CN106636610A - Time-and-furnace-length-based double-dimensional stepping type heating curve optimizing setting method of heating furnace - Google Patents

Abstract

The invention discloses a time-and-furnace-length-based double-dimensional stepping type heating curve optimizing setting method of a heating furnace. According to the method, according to the heating technique requirement of a plurality of steel blank sequences inside the furnace and the steel blank state, the temperature set value inside the furnace is subjected to optimization of the two dimensions including the time and the furnace length; compared with traditional optimization of the single set value in the length direction of a heating furnace based on a single steel blank, the different heating requirements and the front-back rank of a plurality of to-be-heated steel blanks are considered, and the time sequence value of each furnace temperature set point is further optimized, namely, through dynamic optimization of the furnace temperature setting time sequence, accurate control over differentiation heating technique requirements of the front and back steel blanks is achieved, the furnace temperature dynamic adjusting ability of the heating furnace is sufficiently used, differentiation heating of the steel blanks is achieved, and the effects that the heating quality of the steel blanks is improved, energy consumption and the oxidation burning loss are lowered, and the service life of the heating furnace is prolonged are achieved.

Description

Time and furnace length based two-dimensional stepping heating furnace temperature rise curve optimal setting method

Technical Field

The invention relates to the technical field of heating furnace optimization control in steel production, in particular to a method for optimally setting a heating curve of a two-dimensional stepping heating furnace based on time and furnace length.

Background

At present, with the aggravation of market competition, the steel heating and rolling process is changed from the traditional large-batch production according to a plan to a small-batch flexible production mode of multiple steel types, and the traditional heating furnace control mode based on the large-batch production mode cannot adapt to flexible and changeable flexible manufacturing production requirements.

The billet heating furnace is production equipment for heating a primary rolled billet or a continuous cast billet to a certain temperature distribution so as to facilitate rolling by a roughing mill, the steel industry is a big household of energy consumption, wherein the energy consumption of the heating furnace only accounts for 25% of the total energy consumption of the steel industry, the heating efficiency of the heating furnace is improved, the energy consumption is reduced, and the billet heating furnace has important significance for energy conservation of the whole steel industry, and especially, along with the development of modern rolling mills to continuous, large, high-speed, high-precision and multi-variety directions, higher and higher requirements are provided for the heating quality of the billet. However, the heating furnace is a typical complex industrial object, including thermodynamic, chemical and physical processes, which is essentially a complex industrial production object with typical characteristics of multivariable, time-varying, non-linear, strong coupling, large inertia and pure hysteresis.

The purpose of the heating furnace control is to set the furnace temperature of the heating furnace section according to the rolling rhythm of the rolling mill, so that the steel billet is fully heated in the furnace, the tapping temperature and the soaking temperature of the steel billet meet the rolling requirement when the steel billet is tapped, and the consumed fuel is required to be as small as possible. The temperature of the steel billet in the heating furnace is usually controlled by controlling the temperature of the section of the heating furnace at present. In the past, under the era background that the steel productivity can not meet the consumption requirements of China for a long time, steel metallurgy enterprises generally adopt a heating furnace output type high-firing method, and fire steel according to the upper limit of the heating temperature of a billet on the premise of sufficient fuel so as to realize multi-firing and quick-firing, ensure the matching of the production capacity of a heating furnace and the production capacity of a rolling mill as much as possible, and improve the steel output. Therefore, the oxidation burning loss of the steel billet is serious, huge energy consumption waste of the heating furnace is caused, and the service life of the heating furnace is shortened. In recent years, with the turning of the supply and demand relationship of steel productivity in China, the optimization control problem of the heating furnace becomes a hot point problem which is commonly concerned by the industrial and academic industries at home and abroad due to the requirements on reducing the production cost, improving the production efficiency and improving the product quality.

For the temperature rise process of the steel billet, each steel billet uniquely corresponds to an optimal temperature rise curve in theory, and the best heating performance of the steel billet can be ensured and the least heat quantity of the steel billet is consumed only if the temperature of the steel billet rises along the temperature rise curve. The heating furnace is used for heating the steel billet, the temperature rise process of the steel billet is provided by the furnace temperature of the heating furnace, so that the furnace temperature of each section of the heating furnace has corresponding furnace temperature distribution in order to give corresponding temperature distribution to the steel billet. It can be said that the optimal temperature rise curve of the steel billet corresponds to the optimal furnace temperature distribution curve of the heating furnace, so that the optimal furnace temperature distribution curve of the heating furnace needs to be found, and the optimal furnace temperature distribution can be realized through the combustion control system of the heating furnace by the optimal furnace temperature distribution curve. The purpose of optimally controlling the heating furnace is to find the optimal furnace temperature value within the allowable furnace temperature range of each furnace section, that is, the optimal furnace temperature set value, so as to heat the steel billet meeting the requirements with the least energy consumption.

The optimal furnace temperature setting is a typical optimal decision problem, and according to the known working conditions of the specification, the type, the target tapping temperature, the charging temperature, the rolling rhythm and the like of the steel billet, under the production process constraints of a heating furnace process model, the upper limit of the temperature difference of the section of the steel billet, the upper limit of the temperature difference of the section when the steel is discharged, the temperature interval of the surface of the steel when the steel is discharged, the tapping time constraint of the steel billet and the like, the furnace temperature of each section is set, so that the steel billet is heated to the proper temperature at the proper time, and the energy consumption is minimum.

For the problem of furnace temperature setting of the heating furnace, related patents and documents in China also provide some solutions, such as:

in patent publication CN103697712A, "dynamic heating furnace temperature control method based on time sensitivity", the shortest furnace time is dynamically set according to the discharge rhythm, the heating capacity of the heating furnace and the maximum value of the walking beam movement information, and the temperature sensitivity value of the steel billet furnace is calculated according to the shortest furnace time.

In the patent publication CN103146905A, "a method for deciding furnace temperature of heating furnace based on optimized heating curve of steel blank", based on the heat transfer mechanism in the heating furnace, the furnace temperature corresponding to the current position of a steel blank is corrected by virtual heating based on the optimized heating curve of steel blank, and then the furnace temperatures corresponding to all steel blanks in each control section of the heating furnace are weighted and averaged, so as to calculate and obtain the decision furnace temperature of each control section of the heating furnace.

In the CN105018718A patent, "a heating furnace process furnace temperature control method based on heat load distribution", based on the existing "material temperature model", the device status and the product process are organically combined, and by establishing a heat load balance adjustment model, the product temperature deviation is distributed to the control section where the typical product is located, and the process furnace temperature of the heating furnace combustion is compensated and corrected, so as to improve the control accuracy of the process furnace temperature, improve the heating quality of the product, and reduce the energy waste.

In the CN104894362A patent, "a method for setting furnace temperature of heating furnace for mixed loading of cold and hot steel billets", a mean value of entering temperature of all steel billets in a certain heating section and a tapping temperature calculation module are calculated to obtain a tapping temperature value, and the furnace temperature setting of the heating section is compensated together, so as to optimize the furnace temperature setting of each heating section to stabilize the tapping temperature during mixed loading of cold and hot steel billets.

