CN102609784A - Collocation method of iron ores of steel raw material field - Google Patents

Abstract

The invention provides a collocation method of iron ores of a steel raw material field and belongs to the field of a metallurgical industry. In the optimized collocation method disclosed by the invention, the input collocations of various types of the iron cores and strict component control requirements of the iron cores are comprehensively considered so as to assist collocation planning personnel to find out the best storage way, thus improving the production and storage quality of the iron ores, and realizing the purposes of guaranteeing the supply of the iron ores, reducing the transportation energy consumption, preventing the iron ores from being mixed, guaranteeing the quality of the iron ores, improving the quality of a subsequent product and on the like. By comparing an artificial collocation result of actual data with an optimized result, on the premise that the input executing rate of the iron cores is improved by 2.7%, a target function can be obviously reduced by 12.79% by comparing an optimized collocation scheme with an artificial collocation scheme, so that the consumption of the transportation energy loss is effectively reduced, the storage quality of the iron ores is improved and the quality of subsequent products is further improved.

Description

A kind of iron and steel raw material field iron ore collocation method

Technical field

The invention belongs to metallurgical industry field, relate to iron and steel raw material logistics information technical field, specially refer to a kind of iron and steel raw material field iron ore collocation method.

Background technology

The stock yard of iron and steel enterprise is the place that is used for depositing the required diffusing shape raw materials such as iron ore of production units such as blast furnace, sintering.Raw material (mainly the being iron ore) kind that stock yard is stacked has hundreds of, by its separately rule with the form air storage of stockpile in different material bars.Each material bar all has corresponding reclaimer, stacker to be its service.Stock yard utilizes stacker and reclaimer equipment that the position is extremely reasonably stored up in the raw material configuration, and for production processes such as blast furnace and sintering are carried out feed, the raw material schematic layout pattern in the stock yard is as shown in Figure 1.

The configuration of the raw material of stock yard is that raw material various in style with the required preparation of stock yard is according to its physicochemical characteristic and stock ground stock's present situation; Rule according to the stock ground configuration; Be assigned to rational memory location (storage position), make full use of the stock ground storage space, rationally utilize stable, the reduction of various device and resource, assurance material quality to produce targets such as storage expenses and energy resource consumption with realization.Because the raw material of stock yard storage comprises a plurality of kinds such as fine ore, lump ore, pelletizing, sintering deposit, second-time breakage material and mixing material; The physicochemical property significant difference; Therefore the configuration meeting of rationally carrying out raw material significantly reduces stock ground operating cost; Guarantee the stay in grade of feed stock for blast furnace, it is stable, efficient to ensure that ironmaking is produced.

The raw material input layoutprocedure of present domestic iron and steel raw material field is all carried out job-shop by manual work according to experience; Though the operating personnel has accumulated considerable experience; Also obtained very big effect; The feed time undulatory property is big, inlet amount is big, the turnaround time is fast, the stock ground essential information dynamically updates soon but because the iron and steel enterprise's raw material charging and the mode of production have; And operation steps is complicated, strict, the characteristics such as constraint condition is many, randomness height of material composition protection, makes that raw material input configuration is the very loaded down with trivial details time-consuming and work that need constantly repeat, and manual work is simultaneously worked out when planning can't take into account many-sided requirements such as production capacity, raw materials quality, logistics balance and need satisfaction simultaneously; Usually can't take all factors into consideration the charging situation in present and following a period of time; In real time stock ground layout plan is adjusted, usually taked interim expedient processing, therefore depend merely on manual method and be difficult to rational arrangement is made in the raw material configuration of marching into the arena in batches.

Summary of the invention

For overcoming the deficiency that existing method exists, the present invention proposes a kind of iron ore collocation method that is applied to iron and steel enterprise's stock yard, reduces the purpose of producing storage expenses, reduction energy resource consumption and ensureing stove iron ore stay in grade to reach.

Technical scheme of the present invention is achieved in that a kind of iron ore collocation method that is applied to iron and steel enterprise's stock yard of the present invention, may further comprise the steps:

Step 1: obtain and wait to import iron ore data, stock yard storage bit data and stock yard service data, confirm the technology characteristics of iron ore input configuration, and adopt mathematical model to describe its technological process;

Wherein, describedly wait to import the iron ore data and comprise: iron ore lot number, iron ore kind, iron ore weight, chemical constitution, iron ore time of arrival, heap are high; Described stock yard storage bit data comprises: a storage position position coordinates, storage bit length, available storage position last time canned data, with a storage position canned data, input equipment work capacity; Described stock yard service data comprises: stock yard storage inventory, iron ore plan of needs, Plant maintenance plan;

Described mathematical model, it is following to set up process:

Step 1-1: confirm that model parameter is: in time prospect phase T, iron ore set to be disposed is N, and the set of material bar is S in the stock yard, and total K among the material bar j _jIndividual available storage position constitutes available storage position set K; When iron ore i is assigned to k available storage of material bar j position, need to consume levelling of the land, material empties and preparation work expense such as equipment prerun, also can produce simultaneously by iron ore is mixed and put the iron ore failure costs that causes, its cost coefficient is respectively c _IjkAnd p _IjkCan produce corresponding operating cost when opening material bar j, its cost coefficient is f _jWhen store the storage position when be assigned to different material bars with the lot number iron ore on, can cause the operations such as path conversion of logistics equipment, the equipment loss that produced this moment and the coefficient of energy consumption cost are qi;

