CN105787509A - Iron mine blending process pre-proportion method - Google Patents
Iron mine blending process pre-proportion method Download PDFInfo
- Publication number
- CN105787509A CN105787509A CN201610105488.3A CN201610105488A CN105787509A CN 105787509 A CN105787509 A CN 105787509A CN 201610105488 A CN201610105488 A CN 201610105488A CN 105787509 A CN105787509 A CN 105787509A
- Authority
- CN
- China
- Prior art keywords
- groove
- raw material
- class
- blending process
- iron mine
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/2431—Multiple classes
-
- 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—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/04—Manufacturing
-
- 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
Abstract
The present invention relates to the field of data processing, clustering algorithm and combination optimization technology, especially to an iron mine blending process pre-proportion method. The present invention provides an iron mine blending process pre-proportion method based on the clustering algorithm and the combination optimization. The method comprises the steps: performing preprocessing operation of raw material component data; completing clustering operation once according to the content of the silicon in the raw material components, namely dividing the raw material components into N large classes through adoption of a Canopy-Kmeans clustering method; performing the secondary clustering according to the content of the silicon through adoption of the K-means algorithm; taking constraint conditions such as chemical element indexes and the like into account, and performing classification result processing through adoption of the merging and decomposition methods; and determining an optimal blanking order through a combination optimization idea, and obtaining a final proportion scheme. The iron mine blending process pre-proportion method saves calculation time consuming, ensures the errors of blending material chemical components of each class as small as possible, and is small in the variance fluctuation range of the blending material chemical components between classes and high in precision.
Description
Technical field
The present invention relates to the technical fields such as data process, clustering algorithm, Combinatorial Optimization, particularly relate to a kind of iron mine and mixed
The pre-distribution of journey.
Background technology
At present, steel production in China industry is all constantly promoting in the demand of iron and steel production quantity and quality, arrives
2014 only, and the iron and steel output share of China has occupied the 45% of the whole world.But owing to starting late, China is in terms of iron and steel production
Technology still and between developed country, there is bigger gap.And stock yard blending process, the first work produced as iron and steel
Sequence, it is according to windrow mission requirements monthly, by limited mixing facility, by bigger to multiple source difference and component difference
Iron-bearing mineral is uniformly mixed by different ratio so that various iron ore raw material was finished and ensured that the whole month appoints the whole moon to factory
The chemical composition of what class's mixing material is uniform and stable, and i.e. per tour mixing material is in the fluctuation range that enterprise and state quota require, and
The chemical composition variance fluctuation mixing material between class and class is the least, and the production technology meeting follow-up sintering and ironmaking use is wanted
Ask, and then obtain final proportion scheme.It is critical only that of stock yard blending process determines the thing being total to groove dispensing in blending process
Material combination and their blanking order.And iron-bearing mineral to be mixed is of a great variety, mineralogical composition differs greatly, and
Dosage bunker limited amount, it is ensured that in the whole month, the chemical composition of per tour mixing material is uniform and stable extremely difficult, the most still lacks full
The mixing dosage computational methods of foot requirement of real-time.Can blending process proportioning calculate and promptly and accurately be directly connected in mixing material
Whether the chemical composition such as silicon, ferrum meets subsequent production requirement, and then has influence on production efficiency and the product matter of whole iron and steel enterprise
Amount, therefore needs a kind of real-time computational methods for iron mine blending process badly, and the computational methods in this field are broadly divided at present
The mixing dosage method calculated based on artificial experience and mixing dosage method based on fuzzy theory Modeling Calculation.
Summary of the invention
(1) to solve the technical problem that
The technical problem to be solved in the present invention be solve due to iron ore raw material iron-bearing mineral of a great variety, mineralogical composition is poor
Different relatively big, composition fluctuation is big and dosage bunker limited amount, it is impossible to ensure that the chemical composition of per tour mixing material is asked uniformly in the whole month
Topic.
