF

Info

Publication number
CN108469797B
CN108469797B CN201810399311.8A CN201810399311A CN108469797B CN 108469797 B CN108469797 B CN 108469797B CN 201810399311 A CN201810399311 A CN 201810399311A CN 108469797 B CN108469797 B CN 108469797B
Authority
CN
China
Prior art keywords
ore
ball mill
feeding amount
granularity
ore grinding
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.)
Active
Application number
CN201810399311.8A
Other languages
Chinese (zh)
Other versions
CN108469797A (en
Inventor
高宪文
郝得智
王明顺
佟俊霖
张鼎森
刘博健
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.)
Northeastern University China
Original Assignee
Northeastern University China
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 Northeastern University China filed Critical Northeastern University China
Priority to CN201810399311.8A priority Critical patent/CN108469797B/en
Publication of CN108469797A publication Critical patent/CN108469797A/en
Application granted granted Critical
Publication of CN108469797B publication Critical patent/CN108469797B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B19/00—Programme-control systems
    • G05B19/02—Programme-control systems electric
    • G05B19/418—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS], computer integrated manufacturing [CIM]
    • G05B19/41885—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS], computer integrated manufacturing [CIM] characterised by modeling, simulation of the manufacturing system
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B2219/00—Program-control systems
    • G05B2219/30—Nc systems
    • G05B2219/32—Operator till task planning
    • G05B2219/32339—Object oriented modeling, design, analysis, implementation, simulation language
    • 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
    • Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]

Abstract

The invention provides a neural network and evolutionary computation-based ore grinding process modeling method, and relates to the technical field of iron ore grinding. Firstly, establishing a case base, and searching out reasonable ore feeding amount of the ball mill from the case base by adopting a case searching method; establishing a mathematical model of the ball mill in the ore grinding process by a neural network method, and establishing a relation between the ore feeding amount and the water feeding amount of the ball mill and the ore grinding effect; the method comprises the steps of taking the maximum specific productivity of the ball mill and the optimal particle size distribution of ground ore as optimization targets, determining constraint conditions by combining actual working conditions, obtaining a group of non-inferior solution sets through a non-inferior sorting genetic algorithm with an elite strategy, and deciding an optimal solution by adopting a TOPSIS algorithm. The ore grinding process modeling method based on the neural network and the evolutionary computation calculates reasonable ore feeding amount and water feeding amount, increases the processing efficiency of the ball mill on the basis of ensuring the ore granularity, and improves the stability, reliability and economy of the ore grinding production process.