Step 1-2: utilize the parameter of step 1-1, set up mathematical model, its objective function is following:

$\mathrm{Minimize}\underset{\text{i}}{\mathrm{\Σ}}\underset{\text{j}}{\mathrm{\Σ}}\underset{k}{\mathrm{\Σ}}{c}_{\mathrm{ijk}}{x}_{\mathrm{ijk}}+\underset{i}{\mathrm{\Σ}}\underset{j}{\mathrm{\Σ}}\underset{k}{\mathrm{\Σ}}{p}_{\mathrm{ijk}}{y}_{\mathrm{ijk}}+\underset{i}{\mathrm{\Σ}}{f}_j{z}_j+\underset{i}{\mathrm{\Σ}}\mathrm{qi}(\underset{j}{\mathrm{\Σ}}{\∂}_{\mathrm{ij}}-1)---\left(1\right)$

In the formula, x _Ijk, z _j,

And y _IjkBe decision variable, and its value is:

y _Ijk>=0, iron ore i is assigned to the weight of k storage position storage of material bar j;

When being assigned to k storage of material bar j, first of objective function (1) expression iron ore i empty and the energy that preparation work consumed such as equipment prerun and equipment loss expense etc. by levelling of the land, material; Second expression iron ore i is assigned to when expecting k storage position of bar j by the mixed iron ore failure costs that causes of putting of iron ore; The 3rd is the expense of offering material bar j, comprises that the material bar administers and maintains stacker-reclaimer opex in expense and the material bar; The 4th equipment loss and the energy consumption cost that the logistics equipment path conversion that causes when being assigned to different material bar j for a kind of iron ore i is produced;

Step 1-3: confirm the described bound for objective function of step 1-2:

(1) to arbitrary available storage position, all deposit the length (promptly piling up ability) that the shared accumulation length summation of iron ore can not exceed said storage position, and the formula of storage potential energy power constraint is following:

$\begin{array}{cc}\underset{i}{\mathrm{\Σ}}{\mathrm{\ρ}}_i{y}_{\mathrm{ijk}}+D(\underset{i}{\mathrm{\Σ}}{x}_{\mathrm{ijk}}-1)\≤l_{\mathrm{jk}}& \∀j\∈S,k\∈K\end{array}---\left(2\right)$

In the formula, ρ i is the required accumulation length that takies of iron ore i of the unit weight that draws according to known accumulation formula, l _JkBe the length of k storage position of material bar j, D is a required safe distance of isolating between two stockpiles;

(2) when arbitrary material bar is deposited iron ore, the weight of every heap iron ore can not surpass the maximum weight restriction of the single stockpile of this material bar, and the formula of stockpile weight limits constraint is following:

$\begin{array}{cc}{y}_{\mathrm{ijk}}\≤W_j& \∀i\∈N,j\∈S,k\∈K\end{array}----\left(3\right)$

In the formula, W _jUpper weight limit for the arbitrary stockpile of j material bar;

(3) to arbitrary material bar j, iron ore weight sum to be imported can not surpass the work capacity of this material bar support equipment, and the formula of implement ability constraint is following:

$\begin{array}{cc}\underset{i}{\mathrm{\Σ}}\underset{k}{\mathrm{\Σ}}{y}_{\mathrm{ijk}}\≤A_j& \∀j\∈S\end{array}---\left(4\right)$

In the formula, A _jThe work capacity upper limit for material bar j support equipment;

(4) guarantee that institute remains to be imported iron ore and all has been assigned to and stores in the stock ground, the formula of iron ore receiving-transmitting balance constraint is following:

$\begin{array}{cc}\underset{j}{\mathrm{\Σ}}\underset{k}{\mathrm{\Σ}}{y}_{\mathrm{ijk}}=Y_i& \∀i\∈N\end{array}---\left(5\right)$

(5) except manufacturability constraint (2)-(5), also need satisfy the variable logicality constraint of model itself, mainly comprise:

$\begin{array}{cc}{y}_{\mathrm{ijk}}\≤Y_i{x}_{\mathrm{ijk}}& \∀i\∈N,j\∈S,k\∈K\end{array}----\left(6\right)$

$\begin{array}{cc}\underset{i}{\mathrm{\Σ}}\underset{k}{\mathrm{\Σ}}{x}_{\mathrm{ijk}}\≤K_j{z}_j& \∀j\∈S\end{array}---\left(7\right)$

$\begin{array}{cc}\underset{k}{\mathrm{\Σ}}{x}_{\mathrm{ijk}}\≤K_j{\∂}_{\mathrm{ij}}& \∀i\∈N,j\∈S\end{array}---\left(8\right)$

Wherein, constraint (6) expression is assigned to the weight y of k storage position of material bar j as iron ore i _IjkGreater than 0 o'clock, x _IjkMust be 1, thereby guarantee that iron ore i is assigned to k assigning process that stores up the position of material bar j and has taken place; Constraint (7) has guaranteed that material bar j must open, wherein K when iron ore i is assigned to k storage position of material bar j _jAvailable storage bit quantity for material bar j; Constraint (8) has guaranteed that when iron ore i is assigned to k storage position of material bar j be x _IjkBe 1 o'clock, the inevitable material bar j that has been assigned to of iron ore i, promptly Must be 1;

Step 2: to the input iron ore store up the position pre-configured, method is:

Step 2.1: with all available storages positions form available storage position gather U={ (j, k) | j ∈ S, k ∈ K} arranges the iron ore that remains to be imported according to required storage bit length from big to small, forms alternative iron ore sequence N;

Step 2.2: (j k), picks out from N and satisfies ρ from U, to choose an available storage position wantonly _iY _i≤l _JkIron ore subsequence N _Jk, preferentially select first iron ore i in the sequence ^*, if do not satisfy

Then from N _JkMiddle deletion iron ore i ^*And selection N _JkMiddle next bit iron ore is right

Judge; If satisfy

Then with iron ore i ^*Put into this storage position and delete iron ore i from N ^*If, should storage position (j k) has not utilized, with residue storage position as new available storage position (j k) is updated among the U, otherwise from U, delete and store up (j, k); If (j k), can't pick out and satisfy ρ for available storage position _iY _i≤l _JkAnd

The iron ore subsequence, should store up the position (j k) joins in the set B of storage position, deletion storage position from U (j, k);

Step 2.3:, forward step 2.6 to if N=is Φ; Otherwise, forward step 2.4 to;

Step 2.4: if U=Φ forwards step 2.5 to; Otherwise, forward step 2.2 to;

Step 2.5: by the length scale descending sort, filling up these available storage positions successively with order is principle, and iron ore among the N is divided into a plurality of heaps, is assigned among the B and stores up in the position, all has been assigned with up to iron ore with available storage position among the B;

Step 2.6: above-mentioned iron ore storage position allocative decision is stored as an initial scheme individuality, if it is individual to have generated P initial scheme, pre-configured end; Otherwise all data of initialization, and forward step 2.1 to;

Step 3: the pre-configured result in storage position to iron ore is optimized adjustment: the P that is generated in the step 2 initial individual scheme formed an initial scheme population, utilizes difference algorithm that the pre-configured result in storage position is optimized adjustment, and method is:

Step 3.1: each individuality in the previous generation scheme population is all carried out mutation operation; Promptly from previous generation scheme population, generate the individual entering of new scheme population of future generation; Its variation strategy does; In the G time iterative process,, be designated as

i=1 for the individual i of scheme; 2; ..., P, then a new individuality

produces according to following formula:

$V_i^{G+1}=X_a^G+F*(X_b^G-X_c^G)$

In the above-mentioned formula, With Be from current population PG, to select 3 to be different from random

Individuality, i.e. a ≠ b ≠ c; F is the self-adaptation mutagenic factor, is used for controlling schemes population diversity and convergence, and: F=β (e ^λ-1), in the formula, β ∈ [0.2,0.6],

Gmax is a greatest iteration algebraically, and G is the algebraically of current iteration;

Step 3.2: the scheme population is carried out interlace operation; Promptly make new and old scheme individuality intercourse partial code according to Crossover Strategy according to newly-generated scheme individual

; Thereby it is individual to form new scheme, and formula is following:

New scheme individuality after the intersection is:

$U_i^{G+1}=(u_{1i}^{G+1},u_{2i}^{G+1},...,u_{\mathrm{Di}}^{G+1})$

In the formula;

is j composition variable in the vector

, and the equation of interlace operation is:

In the formula,

Be vector In j composition variable,

Be vector

In j composition variable, rand (j) is an even distribution probability between [0,1], produces j estimated value of randomizer between [0,1]; The random integers that generate between mbr (i) expression [1, D] are guaranteed with it At least from Obtain a parameter; CR is a crossover probability, CR=a*CR _s, CR wherein _sBe predefined crossover probability, CR _s∈ (0,1) is constant, and a is the evolutionary rate factor, and its expression formula is following:

$a=\frac{f{(G-2)}_{\mathrm{best}}-f(G-1)\mathrm{best}+\mathrm{const}}{f{(G-2)}_{\mathrm{best}}-f\left(G\right)\mathrm{best}+\mathrm{const}}$

In the formula, f (G) best, f (G-1) best and f (G-2) _BestBe respectively G generation, G-1 generation and G-2 and calculate the optimum data desired value of gained after for individual substitution (1) formula; Const is little constant, and making molecule, denominator perseverance is not 0, and guarantees 0＜a≤1;

Step 3.3: the scheme population is selected; Produce follow-on scheme population; Selection strategy is:

compared with

; If the data target value of calculating gained is superior to the latter after the former substitution (1) formula, then just replace

otherwise

reservation in generation with

at G+1;

Step 3.4: individual for G+1 for arbitrary allocation plan in the scheme population, two iron ores in the same material bar are exchanged or two iron ores belonging to different material bars exchange, it is individual to generate new scheme, thus the scheme that obtains is gathered CH; In the time of must guaranteeing the feasible promptly satisfied constraint of exchange (2)-(5); Exchange can be implemented; Scheme individuality and G+1 among the CH are arranged according to the data target value size that (1) formula draws for all individualities in the scheme population from small to large, and select progressively P mutually different individuality formed new G+1 for the scheme population;

Step 3.5: if satisfy stop criterion, algorithm finishes; Otherwise jump to step 3.1;

Step 4: the allocation plan that step 3 is definite is delivered in the production executive system and is carried out.

