CN101344780B - Optimized furnace combination method and system for cold rolling hood type annealing furnace unit - Google Patents

Abstract

A furnace group optimizing method and a system for cold-rolling bell annealing furnace sets belong to the field of steel material processing information technology. The method comprises the following steps of: (1) according to practical technological parameters, performing priority settings on coiled sheets to be annealed; (2) furnace pre-grouping: arranging coiled sheets according to priority penalty values in ascending order; placing the coiled sheet with the minimum priority into a furnace according to the coiled sheet-furnace matching criteria; according to the coiled sheet-coiled sheet matching criteria, selecting, from the residual coiled sheets, coiled sheets that can form a furnace group with the coiled sheet in the furnace sequentially, and when the height reaches the maximum height of the bell furnace, forming a furnace grouping scheme; repeating the procedure until the demanded furnace group scale is formed; (3) adjusting the initial feasible furnace grouping scheme; and (4) using the developed system to upload the obtained optimized furnace grouping scheme to the manufacture execution system for implementation. The method and the system of the invention increase the productivity of the bell furnace set, improve the annealing quantity of the coiled sheets and effectively lower the consumption of energy resources such as hydrogen, nitrogen, water, coal and power, etc.

Description

A kind of cold rolling hood type annealing furnace set optimization furnace combination method

Technical field

The invention belongs to ferrous materials machining information technical field, relate to automatic technology, specially refer to a kind of bell type annealing furnace set optimization furnace combination method.

Background technology

Cold-rolled products are a kind of high value added products, play key effect to improving economic benefit of enterprises, also are important production links of iron and steel enterprise.Thereby how to utilize existing resource to save energy and reduce the cost, reduce production costs and increase economic efficiency, the problem that has become many enterprises and faced.Annealing is one technology of producing cold-rolled products, it be most smelters produce the high-quality cold-rolled products together must be through operation, according to the different mining of material and workpiece size with different temperature retention times, slowly cool off then, purpose is the metal inside tissue to be reached or near equilibrium state, obtaining favorable manufacturability energy and usability, perhaps is further to quench to do to organize to prepare.In the annealing operation of the cold rolling production of iron and steel enterprise, main employing stove group (bell furnace) is an annealing way, its main technique is that 3～4 roll bendings that performance is similar are wrapped on the same table, after heating, insulation, cooling, come out of the stove, whole annealing process approximately will experience 2-3 days time, bell furnace is finished all annealing operations in case igniting cannot stop until coiled sheet annealing.The annealing operation cycle is long, energy consumption is big, resource constraint is many, and row produces complicated, and the height of what and shove charge quality of its batch plays crucial effects for excavating unit production capacity, energy-saving and cost-reducing, speed production rhythm and shortening delivery date.

The group furnace gauge of bell furnace is drawn all by manually working out according to the personal experience at present, though staff planners have drawn at the group furnace gauge and have accumulated considerable experience aspect the establishment, also obtained very big effect, but because the mode of production of cover annealing has cycle length and energy consumption is big, and operation steps complexity, constraint condition is many, characteristics such as randomness height, make that the layout plan is a very loaded down with trivial details and technical job of having relatively high expectations, can't take into account production capacity simultaneously when manually working out plan, annealing quality, therefore many-sided requirements such as logistics balance and customer satisfaction are depended merely on manual method and are difficult to the annealing production of batch coiled sheet is made well, reasonably arrange.So being cold rolling unit, the cover annealing of coil of strip produces " bottleneck ".

Summary of the invention

At existing iron and steel enterprise bell type annealing furnace unit have that the unit production capacity is low, the bell furnace utilization factor is low, the annealing coiled sheet is of low quality and problem such as energy dissipation, the invention provides a kind of bell type annealing furnace unit coiled sheet of waiting to anneal and optimize furnace combination method, and set up a kind of bell-type and moved back the stove unit coiled sheet optimization group stove model of waiting to anneal.

The main task of group stove is to wait in the coiled sheet of annealing at N, (the individual coiled sheet of n≤N) is formed m shove charge unit and is produced, and improves batch to reach, improves the shove charge quality, purposes such as reduction energy loss to select n according to the matching relationship between physical property qualitative difference between them and they and the bell type annealing furnace.This method has been set up optimization group stove model according to practical problems, considered the complicated technology constraint in the actual production process, target is to pursue to make under the prerequisite that guarantees the cover annealing quality that bell furnace unit production capacity maximizes, energy resource consumption minimizes, and guarantees the logistics balance between cover annealing unit and the front and back operation simultaneously.

Suppose to have P idle bell type annealing furnace, N the coiled sheet of waiting for annealing now selected n according to the matching relationship between physical property qualitative difference between them and they and the bell type annealing furnace (n≤N) P heat of individual coiled sheet composition produced.Bell type annealing furnace optimization group stove problem can be described as:

$\min Z=\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}\underset{j=1}{\overset{\mathrm{<figure-callout\; id="1"\; label="P\; C"\; filenames="DEST\_PATH\_H200810012088301E00021.png,DEST\_PATH\_H200810012088301E00032.png"\; state="\{\{state\}\}">P</figure-callout>}}{\mathrm{\&Sigma;}}}{C}_{}{1}_{\mathrm{ij}}{X}_{\mathrm{ij}}+\underset{j=1}{\overset{P}{\mathrm{\&Sigma;}}}{p}_j{Z}_j+\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{f}_i(1-\underset{j=1}{\overset{P}{\mathrm{\&Sigma;}}}{X}_{\mathrm{ij}})+\underset{j=1}{\overset{P}{\mathrm{\&Sigma;}}}\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}\underset{k=3}{\overset{\mathrm{<figure-callout\; id="2"\; label="N\; C"\; filenames="DEST\_PATH\_H200810012088301E00021.png,DEST\_PATH\_H200810012088301E00032.png"\; state="\{\{state\}\}">N</figure-callout>}}{\mathrm{\&Sigma;}}}{C}_{}{2}_{\mathrm{ik}}{X}_{\mathrm{ij}}{X}_{\mathrm{kj}}---\left(1\right)$

Constraint condition:

$\underset{j=1}{\overset{P}{\mathrm{\&Sigma;}}}{X}_{\mathrm{ij}}\&le;1,i=1,...,N---\left(2\right)$

$\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{g_{i}X}_{\mathrm{ij}}+Z_j=T_j,j=1,...,P---\left(3\right)$

$\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{h_{i}X}_{\mathrm{ij}}\&le;H_j,j=1,...,P---\left(4\right)$

Z _j≥0，j＝1，…，P (5)

X _ij∈{0，1}，i＝1，…，N；j＝1，…，P (6)

In the objective function (1)

The coupling expense of expression coiled sheet and bell furnace is used for guaranteeing the quality of match of coiled sheet and bell furnace;

The rejection penalty of expression bell furnace residual capacity summation is used for guaranteeing the utilization factor of bell furnace;

Expression is used for guaranteeing that to the rejection penalty of selected coiled sheet not total batch can reach maximum horizontal;

Expression be coupling expense between the coiled sheet, be used for guaranteeing to be placed in the quality of the coiled sheet coupling in the same stove.

The constraint condition of model:

Coiled sheet of constraint (2) expression can only be arranged at most in the stove; Constraint (3) expression can not surpass the maximum load capacity of stove with the coiled sheet general assembly (TW) of j buttress; The overall height of the coiled sheet of j buttress of constraint (4) expression can not surpass the maximum height of stove; The residual capacity of constraint (5) expression heat can not be for negative.The span of constraint (6) expression decision variable.

Decision variable

Model parameter

1) Z _jThe residual capacity an of-Di j heat;

2) N-compiles the coiled sheet number of heat;

3) P-heat number, P=N _NHB+ N _NHS+ N _HHB+ N _HHS

4) T _jThe heat size an of-Di j heat;

5) p _jThe additional expense coefficient of-Di j heat residual capacity;

6) g _iThe weight an of-Di i coiled sheet;

7) h _iThe height an of-Di i coiled sheet;

8) f _iThe not selected caused additional expense coefficient an of-Di i coiled sheet;

9) H _jThe maximum carrying height an of-Di j stove;

10) C1 _IjThe coupling expense of-coiled sheet i and bell furnace j, it is mainly concerned with the importance (being the priority punishment of coiled sheet) of coiled sheet and the annealing curve matching degree and the specification matching degree of coiled sheet and bell furnace;

11) C2 _Ik-be placed in interior coiled sheet i of same stove and the coupling expense of k, be to the coiled sheet similarity measurement in the same stove,

Relate generally to the following aspects:

(1) the annealing curve rejection penalty coefficient between the coiled sheet, annealing curve are the control temperature (observed temperature) of bell furnace in the actual production and the function of time, and the embodiment of the amount of annealing process just is the foundation of coiled sheet annealing heating process.The coiled sheet that annealing curve is close is placed on to be processed the annealing quality that can better guarantee the bell furnace coiled sheet and reduces energy loss in the same stove.