In the CN100507027C patent, "method for dynamically setting and controlling furnace temperature of hot rolling heating furnace", a billet temperature prediction model is used to forward recurrently calculate the temperature of the end of the section of the billet, dynamically calculate the target temperature of the end of each furnace section of the billet according to the moving distance of the billet, calculate the required furnace temperature of the current section of the billet, and perform expert experience weighting setting by considering the difference of all billets in the current section.

In the patent CN102433428B entitled "method for controlling furnace temperature during heating process of heating furnace billet", an adaptive differential evolution algorithm is used to determine an optimal control scheme according to the heat transfer characteristics in the heating furnace based on the constraints of the heating quality requirement of the billet and the safety of production equipment, so as to obtain the temperature setting of the heating furnace and realize the temperature control during the heating process of the billet.

Patent No. CN103952529B, "a furnace temperature optimization method based on thermal balance for a walking beam furnace", establishes a relational expression between furnace temperature, steel temperature, heat supply and heat loss by using the relationship of heat balance in the furnace, and calculates the optimal furnace temperature of each section along the furnace length direction according to the furnace type, billet specification, type, target tapping temperature, charging temperature, rolling rhythm and other working conditions, so that the billet can be heated to an appropriate temperature and an allowable temperature difference of the cross section within a specified time with minimum energy consumption.

In addition, Kirenjie et al, in the furnace temperature decision method in mathematical model control of continuous heating furnace (steel, 1995, 30 (1): 67-71), Yun jun et al, in the optimization setting model and software development of step heating furnace (control and decision, 1997,12 (4): 369 + 372), Zhang guardian et al, in the multi-objective fuzzy optimization method of furnace temperature system of heating furnace (Chinese non-ferrous metals academic newspaper, 1998, 8 (2): 715 + 717), also proposed the optimization setting method of furnace temperature.

The above document proposes a furnace temperature optimization setting model based on heating experience of a heating furnace engineer and an operator according to actual conditions of a production site, and reserves a mechanism model based on a billet prediction model for use when conditions are met. And the furnace temperature optimization setting model determines the optimized furnace temperature setting value of each section by inquiring the database and searching the knowledge base according to the field working condition signals acquired in real time. The discharged steel billets heated under the furnace temperature system can meet the quality requirement of rolled steel, and the energy consumption can be reduced to a certain extent, thereby achieving the purpose of the computer optimization control of the heating furnace.

In the method and the document disclosed above, different methods are mainly used for researching the furnace temperature optimization problem of the heating furnace from different sides, and a certain effect is obtained, however, in the optimization process of the method, the heating process of the steel billet in the furnace is regarded as a slow temperature rise process in constant temperature fields of different heating furnace sections, and the furnace temperature is regarded as a means for changing the temperature of the steel billet to optimize the set value, so that the optimization is based on the steady-state working condition. In fact, the working condition of the heating furnace often changes in various ways during the production process of the heating furnace. Variations in the type, gauge, entry temperature and rolling pace of the billet can cause the temperature profile of the billet to deviate from the ideal heating profile. Strictly speaking, the steady-state working condition does not exist in the actual production, the steady-state heating furnace temperature optimization method can only ensure the minimum heating quality of the steel billet and the minimum energy consumption of the heating furnace under approximate conditions and average meanings, meanwhile, the change of the furnace temperature is a relatively slow process, the requirements of the rolling and heating processes of the steel billet in the same furnace section are often different, and the method cannot effectively utilize the change processes of the furnace temperatures in different sections to carry out differential heating on the steel billet, namely the heating control capability of heating equipment is not fully utilized to meet the requirements of the differential rolling processes of the steel billet. Therefore, the furnace temperature change process, the dynamic process of the heating furnace heating in the furnace and the requirements of the billet rolling process must be considered comprehensively, and the furnace temperature setting of the heating furnace when the working condition of the heating furnace changes is optimized under the condition that various heating process constraints of the heating furnace are met, so that the heating quality and the energy consumption of the billet can be guaranteed to be minimized under various production working conditions of the heating furnace, and the full-furnace dynamic optimization of the furnace temperature of the heating furnace is really realized.

The realization of the full-furnace dynamic optimization of the furnace temperature of the heating furnace mainly needs to solve the following technical difficulties:

(1) the basis of the dynamic real-time optimal decision of the furnace temperature system is the requirement of the billet yield and the billet temperature distribution of the whole furnace, but the temperature distribution of the billet in the furnace cannot be continuously and comprehensively measured technically in the actual production at present, and an accurate mathematical model of the billet in the heating furnace needs to be established to realize the accurate prediction of the temperature rise process of the billet in the heating furnace.

(2) The change process of the furnace temperature, the dynamic process of the heating furnace in the furnace and the requirement of the rolling process of the steel billet are comprehensively considered, the single-point temperature optimization of the traditional heating furnace is changed into the curve optimization problem of the temperature changing along with the time, namely the traditional single-dimensional curve optimization along the furnace length direction is changed into the matrix optimization problem along the furnace length and the time, and the solving difficulty is greatly increased. The full-furnace dynamic optimization of the furnace temperature of the heating furnace aims to provide a curve of the furnace temperature set value of each section of the heating furnace changing along with time, the curve is a nonlinear dynamic optimization model, and the maximum value principle in optimal control and a conventional mathematical programming method are difficult to provide a whole curve of the furnace temperature change of each section of the heating furnace in the switching process through one-time optimization calculation.

Disclosure of Invention

In view of the defects of low heating quality of steel billets, high energy consumption, high oxidation burning loss and the like caused by the fact that the prior method for optimally setting the furnace temperature of the heating furnace in the steel industry mostly adopts steady-state working conditions to optimize, fails to meet the rolling process requirements of the steel billets with different specifications and different purposes, provides a method for optimally setting a heating curve of a two-dimensional stepping heating furnace based on time and furnace length, optimizes the set value of the temperature in the furnace simultaneously in two dimensions of time and furnace length according to the heating process requirements (target tapping temperature and section temperature difference limit) and the state of the steel billets (specification, variety and charging temperature of the billet), compared with the traditional method for optimizing a single set value in the length direction of the heating furnace based on a single steel billet, the method provided by the invention considers the different heating requirements and the front-back sequencing of a plurality of steel billet sequences to be heated, and optimizes the time sequence value of each set point of the furnace, the method realizes the accurate control of the differential heating process requirements of the front and the rear steel billets through the dynamic optimization of the time sequence set by the furnace temperature, fully utilizes the dynamic furnace temperature adjusting capability of the heating furnace, realizes the differential heating of the steel billets, and achieves the effects of improving the heating quality of the steel billets, reducing the energy consumption, reducing the oxidation burning loss and prolonging the service life of the heating furnace.

In order to achieve the above purpose, the technical solution for solving the technical problem is as follows:

a method for optimally setting a temperature rise curve of a two-dimensional walking beam furnace based on time and furnace length mainly comprises an off-line analysis step and an on-line control step, wherein:

the off-line analysis step mainly collects parameters related to furnace temperature control, and establishes a billet heating model based on the parameters;

the online control step is carried out with a predetermined control period delta t_cAnd triggering to realize the real-time optimal setting of the temperature rise curve of the stepping heating furnace.