Advantage of the present invention: the method for distributing rationally of the present invention; Take all factors into consideration the input configuration of multiple iron ore and strict components of iron ore control requirement, can help to dispose staff planners and find out best storage mode, thus the production storage quality of raising iron ore; Realize ensureing ferrolite supply; Reduce the transportation energy consumption, prevent the generation of iron ore batch mixing situation, thereby guarantee that the iron ore quality improves purposes such as its follow-up product quality.The comparison of human configuration result and Optimization result of the present invention through real data; Can draw; Improved under 2.7% the prerequisite in iron ore input implementation rate, compared with the human configuration scheme and can significantly objective function be reduced by 12.79% by the allocation plan that optimization of the present invention obtains.Reduce the transportation energy loss effectively, improved the storage quality of iron ore, and then improved the quality of its subsequent product.

Description of drawings

Fig. 1 is the stock yard schematic layout pattern of iron and steel raw material field iron ore collocation method among the embodiment;

Fig. 2 is the general flow chart of iron and steel raw material of the present invention field iron ore collocation method;

Fig. 3 is the mutation operation synoptic diagram of iron and steel raw material of the present invention field iron ore collocation method;

Fig. 4 is the interlace operation process flow diagram of iron and steel raw material of the present invention field iron ore collocation method;

Fig. 5 (a) moves synoptic diagram for same material bar two exchanges of iron and steel raw material of the present invention field iron ore collocation method;

Fig. 5 (b) moves synoptic diagram for difference material bar two exchanges of iron and steel raw material of the present invention field iron ore collocation method;

Fig. 6 is the artificial method of planning of iron and steel raw material of the present invention field iron ore collocation method and The comparison of evaluation indexes of the present invention figure as a result.

Embodiment

Below in conjunction with accompanying drawing and embodiment the present invention is done further detailed explanation.

The iron ore input collocation method of present embodiment adopts C Plus Plus to realize; And be embedded in the visual iron ore input configuration optimization system that is developed by Microsoft Visual Studio 6.0, this system based on enterprise information platform (ERP system of enterprise, data warehouse) and Microsoft SQL Server2000 data system.

Present embodiment is distributed rationally in stock yard input on the integrated management system of method and is realized that described system comprises: data download module, data management module, stock yard dynamic monitoring module, allocation plan automatically-generating module, allocation plan image display module and results modification and go up transmission module.

In the present embodiment, 23 kinds of iron ores to be disposed are assigned to 11 iron ores stackings expect to store in the bars, as shown in Figure 1, adopt the iron ore collocation method that is applied to iron and steel enterprise's stock yard, its flow process is as shown in Figure 2, may further comprise the steps:

Step 1: the user gets into the succession management system that method is distributed in the stock yard input rationally; At first get into data download module; System can download iron ore data, stock yard storage bit data and stock yard service data from the management system or the database of iron and steel enterprise, comprises iron ore lot number, iron ore kind, iron ore weight, chemical constitution, iron ore clitter height, a storage position position coordinates, storage bit length, available storage position last time canned data, stock yard storage inventory, iron ore plan of needs and Plant maintenance plan;

Information after downloading successfully will be presented in the download interface of system with form and figure dual mode, and according to institute's data downloaded, system can be its allocation models and parameter according to the stock ground configuration rule automatically, and is as shown in table 1:

Table 1 is an iron ore information table to be disposed

Sequence number	The name of an article	The material amount	Plan is from ETB expected time of berthing	The alongside time	Heap is high
						1	OHP-N	11883	0616.1430	0615.2200	130
2	PUA-N	10850	0616.1430	0615.2200	124
						3	PBR-N	16273	0618.0400	0617.1500	124
4	OHY-S	10300	0618.1900	0618.0500	130
						5	ORG-N	9858	0618.1900	0618.0500	124
6	PBR-N	16000	0619.1200	0618.2300	130
						7	OHP-S	34000	0620.0500	0619.1300	130
8	OCS-S	21527	0621.1800	0620.0600	130

9	ORG-N	23120	0621.1800	0620.0600	130
						10	ONE-N	96833	0622.0530	0618.1600	130
11	OYD-S	107985	0614.1200	0612.1200	130
						12	PLF-N	16426	0619.0300	0618.1200	124
13	OHP-N	21000	0620.0800	0619.1300	130
						14	OHY-S	21000	0620.0800	0619.1300	130
15	OHY-S	37827	0619.1100	0616.0300	130
						16	ORG-N	6000	0621.0530	0620.0430	130
17	PUK-N	10000	0621.0530	0620.0430	124
						18	MDM-N	3547	0617.2200	0616.2300	124
19	MGD-S	3210	0618.0400	0617.2300	124
						20	MDM-N	4303	0618.1300	0618.0500	124
21	MLC-C	4764	0618.2200	0618.0400	124
						22	MDM-S	7885	0619.0700	0618.2300	124
23	MLC-N	89700	0619.1600	0617.0800	124

Step 1-1: the parameter that incidence relation between production data and process data and data is mapped as mathematical model: in time prospect phase T, iron ore set to be disposed is N, and the set of material bar is S in the stock yard, and total K among the material bar j _jIndividual available storage position constitutes available storage position set K; When iron ore i is assigned to k available storage of material bar j position, need to consume fixing preparation work cost reach with the relevant iron ore of branch dosage lose cost, its cost coefficient is respectively c _IjkAnd p _IjkCan produce corresponding operating cost when opening material bar j, its cost coefficient is f _jWhen store the storage position when be assigned to different material bars with the lot number iron ore on, can cause the operations such as path conversion of logistics equipment, the coefficient of the correlative charges that produced this moment is qi;