(2) the thickness rejection penalty coefficient between the thickness of coiled sheet, because thin coiled sheet needs cold stoking, and thick coiled sheet does not need, therefore when shove charge as far as possible with coil sheet and the separately shove charge of slab volume, the annealing quality that can guarantee coiled sheet so effectively prevents the situation of burning and underburnt.

(3) the smooth finish difference rejection penalty coefficient of coiled sheet, when annealing production, through the coiled sheet smooth surface of degreasing, and there is impurity on unskimmed coiled sheet surface, impurity might be attached on the clean coiled sheet when heating in same stove, can cause coiled sheet the situation on carbonization limit to occur like this.This parameter is exactly to guarantee as far as possible and will separate with degreasing and unskimmed coiled sheet, reduces the generation of carbonization limit situation, reduces loss.

This optimization furnace combination method mainly may further comprise the steps:

The first step:, the coiled sheet of waiting for annealing is carried out the priority setting according to the actual process operational factor

The actual process operational factor refers to when coiled sheet priority is provided with, and each corresponding penalty value of preferred term institute is fixed condition really.Be made as the punishment of coiled sheet i priority, establish P _jBe the punishment of coiled sheet i priority, mainly consider following 5 preferred term factors:

(1) priority punishment at delivery date P _Ij ¹: the determined last delivery availability of production contract that refers to the coiled sheet place delivery date.Therefore the sincerity that is related to enterprise delivery date should guarantee delivering goods on schedule of coiled sheet as far as possible.

(2) collection crowd production priority punishment P _Ij ²: have the coiled sheet of collection batch code need collect batch processing, that is to say to put together as far as possible anneals.

(3) singularity priority punishment P _Ij ³: because the special coiled sheet of specification, material character has certain limitation on producing,, should pay the utmost attention to special coiled sheet and preferentially make it arrange into plan as far as possible so be pressed in situation in the storehouse in order to prevent a large amount of specific panel convolution.

(4) contract integrality priority P _Ij ⁴: collection is joined the cycle and is referred to coiled sheet in the same contract, finishes the time that whole operations are waited for delivery from first coiled sheet, finishes all process steps to a last coiled sheet and waits for that the time of delivery finishes, and is called during this period of time to collect and joins the cycle.Different coiled sheets in the same contract are not often at a time on same operation, and this just causes some coiled sheet to finish all process steps earlier waiting for delivery together after other coiled sheets are finished all process steps in the storehouse, therefore should dwindle collection as far as possible and join the cycle contract of guarantee integrality.

(5) coiled sheet punishes P in storehouse time priority level _Ij ⁵: refer to coiled sheet in the storehouse time and arrive the time in the preceding storehouse of bell furnace and the time interval of current layout planned time, should dwindle this time interval as far as possible, prevent that coiled sheet from bell furnace preceding storehouse accumulation, bringing pressure to it.

The pairing penalty value of each preferred term of coiled sheet according to ascending series arrangement and form the preferred term sequence, is used P _i(t) penalty value of t preferred term in the preferred term sequence of i coiled sheet of expression, t is smaller or equal to the number of total preferred term.If simply the value of each prioritized item is added up, it is very big that the priority penalty value will become, and work as prioritized item more for a long time, should cannot not become preferentially on the contrary by preferential coiled sheet, therefore the importance of coiled sheet for fear of this situation and when reflecting the actual sets stove, introduced a discounting factor beta among the present invention and adjusted factor-alpha, at this moment

P _i＝f(5)；f(t)＝f(t-1)β+αP _i(t)；f(1)＝P _i(1)。

Second step: organize stove in advance, according to coiled sheet priority, matching criteria between coiled sheet and coiled sheet and stove matching criteria generate the initial feasible group of furnace gauge of bell type annealing furnace and draw

An at first optional idle stove is selected the coiled sheet of all these stoves of can packing into according to coiled sheet and bell furnace matching criteria, and with them according to priority penalty value P _iAscending order is arranged, the coiled sheet of equal priority is according to the descending sort of weight, form alternative coiled sheet sequence, because the priority penalty value is more little, then coiled sheet is just preferential more, in the time of guaranteeing that like this each stove begins, the current alternative coiled sheet sequence medium priority the highest (priority penalty value minimum) and the coiled sheet of weight maximum are at first put in the stove.From alternative coiled sheet sequence, choose the coiled sheet that to form a stove with existing coiled sheet in the stove successively according to matching criteria between coiled sheet then, then put it in the stove, when highly reaching the maximum height of bell furnace, automatically form a group stove scheme, this step repeatedly can form P group stove scheme by nature then.

The priority method refers to be estimated the management attribute of each coiled sheet, will not arrange coiled sheet according to the ascending order of evaluation of estimate then and sorts, and sort based on this, generation group stove scheme.

When generating initial feasible group of stove scheme, propose following matching criteria and judge between coiled sheet and coiled sheet and coiled sheet and the stove whether mate:

Matching criteria between coiled sheet: if the annealing curve of two coiled sheet i and k is identical or shape is similar, two coiled sheet thickness all belong to same type (thin material or thick material), and the surface smoothness of coiled sheet is identical, judges that also two coiled sheets can be placed in the same stove.

Coiled sheet and stove matching criteria: if stove j can process the annealing curve of coiled sheet i, and the external diameter of coiled sheet i judges then that less than the external diameter of stove j coiled sheet i and stove j are complementary.

Algorithm steps:

If S represents alternative coiled sheet set, SS represents to select the coiled sheet set, and SR represents to remain the coiled sheet set.

Step1: choose an idle stove, according to stove and coiled sheet matching criteria, select can with the coiled sheet of this stove coupling,

And coiled sheet arranged according to priority punishment ascending order, priority identical according to the weight descending sort;

Step2: first unlabelled coiled sheet is placed in the stove in the selected coiled sheet sequence;

Step3: the next coiled sheet that is not labeled among the selected S, use matching criteria judgement between coiled sheet, if can form a buttress, then organize buttress, this coiled sheet of mark with selecting coiled sheet; If can not form a buttress, select next unmarked coiled sheet to judge.Current planning coiled sheet number equals at 4 o'clock, repeats Step3; Otherwise, forward Step4 to;

Step4: judge whether to also have idle stove, then forward Step2 to if having, otherwise forward Step5 to;

Step5: the coiled sheet of mark is put into SS, the coiled sheet that is not labeled is put into SR, algorithm finishes.

The process flow diagram of this method as shown in Figure 3.

The 3rd step: according to the objective function in the optimization group stove model, after the application enhancements strategy is adjusted formation bell furnace initial set stove scheme to initial feasible group of stove scheme, planned outcome still has certain room for improvement, adopt comparatively popular at present tabu search algorithm as framework both at home and abroad, characteristics structure according to the actual production of bell furnace unit improves neighborhood, cooperates search strategies such as becoming neighborhood, degree of deepening that initial feasible group of stove scheme improved.

(1) neighbour structure:

According to the characteristics of problem, proposed four kinds of search neighborhoods and be respectively: arranged coiled sheet and exchanged neighborhood in twos, two coiled sheets that are about to distribute to different stoves exchange, and this neighborhood search method flow diagram as shown in Figure 4; Row has not arranged coiled sheet two and has changed a neighborhood, is about to one and has arranged coiled sheet and shift out from affiliated stove, and do not arrange coiled sheet with two and put into, and this neighborhood search method flow diagram as shown in Figure 5; Row has not arranged coiled sheet and has exchanged neighborhood one to one, is about to one and has arranged coiled sheet and shift out from affiliated stove, and put into one and do not arrange coiled sheet, and this neighborhood search method flow diagram as shown in Figure 6; Degree of deepening mixes the exchanging chain neighborhood, this neighborhood is by introducing the notion of virtual coiled sheet and virtual stove, problem is converted into the collection partition problem, being about to all does not arrange coiled sheet and packs in the virtual stove according to the shove charge rule, and in each actual furnaces, add a virtual coiled sheet, so just, all coiled sheets are all packed in the stove, carry out swap operation on this basis again, two coiled sheets that are about to belong to different stoves exchange, and after the several times exchange, have just formed one and have mixed exchanging chain, when the exchange number of times is uncertain, this neighborhood is called degree of deepening and mixes the exchanging chain neighborhood, and as shown in Figure 7, this neighborhood search method flow diagram as shown in Figure 8.