Further, the parameters related to furnace temperature control in the offline analysis step include:

collecting all steel types produced by a heating furnace, and marking as Gra;

for each steel grade g ∈ Gra, the thermophysical parameters including the density rho of the steel grade at each temperature are obtained by looking up a table_gThermal capacity C_gAnd thermal conductance K_g；

The characteristic parameters of the heating furnace comprise the number N of furnace zones of the heating furnace and the length L of each furnace zone_nN1, 2, N, effective heat transfer coefficient from furnace gas in each furnace zone to billetAnd heat loss coefficient η of watermark in each furnace zone_n,n＝1,2,...,N。

Further, the off-line analysis step establishes a discrete billet heating model capable of predicting the heating condition of the billet in the heating furnace, the model only considers the temperature gradient in the billet thickness direction, and approximately considers that the temperature of the billet layer at the same thickness position is equal everywhere, therefore, the temperature distribution of the billet becomes a function related to the thickness position x and the time t, and the following heat transfer equation set is established:

wherein, the first equation in the above formula represents the heat conduction process in the billet; and the second equation represents the heat transfer process of the heating furnace to the lower surface of the steel billet, wherein T^fThe furnace temperature value of the furnace zone in which the steel billet is positioned, h_bThe equivalent heat transfer coefficient of the lower surface is related to the characteristics of the heating furnace, the watermark parameter, the furnace temperature and the billet temperature; the third equation represents the heat transfer process of the heating furnace to the upper surface of the steel billet, wherein h_tIs the upper surface equivalent heat transfer coefficient, the parameter is related to the heating furnace characteristic, the furnace temperature and the billet temperature; the fourth equation represents the initial temperature value of the steel billet;

performing position discretization treatment on the equation set, wherein the steel billet is decomposed into m parts along the x direction to obtain a steel billet temperature distribution vector T_s(k) As shown in the following formula:

wherein, T₁(k) Corresponding to the temperature, T, of the upper surface block of the steel billet_m(k) The temperature of the surface block under the steel billet;

at the same time, the time discretization is carried out, assuming that the current time is k, for a given billet s, the steel type is g, the length is len, the width is wdt, the thickness is thk, and the current billet temperature distribution is T_s(k) A billet moving speed v and a current position p of the billet in the heating furnace_s(k) And furnace temperature values of the furnace zonesBased on the information, the temperature distribution T of the steel billet at the k +1 moment can be calculated and obtained through the model_s(k +1) by Mod_sRepresenting the relation of the model, and establishing the following expression:

wherein, because the steel grade, length, width, thickness, thermophysical parameters of the steel billet s, the length of each furnace zone of the heating furnace, the effective heat transfer coefficient and the heat loss coefficient of the watermark are constants, the variables in the formula (1) are reserved, which can be abbreviated as:

further, assuming that the current control period is k and the set of all billet numbers in the heating furnace is Slab, the online control step includes the following specific steps:

step S1: starting a control period, collecting data and updating a temperature distribution calculation value and an average moving speed;

step S2: correcting the temperature distribution calculation value of the steel billet at the outlet of each furnace area;

step S3: judging whether a new billet enters the heating furnace or not, and if so, calculating the optimal temperature rise curve;

step S4: judging whether the optimal temperature rise curves of all the steel billets need to be updated or not, and if so, updating;

step S5: solving a multi-objective optimization problem by adopting an extreme value optimization algorithm to obtain the optimal set value of the furnace temperature of each furnace area;

step S6: and (5) transmitting the optimal set value of the furnace temperature of each furnace zone obtained by the calculation of the S5 to a PLC (programmable logic controller), and finishing the calculation of the control period.

Further, step S1 specifically includes:

step S11: collecting the current moving speed v (k) of the billet in the heating furnace, and updating the positions of all the billets in the heating furnace, namely for all the billetsp_s(k)＝p_s(k-1)+v(k)·Δt_cFurther judging the furnace area n of the heating furnace where the billet s is;

step S12: according to the furnace temperature value of the furnace zoneTemperature distribution T of steel billet s calculated at last moment_s(k-1), the current billet moving speed in the heating furnace v (k), and the billet heating model Mod_sAnd calculating the temperature distribution T of the steel billet s at the current moment_s(k)；

Step S13: meanwhile, according to the moving speed values v (k), v (k-1) and v (k-2) … v (k-q) of the latest control cycles, wherein q is a preset value, the average moving speed of the steel billet in the furnace in the latest period of time is calculated

Further, step S2 specifically includes:

step S21: sequentially checking whether a steel billet passes through the lower part of each temperature detection device;

step S22: supposing that a steel billet s just passes through a certain temperature detection device and the detected surface temperature value is T_s ^SF(k) Then use T_s ^SF(k) Calculated value T for temperature distribution_s(k) And (6) correcting.

Further, in step S22, the temperature values at different positions of the billet are corrected as follows:

temperature T of upper surface of steel billet_m(k)＝T_s ^SF(k)；

Internal and lower surface temperature T of steel billet_i(k)＝T_i(k)+(T_s ^SF(k)-T_m(k))，i＝1,2,..,m-1。

Further, step S3 specifically includes:

step S31: judging whether a new billet enters the heating furnace, if so, adding the billet number s into the Slab, and placing p_s(k)＝0，T_s(k)＝T₀；

Step S32: according to average moving speedTarget tapping temperatureHeating model Mod for billet_sThe optimization problem is solved by combining the factors of the tapping temperature deviation, the section temperature deviation and the maximum temperature rise amplitude, and the optimal temperature rise curve hc of the billet is obtained by calculation_sThe temperature curve specifies the target temperature of the billet at each time in the heating furnace, and the average moving speed corresponding to the temperature rise curve of the billet is recorded

Further, step S4 specifically includes:

step S41: for all billets contained in the Slab set, i.e.Judge one by one whether to satisfyΔ v is a predetermined threshold value;

step S42: if it is satisfied withThe difference between the actual moving speed of the billet and the average moving speed when the optimal temperature rise curve is generated is large, and the working condition is changed greatly, so that the optimal temperature rise curve needs to be recalculated.

Further, step S42 specifically includes:

according to the current average moving speedCurrent billet position p_s(k) The calculated value T of the current temperature distribution of the steel billet_s(k) Target tapping temperatureHeating model Mod for billet_sRecalculating to obtain the optimal temperature rise curve of the billet by combining the factors of the tapping temperature deviation, the section temperature deviation and the maximum temperature rise amplitude, and updating hc_sAnd simultaneously updating the average moving speed corresponding to the temperature rise curve of the billet

Further, in step S5, the following multi-objective optimization problem is first established:

setting an optimization target: firstly, considering an arbitrary billet s in a heating furnace, in a given optimization time domain P, the calculated temperature-rising curve is close to the optimal temperature-rising curve as much as possible, namely the minimum deviation is realized, namely:

wherein, T_s ^ARepresents the calculated average temperature of the steel slab, w_κThe method comprises the steps that the weighting coefficient of each period deviation in a time domain is controlled, all steel billets in a heating furnace are considered, an optimization target shown in a formula (3) needs to be established for each steel billet, and a multi-objective optimization problem is formed together;

contract decision variables: furnace temperature set value of each furnace zonen＝1,2,...,N；

Determining a constraint condition: the relationship between the furnace temperature set value of each furnace zone and the temperature-rise curve of the billet steel calculation can be obtained according to the formula (2):

For s∈Slab，κ＝0,1,...,P-1,n＝1,2,...,N

wherein p is_s(k + k) is the position to which the billet moves at the current average speed at the moment of k + k, and is calculated by adopting the following formula:

For s∈Slab，κ＝0,1,...,P-1

further, if it is necessary to obtain a billet average temperature by averaging the billet temperature distributions of the respective periods in the control time domain and the average calculation function is expressed by a symbol, then:

For s∈Slab，κ＝1,2,...,P

and executing an extreme value optimization algorithm to obtain the optimal set value of the furnace temperature of each furnace zone.