Step 1-2: iron ore is imported the technic index of optimizing in the layoutprocedure be mapped as the mathematical model objective function:

$\underset{\text{i}}{\mathrm{\Σ}}\underset{\text{j}}{\mathrm{\Σ}}\underset{k}{\mathrm{\Σ}}{c}_{\mathrm{ijk}}{x}_{\mathrm{ijk}}+\underset{i}{\mathrm{\Σ}}\underset{j}{\mathrm{\Σ}}\underset{k}{\mathrm{\Σ}}{p}_{\mathrm{ijk}}{y}_{\mathrm{ijk}}+\underset{i}{\mathrm{\Σ}}{f}_j{z}_j+\underset{i}{\mathrm{\Σ}}q(\underset{j}{\mathrm{\Σ}}{\∂}_{\mathrm{ij}}-1)$

Step 1-3: the objective mathematical model constraint condition that is mapped as of process technology limit of iron ore being imported layoutprocedure;

Step 2: formulate the storage position pre-deployment of waiting to import iron ore, be constrained to foundation, do not consider the optimization of iron ore input layoutprocedure each item technic index, form preliminary feasible pre-deployment, may further comprise the steps with basic technology:

Step 2-1: with all available storages positions form available storage position gather U={ (j, k) | j ∈ S, k ∈ K} arranges the iron ore that remains to be imported according to required storage bit length from big to small, forms alternative iron ore sequence N, carries out predistribution according to the following steps:

Step 2-2: (j k), picks out from N and satisfies ρ from U, to choose an available storage position wantonly _iY _i≤l _JkIron ore subsequence N _Jk, preferentially select first iron ore i in the sequence ^*, if do not satisfy

Then from N _JkMiddle deletion iron ore i ^*And selection N _JkMiddle next bit iron ore is right Judge; If satisfy

Then with iron ore i ^*Put into this storage position and delete iron ore i from N ^*If, should storage position (j k) has not utilized, with residue storage position as new available storage position (j k) is updated among the U, otherwise from U, delete and store up (j, k); If (j k), can't pick out and satisfy ρ for available storage position _iY _i≤l _JkThe iron ore subsequence, should store up the position (j k) joins in the set B of storage position, deletion storage position from U (j, k);

Step 2-3:, forward step 3.6 to if N=is Φ; Otherwise, forward step 3.4 to;

Step 2-4: if U=Φ forwards step 3.5 to; Otherwise, forward step 3.2 to;

Step 2-5: by the length scale descending sort, filling up these available storage positions successively with order is principle, and iron ore among the N is divided into a plurality of heaps, is assigned among the B and stores up in the position, all has been assigned with up to iron ore with available storage position among the B;

Step 2-6: above-mentioned iron ore storage position allocative decision is stored as an initial scheme individuality.If generated 50 initial scheme individualities, pre-configured end; Otherwise all data of initialization, and forward step 3.1 to;

Step 3: the mathematical model of being set up with step 1 is a foundation; Adopt difference algorithm that the storage position pre-deployment that step 2 obtains is improved; Carry out the layout of iron ore input allocation plan; And in this process re-optimization input configuration operation process each item technic index, the iron ore that obtains meeting iron ore input actual job target and process constraint is imported allocation plan;

Step 3.1: each individuality in the previous generation scheme population is all carried out mutation operation, promptly from previous generation scheme population, generate the individual entering of new scheme population of future generation.Its variation strategy does; In the G time iterative process; For the individual i of scheme; Be designated as

i=1,2 ...; 50, then a new individuality

produces according to following formula:

$V_i^{G+1}=X_a^G+F*(X_b^G-X_c^G)$

In the above-mentioned formula, With Be from current population PG, to select 3 to be different from random

Individuality, i.e. a ≠ b ≠ c; F is the self-adaptation mutagenic factor, is used for controlling schemes population diversity and convergence, and: F=β (e ^λ-1), in the formula, β ∈ [0.2,0.6],

Gmax is a greatest iteration algebraically, and G is the algebraically of current iteration, and mutation process is as shown in Figure 3;

Step 3.2: the scheme population is carried out interlace operation; Promptly make new and old scheme individuality intercourse partial code according to Crossover Strategy, thereby it is individual to form new scheme according to newly-generated scheme individual .New scheme individuality after note is intersected is:

$U_i^{G+1}=(u_{1i}^{G+1},u_{2i}^{G+1},...,u_{\mathrm{Di}}^{G+1})$

Wherein

is j composition variable in the vector , and the equation of interlace operation is:

In the above-mentioned formula, rand (j) is an even distribution probability between [0,1], produces j estimated value of randomizer between [0,1]; The random integers that generate between mbr (i) expression [1, D] are guaranteed with it

At least from Obtain a parameter; CR is a crossover probability, CR=a*CR _s, CR wherein _sBe predefined crossover probability, CR _s∈ (0,1) is constant, and a is the evolutionary rate factor, and its expression formula is following:

$a=\frac{f{(G-2)}_{\mathrm{best}}-f(G-1)\mathrm{best}+\mathrm{const}}{f{(G-2)}_{\mathrm{best}}-f\left(G\right)\mathrm{best}+\mathrm{const}}$

In the formula, f (G) best, f (G-1) best and f (G-2) _BestBe respectively G generation, G-1 generation and G-2 and calculate the optimum data desired value of gained after for individual substitution (1) formula; Const is little constant, and making molecule, denominator perseverance is not 0, and guarantees 0＜a≤1, and the intersection process is as shown in Figure 4;