Description

Neural network and evolutionary computation based ore grinding process modeling method
Technical Field
The invention relates to the technical field of iron ore grinding, in particular to a neural network and evolutionary computation-based ore grinding process modeling method.
Background
The ore grinding operation is a key link in the metal ore dressing process flow, the ore grinding effect directly influences the ore dressing effect, meanwhile, the ore grinding operation is also a main energy consumption and material consumption unit in the ore dressing process flow, and how to control the optimized operation of the process is one of the keys of the whole ore dressing process. The factors influencing the ore grinding process are more, the influence process of each factor is complex, and the mapping relation between ore grinding stones and ore grinding is difficult to determine. At present, the ore grinding operation mostly depends on manual experience to perform tentative adjustment, so that the parameters of the ball mill are difficult to be adjusted accurately in time according to different ore properties, the variation fluctuation of the granularity of the ground ore is increased, the difficulty of the subsequent ore dressing process is improved, and the ore dressing result is influenced.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides a neural network and evolutionary computation-based ore grinding process modeling method, which realizes modeling of an ore grinding process and further obtains operation guidance of the ore grinding process.
A grinding process modeling method based on neural network and evolutionary computation comprises the following steps:
step 1: collecting historical production data in the ore grinding production process, wherein the historical production data comprises the water supply amount and overflow granularity of the ball mill during the ball mill;
step 2: collecting historical data of the reaction ore property and the optimal beneficiation granularity; the historical data of the reaction ore properties comprises ore grade, feed granularity, pipe refined iron content, iron oxide content and magnetic iron content;
and step 3: the method of case retrieval is adopted, the ore feeding amount of the ball mill is decided according to the ore property and the optimal ore granularity, and the specific method is as follows:
3.1, constructing a case library according to the historical data of the reaction ore property, the optimal ore dressing granularity and the historical data of the ball mill during the operation, which are collected in the step 2; wherein, the ore grade, the feeding granularity, the content of refined iron in the tube, the content of ferric oxide, the content of magnetic iron and the optimal ore dressing granularity are described by case, and the feeding amount of the ball mill is solved by case;
step 3.2: obtaining real-time data of ore properties as a new case, and performing case retrieval on the new case description and historical cases recorded in a case library to obtain a group of similarity values;
step 3.3: arranging the obtained similarity values from large to small, and selecting the cases in the case base corresponding to the first n similarity values as reference cases of the current working condition by adopting a cross validation method;
step 3.4: reusing cases according to reference cases retrieved from a case library to obtain a new case solution which is used as the ore feeding amount of the current working condition;
and 4, step 4: according to the collected historical data of the ore grinding production process, fitting the relation among ore feeding amount, water feeding amount, overflow granularity and ore grinding granularity distribution of the ball mill by a neural network algorithm, wherein the specific method comprises the following steps:
step 4.1: standardizing the ore feeding amount, the water feeding amount, the overflow granularity and the ore grinding granularity of the ball mill by adopting a z-score method;
step 4.2: randomly sequencing the normalized ore feeding amount, water feeding amount, overflow granularity and ore grinding granularity data of the ball mill, taking the first seventy-five percent of the data as training data, and taking the last twenty-five percent as test data;
step 4.3: fitting the relation between the overflow granularity and the ore grinding granularity and the ore feeding amount and the water feeding amount by adopting a neural network algorithm, testing the fitting result, if the fitting error is less than 2% and the testing error is less than 4%, storing the fitted relation, and otherwise fitting the relation between the overflow granularity and the ore grinding granularity and the ore feeding amount and the water feeding amount again;
and 5: according to the requirements of an ore grinding production process, setting the maximum specific productivity of the ball mill as a first optimization target, setting the optimal ore grain size distribution ground by the ball mill as a second optimization target, and setting the saturation degree of the ball mill, the ore grinding concentration of the ball mill, the ore feeding amount and the water feeding amount as constraint conditions by combining an actual production flow;
step 6: calculating a non-inferior solution set of the water supply amount by a non-inferior sorting genetic algorithm (namely NSGA-II) with an elite strategy, and deciding the optimal solution of the water supply amount of the ball mill according to the ore property by adopting a TOPSIS decision method to serve as an operation guide in the ore grinding production process.
According to the technical scheme, the invention has the beneficial effects that: according to the ore grinding process modeling method based on the neural network and the evolutionary computation, operation guidance of each operation parameter in the ore grinding process is given through optimization computation according to the ore grinding property, the real-time performance, the reliability and the economical efficiency of the ore grinding production process can be improved, the adjusting time is shortened, the ore grinding granularity is stabilized, and meanwhile, the operation can be guided when experts are not present, so that the ore grinding production efficiency is improved.
Drawings
Fig. 1 is a flowchart of a modeling method for a grinding process based on a neural network and evolutionary computation according to an embodiment of the present invention.
Detailed Description
The following detailed description of embodiments of the present invention is provided in connection with the accompanying drawings and examples. The following examples are intended to illustrate the invention but are not intended to limit the scope of the invention.
A neural network and evolutionary computation based ore grinding process modeling method, as shown in fig. 1, includes the following steps:
step 1: collecting historical production data in the ore grinding production process, wherein the historical production data comprises the water supply amount and the overflow granularity of the ball mill when the ball mill stands;
in the present example, part of the history data of the ball mill stand time, the ball mill feed water amount, and the overflow particle size are shown in table 1:
TABLE 1 partial historical production data sheet for a grinding process
Figure BDA0001645318890000031
Step 2: collecting historical data of the reaction ore property and the optimal beneficiation granularity; the historical data of the reaction ore properties comprises ore grade, feed granularity, pipe refined iron content, iron oxide content and magnetic iron content;
in this example, the ore grade, feed particle size-12 mm content, tube fine iron content, ferrous oxide, magnetic iron content, and the-200 mesh content corresponding to the optimum ore dressing particle size are shown in table 2:
TABLE 2 Ore Property History data sheet
Figure BDA0001645318890000032
And step 3: the method of case retrieval is adopted, the ore feeding amount of the ball mill is decided according to the ore property and the optimal ore granularity, and the specific method is as follows:
3.1, constructing a case library according to the historical data of the reaction ore property, the optimal ore dressing granularity and the historical data of the ball mill during the operation, which are collected in the step 2; wherein, the ore grade, the feeding granularity, the content of refined iron in the tube, the content of ferric oxide, the content of magnetic iron and the optimal ore dressing granularity are used as case description, and the feeding amount of the ball mill is used as case solution;
by combining historical data, production field investigation and theoretical analysis of the ore grinding process, and aiming at the working condition of ore grinding type fluctuation, the ore property data can directly reflect the grindability of ore, has direct influence on the ore feeding amount of the ball mill, and has large proportion.
The structure of the cases in the case library consists of case description and case solution, wherein the case description is the grade f of the total iron of the ore1Feed particle size f2Content of refined iron in tube f3Magnetic iron content f4Ferrous oxide content f5Optimum beneficiation particle size f6Composition is carried out; case solution to ore feeding quantity j of floating ball mill1. Therefore, the case based case reasoning technique is shown by the following formula:
Ck＝{Fk，JK}
wherein, CkThe k-th case in the case base is 1, 2, … and m, wherein m is the number of cases in the case base; fk＝{fk，1，fk，2，fk，3，fk，4，fk，5，fk，6The description is the characteristic description of the kth case; j. the design is a squareK＝{jk，1The k case solution is the case solution of the k case.
Step 3.2: according to the obtained real-time ore property data, carrying out case retrieval on the new case description and the historical cases recorded in the case base to obtain a group of similarity values, wherein the similarity values are shown in the following formula:
Figure BDA0001645318890000041
wherein, ω isiFor case feature attribute weight, SIM (f)i，fk.i) Case description f for the current operating modeiAnd the kth in case libraryCase description f corresponding to a casek.iThe similarity of (c) is defined as shown in the following formula:
Figure BDA0001645318890000042
the real-time data obtained for the ore properties in this example are shown in table 3:
TABLE 3 real-time data of ore properties
Figure BDA0001645318890000043
Step 3.3: arranging the obtained similarity values from large to small, and selecting the cases in the case base corresponding to the first n similarity values as reference cases of the current working condition by adopting a cross validation method;
the choice of the value of n has a significant impact on the outcome of the algorithm; the small value of n means that only the historical cases closer to the new case will have an effect on the new case solution, but overfitting is easy to happen; if the value of n is larger, the estimation error of learning can be reduced, but the approximation error of learning is increased, and the historical case far away from the new case can also act on the prediction, so that the prediction is wrong. In practical applications, a smaller value is generally selected for the n value, and a cross-validation method is usually adopted to select the optimal n value. In this embodiment, n is 3.
Step 3.4: and reusing the cases according to the reference cases retrieved from the case library to obtain a new case solution which is used as the ore feeding amount of the ball mill under the current working condition.
According to the reference cases retrieved from the case base, the weighted average value of each case solution of the n reference cases is obtained, wherein the weighting coefficient is the similarity between each reference case and the new case, thereby completing case reuse and obtaining the new case solution
Figure BDA0001645318890000051
Taking the result as the ore feeding amount calculation result of the ball mill under the current working condition, wherein J is a new case solution, JkFor referenceExample case solutions.
And 4, step 4: according to the collected ore grinding production process historical data, fitting the relationship between ore feeding amount and water feeding amount of the ball mill and the relationship between overflow particle size and ore grinding particle size distribution through a BP neural network algorithm, wherein the specific method comprises the following steps:
step 4.1: the Z-score method is adopted to carry out standardization treatment on the ore feeding amount, the water feeding amount, the overflow granularity and the ore grinding granularity of the ball mill.
Step 4.2: randomly sequencing the normalized ore feeding amount, water feeding amount, overflow granularity and ore grinding granularity data of the ball mill, taking the first seventy-five percent of the data as training data, and taking the last twenty-five percent as test data;
step 4.3: fitting the relation between the overflow granularity and the ore grinding granularity and the ore feeding amount and the water feeding amount by adopting a BP neural network algorithm, testing the fitting result, if the fitting error is less than 2% and the testing error is less than 4%, storing the fitted relation, and otherwise fitting the relation between the overflow granularity and the ore grinding granularity and the ore feeding amount and the water feeding amount again;
taking the ore feeding quantity Fm and the water feeding quantity Ms of the ball mill as input data of a neural network model, taking the overflow granularity R and the ore grinding granularity distribution coefficient m' as output data of the neural network model, and obtaining the fitting results of the overflow granularity, the ore grinding granularity, the ore feeding quantity and the water feeding quantity through the fitting of the neural network model, wherein the fitting results are shown in the following formula:
(R，m)＝f(Fm，Ms)
from RR (rosin Rammler) distribution formula:
Figure BDA0001645318890000053
wherein: r is the percentage of particles larger than the particle diameter dpIn this example, d is 0.074mm, i.e., R is-200 mesh, and d is the characteristic particle diameterp＝0.054mm。
Further obtaining the particle size distribution coefficient, which is shown in the following formula:
Figure BDA0001645318890000052
and 5: according to the requirements of an ore grinding production process, setting the maximum specific productivity of the ball mill as a first optimization target, setting the optimal ore grain size distribution ground by the ball mill as a second optimization target, and setting the saturation degree of the ball mill, the ore grinding concentration of the ball mill, the ore feeding amount and the water feeding amount as constraint conditions by combining an actual production flow;
in this embodiment, the constraint conditions given by combining the working conditions corresponding to the field conditions are shown in table 4:
TABLE 4 constraint condition table corresponding to field situation
Figure BDA0001645318890000061
Step 6: calculating a non-inferior solution set of the water supply amount by a non-inferior sorting genetic algorithm (namely NSGA-II) with an elite strategy, and deciding the optimal solution of the water supply amount of the ball mill according to the ore property by adopting a TOPSIS decision method to serve as an operation guide in the ore grinding production process.
In this example, the obtained non-inferior solution set of the feed water amount is shown in table 5:
TABLE 5 optimized non-inferior solution set Table
Figure BDA0001645318890000062
In this embodiment, the finally obtained operation guidance is: the feeding amount of the ball mill is 301.24t/h, and the water feeding amount is 30.86 t/h.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; such modifications and substitutions do not depart from the spirit of the corresponding technical solutions and scope of the present invention as defined in the appended claims.

Claims (2)

1. A ore grinding process modeling method based on neural network and evolutionary computation is characterized in that: the method comprises the following steps:
step 1: collecting historical production data in the ore grinding production process, wherein the historical production data comprises the water supply amount and overflow granularity of the ball mill during the ball mill;
step 2: collecting historical data of the reaction ore property and the optimal beneficiation granularity; the historical data of the reaction ore properties comprises ore grade, feed granularity, pipe refined iron content, iron oxide content and magnetic iron content;
and step 3: determining the ore feeding amount of the ball mill according to the ore property and the optimal ore granularity by adopting a case retrieval method;
and 4, step 4: fitting the relation between ore feeding amount and water feeding amount of the ball mill and the relation between overflow granularity and ore grinding granularity distribution through a neural network algorithm according to the collected historical data of the ore grinding production process;
and 5: according to the requirements of an ore grinding production process, setting the maximum specific productivity of the ball mill as a first optimization target, setting the optimal ore grain size distribution ground by the ball mill as a second optimization target, and setting the saturation degree of the ball mill, the ore grinding concentration of the ball mill, the ore feeding amount and the water feeding amount as constraint conditions by combining an actual production flow;
step 6: calculating a non-inferior solution set of the water supply amount through a non-inferior sorting genetic algorithm with an elite strategy, and deciding an optimal solution of the water supply amount of the ball mill according to the ore property by adopting a TOPSIS (technique for order preference by similarity to similarity) decision-making method to serve as an operation guide in the ore milling production process;
3.1, constructing a case library according to the historical data of the reaction ore property, the optimal ore dressing granularity and the historical data of the ball mill during the operation, which are collected in the step 2; wherein, the ore grade, the feeding granularity, the content of refined iron in the tube, the content of ferric oxide, the content of magnetic iron and the optimal ore dressing granularity are described by case, and the feeding amount of the ball mill is solved by case;
step 3.2: obtaining real-time data of ore properties as a new case, and performing case retrieval on the new case description and historical cases recorded in a case library to obtain a group of similarity values;
step 3.3: arranging the obtained similarity values from large to small, and selecting the cases in the case base corresponding to the first n similarity values as reference cases of the current working condition by adopting a cross validation method;
step 3.4: and reusing the cases according to the reference cases retrieved from the case base to obtain a new case solution which is used as the ore feeding amount of the current working condition.
2. The ore grinding process modeling method based on neural network and evolutionary computation of claim 1, characterized in that: the specific method of the step 4 comprises the following steps:
step 4.1: standardizing the ore feeding amount, the water feeding amount, the overflow granularity and the ore grinding granularity of the ball mill by adopting a z-score method;
step 4.2: randomly sequencing the normalized ore feeding amount, water feeding amount, overflow granularity and ore grinding granularity data of the ball mill, taking the first seventy-five percent of the data as training data, and taking the last twenty-five percent as test data;
step 4.3: and fitting the relation between the overflow granularity and the ore grinding granularity and the relation between the ore feeding amount and the water feeding amount by adopting a neural network algorithm, testing the fitting result, if the fitting error is less than 2% and the testing error is less than 4%, storing the fitted relation, and otherwise, fitting the relation between the overflow granularity and the ore grinding granularity and the relation between the ore feeding amount and the water feeding amount again.
CN201810399311.8A 2018-04-28 2018-04-28 Neural network and evolutionary computation based ore grinding process modeling method Active CN108469797B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810399311.8A CN108469797B (en) 2018-04-28 2018-04-28 Neural network and evolutionary computation based ore grinding process modeling method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810399311.8A CN108469797B (en) 2018-04-28 2018-04-28 Neural network and evolutionary computation based ore grinding process modeling method

Publications (2)

Publication Number Publication Date
CN108469797A CN108469797A (en) 2018-08-31
CN108469797B true CN108469797B (en) 2020-09-29

Family

ID=63263867

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810399311.8A Active CN108469797B (en) 2018-04-28 2018-04-28 Neural network and evolutionary computation based ore grinding process modeling method

Country Status (1)

Country Link
CN (1) CN108469797B (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109569887A (en) * 2018-11-23 2019-04-05 鞍钢集团矿业有限公司 A kind of floatation of iron ore dosing autocontrol method based on tailings grade
CN110270414B (en) * 2019-05-30 2021-02-09 宜春钽铌矿有限公司 Method for online real-time detection of ore properties in ore grinding process
CN110378799B (en) * 2019-07-16 2022-07-12 东北大学 Alumina comprehensive production index decision method based on multi-scale deep convolution network
CN112598618B (en) * 2020-11-16 2023-11-17 鞍钢集团矿业有限公司 Image recognition technology-based ore feeding amount prediction method for mill

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2008040682A (en) * 2006-08-03 2008-02-21 Matsushita Electric Works Ltd Abnormality monitoring device
CN106292292A (en) * 2016-10-17 2017-01-04 鞍钢集团矿业有限公司 The floatation of iron ore dosing Optimal Setting method and system of case-based reasioning
CN106406257A (en) * 2016-10-17 2017-02-15 鞍钢集团矿业有限公司 Iron ore flotation concentrate grade soft measurement method and system based on case-based reasoning
CN107133723A (en) * 2017-04-18 2017-09-05 东北大学 It is a kind of based on the ore dressing overall target Forecasting Methodology with mineral products property
CN107145970A (en) * 2017-04-18 2017-09-08 东北大学 One kind is based on the maximized milling ore Optimization Ore Matching method of Utilization Rate of Mineral Resources

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8598980B2 (en) * 2010-07-19 2013-12-03 Lockheed Martin Corporation Biometrics with mental/physical state determination methods and systems
CN104134120B (en) * 2014-07-30 2017-05-24 东北大学 System and method for monitoring ore-dressing production indexes
CN107330451B (en) * 2017-06-16 2020-06-26 西交利物浦大学 Clothing attribute retrieval method based on deep convolutional neural network

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2008040682A (en) * 2006-08-03 2008-02-21 Matsushita Electric Works Ltd Abnormality monitoring device
CN106292292A (en) * 2016-10-17 2017-01-04 鞍钢集团矿业有限公司 The floatation of iron ore dosing Optimal Setting method and system of case-based reasioning
CN106406257A (en) * 2016-10-17 2017-02-15 鞍钢集团矿业有限公司 Iron ore flotation concentrate grade soft measurement method and system based on case-based reasoning
CN107133723A (en) * 2017-04-18 2017-09-05 东北大学 It is a kind of based on the ore dressing overall target Forecasting Methodology with mineral products property
CN107145970A (en) * 2017-04-18 2017-09-08 东北大学 One kind is based on the maximized milling ore Optimization Ore Matching method of Utilization Rate of Mineral Resources

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
《基于NSGA-Ⅱ算法的磨矿过程稳态优化》;陈宝宇;《中国优秀硕士学位论文全文数据库 工程科技Ⅰ辑》;20120615;第2章、第4.2.2节 *

Also Published As

Publication number Publication date
CN108469797A (en) 2018-08-31

Similar Documents

Publication Publication Date Title
CN108469797B (en) Neural network and evolutionary computation based ore grinding process modeling method
CN110378799B (en) Alumina comprehensive production index decision method based on multi-scale deep convolution network
CN110245850B (en) Sintering process working condition identification method and system considering time sequence
CN104375478B (en) A kind of method and device of Rolling production process product quality on-line prediction and optimization
Li et al. Data-driven multiobjective optimization for burden surface in blast furnace with feedback compensation
CN106845012B (en) A kind of blast furnace gas system model subordinating degree function based on multiple target Density Clustering determines method
CN108846526A (en) A kind of CO2 emissions prediction technique
CN112100745A (en) Automobile girder steel mechanical property prediction method based on LDA theory
CN110991495A (en) Method, system, medium, and apparatus for predicting product quality in manufacturing process
CN110246547B (en) Ore blending optimization method in sintering process
CN113296561A (en) Control system and control method for pelletizer
CN116070750A (en) Multi-target robust optimization decision method for comprehensive production index of mineral separation day
CN100370453C (en) Rare-earth cascade extraction separation component content soft measuring method
CN110502781A (en) A kind of ferroalloy production blending optimization method based on priori knowledge
CN107016454B (en) A kind of heterogeneous ore deposit ball mill crushing distribution function Forecasting Methodology
CN114871000B (en) Flotation dosing self-adaptive regulation and control method
CN116305885A (en) Method, device and storage medium for predicting or controlling iron loss of silicon steel
CN115394381A (en) High-entropy alloy hardness prediction method and device based on machine learning and two-step data expansion
CN115994784A (en) Price determination model and construction method thereof
CN115457337A (en) Image classification method containing fine-grained noise based on label distribution learning
CN115169453A (en) Hot continuous rolling width prediction method based on density clustering and depth residual error network
CN113962150A (en) Method and system for predicting granularity of sintering mixture
CN115186900A (en) Dynamic blast furnace gas production prediction method and system suitable for multiple working condition types
CN111881993B (en) Operation mode multilayer hierarchical matching optimization method for copper matte converting process
CN115049123A (en) Prediction method for silicon content of molten iron in blast furnace based on GA-XGboost model

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
GR01 Patent grant
GR01 Patent grant