Further, an extremum optimization algorithm is executed, and the specific steps are as follows:

step A: initializing, defining a solution warehouse with a fixed size, making the solution warehouse empty, and setting the iteration number iter to be 0;

and B: randomly generating an initial solution

And C: for each variable of the current solutionN sequentially performing mutation, namely when a variable is mutated, keeping the other components unchanged to obtain N neighborhood solutions;

step D: carrying out dominant sorting on N neighborhood solutions so as to obtain a dominant sorting number r of the N neighborhood solutions_j∈[0,N-1]，j∈{1,...,N}；

Step E: let each variable beOf the fitness value lambda_jEqual to the dominant rank of its corresponding neighborhood solution, i.e., λ_j＝r_j，j∈{1,...,N}；

Step F: if the fitness of only one variable is 0, the variable is considered as the worst variable in the current solution; if more than two variables with the fitness of 0 exist, a diversity maintenance mechanism is used to determine which is the worst variable in the current solution, and the worst variable is assumed to beIts fitness lambda_wBy variation of 0Producing a new solution as S_w，w∈{1,...,N}；

Step G: unconditionally using S_wBy replacing the current individual S, i.e. S_w→S；

Step H: judging whether S is added into a solution warehouse or not;

step I: if the iteration times reach the preset termination iteration times, entering a step J, otherwise, making iter equal to iter +1, and entering a step C;

step J: and selecting any one solution stored in the solution warehouse as the solution output of the multi-objective optimization problem.

Further, in step H, Pareto dominant comparison is performed on S and all solutions in the solution warehouse:

if the solution dominated by S exists in the solution warehouse, deleting the solution and adding S into the solution warehouse;

if at least one solution dominance S exists in the solution warehouse, not updating the solution warehouse;

if the S and any existing solution in the solution warehouse are not mutually independent, updating the solution warehouse according to the following three conditions;

a) if the solution warehouse is not full, adding S into the solution warehouse;

b) if the solution warehouse is full and S is located in the most dense place in the archive, then S is not added to the solution warehouse;

c) if the solution warehouse is full and S is not located at the most dense place in the solution warehouse, S is used to replace the non-inferior solutions located at the most dense place in the solution warehouse.

Due to the adoption of the technical scheme, compared with the prior art, the invention has the following advantages and positive effects:

1. the invention optimizes the set value of the temperature in the furnace in two dimensions of time and furnace length, sets an optimized time domain P in the aspect of time dimension, aims to reduce the deviation between the calculated heating curve and the optimal heating curve of each steel billet in the period of time as much as possible, establishes respective optimized targets aiming at different characteristics of each steel billet in the heating furnace in the aspect of furnace length dimension, and jointly forms a multi-target optimization problem. By solving the optimization problem of the two dimensions of the time and the furnace length, the furnace temperature dynamic optimization control is realized, and the requirements of the billet differential heating process in the furnace are considered;

2. the invention adopts a multi-objective optimization method based on an extreme value optimization principle, the method has multi-constraint condition processing capacity, multi-objective optimization capacity and rapid optimization searching capacity in a complex optimization space, and meets the real-time requirement of online furnace temperature control while generating a better furnace temperature setting solution;

3. according to the method, the optimal temperature rising curve is generated when the steel billets enter the furnace according to the characteristics of the steel billets and is used as a target curve tracked in the heating process of the steel billets in the furnace, and the optimal temperature rising curve is updated only when the moving speed of the steel billets is changed greatly, so that the huge calculation load caused by the fact that the optimal temperature rising curve needs to be calculated in each control period in the traditional method is avoided;

4. in the invention, the temperature distribution model calculation value of the steel billet is corrected at the outlet position of each furnace zone of the heating furnace, the mechanism can eliminate the deviation between the accumulated deviation and the real system due to multi-step model calculation in the furnace zone, and the accurate initial temperature distribution state of the steel billet entering the next furnace zone is obtained, thereby improving the optimization control precision.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. It is obvious that the drawings in the following description are only some embodiments of the invention, and that for a person skilled in the art, other drawings can be derived from them without inventive effort. In the drawings:

FIG. 1 is a coordinate axis system of a one-dimensional temperature model of a billet in the present invention;

FIG. 2 is a typical optimal temperature ramp curve for the present invention;

FIG. 3 is a flow chart of each control cycle in the present invention;

FIG. 4 is a graph showing the variation of the target tapping temperature of six continuous steel billets in accordance with the present invention;

FIG. 5 is a temperature rise curve of six billets in accordance with the present invention;

FIG. 6 is a furnace temperature setting curve of the soaking zone of the heating furnace according to the invention.

Detailed Description

While the embodiments of the present invention will be described and illustrated in detail with reference to the accompanying drawings, it is to be understood that the invention is not limited to the specific embodiments disclosed, but is intended to cover various modifications, equivalents, and alternatives falling within the scope of the invention as defined by the appended claims.

With the development of industrial scale, at present, a continuous heating furnace represented by a stepping heating furnace is widely applied to rolling production, and different from a traditional periodic heating furnace, the continuous heating furnace adopts a continuous feeding and continuous discharging mode, so that the production continuity between the heating furnace and a rolling mill is improved, and the production efficiency is greatly improved. In actual production, a heating furnace yield type high-firing method is generally adopted for setting and controlling the temperature of the heating furnace, namely, steel is fired according to the upper limit of the heating temperature of a billet on the premise of sufficient fuel, so that multi-firing quick firing is realized, the matching of the production capacity of the heating furnace and the production capacity of a rolling mill is ensured as much as possible, and the yield of steel is improved. However, with the increasingly fierce international competition of steel production, the traditional continuous large-batch production gradually changes to a flexible small-batch and multi-variety manufacturing mode according to the operation mode of planned production, the requirements of different users and application occasions are different, the requirements of the processing and rolling processes of the steel billets are different, and the traditional furnace temperature control method adopting a fixed temperature rise setting curve not only causes serious oxidation burning loss of the steel billets, but also causes huge energy consumption waste of the heating furnace, and shortens the service life of the heating furnace. Therefore, the production of the heating furnace is urgently required to be capable of carrying out real-time dynamic optimization of furnace temperature setting of the heating furnace according to the rolling process requirement of processing the billet steel, and lean control of the production process of the heating furnace is realized.