(2) technical scheme
In order to solve above-mentioned technical problem, the invention provides the pre-distribution of a kind of iron mine blending process, it includes
Following step:
Step one: according to the various iron ore raw material compositional datas in this month to factory, carries out pretreatment operation, gets rid of silicon, ferrum unit
Element component difference is relatively greatly or its physicochemical characteristics cannot be total to the special material of groove, obtains raw material collection to be clustered；
Step 2: the raw material collection to be clustered obtaining step one uses Canopy-Kmeans clustering algorithm once to gather
Generic operation.The division of big class is carried out so that it is be automatically divided into rational N number of big class, reach primary element according to primary constituent content
Balance；
Step 3: use K-Means algorithm to carry out secondary cluster the once cluster raw material set obtained by step 2
Operation.I.e. again cluster according to minor element content in material composition contained by each big apoplexy due to endogenous wind and realize group division, full
On the premise of the primary element of foot is in a basic balance, the composition reaching each little apoplexy due to endogenous wind minor element approximates substantially；
Step 4: the thought merging each little class a operation obtained in step 3 and decomposing processes.Consider reality
In production, dosage bunker number, mixing expect each element index, number and the total siccative amount etc. of arranging an order according to class and grade constraints, to gained cluster result and spy
Different raw material merges and is automatically brought into operation with decomposition etc., determines final clustering schemes, it is ensured that all kinds of middle raw material types, quantity meet real
Border production requirement, i.e. guarantees that finally clustering number is less than, equal to available dosage bunker number, every class raw material type, the raw material that can be total to groove
The sum upper limit, meets other simultaneously and produces constraint requirements.Thus, every class raw material participates in dispensing in a dosage bunker, to common groove
Plurality of raw materials determine blanking precedence in groove at random, and then generate initial proportion scheme；
Step 5: use combined optimization method that the initial ingredients scheme obtained in step 4 carries out common groove raw material blanking
The optimization of sequence, determines final proportion scheme.The shadow of the factor such as consider that raw material type is various, composition fluctuation is many compared with big and constraints
Ring, the method for exhaustion in each raw material transport Combinatorial Optimization in common groove determined to the priority ordering of each class groove dispensing altogether,
The mixing material composition making any order of classes or grades at school of the whole moon fluctuates in the range of the minimal ripple of plan component target, obtains final joining
Material scheme.
In step one, the process of raw material compositional data pretreatment is:
Silicon, ferrum element component content are differed greatly or plan the raw material that proportioning is bigger, because being total to unclassified stores
Groove, needs to get rid of, and stops it to join in cluster sample；Physicochemical characteristics for raw material such as toughness own, aqueous
Amount etc. causes being total to groove, is distinguished by this kind of material according to practical production experience and feedstock assay result and gets rid of, hindering
During only it joins cluster sample.
Raw material set to be clustered is used the process that Canopy-Kmeans clustering algorithm once clusters by step 2
For:
According to constituent content difference primary in material composition, use Canopy-Kmeans clustering technique by raw material to be clustered
Collection is automatically divided into rational N number of big class, reaches the balance of primary constituent content, it is to avoid artificial appointment number of clusters and drop
The robustness of low method；Feedstock property and blanking that such as different iron-smelters are used require it is different, need to be according to its raw material
The feature of compositional data carries out intelligence choose quantity and the initial point of cluster, uses Canopy-Kmeans clustering technique to overcome
The shortcoming that traditional K-Means method is absorbed in local optimum when number of clusters and initial point are chosen.Specifically comprise the following steps that
The pretreated Ge Ru factory primary constituent content of iron mine material is stored in data set, selects two threshold values T1And T2,
Wherein T1>T2.1 P of arbitrary extracting again, calculates the distance of P point and Canopy, if P point is in distance T of certain Canopy1With
In, then P point is added this Canopy；If this P point is once in distance T of certain Canopy2Within, then P from data set
Deleting, repeat above-mentioned two steps until data set is sky, final number N adding up Canopy class, i.e. by Canopy clustering method
Legacy data is divided into N number of big class.
Step 3 to once clustering raw material collection and using K-Means algorithm to carry out the process of secondary cluster operation is:
By once clustering in the N drawn big class material, at each big apoplexy due to endogenous wind, minor element is carried out the K-Means in class and gather
Class, is divided into two groups by the material in big class, makes minor element composition gap in the material that primary element approximates diminish, by institute
Two the group data divided temporarily are assigned in common groove memory space, to reach the mesh that each apoplexy due to endogenous wind minor element approximates substantially
's；
Step 4, the thought merging each little class a operation obtained in step 3 and decomposing processes.Consider reality
In production, dosage bunker number, mixing expect each element index, number and the total siccative amount etc. of arranging an order according to class and grade constraints, to gained cluster result and spy
Different raw material merges and is automatically brought into operation with decomposition etc., determines final clustering schemes, it is ensured that all kinds of middle raw material types, quantity meet real
Border production requirement, i.e. guarantees that finally clustering number is less than, equal to available dosage bunker number, every class raw material type, the raw material that can be total to groove
The sum upper limit, meets other simultaneously and produces constraint requirements.Thus, every class raw material participates in dispensing in a dosage bunker, to common groove
Plurality of raw materials determine blanking precedence in groove at random, and then generate initial proportion scheme.The transformation of note groove altogether is k
Individual:
1., when little apoplexy due to endogenous wind material number is 1, illustrate that its primary element and minor element are big with unclassified stores difference, it is impossible to
Cluster, then put it in single cavity, if but the discharge quantity of this material is less than Mmin, then dispensing can not individually be carried out, it is necessary to find
Material suitable with it carries out common groove process, or when this material purchase volume is less, the characteristic utilizing itself proportioning little adds
It is added to the very end of the train of the material that proportioning is bigger in certain single cavity so that it is the power of influence of bulk chemical element index is minimized；
2., when the material number of big class is between k+1～2k, being divided into two groups after clustering according to minor element has a lot
Kind of combinatory possibility, such as: if k is 3, possible combination have 23,22,13,33, for these may combination, due to
Every class material number is few, it may be considered that be directly total to groove；Other situation, such as 14,4 number upper limits k=having exceeded common groove
3, then it is required for those 4 materials and the most once clusters and assign to 2 altogether in grooves；
3. when the material number of big class is more than 2k kind, then such front k kind material is moved to other common grooves, then sentence
Disconnected big class leftover materials number, if so continuing executing with above step still greater than 2k kind；If less than 2k kind, then jump to
State 1 and 2 steps to process.