Step 3.3: the scheme population is selected, produced follow-on scheme population.Selection strategy is:

compared with

; If the data target value of calculating gained is superior to the latter after the former substitution (1) formula, then just replace

otherwise

reservation in generation with at G+1;

Step 3.4: individual for G+1 for arbitrary allocation plan in the scheme population, two iron ores in the same material bar are exchanged or two iron ores belonging to different material bars exchange, it is individual to generate new scheme, thus the scheme that obtains is gathered CH.In the time of must guaranteeing the feasible promptly satisfied constraint of exchange (2)-(5), exchange just can be implemented.Shown in Figure 4 is that synoptic diagram is moved in two kinds of exchanges under the situation; Scheme individuality and G+1 among the CH are arranged according to the data target value size that (1) formula draws for all individualities in the scheme population from small to large, and 50 mutually different individualities of select progressively are formed new G+1 for the scheme population.

Step 3.5: if satisfy stop criterion, algorithm finishes; Otherwise jump to step 3.1;

Step 4: for the allocation plan that obtains through the algorithm operation in the system; The user can obtain the evaluation of result information of current allocation plan through Function of Evaluation: if dissatisfied to the result; The user can make amendment to the result through form and figure dual mode, till satisfaction; System all can carry out the violation inspection to current result in the each process of revising of user.If satisfied to the result, then carry out uploading of allocation plan, it is uploaded in enterprise's ERP system, formation standard input configuration production plan is issued in the production executive system at last and carries out, and the configuration result that present embodiment finally obtains is as shown in table 2:

The final configuration result table of table 2

Sequence number

The name of an article

The material amount

Plan is from ETB expected time of berthing

Device number

Material bar numbering

The memory location

1

OHP-N

11883

0616.1430

5ST

OE

520-615

2

PUA-N

10850

0616.1430

5ST

OF

235-295

3	PBR-N	16273	0618.0400	13ST	OM	410-530
							4	OHY-S	10300	0618.1900	12ST	OK	435-520
5	ORG-N	9858	0618.1900	12ST	OJ	355-460
							6	PBR-N	16000	0619.1200	12ST	OK	595-640
7	OHP-S	22500	0620.0500	13ST	OL	525-580
							8	OHP-S	11500	0620.0500	13ST	OM	500-630
9	OCS-S	21527	0621.1800	6ST	OH	230-310
							10	ORG-N	23120	0621.1800	12ST	OJ	355-460
11	ONE-N	96833	0622.0530	6ST	OG	110-235
							12	OYD-S	107985	0614.1200	13ST	OL	390-515
13	PLF-N	16426	0619.0300	6ST	OG	515-585
							14	OHP-N	21000	0620.0800	12ST	OJ	260-345
15	OHY-S	21000	0620.0800	12ST	OK	435-520
							16	OHY-S	37827	0619.1100	12ST	OJ	470-540
17	ORG-N	6000	0621.0530	12ST	OJ	355-460
							18	PUK-N	10000	0621.0530	13ST	OL	100-200
19	MDM-N	3547	0617.2200	4ST	OC	445-525
							20	MDM-N	4303	0618.1300	4ST	OC	445-525
21	MGD-S	3210	0618.0400	4ST	OD	540-580
							22	MLC-C	4764	0618.2200	13ST	OM	700-760
23	MDM-S	7885	0619.0700	4ST	OC	530-605
							24	MLC-N	53362	0619.1600	4ST	OC	210-360
25	MLC-N	36338	0619.1600	4ST	OD	330-450

The comparison of human configuration result and Optimization result of the present invention through real data; Can draw; Improved under 2.7% the prerequisite in iron ore input implementation rate; As shown in Figure 6, compare with the human configuration scheme and can significantly objective function be reduced by 12.79% by the allocation plan that optimization of the present invention obtains.Reduce the transportation energy loss effectively, improved the storage efficiency of stock yard, improved the storage quality of iron ore, and then improved the quality of its subsequent product.

Claims (2)

1. iron and steel raw material field iron ore collocation method is characterized in that: may further comprise the steps:

Step 1: obtain and wait to import iron ore data, stock yard field storage bit data and stock yard service data, confirm the technology characteristics of iron ore input configuration, and adopt mathematical model to describe its technological process;

Wherein, describedly wait to import the iron ore data and comprise: iron ore lot number, iron ore kind, iron ore weight, chemical constitution, iron ore time of arrival, heap are high; Described stock yard storage bit data comprises: a storage position position coordinates, storage bit length, available storage position last time canned data, with a storage position canned data, input equipment work capacity; Described stock yard service data comprises: stock yard storage inventory, iron ore plan of needs, Plant maintenance plan;

Described mathematical model, it is following to set up process:

Step 1-1: confirm that model parameter is: in time prospect phase T, iron ore set to be disposed is N, and the set of material bar is S in the stock yard, and total K among the material bar j _jIndividual available storage position constitutes available storage position set K; When iron ore i is assigned to k available storage of material bar j position, need to consume levelling of the land, material empties and preparation work expense such as equipment prerun, also can produce simultaneously by iron ore is mixed and put the iron ore failure costs that causes, its cost coefficient is respectively c _IjkAnd p _IjkCan produce corresponding operating cost when opening material bar j, its cost coefficient is f _jWhen store the storage position when be assigned to different material bars with the lot number iron ore on, can cause the operations such as path conversion of logistics equipment, the equipment loss that produced this moment and the coefficient of energy consumption cost are q _i