(2) search strategy

When initial feasible group of stove scheme adjusted, difference in functionality according to three kinds of neighborhoods, search procedure is divided into three phases, specifically: because " row has not arranged two and changed one " neighborhood is to the improvement effect maximum of batch, therefore at first adopt this neighborhood search, adopting all neighborhoods to move maximum the separating as current of middle objective function improvement separates, next adopt " having arranged not row's exchange one to one " neighborhood to continue the direction search that increases to batch, same employing objective function improves maximum separating as current and separates, if these two kinds of neighborhoods all can't improve target, adopt " having arranged coiled sheet exchanges one to one " neighborhood so, the direction that moves is to make under the prerequisite of not violating constraint that to arrange coiled sheet concentrated as far as possible, for bigger probability is created in the insertion of not arranging coiled sheet next time.Repeat above search step, the result does not still improve after continuous search repeatedly, and then search degree of deepening mixes the exchanging chain neighborhood.

(3), proposed by the invention to initial feasible group of stove project setting method concrete steps following (process flow diagram is seen Fig. 9) based on top neighborhood and search strategy:

Step 1 is chosen an idle stove, according to stove and coiled sheet matching criteria, select can with the coiled sheet of this stove coupling, and coiled sheet arranged according to priority punishment ascending order, priority identical according to the weight descending sort.

First unlabelled coiled sheet is placed in the stove in the selected coiled sheet sequence of step 2.

The next coiled sheet that is not labeled among the selected S of step 3 is used matching criteria judgement between coiled sheet,, then puts it in the stove this coiled sheet of mark if can form a stove with selecting coiled sheet; If can not form a stove, select next unmarked coiled sheet to judge.Current planning coiled sheet number equals at 4 o'clock, repeats Step3; Otherwise, forward Step4 to;

Step 4 judges whether to also have idle stove, then forwards Step2 to if having, otherwise forwards Step5 to;

Step 5 is established F={F _l..., F _PMake up for P the shove charge that obtains, all do not arrange coiled sheet S={ }, calculate current goal functional value Z according to F and S.If current preferably separating is F _Best, corresponding target function value is Z _Best, other is F _Best=F, Z _Best=Z forwards step 6 to.

Step 6 couple F and S search has not been arranged coiled sheet and has been exchanged neighborhood in twos, and the record neighborhood is not preferably separated F ' in the taboo table, upgrade current group of furnace accretion and really gather F, residue coiled sheet S set and taboo table.If F '＜F _Best, F in addition then _Best=F '.

Step 7 is if preferably separate F _BestBe updated, then repeating step 6, otherwise pair set F and S search is not arranged coiled sheet and arranged coiled sheet two and change a neighborhood, choose not preferably to separate F ' in neighborhood in the taboo table, upgrade current group of furnace accretion and really gather F, do not arrange coiled sheet S set and taboo table.If F '＜F _Best, F in addition then _Best=F '.

If step 8 continuous some generations preferably do not separate and are updated, then F and S degree of deepening are mixed the exchanging chain neighborhood search.Record searching to neighborhood preferably separate F ', upgrade current group of furnace accretion and really gather F and do not arrange coiled sheet, if F '＜F in conjunction with S _Best, F in addition then _Best=F '.

It is to satisfy end condition that step 9 is judged, if satisfy, then terminator is exported F as a result _Best, otherwise, forward step 6 to, until satisfying end condition.

The 4th step: use the bell type annealing furnace set optimization group furnace system that the present invention developed and to be uploaded to enterprise's manufacturing execution system according to the resulting optimization group of above-mentioned steps stove scheme, and carry out last resulting optimization group stove scheme by the production automatic control system of bell furnace unit.

The system that a kind of cold rolling hood type annealing furnace set optimization furnace combination method adopts, this system comprises following functional module: authorized user login module, raw data download module, raw data administration module, group stove information typing module, static parameter maintenance module, group stove scheme automatically-generating module, group stove scheme display module, group stove scheme modifying module, group stove evaluate alternatives module, group stove scheme are checked transmission module, system configuration module on module, the group stove scheme result in violation of rules and regulations, and each functions of modules is as follows:

1, authorized user login module

Only to open through the planner who authorizes, the input username and password, and be verified and just can enter system after correct;

2, raw data download module

Institute's data downloaded is imported in the database, and data item mainly comprises: state, plate reel number, thickness, width, weight, tapping mark, the trade mark, material group, additional number, external diameter, annealing curve, delivery date, contract number, degree of sorting;

3, raw data administration module

Raw data is managed: increase, deletion, revise or coiled sheet is sorted according to a certain attribute of coiled sheet, but to domestic consumer openly increase, deletion and revise authority;

4, group stove information typing module

The user can import in this module and comprise that group stove scale, preformed plate volume information and coiled sheet mix the group stove programming parameter of the information of taking, bell furnace group stove scheme kind is divided into four kinds: the big stove group of nitrogen hydrogen stove scheme, nitrogen hydrogen port group stove scheme, the big stove group of perhydro stove scheme and perhydro port group stove scheme, group stove scale is the quantity of every kind of group stove of required layout scheme, is expressed as N respectively _NHB, N _NHS, N _HHBAnd N _HHS, reserving coiled sheet is exactly to refer to that some coiled sheet does not participate in the establishment of current group stove scheme in the layout plan;

5, static parameter maintenance module

This module comprises expense punishment table and two parts of data screening condition configuration, wherein expense punishment table comprises collocation punishment between every priority punishment of the every weight of objective function, coiled sheet and coiled sheet, and wherein the configuration of data screening condition comprises degreasing condition, delivery date option and three parts of priority option;

6, group stove scheme automatically-generating module

This module mainly comprises two submodules: group stove process prescription submodule and optimization group stove module, optimization group stove model and bell furnace are optimized furnace combination method and are embedded in respectively in group stove process prescription submodule and the optimization group stove module, when generation group furnace gauge is drawn, system at first is delivered to the real-time production data information of downloading in the raw data download module in the group stove process prescription submodule, and optimizing application group stove module is found the solution;

7, group stove scheme display module

For the group stove scheme that generates, interface the first half is an X-Y scheme form result displayed picture, interface the latter half is the form of its corresponding data, the user can and not arrange between the coiled sheet pattern four kinds of group stove schemes switches, the coiled sheet in the acquisition group stove scheme and row advance to organize the coiled sheet information of stove scheme;

8, group stove scheme modifying module

This module can be edited in the graphic result of display module, mobile coiled sheet or coiled sheet arranged the coiled sheet pattern and do not arranging between the coiled sheet pattern and to move;

9, group stove evaluate alternatives module

After group stove scheme generated, this module can provide the group furnace accretion fruit of algorithm automatic editing and correlative value and the objectives functional value of group furnace accretion fruit under given objective function of manual layout;

10, group stove scheme is checked module in violation of rules and regulations

For each result, system can be automatically carries out violation inspection according to the group furnace gauge journey of planing to it, if then can demonstrate violation information in violation of rules and regulations;

11, transmission module on the group stove scheme

When the user is satisfied to group stove scheme, the result can be uploaded to enterprise's ERP system;

12, system configuration module

The user can be configured system: calculating parameter, server, and calculating parameter comprises: objective function weight and penalty value; Server configures mainly is address, port, the username and password of server.

Advantage of the present invention and technical progress are tangible:

(1) the present invention is not only applicable to the production of iron and steel enterprise's cover annealing unit, can also be widely used in the batch processing operation that other industry has same nature.The present invention is from the actual production technology of bell furnace, with bell furnace actual sets stove process complete be described as the batch processor scheduling problem that a class has process constraint, and efficient, the feasible optimization furnace combination method of one cover has been proposed according to the characteristics of problem.