Through the initial optimization of dosage bunker material number, make that each dosage bunker can be total to groove material number and meet actual production
Requirement, further, it is contemplated that can be used for the groove number constraint of dispensing in actual production, is respectively processed:
If 1. after above-mentioned initial optimization, the usage quantity of dosage bunker is the most identical, then with enterprise practical dosage bunker quantity
Do not carry out current procedures process；
If 2. after above-mentioned initial optimization, the usage quantity of dosage bunker less than enterprise practical dosage bunker quantity, is then carried out altogether
The fractionation of groove；Treating method for this situation is: chooses k groove altogether, is moved by material maximum for proportioning in k kind material in groove
Go out, put to single cavity, make original k groove altogether be changed into a k-1 groove and a single cavity altogether, make dosage bunker usage quantity increase by 1
Individual, the like, revising with increasing the minimum cost that ensure that original ingredient scheme while dosage bunker；
If 3. after above-mentioned initial optimization, the usage quantity of dosage bunker is more than enterprise practical dosage bunker quantity, then to dispensing
Be combined between groove, mainly have following two exchange combined method:
Single cavity in the proportion scheme preliminarily formed is analyzed, does not consider to carry out the material of single cavity process, and unite
Meter can redistribute the number of material, is ranked up by proportioning by the material that can redistribute, and selects the thing that two kinds of proportionings are minimum
Material groove altogether, makes dosage bunker usage quantity reduce 1；
As described above, statistics can redistribute the number of material, can enter if only 1 can be redistributed in material
Go and be total to groove, then by itself and other 2 groove combinations altogether, work 3 is total to groove and processes, and makes dosage bunker usage quantity reduce 1；In like manner can also be by
Material in single cavity is put into 3 and is total in groove, constitutes 4 and is total to groove, only need to meet less than being total to groove upper limit k.
Step 5, uses combined optimization method that the initial ingredients scheme obtained in step 4 carries out common groove raw material blanking
The optimization of sequence:
The ordering randomly generated according to step 4, adds up the index situation of current allocation result, it may be judged whether meet
The production requirement of current enterprise, and the fluctuation situation of index is analyzed, in ensureing original groove altogether, material information is constant
On the premise of, current all groove materials altogether are carried out fully intermeshing analysis, after arranging by its possible blanking sequencing every time
Material combinations carry out index verification, and this index is made assessment, the optimum combination mode of last index for selection assessment, make to work as
Front index reaches global optimum, obtains the optimal solution of material blanking order in dosage bunker so that the whole moon to factory, various iron mine was former
Material is finished and is ensured that the chemical composition of any class of the whole month mixing material is uniform and stable, i.e. per tour mixing material refers in enterprise and country
In the fluctuation range that mark requires, and the chemical composition variance fluctuation mixing material between class and class is the least, and then obtains final
Proportion scheme.
This patent, by combining clustering algorithm and combined optimization method, generates mixing dosage scheme quick, intelligently.At meter
Calculating time-consuming aspect, this patent whole process i.e. can get final proportion scheme in 1～2 second, and most of factory of China manually calculates time-consuming 2
～4 hours, within even 5 hours, all cannot try to achieve satisfactory solution；In terms of precision, the present invention can make per tour index all raw in enterprise
Produce in the fluctuation range required, ensure that between class and class, the variance fluctuation of mixing material chemical composition is the least simultaneously, and artificial
Calculate and be difficult to meet all indexs simultaneously, the situation that chemical component fluctuation is bigger even occurs.This patent utilizes clustering algorithm big
Width reduces under solution room, and constraint premise multiple in considering actual production, and the thought in conjunction with Combinatorial Optimization makes to gather
Result after class reaches global optimum, and then effectively instructs actual production process.It is the most most of that the present invention solves China
Factory's pre-dispensing link still uses the artificial present situation calculating proportioning, and automatization large-scale to commercial production produces to promote and make
With, it is effectively improved speed and the precision of pre-dispensing, lays good basis for enterprise's subsequent production link.