The invention aims at the problems and discloses a method for optimally setting a temperature-rising curve of a two-dimensional walking beam furnace based on time and furnace length, which mainly comprises an off-line analysis step and an on-line control step, wherein:

the off-line analysis step mainly collects parameters related to furnace temperature control, and establishes a billet heating model based on the parameters;

wherein the parameters related to furnace temperature control in the offline analysis step include:

collecting all steel types produced by a heating furnace, and marking as Gra;

for each steel grade g ∈ Gra, the thermophysical parameters including the density rho of the steel grade at each temperature are obtained by looking up a table_gThermal capacity C_gAnd thermal conductance K_gEtc.;

the characteristic parameters of the heating furnace comprise the number N of furnace zones of the heating furnace and the length L of each furnace zone_nN1, 2, N, effective heat transfer coefficient from furnace gas in each furnace zone to billetN1, 2, N and heat loss coefficient of each furnace zone watermark η_nN is 1,2,.., N, etc.

Further, the off-line analysis step establishes a discrete billet heating model which can predict the heating condition of the billet in the heating furnace, and the invention adopts a billet one-dimensional temperature model, as shown in fig. 1, which only considers the temperature gradient in the billet thickness direction, but approximately considers that the temperature of the billet layer at the same thickness position is equal everywhere, so that the temperature distribution of the billet becomes a function related to the thickness position x and the time t, and establishes the following heat transfer equation set:

wherein the first equation in the above equation represents the heat inside the steel slabConducting; and the second equation represents the heat transfer process of the heating furnace to the lower surface of the steel billet, wherein T^fThe furnace temperature value of the furnace zone in which the steel billet is positioned, h_bThe equivalent heat transfer coefficient of the lower surface is related to parameters such as heating furnace characteristics, watermark parameters, furnace temperature, billet temperature and the like; the third equation represents the heat transfer process of the heating furnace to the upper surface of the steel billet, wherein h_tThe equivalent heat transfer coefficient of the upper surface is related to parameters such as the characteristic of the heating furnace, the temperature of the steel billet and the like; the fourth equation represents the initial temperature value of the steel billet;

performing position discretization treatment on the equation set, wherein the steel billet is decomposed into m parts along the x direction to obtain a steel billet temperature distribution vector T_s(k) As shown in the following formula:

wherein, T₁(k) Corresponding to the temperature, T, of the upper surface block of the steel billet_m(k) The temperature of the surface block under the steel billet;

at the same time, the time discretization is carried out, assuming that the current time is k, for a given billet s, the steel type is g, the length is len, the width is wdt, the thickness is thk, and the current billet temperature distribution is T_s(k) A billet moving speed v and a current position p of the billet in the heating furnace_s(k) And furnace temperature values of the furnace zonesN is 1,2, the billet temperature distribution T at the moment k +1 can be calculated and obtained through the model based on the information_s(k +1) by Mod_sRepresenting the relation of the model, and establishing the following expression:

wherein, because the steel grade, length, width, thickness, thermophysical parameters of the steel billet s, the length of each furnace zone of the heating furnace, the effective heat transfer coefficient and the heat loss coefficient of the watermark are constants, the variables in the formula (1) are reserved, which can be abbreviated as:

T_s(k+1)＝Mod_s(T_s(k),p_s(k),v,T_n ^f) (2)

the online control step is carried out with a predetermined control period delta t_cAnd triggering to realize the real-time optimal setting of the temperature rise curve of the stepping heating furnace. In this embodiment, the online control module mainly performs the following four functions:

(1) the method comprises the following steps of (1) steel billet position tracking, namely calculating the positions of all steel billets in a heating furnace in real time and detecting whether steel billets enter or leave the heating furnace;

(2) calculating and updating the optimal temperature-rising curve of the steel billet, namely calculating the optimal temperature-rising curve of each steel billet when the steel billet enters a heating furnace, and recalculating the optimal temperature-rising curve under the condition of large change of working conditions (large change of moving speed of the steel billet);

(3) correcting the temperature of the steel billet, namely correcting the model calculation temperature value of the steel billet by utilizing the steel billet temperature detection information at the outlet of each furnace area;

(4) and optimizing the furnace temperature set value of each furnace zone, and generating the furnace temperature set value of each furnace zone by taking the aim of minimizing the deviation between the actual temperature rise process and the optimal temperature rise curve of all the steel billets in the control time domain.

As shown in fig. 3, further, assuming that the current control period is k and the set of all billet numbers in the heating furnace is Slab, the online control step includes the following specific steps:

step S1: starting a control period, collecting data and updating a temperature distribution calculation value and an average moving speed;

step S2: correcting the temperature distribution calculation value of the steel billet at the outlet of each furnace area;

step S3: judging whether a new billet enters the heating furnace or not, and if so, calculating the optimal temperature rise curve;

step S4: judging whether the optimal temperature rise curves of all the steel billets need to be updated or not, and if so, updating;

step S5: solving a multi-objective optimization problem by adopting an extreme value optimization algorithm to obtain the optimal set value of the furnace temperature of each furnace area;

step S6: and (4) issuing the optimal set value of the furnace temperature of each furnace zone calculated in the step (S5) to a Programmable Logic Controller (PLC), and finishing the calculation of the control period.

In an embodiment, step S1 further includes:

step S11: collecting the current moving speed v (k) of the billet in the heating furnace, and updating the positions of all the billets in the heating furnace, namely for all the billetsp_s(k)＝p_s(k-1)+v(k)·Δt_cFurther judging the furnace area n of the heating furnace where the billet s is;

step S12: according to the furnace temperature value of the furnace zoneTemperature distribution T of steel billet s calculated at last moment_s(k-1), the current billet moving speed in the heating furnace v (k), and the billet heating model Mod_sAnd calculating the temperature distribution T of the steel billet s at the current moment_s(k)；

Step S13: meanwhile, according to the moving speed values v (k), v (k-1) and v (k-2) … v (k-q) of the latest control cycles, wherein q is a preset value, the average moving speed of the steel billet in the furnace in the latest period of time is calculated

In an embodiment, step S2 further includes:

step S21: sequentially checking whether a steel billet passes through the lower part of each temperature detection device;

step S22: assuming that a steel billet s just passes through a certain temperature detection device and the detected surface temperature value isThen utilizeCalculated value T for temperature distribution_s(k) And (6) correcting.

Further, in step S22, the temperature values at different positions of the billet are corrected as follows:

temperature of upper surface of steel billet

Internal and lower surface temperature of steel billet

In an embodiment, step S3 further includes:

step S31: judging whether a new billet enters the heating furnace, if so, adding the billet number s into the Slab, and placing p_s(k)＝0，T_s(k)＝T₀；

Step S32: according to average moving speedTarget tapping temperatureHeating model Mod for billet_sThe optimization problem is solved by combining the factors of the tapping temperature deviation, the section temperature deviation and the maximum temperature rise amplitude, and the optimal temperature rise curve hc of the billet is obtained by calculation_sA typical optimum temperature rise curve is shown in FIG. 2, which determines a target temperature of a billet at each time in a heating furnace, and an average moving speed of the billet corresponding to the temperature rise curve is recorded

In an embodiment, step S4 further includes:

step S41: for all billets contained in the Slab set, i.e.Judge one by one whether to satisfyΔ v is a predetermined threshold value;

step S42: if it is satisfied withThe difference between the actual moving speed of the billet and the average moving speed when the optimal temperature rise curve is generated is large, and the working condition is changed greatly, so that the optimal temperature rise curve needs to be recalculated.