(2) the present invention quantizes actual bell furnace unit production target, under the prerequisite that the structure of model parameter is analyzed, provide the plan of establishment and the computing method of concrete parameter, made model parameter and objective function can reflect actual group stove situation.In tabu search algorithm, incorporated degree of deepening and become the neighborhood search strategy, and introduced effective search neighborhood according to bell furnace group stove characteristics.Test findings shows, the realistic condition of production of bell furnace group stove process prescription method proposed by the invention, and preferably separating of being drawn of group stove optimization method satisfy the actual production constraint, is better than artificial row and produces the result.

(3) manually organize furnace accretion is really optimized the result with the present invention detailed comparison by actual 6 groups of data, can draw, under the prerequisite of batch raising 5.77% (batch result relatively sees Figure 10), compare and significantly objective function to be reduced by 10.79% with the group stove scheme that artificial row produces by the group stove scheme that optimization of the present invention obtains.Improve the production capacity of bell furnace unit effectively, improved the annealing quality of coiled sheet, and then reduced the loss of hydrogen, nitrogen, water, coal, electric equal energy source.(the objective function result relatively sees Figure 11)

(2) the bell type annealing furnace optimization group stove software systems that the present invention developed, make and be accurate to the precision of group stove scheme layout " day " can be from artificial group of stove the time " hour ", can satisfy accurate more production operation requirement, improve the overall operation efficient of bell type annealing furnace unit.

Description of drawings

Fig. 1 is a bell type annealing furnace furnace combination method process flow diagram proposed by the invention,

Fig. 2 produces the system operation process flow diagram for the bell type annealing furnace optimization row that the present invention developed,

Fig. 3 is the initial feasible scheme production method of a bell type annealing furnace group stove process flow diagram proposed by the invention,

Fig. 4 exchanges the neighborhood search method flow diagram in twos for the coiled sheet of row proposed by the invention,

Fig. 5 changes an exchange neighborhood search method flow diagram for not row proposed by the invention has arranged coiled sheet two,

Fig. 6 exchanges the neighborhood search method flow diagram one to one for not row proposed by the invention has arranged coiled sheet,

Fig. 7 mixes exchanging chain neighborhood synoptic diagram for degree of deepening proposed by the invention,

Fig. 8 mixes exchanging chain neighborhood search procedure figure for degree of deepening proposed by the invention,

Fig. 9 proposed by the invention adjusts the scheme process flow diagram to initial feasible scheme,

Figure 10 compares synoptic diagram for using optimization furnace combination method proposed by the invention and manual method batch result,

Figure 11 compares synoptic diagram for using optimization furnace combination method proposed by the invention and manual method objective function result,

Figure 12 downloads the interface for the optimization group furnace system data that the present invention developed,

Figure 13 is the optimization group furnace system initial information inputting interface that the present invention developed,

Figure 14 is the optimization group furnace system result data interface that the present invention developed,

Figure 15 manually adjusts the interface for the optimization group furnace system that the present invention developed,

Figure 16 estimates the interface for the optimization group furnace system that the present invention developed.

Embodiment

The enforcement of system of the present invention needs as lower device: at least one PC; At least one cable interface or optical cable interface or special telephone line interface; At least one router.These equipment are placed on the logistics chamber or the Plan Office of establishment hot rolling plan.Be connected in enterprise's ERP system after forming a small-sized LAN (Local Area Network) by these equipment.System of the present invention is installed in PC, and is connected to the front end of the production automatic control system of bell-type unit by network, router.This software is support platform with Windows operating system, and Microsoft SQL Server 2000 Database Systems at first are installed, and system of the present invention is installed then, and the initialization information configured port is set, and installs and finishes.

System of the present invention operates according to the following steps:

(1) after the user starts bell furnace optimization group furnace system, at first imports username and password, if the user then can not enter optimization group furnace system for the disabled user; If the user is a legal users, then can enter optimization group furnace system smoothly.

(2) user utilizes job number of oneself and password to enter data-triggered picture in enterprise's ERP system, carries out bell furnace and waits to arrange the functional module that coiled sheet information generates.This moment, enterprise's ERP system can generate this data file (with the form storage of text) under the assigned catalogue in the ftp server of assigned ip address.

(3) user enters data and downloads the interface in optimization group furnace system, carries out the download of the preceding storehouse of bell furnace unit coiled sheet information.The coiled sheet information field of downloading comprises: state, plate reel number, thickness, width, weight, tapping mark, the trade mark, material group, additional number, external diameter, annealing curve, delivery date, contract number, degree of sorting.The download of coiled sheet information is to utilize FTP order to download data file with the form storage of text from the ftp server of appointment in this machine, utilizes SQL statement that the slab information in the text is read in the SQL Server 2000 data of database tables then.Coiled sheet information after downloading successfully will be presented in the download interface of system (Figure 12).

(4) after data were downloaded and finished, the user need import the capacity of every class bell furnace plan, the mixed information (Figure 13) such as putting and reserve coiled sheet of setting up.Behind the input plan information, the automatic systematic function of execution group stove scheme (optimizing furnace combination method) based on bell furnace optimization group stove model proposed by the invention and bell furnace.

(5) the group furnace accretion fruit (being the group stove scheme that algorithm obtains) that generates automatically for system, every penalty value that the user can really compare and check objective function by Function of Evaluation and artificial group furnace accretion, if it is dissatisfied to the result, then the user can be in graphic editor makes amendment (comprising deletion, interpolation, replacement of coiled sheet etc.) to the result, till satisfaction; System all can carry out violation inspection to current result in the each process of revising of user.If satisfied to the result, uploading of then execution group stove scheme uploads to it in enterprise's ERP system.The group furnace accretion fruit that uploads to enterprise's ERP system is (with the heat (batch) number series arrangement, each heat (batch) number comprises plate reel number wherein) be to be stored on the assigned catalogue of this machine with the text form earlier equally, its filename with appointment is uploaded in the assigned catalogue of ftp server of enterprise's ERP system appointment by FTP order then, and then ftp server uploads to the result ERP system (Figure 14-26) of enterprise thus.

(6) enterprise's ERP system according to the group stove scheme that obtains (with the heat (batch) number series arrangement, each heat (batch) number comprises plate reel number wherein) the standard group furnace gauge of the data formation enterprise ERP system of taking-up correspondence is drawn from relevant tables of data, is issued to the last group stove scheme of execution in the system that automatically performs of producing at last.

Embodiment

Example one: this embodiment carries out the shove charge combination at 20 coiled sheets and 3 bell type annealing furnaces, and wherein the plate reel number is a unique identification coiled sheet index entry, carries out the shove charge combination according to technological requirement, and every stove is adorned 4 coiled sheets at most.

Reel number	Thickness	Volume is heavy	Width	Annealing curve	Contract number	External diameter	Delivery date	Degree of sorting
									1	1.264	28.11	1512	01	L750739B11	1835	20071231	2
2	2.522	32.21	1511	01	LH10432A07	1958	20071205	3
									3	0.599	25.4	1551	11	N770465B02	1761	20071117	3
4	1.213	28.58	1262	02	L761656D02	2029	20071130	2
									5	0.807	34.123	1525	64	L780834A01	1992	20071130	4
6	1.01	34.629	1811	64	L761656G03	1852	20071130	4
									7	2.533	32.24	1511	01	LH10434A02	1960	20071205	3
8	1.01	40.599	1811	64	L761656G03	1992	20071130	4
									9	0.757	26.954	1292	64	L761564A01	1925	20071130	4
10	0.706	18.34	1015	02	L734576B01	1821	20071130	2
									11	0.807	19.83	1015	01	L761621B01	1902	20071130	2
12	1.213	22.72	1262	02	L761656D02	1813	20071130	2
									13	0.706	16.57	1125	02	L771687A02	1663	20071130	2
14	2.522	32.2	1511	01	LH10432A07	1945	20071205	3
									15	0.706	24.955	1185	04	L780826D01	1938	20071130	2
16	0.807	22.93	1185	02	L771675A04	1889	20071130	2
									17	0.909	23.47	1182	02	L734576B03	1898	20071130	2
18	0.757	26.614	1292	64	L761564A01	1915	20071130	4
									19	0.807	33.853	1525	64	L780834A01	1986	20071130	4
20	0.599	29.075	1465	11	N731327A04	1893	20071231	3

The concrete operations step is as follows:

The first step: the user at first imports username and password after starting bell furnace optimization group furnace system, if the user then can not enter optimization group furnace system for the disabled user; If the user is a legal users, then can enter optimization group furnace system smoothly.