Step 1-2: utilize the parameter of step 1-1, set up mathematical model, its objective function is following:

$\mathrm{Minimize}\underset{\text{i}}{\mathrm{\Σ}}\underset{\text{j}}{\mathrm{\Σ}}\underset{k}{\mathrm{\Σ}}{c}_{\mathrm{ijk}}{x}_{\mathrm{ijk}}+\underset{i}{\mathrm{\Σ}}\underset{j}{\mathrm{\Σ}}\underset{k}{\mathrm{\Σ}}{p}_{\mathrm{ijk}}{y}_{\mathrm{ijk}}+\underset{i}{\mathrm{\Σ}}{f}_j{z}_j+\underset{i}{\mathrm{\Σ}}\mathrm{qi}(\underset{j}{\mathrm{\Σ}}{\∂}_{\mathrm{ij}}-1)---\left(1\right)$

In the formula, x _Ijk, z _j,

And y _IjkBe decision variable, and its value is:

y _Ijk>=0, iron ore i is assigned to the weight of k storage position storage of material bar j;

When being assigned to k storage of material bar j, first of objective function (1) expression iron ore i empty and the energy that preparation work consumed such as equipment prerun and equipment loss expense etc. by levelling of the land, material; Second expression iron ore i is assigned to when expecting k storage position of bar j by the mixed iron ore failure costs that causes of putting of iron ore; The 3rd is the expense of offering material bar j, comprises that the material bar administers and maintains stacker-reclaimer opex in expense and the material bar; The 4th equipment loss and the energy consumption cost that the logistics equipment path conversion that causes when being assigned to different material bar j for a kind of iron ore i is produced;

Step 1-3: confirm the described bound for objective function of step 1-2:

(1) to arbitrary available storage position, all deposit the length (promptly piling up ability) that the shared accumulation length summation of iron ore can not exceed said storage position, and the formula of storage potential energy power constraint is following:

$\begin{array}{cc}\underset{i}{\mathrm{\Σ}}{\mathrm{\ρ}}_i{y}_{\mathrm{ijk}}+D(\underset{i}{\mathrm{\Σ}}{x}_{\mathrm{ijk}}-1)\≤l_{\mathrm{jk}}& \∀j\∈S,k\∈K---\left(2\right)\end{array}$

In the formula, ρ i is the required accumulation length that takies of iron ore i of the unit weight that draws according to known accumulation formula, l _JkBe the length of k storage position of material bar j, D is a required safe distance of isolating between two stockpiles;

(2) when arbitrary material bar is deposited iron ore, the weight of every heap iron ore can not surpass the maximum weight restriction of the single stockpile of this material bar, and the formula of stockpile weight limits constraint is following:

$\begin{array}{cc}{y}_{\mathrm{ijk}}\≤W_j& \∀i\∈N,j\∈S,k\∈K\end{array}----\left(3\right)$

In the formula, W _jUpper weight limit for the arbitrary stockpile of j material bar;

(3) to arbitrary material bar j, iron ore weight sum to be imported can not surpass the work capacity of this material bar support equipment, and the formula of implement ability constraint is following:

$\begin{array}{cc}\underset{i}{\mathrm{\Σ}}\underset{k}{\mathrm{\Σ}}{y}_{\mathrm{ijk}}\≤A_j& \∀j\∈S---\left(4\right)\end{array}$

In the formula, A _jThe work capacity upper limit for material bar j support equipment;

(4) guarantee that institute remains to be imported iron ore and all has been assigned to and stores in the stock ground, the formula of iron ore receiving-transmitting balance constraint is following:

$\begin{array}{cc}\underset{j}{\mathrm{\Σ}}\underset{k}{\mathrm{\Σ}}{y}_{\mathrm{ijk}}=Y_i& \∀i\∈N\end{array}---\left(5\right)$

(5) except manufacturability constraint (2)-(5), also need satisfy the variable logicality constraint of model itself, mainly comprise:

$\begin{array}{cc}{y}_{\mathrm{ijk}}\≤Y_i{x}_{\mathrm{ijk}}& \∀i\∈N,j\∈S,k\∈K\end{array}----\left(6\right)$

$\begin{array}{cc}\underset{i}{\mathrm{\Σ}}\underset{k}{\mathrm{\Σ}}{x}_{\mathrm{ijk}}\≤K_j{z}_j& \∀j\∈S\end{array}---\left(7\right)$

$\begin{array}{cc}\underset{k}{\mathrm{\Σ}}{x}_{\mathrm{ijk}}\≤K_j{\∂}_{\mathrm{ij}}& \∀i\∈N,j\∈S\end{array}---\left(8\right)$

Wherein, constraint (6) expression is assigned to the weight y of k storage position of material bar j as iron ore i _IjkGreater than 0 o'clock, x _IjkMust be 1, thereby guarantee that iron ore i is assigned to k assigning process that stores up the position of material bar j and has taken place; Constraint (7) has guaranteed that material bar j must open, wherein K when iron ore i is assigned to k storage position of material bar j _jAvailable storage bit quantity for material bar j; Constraint (8) has guaranteed that when iron ore i is assigned to k storage position of material bar j be x _IjkBe 1 o'clock, the inevitable material bar j that has been assigned to of iron ore i, promptly

Must be 1;

Step 2: to the input iron ore store up the position pre-configured, method is:

Step 2.1: with all available storages positions form available storage position gather U={ (j, k) | j ∈ S, k ∈ K} arranges the iron ore that remains to be imported according to required storage bit length from big to small, forms alternative iron ore sequence N;

Step 2.2: (j k), picks out from N and satisfies ρ from U, to choose an available storage position wantonly _iY _i≤l _JkIron ore subsequence N _Jk, preferentially select first iron ore i in the sequence ^*, if do not satisfy

Then from N _JkMiddle deletion iron ore i ^*And selection N _JkMiddle next bit iron ore is right

Judge; If satisfy

Then with iron ore i ^*Put into this storage position and delete iron ore i from N ^*If, should storage position (j k) has not utilized, with residue storage position as new available storage position (j k) is updated among the U, otherwise from U, delete and store up (j, k); If (j k), can't pick out and satisfy ρ for available storage position _iY _i≤l _JkAnd The iron ore subsequence, should store up the position (j k) joins in the set B of storage position, deletion storage position from U (j, k);

Step 2.3:, forward step 2.6 to if N=is Φ; Otherwise, forward step 2.4 to;

Step 2.4: if U=Φ forwards step 2.5 to; Otherwise, forward step 2.2 to;

Step 2.5: by the length scale descending sort, filling up these available storage positions successively with order is principle, and iron ore among the N is divided into a plurality of heaps, is assigned among the B and stores up in the position, all has been assigned with up to iron ore with available storage position among the B;

Step 2.6: above-mentioned iron ore storage position allocative decision is stored as an initial scheme individuality, if it is individual to have generated P initial scheme, pre-configured end; Otherwise all data of initialization, and forward step 2.1 to;

Step 3: the pre-configured result in storage position to iron ore is optimized adjustment: the P that is generated in the step 2 initial individual scheme formed an initial scheme population, utilizes difference algorithm that the pre-configured result in storage position is optimized adjustment, and method is:

Step 4: the allocation plan that step 3 is definite is delivered in the production executive system and is carried out.

2. iron and steel raw material according to claim 1 field iron ore collocation method is characterized in that: the described difference algorithm that utilizes of step 3 is optimized adjustment to the pre-configured result in storage position, and method is:

Step 3.1: each individuality in the previous generation scheme population is all carried out mutation operation; Promptly from previous generation scheme population, generate the individual entering of new scheme population of future generation; Its variation strategy does; In the G time iterative process; For the individual i of scheme; Be designated as i=1,2 ...; P, then a new individuality

produces according to following formula:

$V_i^{G+1}=X_a^G+F*(X_b^G-X_c^G)$

In the above-mentioned formula,

With

Be from current population P _GIn select 3 to be different from random Individuality, i.e. a ≠ b ≠ c; F is the self-adaptation mutagenic factor, is used for controlling schemes population diversity and convergence, and: F=β (e ^λ-1), in the formula, β ∈ [0.2,0.6],

Gmax is a greatest iteration algebraically, and G is the algebraically of current iteration;

Step 3.2: the scheme population is carried out interlace operation; Promptly make new and old scheme individuality intercourse partial code according to Crossover Strategy according to newly-generated scheme individual

; Thereby it is individual to form new scheme, and formula is following:

New scheme individuality after the intersection is:

$U_i^{G+1}=(u_{1i}^{G+1},u_{2i}^{G+1},...,u_{\mathrm{Di}}^{G+1})$

In the formula;

is j composition variable in the vector

, and the equation of interlace operation is:

In the formula,

Be vector

In j composition variable, Be vector

In j composition variable, rand (j) is an even distribution probability between [0,1], produces j estimated value of randomizer between [0,1]; The random integers that generate between mbr (i) expression [1, D] are guaranteed with it

At least from

Obtain a parameter; CR is a crossover probability, CR=a*CR _s, CR wherein _sBe predefined crossover probability, CR _s∈ (0,1) is constant, and a is the evolutionary rate factor, and its expression formula is following:

$a=\frac{f{(G-2)}_{\mathrm{best}}-f{(G-1)}_{\mathrm{best}}+\mathrm{const}}{f{(G-2)}_{\mathrm{best}}-f{\left(G\right)}_{\mathrm{best}}+\mathrm{const}}$

In the formula, f (G) best, f (G-1) best and f (G-2) _BestBe respectively G generation, G-1 generation and G-2 and calculate the optimum data desired value of gained after for individual substitution (1) formula; Const is little constant, and making molecule, denominator perseverance is not 0, and guarantees 0＜a≤1;

Step 3.3: the scheme population is selected; Produce follow-on scheme population; Selection strategy is:

compared with

; If the data target value of calculating gained is superior to the latter after the former substitution (1) formula, then just replace

otherwise

reservation in generation with

at G+1;

Step 3.4: individual for G+1 for arbitrary allocation plan in the scheme population, two iron ores in the same material bar are exchanged or two iron ores belonging to different material bars exchange, it is individual to generate new scheme, thus the scheme that obtains is gathered CH; In the time of must guaranteeing the feasible promptly satisfied constraint of exchange (2)-(5); Exchange can be implemented; Scheme individuality and G+1 among the CH are arranged according to the data target value size that (1) formula draws for all individualities in the scheme population from small to large, and select progressively P mutually different individuality formed new G+1 for the scheme population;

Step 3.5: if satisfy stop criterion, algorithm finishes; Otherwise jump to step 3.1.

Images