Further, step S42 specifically includes:

according to the current average moving speedCurrent billet position p_s(k) The calculated value T of the current temperature distribution of the steel billet_s(k) Target tapping temperatureHeating model Mod for billet_sRecalculating to obtain the optimal temperature rise curve of the billet by combining the factors of the tapping temperature deviation, the section temperature deviation and the maximum temperature rise amplitude, and updating hc_sAnd simultaneously updating the average moving speed corresponding to the temperature rise curve of the billet

In the specific embodiment, in step S5, the following multi-objective optimization problem is first established:

setting an optimization target: firstly, considering an arbitrary billet s in a heating furnace, in a given optimization time domain P, the calculated temperature-rising curve is close to the optimal temperature-rising curve as much as possible, namely the minimum deviation is realized, namely:

wherein,represents the calculated average temperature of the steel slab, w_κThe method comprises the steps that the weighting coefficient of each period deviation in a time domain is controlled, all steel billets in a heating furnace are considered, an optimization target shown in a formula (3) needs to be established for each steel billet, and a multi-objective optimization problem is formed together;

contract decision variables: furnace temperature set value of each furnace zonen＝1,2,...,N；

Determining a constraint condition: the relationship between the furnace temperature set value of each furnace zone and the temperature-rise curve of the billet steel calculation can be obtained according to the formula (2):

For s∈Slab，κ＝0,1,...,P-1,n＝1,2,...,N

wherein p is_s(k + k) is the position to which the billet moves at the current average speed at the moment of k + k, and is calculated by adopting the following formula:

For s∈Slab，κ＝0,1,...,P-1

further, if it is necessary to obtain a billet average temperature by averaging the billet temperature distributions of the respective periods in the control time domain and the average calculation function is expressed by a symbol, then:

For s∈Slab，κ＝1,2,...,P

and executing an extreme value optimization algorithm to obtain the optimal set value of the furnace temperature of each furnace zone.

Further, an extremum optimization algorithm is executed, and the specific steps are as follows:

step A: initializing, defining a solution warehouse with a fixed size, making the solution warehouse empty, and setting the iteration number iter to be 0;

and B: randomly generating an initial solution

And C: for each variable of the current solution1,2, nyPerforming mutation again, namely keeping the rest components unchanged when a variable is mutated to obtain N neighborhood solutions;

step D: carrying out dominant sorting on N neighborhood solutions so as to obtain a dominant sorting number r of the N neighborhood solutions_j∈[0,N-1]，j∈{1,...,N}；

Step E: let each variable beOf the fitness value lambda_jEqual to the dominant rank of its corresponding neighborhood solution, i.e., λ_j＝r_j，j∈{1,...,N}；

Step F: if the fitness of only one variable is 0, the variable is considered as the worst variable in the current solution; if more than two variables with the fitness of 0 exist, a diversity maintenance mechanism is used to determine which is the worst variable in the current solution, and the worst variable is assumed to beIts fitness lambda_wBy variation of 0Producing a new solution as S_w，w∈{1,...,N}；

Step G: unconditionally using S_wBy replacing the current individual S, i.e. S_w→S；

Step H: judging whether S is added into a solution warehouse or not;

further, in step H, Pareto dominant comparison is performed on S and all solutions in the solution warehouse:

if the solution dominated by S exists in the solution warehouse, deleting the solution and adding S into the solution warehouse;

if at least one solution dominance S exists in the solution warehouse, not updating the solution warehouse;

if the S and any existing solution in the solution warehouse are not mutually independent, updating the solution warehouse according to the following three conditions;

a) if the solution warehouse is not full, adding S into the solution warehouse;

b) if the solution warehouse is full and S is located in the most dense place in the archive, then S is not added to the solution warehouse;

c) if the solution warehouse is full and S is not located at the most dense place in the solution warehouse, S is used to replace the non-inferior solutions located at the most dense place in the solution warehouse.

Step I: if the iteration times reach the preset termination iteration times, entering a step J, otherwise, making iter equal to iter +1, and entering a step C;

step J: and selecting any one solution stored in the solution warehouse as the solution output of the multi-objective optimization problem.

Because the method provided by the invention optimizes the temperature set value in the furnace in two dimensions of time and length, and simultaneously, the method takes the temperature deviation of all steel billets as an optimization target, the number of the optimized targets and optimized variables is greatly increased, the traditional deterministic method is difficult to obtain a solution which can meet the constraint in limited time, and is also difficult to accept or reject the advantages and disadvantages among a plurality of targets, and the method adopts a multi-target extremum optimization algorithm to solve, thereby effectively solving the problems.

The following takes the practical production process as an example to illustrate the beneficial effects obtained by the present invention:

the method for optimally setting the temperature rise curve of the stepping heating furnace based on the double coordinate axes of time and furnace length is adopted to perform simulation operation on a certain heating furnace, the stepping heating furnace is provided with four furnace areas of a preheating section, a first heating section, a second heating section and a soaking section, the temperature of the steel billet entering the furnace is 0 ℃, and the temperature of the steel billet leaving target is distributed between 1050 ℃ and 1150 ℃ according to different process requirements of various steel billet steel types. Counting 20 steel billets produced by the heating furnace in a period of time, and comparing the temperature control effect of the steel billets after the original setting method and the optimized setting method, wherein the results are shown in the following table:

	average steel billet tapping temperature deviation	Mean cross-sectional temperature difference
			Original setting method	18.23℃	6.47℃
Optimization setting method	7.54℃	5.79℃

Wherein, for any billet s at the moment k, the temperature difference of the cross section of the billet sThe difference between the temperature of the upper surface and the lower surface of the steel billet and the temperature of the center thereof is defined as follows:

this index reflects the degree of uniformity of heating inside the billet. Simulation results show that compared with the original setting method, the optimized setting method can greatly reduce the temperature difference of the discharged billet and the temperature difference of the cross section.

The following description will be made of an exemplary continuous production process as an example to show the advantageous effects of the present invention:

as shown in fig. 4, the heating furnace continuously heats six steel billets, the target tapping temperatures of the steel coils 1 and 2 are 1050 ℃, the target tapping temperatures of the steel coils 3 and 4 are 1110 ℃, and the target tapping temperatures of the steel coils 5 and 6 are 1080 ℃. The heating and temperature rising effects of the steel billets under the control of the original setting method and the method of the invention are shown in figure 5, and it can be seen that under the control of the original setting method, the steel billets 1,2, 5 and 6 are over-burnt to a certain extent, and the steel billets 3 and 4 are under-heated, because the target tapping temperatures of the steel billets are greatly different, the traditional one-dimensional curve optimization setting method along the furnace length direction cannot realize the rapid and dynamic adjustment of the furnace temperature setting according to the difference of the steel billets. Under the control of the setting method, the outlet temperatures of the six steel billets can reach the vicinity of the target tapping temperature. According to specific principle analysis, the furnace temperature setting change curve of the soaking section of the heating furnace is shown in figure 6, as the target tapping temperature of the steel billets 3 and 4 reaches 1110 ℃, the temperature is obviously improved compared with 1050 ℃ of the steel billets 1 and 2, the method can realize rapid dynamic adjustment of the furnace temperature setting value in advance according to the target tapping temperature change of the steel billets in the future, the dynamic adjusting capability of the heating furnace is fully utilized, and compared with the traditional method for optimally setting the single-dimensional curve along the furnace length direction, the method is lagged and slow, so that the method achieves better control effect.