Second step: the user utilizes own job number and password to enter data-triggered picture in enterprise's ERP system, carries out the functional module that bell furnace waits to arrange the generation of coiled sheet information.This moment, enterprise's ERP system can generate this data file (with the form storage of text) under the assigned catalogue in the ftp server of assigned ip address.

The 3rd step: the user enters data and downloads the interface in optimization group furnace system, carries out the download of the preceding storehouse of bell furnace unit coiled sheet information.The coiled sheet information field of downloading comprises: state, plate reel number, thickness, width, weight, tapping mark, the trade mark, material group, additional number, external diameter, annealing curve, delivery date, contract number, degree of sorting.The download of coiled sheet information is to utilize FTP order to download data file with the form storage of text from the ftp server of appointment in this machine, utilizes SQL statement that the coiled sheet information in the text is read in the SQL Server 2000 data of database tables then.Coiled sheet information after downloading successfully will be presented in the download interface of system.

The 4th step: bell furnace group stove model is carried out following parameter setting:

Sequence number	Fee item	Symbol	Concrete parameter value	Remarks
						1	The coiled sheet number	N		20
2	The heat number	P		3	2 in the big stove of NH, one in the big stove of HH
						3	Heat size	T j	155 99	Big stove 155 ports 99
4	The residual capacity expense	p j	0.5/t+0.01/mm
					5	Do not arrange the coiled sheet expense	f	i	60
6	The maximum carrying height of stove	H j	4700

1. the priority rejection penalty P of coiled sheet i _i:

P _iBe the punishment of coiled sheet priority, mainly consider following 5 preferred term factors:

(1) P1 _i ¹Expression priority punishment at delivery date: refer to the last delivery availability that the production contract of this coiled sheet is determined delivery date, any product all is to be purpose with the sale, and whether can deliver goods on schedule is related to the sincerity of producer, and this index is exactly most important naturally.

${P1}_{\mathrm{<figure-callout\; id="1"\; label="i"\; filenames="DEST\_PATH\_H200810012088301E00021.png,DEST\_PATH\_H200810012088301E00032.png"\; state="\{\{state\}\}">i</figure-callout>}}^1=\left\{\begin{array}{cc}20,& A<6\\ 0,& 6\&le;A\&le;8\\ 60,& A>8\end{array}\right.$

Wherein, the current date when A represents to work out plan and the time interval at delivery date, three intervals of A are the actual motion technological parameter of preferred term at delivery date.

(2) P1 _i ²Expression material group/degree of sorting priority punishment.Material group refers to the machining path of material, and material group is that the coiled sheet of " 25 ", " 2C " has the collection batch code need collect batch processing, belongs to second priority; Degree of sorting refers to the degree of treatment of material surface, and degree of sorting is that 05 plate need collect batch processing for the coiled sheet of " 4 ", so belong to second priority equally.

${P1}_{\mathrm{<figure-callout\; id="2"\; label="i"\; filenames="DEST\_PATH\_H200810012088301E00021.png,DEST\_PATH\_H200810012088301E00032.png"\; state="\{\{state\}\}">i</figure-callout>}}^2=\left\{\begin{array}{cc}11,& \mathrm{coi}{l}_i\&Element;\mathrm{<figure-callout\; id="60"\; label="SF"\; filenames="DEST\_PATH\_H200810012088301E00041.png"\; state="\{\{state\}\}">SF</figure-callout>}\\ 60,& \mathrm{Normal}\end{array}\right.$

Wherein, SF represents to collect the coiled sheet set of batch processing, and Normal is common coiled sheet.

(3) P1 _i ³The singularity priority punishment of expression coiled sheet.Because thickness has certain limitation less than the thin material of 0.65mm on producing, so, should pay the utmost attention to thin material and make that as far as possible approaching material row advances plan in order to prevent a large amount of thin situations of expecting to overstock in the storehouse.In like manner, some have the coiled sheet of special tapping mark and special annealing curve, because amount is fewer, are difficult to find the coiled sheet of collocation with it, so can arrange into plan in case find this class coiled sheet, will pay the utmost attention to.

${P1}_{\mathrm{<figure-callout\; id="3"\; label="i"\; filenames="DEST\_PATH\_H200810012088301E00021.png,DEST\_PATH\_H200810012088301E00032.png"\; state="\{\{state\}\}">i</figure-callout>}}^3=\left\{\begin{array}{cc}33,& {\mathrm{Curv}}_i\&Element;\mathrm{Spec}\_\mathrm{<figure-callout\; id="18"\; label="C"\; filenames="DEST\_PATH\_H200810012088301E00021.png"\; state="\{\{state\}\}">C</figure-callout>}\\ 18,& {\mathrm{Mark}}_i\&Element;\mathrm{Spec}\_M\\ 45,& T{\mathrm{hick}}_i\&le;0.65\\ 60,& \mathrm{Normal}\end{array}\right.$

Wherein, Curv _iThe annealing curve of representing i coiled sheet, Spec_C={11,12,13,34,35,36,66,67,68} represents the set of special annealing curve, Mark _iThe tapping mark of representing i coiled sheet, Spec_M={ " DQ0240D1 ", " IP1240D1 ", " DU3340D1 ", " DT3141D1 ", " DT3480D1 ", " DU3541D1 " } set of the special tapping mark of expression, Thick _iThe thickness of representing i coiled sheet, when coiled sheet thickness during smaller or equal to Low_Th, coiled sheet is thin material, Normal is common coiled sheet.

(4) P1 _i ⁴Expression contract integrality priority.Collection is joined the cycle and is referred to coiled sheet in the same contract, finishes the time that whole operations are waited for delivery from first coiled sheet, finishes all process steps to a last coiled sheet and waits for that the time of delivery finishes, and is called during this period of time to collect and joins the cycle.Different coiled sheets in the same contract are not often at a time on same operation, this just causes some coiled sheet to finish all process steps earlier waiting for delivery together after other coiled sheets are finished all process steps in the storehouse, can bring pressure to the stock like this, therefore we consider the integrality and the operation amount of owing of contract, it as the 4th priority.

${P1}_{\mathrm{<figure-callout\; id="4"\; label="i"\; filenames="DEST\_PATH\_H200810012088301E00021.png,DEST\_PATH\_H200810012088301E00032.png"\; state="\{\{state\}\}">i</figure-callout>}}^4=\left\{\begin{array}{cc}34,& {\mathrm{Per}}_i\&GreaterEqual;4\\ 35,& W_i\&GreaterEqual;{\mathrm{Loss}}_i\\ 60,& \mathrm{Normal}\end{array}\right.$

Wherein, Per _iThe standard of expression coiled sheet i is sent out throughput, and Up_Per represents the accurate upper limit of sending out throughput of contract, W _iThe weight of expression coiled sheet i, Loss _iThe operation amount of owing of expression coiled sheet i, Normal represents common coiled sheet.

(5) P1 _i ⁵Punishing of expression coiled sheet in storehouse time priority level.Refer to coiled sheet in the storehouse time and arrive the time in the preceding storehouse of bell furnace and the time interval of current layout planned time, should dwindle this time interval as far as possible, prevent that coiled sheet from piling up in the preceding storehouse of bell furnace, bring pressure to it, thus material the storehouse time as the 5th priority.

${P1}_{\mathrm{<figure-callout\; id="5"\; label="i"\; filenames="DEST\_PATH\_H200810012088301E00021.png"\; state="\{\{state\}\}">i</figure-callout>}}^5=\left\{\begin{array}{cc}20,& d_i\&GreaterEqual;5\\ 60,& d_i<5\end{array}\right.$

Wherein, d _iRepresent i coiled sheet in the storehouse time, the D display plate twists in the storehouse time tolerance upper limit.