(3) beneficial effect
The technique scheme of the present invention has the advantage that the pre-dispensing side of the iron mine blending process that the present invention provides
Method, saves and calculates time-consumingly, it is ensured that per tour mixing material chemical composition error is the least, and mixing material chemical composition between class and class
Variance fluctuation range little, precision is higher.
Accompanying drawing explanation
Fig. 1 is the step schematic diagram of the pre-distribution of embodiment of the present invention iron mine blending process；
Fig. 2 is pre-distribution intelligent clustering and the combinatorial optimization algorithm flow process of embodiment of the present invention iron mine blending process
Figure；
Fig. 3 is the pre-distribution Canopy algorithm principle figure of embodiment of the present invention iron mine blending process；
Fig. 4 is the pre-distribution Canopy-Kmeans clustering algorithm flow chart of embodiment of the present invention iron mine blending process.
Detailed description of the invention
For making the purpose of the embodiment of the present invention, technical scheme and advantage clearer, below in conjunction with the embodiment of the present invention
In accompanying drawing, the technical scheme in the embodiment of the present invention is clearly and completely described, it is clear that described embodiment is
A part of embodiment of the present invention rather than whole embodiments.Based on the embodiment in the present invention, ordinary skill people
The every other embodiment that member is obtained on the premise of not making creative work, broadly falls into the scope of protection of the invention.
Embodiment 1
As shown in Figure 1 to 4, the pre-distribution of a kind of iron mine blending process proposed in the present invention,
Step one: according to the various iron ore raw material compositional datas in this month to factory, carries out pretreatment operation, gets rid of silicon, ferrum unit
Element component difference is relatively greatly or its physicochemical characteristics cannot be total to the special material of groove, obtains raw material collection to be clustered；
Step 2: the raw material collection to be clustered obtaining step one uses Canopy-Kmeans clustering algorithm once to gather
Generic operation.The division of big class is carried out so that it is be automatically divided into rational N number of big class, reach primary element according to primary constituent content
Balance；
Step 3: use K-Means algorithm to carry out secondary cluster the once cluster raw material set obtained by step 2
Operation.I.e. again cluster according to minor element content in material composition contained by each big apoplexy due to endogenous wind and realize group division, full
On the premise of the primary element of foot is in a basic balance, the composition reaching each little apoplexy due to endogenous wind minor element approximates substantially；
Step 4: the thought merging each little class a operation obtained in step 3 and decomposing processes.Consider reality
In production, dosage bunker number, mixing expect each element index, number and the total siccative amount etc. of arranging an order according to class and grade constraints, to gained cluster result and spy
Different raw material merges and is automatically brought into operation with decomposition etc., determines final clustering schemes, it is ensured that all kinds of middle raw material types, quantity meet real
Border production requirement, i.e. guarantees that finally clustering number is less than, equal to available dosage bunker number, every class raw material type, the raw material that can be total to groove
The sum upper limit, meets other simultaneously and produces constraint requirements.Thus, every class raw material participates in dispensing in a dosage bunker, to common groove
Plurality of raw materials determine blanking precedence in groove at random, and then generate initial proportion scheme；
Step 5: use combined optimization method that the initial ingredients scheme obtained in step 4 carries out common groove raw material blanking
The optimization of sequence, determines final proportion scheme.The shadow of the factor such as consider that raw material type is various, composition fluctuation is many compared with big and constraints
Ring, the method for exhaustion in each raw material transport Combinatorial Optimization in common groove determined to the priority ordering of each class groove dispensing altogether,
The mixing material composition making any order of classes or grades at school of the whole moon fluctuates in the range of the minimal ripple of plan component target, obtains final joining
Material scheme.
Known parameters is set and requires index fluctuation range；
Anticipated total amount 92130 tons, average per tour blanking 4500 tons, set the groove number that can be used for dispensing as 11, Yi Jiyuan
Dish blanking velocity upper and lower bound；
Every content composition fluctuation range: primary element within ± 0.2%, minor element within ± 0.62% and
Other component requirements, between class and class, chemical element fluctuation is the least.