The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present invention are included in the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims (13)

1. A method for optimally setting a temperature rise curve of a two-dimensional walking beam furnace based on time and furnace length is characterized by mainly comprising an off-line analysis step and an on-line control step, wherein:

the off-line analysis step mainly collects parameters related to furnace temperature control, and establishes a billet heating model based on the parameters;

the online control step is carried out with a predetermined control period delta t_cAnd triggering to realize the real-time optimal setting of the temperature rise curve of the stepping heating furnace.

2. The method for optimally setting the temperature-rise curve of the two-dimensional walking beam furnace based on time and furnace length as claimed in claim 1, wherein the parameters related to the furnace temperature control in the offline analysis step comprise:

collecting all steel types produced by a heating furnace, and marking as Gra;

for each steel grade g ∈ Gra, the thermophysical parameters including the density rho of the steel grade at each temperature are obtained by looking up a table_gThermal capacity C_gAnd thermal conductance K_g；

The characteristic parameters of the heating furnace comprise the number N of furnace zones of the heating furnace and the length L of each furnace zone_nN1, 2, N, effective heat transfer coefficient from furnace gas in each furnace zone to billetAnd heat loss coefficient η of watermark in each furnace zone_n,n＝1,2,...,N。

3. The method of claim 2, wherein the step of off-line analysis is performed to create a discrete slab heating model capable of predicting the heating condition of the slab in the furnace, the model only considers the temperature gradient in the slab thickness direction and approximately considers the slab layer temperature at the same thickness position to be equal everywhere, so that the slab temperature distribution becomes a function of the thickness position x and the time t, and the following set of heat transfer equations is created:

$\left\{\begin{array}{c}\frac{\∂T(x,t)}{\∂t}=\frac{1}{C_g\left(T\right){\mathrm{\ρ}}_g\left(T\right)}\frac{\∂}{\∂x}\left(K_g\left(T\right)\frac{\∂T(x,t)}{\∂x}\right)\\ \frac{\∂T(x,t)}{\∂t}{|}_{x=0}=\frac{-h_b}{K_g\left(T\right)}(T^f-T(0,t\left)\right)\\ \frac{\∂T(x,t)}{\∂t}{|}_{x=d}=\frac{-h_t}{K_g\left(T\right)}(T^f-T(d,t\left)\right)\\ T(x,0)=T_0\end{array}\right.$

wherein, the first equation in the above formula represents the heat conduction process in the billet; and the second equation represents the heat transfer process of the heating furnace to the lower surface of the steel billet, wherein T^fThe furnace temperature value of the furnace zone in which the steel billet is positioned, h_bThe equivalent heat transfer coefficient of the lower surface is related to the characteristics of the heating furnace, the watermark parameter, the furnace temperature and the billet temperature; the third equation represents the heat transfer process of the heating furnace to the upper surface of the steel billet, wherein h_tIs the upper surface equivalent heat transfer coefficient, the parameter is related to the heating furnace characteristic, the furnace temperature and the billet temperature; the fourth equation represents the initial temperature value of the steel billet；

Performing position discretization treatment on the equation set, wherein the steel billet is decomposed into m parts along the x direction to obtain a steel billet temperature distribution vector T_s(k) As shown in the following formula:

$T_s\left(k\right)=\left[\begin{array}{c}{T}_1\left(k\right)\\ T_2\left(k\right)\\ ...\\ T_m\left(k\right)\end{array}\right]$

wherein, T₁(k) Corresponding to the temperature, T, of the upper surface block of the steel billet_m(k) The temperature of the surface block under the steel billet;

at the same time, the time discretization is carried out, assuming that the current time is k, for a given billet s, the steel type is g, the length is len, the width is wdt, the thickness is thk, and the current billet temperature distribution is T_s(k) A billet moving speed v and a current position p of the billet in the heating furnace_s(k) And furnace temperature values of the furnace zonesBased on the information, the temperature distribution T of the steel billet at the k +1 moment can be calculated and obtained through the model_s(k +1) by Mod_sRepresenting the model relationship, the following table is establishedThe expression is as follows:

$T_s(k+1)={\mathrm{Mod}}_s\left(T_s\right(k),p_s(k),v,{T_n}^f,g,len,wdt,thk,{\mathrm{\ρ}}_g,C_g,K_g,L_n,{K_n}^{eq},{\mathrm{\η}}_n),n=1,2,...,N---\left(1\right)$

wherein, because the steel grade, length, width, thickness, thermophysical parameters of the steel billet s, the length of each furnace zone of the heating furnace, the effective heat transfer coefficient and the heat loss coefficient of the watermark are constants, the variables in the formula (1) are reserved, which can be abbreviated as:

T_s(k+1)＝Mod_s(T_s(k),p_s(k),v,T_n ^f) (2)

4. the method for optimally setting the temperature-rising curve of the two-dimensional walking beam furnace based on time and furnace length as claimed in claim 3, wherein assuming that the current control period is k and the set of all billet numbers in the furnace is Slab, the online control step comprises the following specific steps:

step S1: starting a control period, collecting data and updating a temperature distribution calculation value and an average moving speed;

step S2: correcting the temperature distribution calculation value of the steel billet at the outlet of each furnace area;

step S3: judging whether a new billet enters the heating furnace or not, and if so, calculating the optimal temperature rise curve;

step S4: judging whether the optimal temperature rise curves of all the steel billets need to be updated or not, and if so, updating;

step S5: solving a multi-objective optimization problem by adopting an extreme value optimization algorithm to obtain the optimal set value of the furnace temperature of each furnace area;

step S6: and (5) transmitting the optimal set value of the furnace temperature of each furnace zone obtained by the calculation of the S5 to a PLC (programmable logic controller), and finishing the calculation of the control period.

5. The method for optimally setting the temperature-rise curve of the two-dimensional walking beam furnace based on time and furnace length as claimed in claim 4, wherein the step S1 specifically comprises:

step S11: collecting the current moving speed v (k) of the billet in the heating furnace, and updating the positions of all the billets in the heating furnace, namely for all the billetsp_s(k)＝p_s(k-1)+v(k)·Δt_cFurther judging the furnace area n of the heating furnace where the billet s is;

step S12: according to the furnace temperature value of the furnace zoneTemperature distribution T of steel billet s calculated at last moment_s(k-1), the current billet moving speed in the heating furnace v (k), and the billet heating model Mod_sAnd calculating the temperature distribution T of the steel billet s at the current moment_s(k)；

Step S13: meanwhile, according to the moving speed values v (k), v (k-1) and v (k-2) … v (k-q) of the latest control cycles, wherein q is a preset value, the average moving speed of the steel billet in the furnace in the latest period of time is calculated

6. The method for optimally setting the temperature-rise curve of the two-dimensional walking beam furnace based on time and furnace length as claimed in claim 4, wherein the step S2 specifically comprises:

step S21: sequentially checking whether a steel billet passes through the lower part of each temperature detection device;

step S22: supposing that a steel billet s just passes through a certain temperature detection device and the detected surface temperature value is T_s ^SF(k) Then use T_s ^SF(k) Calculated value T for temperature distribution_s(k) And (6) correcting.

7. The method of claim 6, wherein the temperature values of different positions of the billet in step S22 are corrected as follows:

temperature T of upper surface of steel billet_m(k)＝T_s ^SF(k)；

Internal and lower surface temperature T of steel billet_i(k)＝T_i(k)+(T_s ^SF(k)-T_m(k))，i＝1,2,..,m-1。

8. The method for optimally setting the temperature-rise curve of the two-dimensional walking beam furnace based on time and furnace length as claimed in claim 4, wherein the step S3 specifically comprises:

step S31: judging whether a new billet enters the heating furnace, if so, adding the billet number s into the Slab, and placing p_s(k)＝0，T_s(k)＝T₀；

Step S32: according to average moving speedTarget tapping temperature T_s ^THeating model Mod for billet_sThe optimization problem is solved by combining the factors of the tapping temperature deviation, the section temperature deviation and the maximum temperature rise amplitude, and the optimal temperature rise curve hc of the billet is obtained by calculation_sThe temperature curve specifies the target temperature of the billet at each time in the heating furnace, and the average moving speed corresponding to the temperature rise curve of the billet is recorded

9. The method for optimally setting the temperature-rise curve of the two-dimensional walking beam furnace based on time and furnace length as claimed in claim 4, wherein the step S4 specifically comprises:

step S41: for all billets contained in the Slab set, i.e.Judge one by one whether to satisfyΔ v is a predetermined threshold value;

step S42: if it is satisfied withThe difference between the actual moving speed of the billet and the average moving speed when the optimal temperature rise curve is generated is large, and the working condition is changed greatly, so that the optimal temperature rise curve needs to be recalculated.

10. The method for optimally setting the temperature-rise curve of the two-dimensional walking beam furnace based on time and furnace length as claimed in claim 9, wherein the step S42 specifically comprises:

according to the current average moving speedCurrent billet position p_s(k) The calculated value T of the current temperature distribution of the steel billet_s(k) Target tapping temperature T_s ^THeating model Mod for billet_sRecalculating to obtain the optimal temperature rise curve of the billet by combining the factors of the tapping temperature deviation, the section temperature deviation and the maximum temperature rise amplitude, and updating hc_sAnd simultaneously updating the average moving speed corresponding to the temperature rise curve of the billet

11. The method for optimally setting the temperature-rise curve of the two-dimensional walking beam furnace based on time and furnace length as claimed in claim 4, wherein in step S5, the following multi-objective optimization problem is firstly established:

setting an optimization target: firstly, considering an arbitrary billet s in a heating furnace, in a given optimization time domain P, the calculated temperature-rising curve is close to the optimal temperature-rising curve as much as possible, namely the minimum deviation is realized, namely:

$\min \underset{\mathrm{\κ}=1}{\overset{P}{\Σ}}{w}_{\mathrm{\κ}}\left|\right|T_s^A(k+\mathrm{\κ})-{\mathrm{hc}}_s(k+\mathrm{\κ})\left|\right|---\left(3\right)$

wherein, T_s ^ARepresents the calculated average temperature of the steel slab, w_κThe method comprises the steps that the weighting coefficient of each period deviation in a time domain is controlled, all steel billets in a heating furnace are considered, an optimization target shown in a formula (3) needs to be established for each steel billet, and a multi-objective optimization problem is formed together;

contract decision variables: furnace temperature set value of each furnace zone

Determining a constraint condition: the relationship between the furnace temperature set value of each furnace zone and the temperature-rise curve of the billet steel calculation can be obtained according to the formula (2):

For s∈Slab，κ＝0,1,...,P-1,n＝1,2,...,N

$T_s(k+\mathrm{\κ}+1)={\mathrm{Mod}}_s\left(T_s\right(k+\mathrm{\κ}),p_s(k+\mathrm{\κ}),{\stackrel{\‾}{v}}_s,T_n^{sp})$

wherein p is_s(k + k) is the position to which the billet moves at the current average speed at the moment of k + k, and is calculated by adopting the following formula:

For s∈Slab，κ＝0,1,...,P-1

$p_s(k+\mathrm{\κ}+1)=p_s(k+\mathrm{\κ})+{\stackrel{\‾}{v}}_s{\mathrm{\Δt}}_c$

further, if it is necessary to obtain a billet average temperature by averaging the billet temperature distributions of the respective periods in the control time domain and the average calculation function is expressed by a symbol, then:

For s∈Slab，κ＝1,2,...,P

T_s ^A(k+κ)＝(T_s(k+κ))。

and executing an extreme value optimization algorithm to obtain the optimal set value of the furnace temperature of each furnace zone.

12. The method for optimally setting the heating curve of the two-dimensional walking beam furnace based on the time and the furnace length as claimed in claim 11, wherein an extremum optimization algorithm is executed, and the method comprises the following specific steps:

step A: initializing, defining a solution warehouse with a fixed size, making the solution warehouse empty, and setting the iteration number iter to be 0;

and B: randomly generating an initial solution

And C: for each variable of the current solutionCarrying out mutation in sequence, namely keeping the rest components unchanged when a variable is mutated to obtain N neighborhood solutions;

step D: carrying out dominant sorting on N neighborhood solutions so as to obtain a dominant sorting number r of the N neighborhood solutions_j∈[0,N-1]，j∈{1,...,N}；

Step E: let each variable beOf the fitness value lambda_jEqual to the dominant rank of its corresponding neighborhood solution, i.e., λ_j＝r_j，j∈{1,...,N}；

Step F: if the fitness of only one variable is 0, the variable is considered as the worst variable in the current solution; if more than two variables with the fitness of 0 exist, a diversity maintenance mechanism is used to determine which is the worst variable in the current solution, and the worst variable is assumed to beIts fitness lambda_wBy variation of 0Producing a new solution as S_w，w∈{1,...,N}；

Step G: unconditionally using S_wReplacing the current individual S, i.e.S_w→S；

Step H: judging whether S is added into a solution warehouse or not;

step I: if the iteration times reach the preset termination iteration times, entering a step J, otherwise, making iter equal to iter +1, and entering a step C;

step J: and selecting any one solution stored in the solution warehouse as the solution output of the multi-objective optimization problem.

13. The method for optimally setting the heating curve of the two-dimensional walking beam furnace based on the time and the furnace length as claimed in claim 12, wherein the Pareto dominant comparison of S with all solutions in the solution warehouse is performed in step H:

if the solution dominated by S exists in the solution warehouse, deleting the solution and adding S into the solution warehouse;

if at least one solution dominance S exists in the solution warehouse, not updating the solution warehouse;

if the S and any existing solution in the solution warehouse are not mutually independent, updating the solution warehouse according to the following three conditions;

a) if the solution warehouse is not full, adding S into the solution warehouse;

b) if the solution warehouse is full and S is located in the most dense place in the archive, then S is not added to the solution warehouse;

c) if the solution warehouse is full and S is not located at the most dense place in the solution warehouse, S is used to replace the non-inferior solutions located at the most dense place in the solution warehouse.