The pairing penalty value of each preferred term of coiled sheet according to ascending series arrangement and form the preferred term sequence, is used P _i(t) penalty value of t preferred term in the preferred term sequence of i coiled sheet of expression, t is smaller or equal to the number of total preferred term.If simply the value of each prioritized item is added up, it is very big that the priority penalty value will become, and work as prioritized item more for a long time, should cannot not become preferentially on the contrary by preferential coiled sheet, therefore the importance of coiled sheet for fear of this situation and when reflecting the actual sets stove, introduced a discounting factor beta among the present invention and adjusted factor-alpha, at this moment

P _i＝f(5)；f(t)＝f(t-1)β+αP _i(t)；f(1)＝P _i(l)。

It is as follows to obtain coiled sheet priority rejection penalty by above-mentioned setting:

Reel number	Thickness	Volume is heavy	Width	Annealing curve	Contract number	External diameter	Delivery date	Degree of sorting	Priority
										8	1.01	40.599	1811	64	L761656G03	1992	20071130	4	0
6	1.01	34.629	1811	64	L761656G03	1852	20071130	4	0
										5	0.807	34.123	1525	64	L780834A01	1992	20071130	4	0
19	0.807	33.853	1525	64	L780834A01	1986	20071130	4	0
										1	1.264	28.11	1512	01	L750739B11	1835	20071124	2	2.4
3	0.599	25.4	1551	11	N770465B02	1761	20071117	3	7.846
										9	0.757	26.954	1292	64	L761564A01	1925	20071130	4	10.186
18	0.757	26.614	1292	64	L761564A01	1915	20071130	4	10.186
										7	2.533	32.24	1511	01	LH10434A02	1960	20071205	3	12.4
2	2.522	32.21	1511	01	LH10432A07	1958	20071205	3	12.4
										14	2.522	32.2	1511	01	LH10432A07	1945	20071205	3	12.4

4	1.213	28.58	1262	02	L761656D02	2029	20071130	2	30.524
										12	1.213	22.72	1262	02	L761656D02	1813	20071130	2	30.524
15	0.706	24.955	1185	04	L780826D01	1938	20071130	2	31.359
										17	0.909	23.47	1182	02	L734576B03	1898	20071130	2	31.359
16	0.807	22.93	1185	02	L771675A04	1889	20071130	2	31.359
										10	0.706	18.34	1015	02	L734576B01	1821	20071130	2	31.359
13	0.706	16.57	1125	02	L771687A02	1663	20071130	2	31.359
										11	0.807	19.83	1015	01	L761621B01	1902	20071130	2	42.4
20	0.599	29.075	1465	11	N731327A04	1893	20071231	3	43.575

2. the punishment of the matching degree between coiled sheet i and the bell furnace j

If C1 _IjBe the punishment of the matching degree between coiled sheet i and the bell furnace j, it is mainly concerned with the importance (being the priority punishment of coiled sheet) of coiled sheet and the annealing curve matching degree and the specification matching degree of coiled sheet and bell furnace;

C1 _ij＝P _i+P2 _ij

P2 _IjBe the punishment of the matching degree between bell furnace and the coiled sheet, Consideration mainly comprises the annealing curve matching degree and the specification matching degree of coiled sheet and bell furnace.

${\mathrm{<figure-callout\; id="2"\; label="P"\; filenames="DEST\_PATH\_H200810012088301E00021.png,DEST\_PATH\_H200810012088301E00032.png"\; state="\{\{state\}\}">P</figure-callout>}2}_{\mathrm{ij}}=\left\{\begin{array}{cc}0,& f({\mathrm{coil}}_i,j)=1\\ +\&infin;,& f({\mathrm{coil}}_i,j)=0\end{array}\right.$

F (coil _i, j) coiled sheet and the judgement of stove matching criteria are used in expression, and whether coiled sheet i and stove j mate, if the coupling rreturn value is 1, otherwise are 0.

3. coiled sheet i in the same stove and the coupling expense C2 of k _Ik, mainly comprise three factors:

(1) C2 _Ik ¹Annealing curve rejection penalty coefficient between-coiled sheet i and the k, annealing curve are the control temperature (observed temperature) of bell furnace in the actual production and the function of time, and the embodiment of the amount of annealing process just is the foundation of coiled sheet annealing heating process.The coiled sheet that annealing curve is close is placed on to be processed the annealing quality that can better guarantee the bell furnace coiled sheet and reduces energy loss in the same stove, annealing curve rejection penalty coefficient is defined as follows:

${\mathrm{<figure-callout\; id="2"\; label="C"\; filenames="DEST\_PATH\_H200810012088301E00021.png,DEST\_PATH\_H200810012088301E00032.png"\; state="\{\{state\}\}">C</figure-callout>}2}_{\mathrm{<figure-callout\; id="1"\; label="ik"\; filenames="DEST\_PATH\_H200810012088301E00021.png,DEST\_PATH\_H200810012088301E00032.png"\; state="\{\{state\}\}">ik</figure-callout>}}^1=\left\{\begin{array}{cc}0,& {\mathrm{Curv}}_i={\mathrm{Curv}}_k;\\ 2.5,& f\left({\mathrm{Curv}}_i\right)=f\left({\mathrm{Curv}}_k\right);\\ +\&infin;,& f\left({\mathrm{Curv}}_i\right)\&NotEqual;f\left({\mathrm{Curv}}_k\right)\end{array}\right.$

Wherein, Curv _i, Curv _kThe annealing curve of representing i coiled sheet and k coiled sheet respectively, f (Curv _i) and f (Curv _k) represent the affiliated type of i coiled sheet and k coiled sheet annealing curve respectively.

(2) C2 _Ik ²Thickness rejection penalty coefficient between the thickness of-coiled sheet i and k, because thin coiled sheet needs cold stoking, and thick coiled sheet does not need, therefore when shove charge as far as possible with coil sheet and the separately shove charge of slab volume, the annealing quality that can guarantee coiled sheet so effectively prevents the situation of burning and underburnt.Thickness rejection penalty coefficient is defined as follows:

${\mathrm{<figure-callout\; id="2"\; label="C"\; filenames="DEST\_PATH\_H200810012088301E00021.png,DEST\_PATH\_H200810012088301E00032.png"\; state="\{\{state\}\}">C</figure-callout>}2}_{\mathrm{<figure-callout\; id="2"\; label="ik"\; filenames="DEST\_PATH\_H200810012088301E00021.png,DEST\_PATH\_H200810012088301E00032.png"\; state="\{\{state\}\}">ik</figure-callout>}}^2=\left\{\begin{array}{cc}+\&infin;,& f\left(T_i\right)\&NotEqual;f\left(T_k\right)\\ 5*(T_i-T_k),& f\left(T_i\right)=f\left(T_k\right)\end{array}\right.$

Wherein, T _iAnd T _kThe thickness of representing coiled sheet i and coiled sheet k respectively, f (T _i) and f (T _k) represent the thickness type of coiled sheet i and coiled sheet k respectively.

(3) C2 _Ik ³The preceding operation difference rejection penalty coefficient of-coiled sheet i and coiled sheet k, when annealing production, coiled sheet smooth surface through degreasing, and there is impurity on unskimmed coiled sheet surface, impurity might be attached on the clean coiled sheet when heating in same stove, can cause coiled sheet the situation on carbonization limit to occur like this.This parameter is exactly to guarantee as far as possible and will separate with degreasing and unskimmed coiled sheet, reduces the generation of carbonization limit situation, reduces loss.Preceding operation difference rejection penalty coefficient is defined as follows:

${\mathrm{<figure-callout\; id="2"\; label="C"\; filenames="DEST\_PATH\_H200810012088301E00021.png,DEST\_PATH\_H200810012088301E00032.png"\; state="\{\{state\}\}">C</figure-callout>}2}_{\mathrm{<figure-callout\; id="3"\; label="ik"\; filenames="DEST\_PATH\_H200810012088301E00021.png,DEST\_PATH\_H200810012088301E00032.png"\; state="\{\{state\}\}">ik</figure-callout>}}^3=\left\{\begin{array}{cc}0,& p_i=p_k;\\ 1.25,& p_i\&NotEqual;p_k\end{array}\right.$

Wherein, p _iAnd p _kThe preceding working procedure sequence number of representing coiled sheet i and coiled sheet k respectively.