Certain month various iron ore raw material data 1 to factory of table 1
After eliminating cannot be total to the special material of groove with unclassified stores, obtain raw material set to be clustered, then it is carried out
Canopy-Kmeans cluster analysis, becomes material rough segmentation in current planning list big class, more each big class utilizes K-Means gather
After class algorithm is finely divided, draw the result through secondary cluster, as shown in table 2.
Table 1.1 material cluster result
Result after clustering secondary merges and resolution process, the i.e. the 3rd class and a kind of material of the 4th class, wherein
No. 11 material proportion amounts are more than MminSingle cavity process can be carried out, but No. 20 materials then can not process by single cavity, can only select and it
Altogether the material of groove, is therefore analyzed original proportioning list, it is ensured that the most close at composition and do not violate physical constraint
Under principle, select groove material, i.e. material 4 the most altogether；
The usage quantity of dosage bunker after initial optimization is carried out statistical analysis, and adding up actually used dosage bunker number is 9,
Less than setting value 11, it is therefore desirable to material in 2 common grooves or 3 groove altogether is carried out resolution process, select No. 1 groove and No. 3 common grooves altogether
Reject wherein proportioning the maximum and put in single cavity, generate initial proportion scheme.
Use Combinatorial Optimization thought Optimization Cutting order, meeting mixing material chemical component fluctuation minimum as far as possible and meeting
On the premise of every constraints, draw the proportioning blanking table of last optimization, as shown in table 3 (wherein dotted line be groove with groove point
Boundary line).
Certain moon of table 1.2 is to the final proportion scheme of factory's iron mine
Embodiment 2
Known parameters is set and requires index fluctuation range；
Anticipated total amount 92130 tons, average per tour blanking 4500 tons, set the groove number that can be used for dispensing as 11, Yi Jiyuan
Dish blanking velocity upper and lower bound；
Every content composition fluctuation range: primary element within ± 0.2%, minor element within ± 0.62% and
Other component requirements, between class and class, chemical element fluctuation is the least.
Certain moon of table 2 is to factory's iron ore raw material inventory 2
Certain moon of table 2.1 is to the final proportion scheme of factory's iron mine
Embodiment 3
Known parameters is set and requires index fluctuation range；
Anticipated total amount 98000 tons, average per tour blanking 5300 tons, set the groove number that can be used for dispensing as 11, Yi Jiyuan
Dish blanking velocity upper and lower bound；
Every content composition fluctuation range: primary element within ± 0.2%, minor element within ± 0.62% and
Other component requirements, between class and class, chemical element fluctuation is the least.
Certain moon of table 3 is to factory's iron ore raw material inventory 3
Certain moon of table 3.1 is to the final proportion scheme of factory's iron mine
Embodiment 4
Known parameters is set and requires index fluctuation range；
Anticipated total amount 92130 tons, average per tour blanking 4500 tons, set the groove number that can be used for dispensing as 11, Yi Jiyuan
Dish blanking velocity upper and lower bound；
Every content composition fluctuation range: primary element within ± 0.2%, minor element within ± 0.62% and
Other component requirements, between class and class, chemical element fluctuation is the least.
Certain moon of table 4 is to factory's iron ore raw material inventory 4
Certain moon of table 4.1 is to the final proportion scheme of factory's iron mine
Embodiment 5
Known parameters is set and requires index fluctuation range；
Anticipated total amount 98000 tons, average per tour blanking 4635 tons, set the groove number that can be used for dispensing as 11, Yi Jiyuan
Dish blanking velocity upper and lower bound；
Every content composition fluctuation range: primary element within ± 0.2%, minor element within ± 0.62% and
Other component requirements, between class and class, chemical element fluctuation is the least.
Certain moon of table 5 is to factory's iron ore raw material inventory 5
Certain moon of table 5.1 is to the final proportion scheme of factory's iron mine
Embodiment 6
Known parameters is set and requires index fluctuation range；
Anticipated total amount 98000 tons, average per tour blanking 5600 tons, set the groove number that can be used for dispensing as 11, Yi Jiyuan
Dish blanking velocity upper and lower bound；
Every content composition fluctuation range: primary element within ± 0.2%, minor element within ± 0.62% and
Other component requirements, between class and class, chemical element fluctuation is the least.
Certain moon of table 6 is to factory's iron ore raw material inventory 6
Certain moon of table 6.1 is to the final proportion scheme of factory's iron mine
Embodiment 7
Known parameters is set and requires index fluctuation range；
Anticipated total amount 92130 tons, average per tour blanking 4500 tons, set the groove number that can be used for dispensing as 11, Yi Jiyuan
Dish blanking velocity upper and lower bound；
Every content composition fluctuation range: primary element within ± 0.2%, minor element within ± 0.62% and
Other component requirements, between class and class, chemical element fluctuation is the least.
Certain moon of table 7 is to factory's iron ore raw material inventory 7
Certain moon of table 7.1 is to the final proportion scheme of factory's iron mine
Through checking, the index of the primary element of per tour and minor element all within enterprise practical claimed range, other yuan
There is the phenomenon that exceeds standard individually in element, but has complied fully with the demand of actual blanking.This result, compared with artificial result of calculation, not only has
There is rapidity, also there is accuracy and operator's universality.
In sum, the present invention proposes the pre-dispensing side of a kind of iron mine blending process based on clustering algorithm and Combinatorial Optimization
Method, its step includes: material composition data prediction operates；Complete once to cluster according to silicon content difference in material composition
Operation, i.e. utilizes Canopy-Kmeans clustering method to be divided into the big class of N；On this basis, according to iron content difference profit
Secondary cluster is carried out with K-means algorithm；Consider the constraintss such as chemical element index, utilize the method merged and decompose to dividing
Class result processes；Subsequently, use Combinatorial Optimization thought to determine Optimum panel cutting order, obtain final proportion scheme, this
The pre-distribution of the iron mine blending process of bright offer, saves and calculates time-consumingly, it is ensured that per tour mixing material chemical composition error to the greatest extent may be used
Can be little, and between class and class, the variance fluctuation range of mixing material chemical composition is little, precision is higher.
Last it is noted that above example is only in order to illustrate technical scheme, it is not intended to limit；Although
With reference to previous embodiment, the present invention is described in detail, it will be understood by those within the art that: it still may be used
So that the technical scheme described in foregoing embodiments to be modified, or wherein portion of techniques feature is carried out equivalent；
And these amendment or replace, do not make appropriate technical solution essence depart from various embodiments of the present invention technical scheme spirit and
Scope.
Claims (9)
1. the pre-distribution of iron mine blending process, it is characterised in that comprise the steps of
Step one: iron ore raw material carries out pretreatment operation, and eliminating elemental composition differs greatly or its physicochemical characteristics cannot
The special material of groove, obtains raw material collection to be clustered altogether；
Step 2: described raw material set to be clustered employing Canopy-Kmeans clustering algorithm is carried out a cluster operation, obtains
Once cluster raw material set, described once cluster raw material collection is combined into the multiple big class of most important chemical element index balance；
Step 3: use K-Means algorithm to carry out secondary cluster operation the raw material set of described once cluster, obtain secondary cluster
Raw material set, described secondary cluster raw material collection is combined in iron ore raw material composition minor element content close to multiple groups；
Step 4: by described secondary cluster raw material set and described special material for dosage bunker number, mixing expect each element index,
The conditions such as number and total siccative amount of arranging an order according to class and grade merge or operation splitting, obtain initial ingredients scheme；
Step 5: use combined optimization method that the initial ingredients scheme obtained in step 4 carries out common groove raw material blanking order
Optimize, determine final proportion scheme.
The pre-distribution of iron mine blending process the most according to claim 1, it is characterised in that: the composition in described step one
The element differed greatly includes silicon, ferrum, calcium constituent.
The pre-distribution of iron mine blending process the most according to claim 1, it is characterised in that: described Canopy-Kmeans
Clustering algorithm carries out the division of big class according to primary elemental silicon content.
The pre-distribution of iron mine blending process the most according to claim 3, it is characterised in that: described Canopy-Kmeans
Clustering algorithm is for selecting two threshold values T1And T2, wherein T1>T2.Solid circles is T1, dotted line circle is T2；1 P of arbitrary extracting again, meter
Calculate the distance of P point and Canopy, if P point is in distance T of certain Canopy1Within, then P point is added this Canopy；As
Really this P point is once in distance T of certain Canopy2Within, then P is deleted from data set, repeat above-mentioned two steps until data set
For sky, final number N adding up Canopy class.
The pre-distribution of iron mine blending process the most according to claim 1, it is characterised in that: each big apoplexy due to endogenous wind is to secondary unit
Plain sheet content carries out K-Means cluster, and the material in big class is divided into two groups.