${\mathrm{<figure-callout\; id="2"\; label="C"\; filenames="DEST\_PATH\_H200810012088301E00021.png,DEST\_PATH\_H200810012088301E00032.png"\; state="\{\{state\}\}">C</figure-callout>}2}_{\mathrm{ik}}={\mathrm{<figure-callout\; id="2"\; label="C"\; filenames="DEST\_PATH\_H200810012088301E00021.png,DEST\_PATH\_H200810012088301E00032.png"\; state="\{\{state\}\}">C</figure-callout>}2}_{\mathrm{<figure-callout\; id="1"\; label="ik"\; filenames="DEST\_PATH\_H200810012088301E00021.png,DEST\_PATH\_H200810012088301E00032.png"\; state="\{\{state\}\}">ik</figure-callout>}}^1+{\mathrm{<figure-callout\; id="2"\; label="C"\; filenames="DEST\_PATH\_H200810012088301E00021.png,DEST\_PATH\_H200810012088301E00032.png"\; state="\{\{state\}\}">C</figure-callout>}2}_{\mathrm{<figure-callout\; id="2"\; label="ik"\; filenames="DEST\_PATH\_H200810012088301E00021.png,DEST\_PATH\_H200810012088301E00032.png"\; state="\{\{state\}\}">ik</figure-callout>}}^2+{\mathrm{<figure-callout\; id="2"\; label="C"\; filenames="DEST\_PATH\_H200810012088301E00021.png,DEST\_PATH\_H200810012088301E00032.png"\; state="\{\{state\}\}">C</figure-callout>}2}_{\mathrm{<figure-callout\; id="3"\; label="ik"\; filenames="DEST\_PATH\_H200810012088301E00021.png,DEST\_PATH\_H200810012088301E00032.png"\; state="\{\{state\}\}">ik</figure-callout>}}^3$

The 5th step:, use bell furnace optimization furnace combination method and find the solution according to the parameter and the bell furnace group stove model of above-mentioned setting:

Step1: S set={ all do not arrange coiled sheet } in addition, and k=0 in addition, j=0.The optional big stove of a stove HH, according to coiled sheet and bell furnace matching criteria, select can with the coiled sheet of this stove coupling, and coiled sheet arranged according to priority punishment ascending order, priority identical according to the weight descending sort, get following data:

Reel number	Thickness	Volume is heavy	Width	Annealing curve	Contract number	External diameter	Delivery date	Degree of sorting	Priority
										8	1.01	40.599	1811	64	L761656G03	1992	20071130	4	0
6	1.01	34.629	1811	64	L761656G03	1852	20071130	4	0
										5	0.807	34.123	1525	64	L780834A01	1992	20071130	4	0
19	0.807	33.853	1525	64	L780834A01	1986	20071130	4	0

9	0.757	26.954	1292	64	L761564A01	1925	20071130	4	0
										18	0.757	26.614	1292	64	L761564A01	1915	20071130	4	2.4
15	0.706	24.955	1185	4	L780826D01	1938	20071130	2	7.846

Step2: first unlabelled coiled sheet is placed in the stove i.e. No. 8 coiled sheets in the selected coiled sheet sequence;

Step3: the next coiled sheet that is not labeled among the selected S, use matching criteria judgement between coiled sheet, if can form a stove with selecting coiled sheet, in the stove of then this coiled sheet being packed into, this coiled sheet of mark; If can not form a stove, select next unmarked coiled sheet to judge.Current planning coiled sheet number equals at 4 o'clock, repeats Step3; Otherwise, forward Step4 to; The initial set stove scheme that obtains being placed in the big stove of HH by this step is as follows:

Reel number	Thickness	Volume is heavy	Width	Annealing curve	Contract number	External diameter	Delivery date	Degree of sorting	Stove number
										8	1.01	40.599	1811	64	L761656G03	1992	20071130	4	1
5	0.807	34.123	1525	64	L780834A01	1992	20071130	4	1
										15	0.706	24.955	1185	4	L780826D01	1938	20071130	2	1

Step4: judge whether to also have idle stove, then forward Step2 to if having, otherwise forward Step5 to; Get under the initial feasible group of stove planning chart by above step:

Reel number	Thickness	Volume is heavy	Width	Annealing curve	Contract number	External diameter	Delivery date	Degree of sorting	Stove number
										8	1.01	40.599	1811	64	L761656G03	1992	20071130	4	1
5	0.807	34.123	1525	64	L780834A01	1992	20071130	4	1
										15	0.706	24.955	1185	4	L780826D01	1938	20071130	2	1
4	1.213	28.58	1262	2	L761656D02	2029	20071130	2	2
										1	1.264	28.11	1512	1	L750739B 11	1835	20071124	2	2
12	1.213	22.72	1262	2	L761656D02	1813	20071130	2	2
										17	0.909	23.47	1182	2	L734576B03	1898	20071130	2	3
16	0.807	22.93	1185	2	L771675A04	1889	20071130	2	3
										3	0.599	25.4	1551	11	N770465B02	1761	20071117	3	3

Step5: establish F={F ₁, F ₂, F ₃Make up for 3 shove charges that obtain, all do not arrange coiled sheet S={ }, F, S and set parameter are input to calculating current goal functional value Z=936.749 in bell furnace group sieve process prescription model.If current preferably separating is F _Best, corresponding target function value is Z _Best, other is F _Best=F, Z _Best=Z forwards Step6 to.

Step6: search has not been arranged coiled sheet and has been exchanged neighborhood (promptly arranged coiled sheet with do not arrange coiled sheet exchange one to one) in twos to F and S, and the record neighborhood is not preferably separated F ' in the taboo table, upgrade current group of furnace accretion and really gather F, residue coiled sheet S set and taboo table.If Z '＜Z _Best, F in addition then _Best=F '.Through searching for to such an extent that the plate reel number is can obtain not preferably separating in the taboo table, current group of stove objective function Z '=917.721 after 4 coiled sheet and plate reel number are 7 coiled sheet exchange.

Step7: if preferably separate F _BestBe updated or be not updated iterations less than setting value, then repeat Step6, otherwise, pair set F and S search is not arranged coiled sheet and has been arranged coiled sheet two and change a neighborhood and (promptly arranged coiled sheet and two and do not arrange coiled sheet and exchange for one, purpose is to significantly improve batch), in neighborhood, choose not and in the taboo table, preferably to separate F ', upgrade current group of furnace accretion and really gather F, do not arrange coiled sheet S set and taboo table.If Z '＜Z _Best, F in addition then _Best=F '.Herein, in continuous 5 generations, be not updated, and then search is not arranged coiled sheet and arranged coiled sheet two and change with neighborhood, arranged coiled sheet coiled sheet 3 and do not arranged and can access not preferably separating in the taboo table, current group of stove objective function Z '=897.802 after coiled sheet 11 and 16 exchanges with two.

Step8: if continuous some generations preferably separate be not updated, then to F and S degree of deepening exchanging chain neighborhood search, record searching to neighborhood preferably separate F ', upgrade current group of furnace accretion and really gather F and do not arrange coiled sheet, if F '＜F in conjunction with S _Best, F in addition then _Best=F '.In continuous herein 5 generations, be not updated, and then searches for degree of deepening exchanging chain neighborhood, and coiled sheet 12 accesses to improve with coiled sheet 7 exchange energies separates current group of stove objective function Z '=861.914.

Step9: judgement is to satisfy end condition, if satisfy, then terminator is exported F as a result _Best, otherwise, forward Step6 to, until satisfying end condition.

Obtaining final shove charge assembled scheme at last is:

Reel number	Thickness	Volume is heavy	Width	Annealing curve	Contract number	External diameter	Delivery date	Degree of sorting	Stove number
										5	0.807	34.123	1525	64	L780834A01	1992	20071130	4	1
19	0.807	33.853	1525	64	L780834A01	1986	20071130	4	1
										9	0.757	26.954	1292	64	L761564A01	1925	20071130	4	1
4	1.213	28.58	1262	2	L761656D02	2029	20071130	2	2
										1	1.264	28.11	1512	1	L750739B11	1835	20071124	2	2
12	1.213	22.72	1262	2	L761656D02	1813	20071130	2	2
										7	2.533	32.24	1511	1	LH10434A02	1960	20071205	3	3
16	2.522	32.21	1511	1	LH10432A07	1958	20071205	3	3
										2	2.522	32.2	1511	1	LH10432A07	1945	20071205	3	3

The 6th step: for the automatic group furnace accretion fruit (being the result that algorithm obtains) that generates of system, every penalty value that the user can compare and check objective function by Function of Evaluation and artificial group stove, if it is dissatisfied to the result, then the user can be in graphic editor makes amendment (comprising deletion, interpolation, replacement of coiled sheet etc.) to the result, till satisfaction; System all can carry out violation inspection to current result in the each process of revising of user.If satisfied to the result, uploading of then execution group stove scheme uploads to it in enterprise's ERP system.The group furnace accretion fruit that uploads to enterprise's ERP system is (with the heat (batch) number series arrangement, each heat (batch) number comprises plate reel number wherein) be to be stored on the assigned catalogue of this machine with the text form earlier equally, its filename with appointment is uploaded in the assigned catalogue of ftp server of enterprise's ERP system appointment by FTP order then, and then ftp server uploads to the result ERP system of enterprise thus.