The pre-distribution of iron mine blending process the most according to claim 1, it is characterised in that: described merging or the tool of decomposition
Body method is as follows, and the transformation of note groove altogether is k；
(1) when little apoplexy due to endogenous wind material number is 1, illustrate that its primary element and minor element are big with unclassified stores difference, it is impossible to poly-
Class, then put it in single cavity, if but the discharge quantity of this material is less than Mmin, then can not individually carry out dispensing, it is necessary to find with
Its suitable material carries out common groove process, or when this material purchase volume is less, utilizes the characteristic that itself proportioning is little to add
The very end of the train of the material that proportioning is bigger in certain single cavity so that it is the power of influence of bulk chemical element index is minimized；
(2) when the material number of big class is between k+1～2k, after clustering according to minor element, it is divided into two groups, if two groups little
Material number in class is tasted k value, can directly be total to groove；If the material number in one of which group exceedes k value, then for
This group group the most once clusters to be assigned in two common grooves；
(3) when the material number of big class is more than 2k kind, then such front k kind material is moved to other common grooves, then judge
Big class leftover materials number, if so continuing executing with above step still greater than 2k kind；If less than 2k kind, then jump to above-mentioned
1 and 2 steps process.
The pre-distribution of iron mine blending process the most according to claim 6, it is characterised in that: described merging or the side of decomposition
Method also comprises the steps；
(1) if the most identical with actual dosage bunker quantity through the usage quantity of above-mentioned initial ingredients scheme dosage bunker, do not enter
Step process before the trade；
(2) if through the usage quantity of above-mentioned initial ingredients scheme dosage bunker less than enterprise practical dosage bunker quantity, then carried out altogether
The fractionation of groove；Treating method for this situation is: chooses k groove altogether, is moved by material maximum for proportioning in k kind material in groove
Go out, put to single cavity, make original k groove altogether be changed into a k-1 groove and a single cavity altogether, make dosage bunker usage quantity increase by 1
Individual, the like, revising with increasing the minimum cost that ensure that original ingredient scheme while dosage bunker；
(3) if through the usage quantity of above-mentioned initial ingredients scheme dosage bunker more than enterprise practical dosage bunker quantity, then to dispensing
It is combined by exchange combined method between groove, obtains final proportion scheme.
The pre-distribution of iron mine blending process the most according to claim 7, it is characterised in that: described exchange combined method has
Two kinds；
Method 1: be analyzed single cavity in final proportion scheme, does not consider to carry out the material of single cavity process, and statistics can
Redistributing the number of material, be ranked up by proportioning by the material that can redistribute, the material selecting two kinds of proportionings minimum is total to
Groove, makes dosage bunker usage quantity reduce 1；
Method 2: add up final proportion scheme and can redistribute the number of material, if only 1 thing in material can be redistributed
Material can carry out common groove, then by itself and the groove combination altogether of other 2 materials, make 3 materials groove altogether and process, make dosage bunker usage quantity reduce
1；Material in single cavity in like manner can also be put into 3 be total in groove, constitute 4 and be total to groove, only need to meet less than being total to groove upper limit k i.e.
Can.
The pre-distribution of iron mine blending process the most according to claim 1, it is characterised in that described combined optimization method
For: add up the index situation of current allocation result, on the premise of material information is constant in ensureing original groove altogether, to current all
Groove material carries out fully intermeshing analysis by its possible blanking sequencing altogether, the material combinations after arrangement every time is carried out index and tests
Card, and this index is made assessment, the optimum combination mode of last index for selection assessment, make current criteria reach global optimum,
Obtain the optimal solution of material blanking order in dosage bunker, and then obtain final proportion scheme.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610105488.3A CN105787509B (en) | 2016-02-25 | 2016-02-25 | A kind of pre- distribution of iron ore blending process |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610105488.3A CN105787509B (en) | 2016-02-25 | 2016-02-25 | A kind of pre- distribution of iron ore blending process |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105787509A true CN105787509A (en) | 2016-07-20 |
| CN105787509B CN105787509B (en) | 2018-12-14 |
Family
ID=56403033
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610105488.3A Active CN105787509B (en) | 2016-02-25 | 2016-02-25 | A kind of pre- distribution of iron ore blending process |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105787509B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106484838A (en) * | 2016-09-30 | 2017-03-08 | 中国南方电网有限责任公司 | Safety inspection java standard library dynamic updating method based on data mining |