The 7th step: the group stove scheme that enterprise's ERP system basis obtains is (with the heat (batch) number series arrangement, each heat (batch) number comprises plate reel number wherein) the standard group furnace gauge of the data formation enterprise ERP system of taking-up correspondence is drawn from relevant tables of data, is issued to the last group stove scheme of execution in the system that automatically performs of producing at last.

The evaluation of result data:

Average weight (t)	Average height (mm)	General assembly (TW) (t)	Working time (s)
				90.33	4303.667	12911	0.32

Example two: adopt the actual marshal data of steel mill's bell furnace unit as concrete objective for implementation, altogether 139 coiled sheets and 26 bell furnaces are carried out the shove charge combination in the present embodiment.

Operation steps is with embodiment one, and the parameter setting sees the following form:

Sequence number	Fee item	Symbol	Concrete parameter value	Remarks
						1	The coiled sheet number	N	139
2	The heat number	P	26	2 in the big stove of NH, one in the big stove of HH
					3	Heat size	T j	155 99	Big stove 155 ports 99
4	The residual capacity expense	p j	0.5/t+0.01/mm
					5	Do not arrange the coiled sheet expense	f	i	60
6	The maximum carrying height of stove	H j	4700

Other parameters settings and solution procedure are with embodiment one.Get following evaluation of result data:

Average weight (t)	Average height (mm)	General assembly (TW) (t)	Working time (s)
				93.51	4301	2431.26	65.21

Claims (4)

1. a cold rolling hood type annealing furnace set optimization furnace combination method is characterized in that satisfying under the production constraint condition, has set up optimization group stove model, optimizes furnace combination method and may further comprise the steps:

(1), the coiled sheet of waiting for annealing is carried out the priority setting according to the actual process operational factor;

(2) organize stove in advance, coiled sheet is arranged according to priority penalty value ascending order, the identical coiled sheet of priority penalty value is according to the weight descending sort, according to coiled sheet and stove matching criteria, the coiled sheet of priority minimum is put into stove, from the residue coiled sheet, choose the coiled sheet that to form a stove with existing coiled sheet in the stove successively according to matching criteria between coiled sheet then, put it in the stove, when highly reaching the maximum height of bell furnace, automatically form a group stove scheme, this step repeatedly forms the initial feasible group of stove scheme of bell type annealing furnace then; Above-described coiled sheet and stove matching criteria are meant: if stove can be annealed according to the annealing curve of a certain coiled sheet, and the external diameter of this coiled sheet judges then that less than the external diameter of stove this coiled sheet and stove are complementary; Matching criteria is meant between described coiled sheet: if the annealing curve of two coiled sheets is identical or shape is similar, two coiled sheet thickness all belong to thin material of same type or thick material, and the surface smoothness of coiled sheet is identical, and then two coiled sheets can be placed in the same stove;

(3) initial feasible group of stove scheme adjusted, according to the objective function in the optimization group stove model, the application enhancements strategy is adjusted initial feasible group of stove scheme; Wherein, described group of stove model is described as:

Constraint condition

Z _j≥0，j＝1，...，P (5)

X _ij∈{0，1}，i＝1，...，N；j＝1，...，P (6)

In the objective function (1)

The coupling expense of expression coiled sheet and bell furnace is used for guaranteeing the quality of match of coiled sheet and bell furnace;

The rejection penalty of expression bell furnace residual capacity summation is used for guaranteeing the utilization factor of bell furnace;

Expression is used for guaranteeing that to the rejection penalty of selected coiled sheet not total batch can reach maximum horizontal;

Expression be coupling expense between the coiled sheet, be used for guaranteeing to be placed in the quality of the coiled sheet coupling in the same stove;

The constraint condition of model:

Coiled sheet of constraint (2) expression can only be arranged at most in the stove; Constraint (3) expression can not surpass the maximum load capacity of stove with the coiled sheet general assembly (TW) of j buttress; The overall height of the coiled sheet of j buttress of constraint (4) expression can not surpass the maximum height of stove; The residual capacity of constraint (5) expression heat can not be for negative, the span of constraint (6) expression decision variable;

Decision variable

Model parameter

1) Z _jThe residual capacity an of-Di j heat;

2) N-compiles the coiled sheet number of heat;

3) P-heat number, P=N _NHB+ N _NHS+ N _HHB+ N _HHS

4) T _jThe heat size an of-Di j heat;

5) p _jThe additional expense coefficient of-Di j heat residual capacity;

6) g _iThe weight an of-Di i coiled sheet;

7) h _iThe height an of-Di i coiled sheet;

8) f _iThe not selected caused additional expense coefficient an of-Di i coiled sheet;

9) H _jThe maximum carrying height an of-Di j stove;

10) C1 _IjThe coupling expense of-coiled sheet i and bell furnace j;

11) C2 _Ik-be placed in coiled sheet i in the same stove and the coupling expense of k;

(4) the group stove scheme that is optimized is carried out last resulting optimization group stove scheme by the production automatic control system of bell furnace unit.

2. according to the described a kind of cold rolling hood type annealing furnace set optimization furnace combination method of claim 1, the priority that it is characterized in that described coiled sheet, preferred term comprises: the delivery date of coiled sheet, the collection of coiled sheet criticize the contract integrality of singularity, coiled sheet of production, coiled sheet and coiled sheet in the storehouse time, priority is quantified as priority penalty value P _i, priority is high more, then priority penalty value P _iMore little, use P respectively _Ij ¹, P _Ij ², P _Ij ³, P _Ij ⁴And P _Ij ⁵The pairing penalty value of each preferred term of expression coiled sheet, and with it according to ascending series arrangement and form the preferred term sequence, use P _i(t) penalty value of t preferred term in the preferred term sequence of i coiled sheet of expression, t is smaller or equal to the number of total preferred term, discounting factor beta and adjustment factor-alpha, at this moment

P _i＝f(5)；

f(t)＝f(t-1)β+αP _i(t)；

f(1)＝P _i(1)。

3. according to the described a kind of cold rolling hood type annealing furnace set optimization furnace combination method of claim 1, it is characterized in that initial feasible group of stove scheme adjusted described in the step (3), be that structure improves neighborhood, cooperate change neighborhood, degree of deepening search strategy that initial feasible group of stove scheme improved, four kinds of search neighborhoods are respectively: arranged coiled sheet and exchanged neighborhood in twos, two coiled sheets that are about to distribute to different stoves exchange; Row has not arranged coiled sheet two and has changed a neighborhood, is about to one and has arranged coiled sheet and shift out from affiliated stove, and do not arrange coiled sheet with two and put into; Row has not arranged coiled sheet and has exchanged neighborhood one to one, is about to one and has arranged coiled sheet and shift out from affiliated stove, and put into one and do not arrange coiled sheet; Degree of deepening mixes the exchanging chain neighborhood, all are not arranged coiled sheet packs in the virtual stove according to the shove charge rule, and in each actual furnaces, add a virtual coiled sheet, so just, all coiled sheets are all packed in the stove, carry out swap operation on this basis, two coiled sheets that are about to belong to different stoves exchange one to one, after the several times exchange, just formed one and mixed exchanging chain, when the exchange number of times was uncertain, this neighborhood was called degree of deepening and mixes the exchanging chain neighborhood.

4. according to the described a kind of cold rolling hood type annealing furnace set optimization furnace combination method of claim 3, it is characterized in that: described when initial feasible group of stove scheme adjusted, search procedure is divided into three phases, at first adopt described not row to arrange coiled sheet two and change a neighborhood search, adopting all neighborhoods to move maximum the separating as current of middle objective function improvement separates, next adopting row to arrange coiled sheet exchanges neighborhood one to one and continues the direction search that increases to batch, same employing objective function improves maximum separating as current and separates, if these two kinds of neighborhoods all can't improve target, employing has been arranged coiled sheet and has been exchanged neighborhood in twos so, current separate to choose mode the same, the result does not still improve after continuous search repeatedly, and then search degree of deepening mixes the exchanging chain neighborhood.

Images