| CN108596375A (en) * | 2018-04-11 | 2018-09-28 | 武汉理工大学 | A kind of equipment manufacture material requirement Design Method and system |
| CN110681855A (en) * | 2018-07-06 | 2020-01-14 | 米亚索乐装备集成（福建）有限公司 | Preparation method of alloy powder |
| CN114913932A (en) * | 2022-07-15 | 2022-08-16 | 北矿智云科技(北京)有限公司 | Batching method, batching device, electronic equipment and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102193502A (en) * | 2011-06-14 | 2011-09-21 | 中南大学 | Comprehensive optimization method for steel sintering secondary burdening |
| CN104561411A (en) * | 2015-01-23 | 2015-04-29 | 武钢集团昆明钢铁股份有限公司 | Blending method capable of effectively improving quality of uniformly-mixed ores |
| CN104649036A (en) * | 2014-12-22 | 2015-05-27 | 天津钢铁集团有限公司 | Piling method for improving stability of uniformly blended material |
| CN105095266A (en) * | 2014-05-08 | 2015-11-25 | 中国科学院声学研究所 | Method and system for clustering optimization based on Canopy algorithm |
-
2016
- 2016-02-25 CN CN201610105488.3A patent/CN105787509B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102193502A (en) * | 2011-06-14 | 2011-09-21 | 中南大学 | Comprehensive optimization method for steel sintering secondary burdening |
| CN105095266A (en) * | 2014-05-08 | 2015-11-25 | 中国科学院声学研究所 | Method and system for clustering optimization based on Canopy algorithm |
| CN104649036A (en) * | 2014-12-22 | 2015-05-27 | 天津钢铁集团有限公司 | Piling method for improving stability of uniformly blended material |
| CN104561411A (en) * | 2015-01-23 | 2015-04-29 | 武钢集团昆明钢铁股份有限公司 | Blending method capable of effectively improving quality of uniformly-mixed ores |
Non-Patent Citations (1)
| Title |
|---|
| 曹跃 等: "Canopy-Kmeans聚类和组合优化的铁矿预配料智能调度", 《控制理论与应用》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106484838A (en) * | 2016-09-30 | 2017-03-08 | 中国南方电网有限责任公司 | Safety inspection java standard library dynamic updating method based on data mining |
| CN108596375A (en) * | 2018-04-11 | 2018-09-28 | 武汉理工大学 | A kind of equipment manufacture material requirement Design Method and system |
| CN108596375B (en) * | 2018-04-11 | 2021-07-27 | 武汉理工大学 | Equipment manufacturing material demand scheme design method and system |
| CN110681855A (en) * | 2018-07-06 | 2020-01-14 | 米亚索乐装备集成（福建）有限公司 | Preparation method of alloy powder |
| CN114913932A (en) * | 2022-07-15 | 2022-08-16 | 北矿智云科技(北京)有限公司 | Batching method, batching device, electronic equipment and storage medium |
| CN114913932B (en) * | 2022-07-15 | 2022-10-14 | 北矿智云科技(北京)有限公司 | Batching method, batching device, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105787509B (en) | 2018-12-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105787509A (en) | Iron mine blending process pre-proportion method | |
| Yang et al. | Evaluation of regional environmental efficiencies in China based on super-efficiency-DEA | |
| Wu et al. | Resource reallocation and target setting for improving environmental performance of DMUs: An application to regional highway transportation systems in China | |
| CN106406099B (en) | Cement batching system and method based on fuzzy matching sum value feedback | |
| CN104680012A (en) | Calculating model for sintering and burdening | |
| CN102622667A (en) | Balance production scheduling method based on multi-product and multi-production-line production mode | |
| Monkiewicz | International technology flows and the technology gap: the experience of Eastern European socialist countries in international perspective | |
| Chang et al. | Integrated planning for design and production in two-stage recycling operations | |
| CN106094751A (en) | The dispatching method of a kind of raw material and device | |
| Moosavi et al. | Optimal extraction sequence modeling for open pit mining operation considering the dynamic cutoff grade | |
| Ackerman | Carbon Embedded in China’s Trade | |
| Badia-Miró et al. | Copper Prices and Heterogeneity: The Chilean and Norwegian Cases, 1850–1913 | |
| CN107918820A (en) | A kind of supply chain order collocation method for considering income | |
| Dong et al. | How carbon emission quotas can be allocated fairly and efficiently among different industrial sectors: the case of Chinese industry. | |
| CN103971179B (en) | TOCh method for solving problem of single-bottle neck product mix optimization based on process control | |
| CN106920004A (en) | A kind of one-dimensional stock-cutting method based on cost dynamic equilibrium | |
| Masoumi et al. | A framework to form balanced project portfolios | |
| Yue et al. | Regional allocation of energy intensity reduction target: The case of Henan province in China | |
| Carlsson et al. | Technological capabilities and international competitiveness in the engineering industries | |
| CN111369129B (en) | Production scheduling method and system for building robot | |
| Jeong et al. | Profitability gains of Korean defense firms: technological progress or cost shifting? | |
| Dulange et al. | Quantification of factors influencing productivity using AHP technique: an approach towards productivity improvement | |
| Rybchuk et al. | Ukraine in the promotion system for exports of high-tech products | |
| Zhu et al. | Order batching optimization in dual zone type warehouse based on genetic Algorithms | |
| Marques Gomes | Oskar Lange’s Economics and the Socialist Economy |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |