CN108694288B - Method for rapidly acquiring set temperatures of walking beam type billet heating furnace under different yields - Google Patents
Method for rapidly acquiring set temperatures of walking beam type billet heating furnace under different yields Download PDFInfo
- Publication number
- CN108694288B CN108694288B CN201810528045.4A CN201810528045A CN108694288B CN 108694288 B CN108694288 B CN 108694288B CN 201810528045 A CN201810528045 A CN 201810528045A CN 108694288 B CN108694288 B CN 108694288B
- Authority
- CN
- China
- Prior art keywords
- neural network
- billet
- value
- heating furnace
- walking beam
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
- 238000010438 heat treatment Methods 0.000 title claims abstract description 64
- 238000000034 method Methods 0.000 title claims abstract description 23
- 229910000831 Steel Inorganic materials 0.000 claims abstract description 37
- 239000010959 steel Substances 0.000 claims abstract description 37
- 238000012549 training Methods 0.000 claims abstract description 30
- 238000013528 artificial neural network Methods 0.000 claims abstract description 27
- 238000003062 neural network model Methods 0.000 claims abstract description 18
- 210000002569 neuron Anatomy 0.000 claims description 15
- 238000004519 manufacturing process Methods 0.000 claims description 5
- 238000010079 rubber tapping Methods 0.000 claims description 5
- 238000010845 search algorithm Methods 0.000 claims description 4
- 238000005265 energy consumption Methods 0.000 abstract description 7
- 238000013461 design Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000004134 energy conservation Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 239000012530 fluid Substances 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/08—Thermal analysis or thermal optimisation
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Data Mining & Analysis (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Computational Linguistics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Control Of Heat Treatment Processes (AREA)
- Heat Treatments In General, Especially Conveying And Cooling (AREA)
Abstract
The invention discloses a method for quickly acquiring set temperatures of a walking beam type billet heating furnace under different yields, which comprises the following steps of: acquiring a plurality of groups of training data of the walking beam type billet heating furnace under different yields, wherein the training data comprises input data and output data; the input data comprises a stepping period, billet heating precision and the hot uniformity of discharged billets; the output data comprises the set temperature of each control area; b, training a BP neural network to obtain i BP neural network models; step C, determining a stepping period corresponding to the target yield, and simultaneously adjusting the hot uniformity degree value of the discharged steel billet to ensure that the heating precision value of the steel billet is in the range of-5 ℃; and D, taking the stepping period, the hot uniformity degree value of the discharged steel billet and the steel billet heating precision value in the step C as input data of the BP neural network model to obtain the set temperature value of each control area corresponding to the target yield. The invention has simple operation process, high heating precision, high energy utilization rate and low energy consumption.
Description
Technical Field
The invention belongs to the technical field of walking beam type steel billet heating furnaces, and particularly relates to a method for quickly acquiring set temperatures of a walking beam type steel billet heating furnace under different yields.
Background
In the prior art, a method for quickly obtaining the set temperatures of the walking beam type billet heating furnace under different production rates does not exist, so the set temperatures of the walking beam type billet heating furnace under different production rates are set according to experience, and therefore, the heating precision is low, the energy utilization rate is low, and the energy consumption is large. In order to meet the urgent needs of the steel industry on the background of energy conservation and consumption reduction, a method for quickly obtaining the set temperature of a walking beam type billet heating furnace under different yields is urgently needed, so that the heating precision and the energy utilization rate are improved, and the energy consumption is reduced.
Disclosure of Invention
The invention aims to provide a method for rapidly acquiring the set temperatures of a walking beam type billet heating furnace under different yields aiming at the defects of the prior art, and the method has the advantages of high heating precision, high energy utilization rate and low energy consumption.
In order to solve the technical problems, the technical scheme adopted by the invention is as follows:
a method for rapidly acquiring set temperatures of a walking beam type billet heating furnace under different yields is characterized by comprising the following steps:
step A, obtaining a plurality of groups of training data of the walking beam type billet heating furnace under different production rates, wherein the training data comprisesInput data and output data; the input data comprises a stepping period WR and billet heating precision delta TdisAnd the thermal uniformity degree delta T of the discharged steel billetmax(ii) a The output data includes set temperatures T of the control regions corresponding to the sets of input dataspi;
B, training the BP neural network by using the training data in the step A to obtain i BP neural network models, wherein i is the number of the control areas;
step C, determining a stepping period corresponding to the target yield, and simultaneously adjusting the input value of the neural network model obtained after training, wherein the adjustment of the input value of the neural network model obtained after training comprises the adjustment of the thermal uniformity degree value of the discharged steel billet, so that the heating precision value of the steel billet is in the range of-5 ℃; and D, taking the stepping period, the hot uniformity degree value of the discharged steel billet and the steel billet heating precision value in the step C as input data of the i BP neural network models in the step B to obtain the set temperature value of each control area corresponding to the target yield.
By the method, the heating process of the heating furnace is simulated by taking the heating process of the steel billet in the walking beam type steel billet heating furnace as a research object and adopting a Computational Fluid Dynamics (CFD) method. The method is equivalent to a 'virtual' experiment of the walking beam type billet heating furnace in a computer, and has the advantages of low cost, short period and wide environment. The CFD technology is introduced into the design of the heating furnace, so that design errors and cost required by experimental research can be reduced, distribution of air flow, temperature and the like in the furnace is predicted, indexes such as billet heating accuracy and the like are optimized and calculated through a neural network algorithm, and the aims of high heating accuracy, high energy utilization rate, high efficiency, low energy consumption and low emission are fulfilled.
Further, in the step A, the billet is heated with the billet heating precision delta TdisAnd the thermal uniformity degree delta T of the discharged steel billetmaxAnd solving a plurality of groups of training data of the walking beam type billet heating furnace under different yields by using a Hooke-Jeeves direct search algorithm.
In order to prepare enough training data for the BP neural network, the invention solves the optimal set temperature data of a series of heating furnaces with different yields by using a Hooke-Jeeves direct search algorithm. Based on the data, the BP neural network is established and trained, and the method has the advantages of fast calculation, low operation cost and high accuracy.
As a preferable mode, in the step B, the BP neural network is a three-layer BP neural network; the input layer of the BP neural network has 3 neurons, and the 3 neurons of the input layer respectively represent a stepping period WR and a billet heating precision delta TdisAnd the thermal uniformity degree delta T of the discharged steel billetmax(ii) a The hidden layer of the BP neural network has 8 neurons; the output layer of the BP neural network is provided with 1 neuron, and the neuron of the output layer correspondingly represents the set temperature T of a control areaspi(ii) a The i BP neural network models obtained by training are as follows:
compared with the prior art, the invention has the advantages of simple operation process, high heating precision, high energy utilization rate and low energy consumption.
Drawings
FIG. 1 is a schematic view of a walking beam type billet heating furnace.
FIG. 2 shows 10 training data sets for a walking beam billet furnace at different production rates.
FIG. 3 is a model diagram of a BP neural network.
FIG. 4 is a graph comparing the predicted value and the actual value of the BP neural network. Wherein, in FIG. 4, (1) is Tsp1The predicted value and the actual value of (2) are compared, and T is shown in FIG. 4sp2The predicted value and the actual value of (2) are compared, and T is shown in (3) of FIG. 4sp3The predicted value and the actual value of (2) are compared, and T is shown in (4) of FIG. 4sp4Comparing the predicted value with the actual value; the abscissa is the step period (in units of s) and the ordinate is the set temperature (in units of ℃); the curve with the triangular points represents the actual values and the curve with the square points represents the predicted values.
FIG. 5 is a graph of Δ T set for different ranges of step periodsmaxThe value is obtained.
FIG. 6 is a comparison graph of the tapping temperature of steel billets in different step periods.
Detailed Description
The method for rapidly acquiring the set temperatures of the walking beam type billet heating furnace shown in figure 1 under different yields by utilizing the method comprises the following steps:
a, acquiring a plurality of groups of accurate training data of the walking beam type billet heating furnace under different yields, wherein the training data comprises input data and output data; the input data comprises a stepping period WR and billet heating precision delta TdisAnd the thermal uniformity degree delta T of the discharged steel billetmax(ii) a The output data includes set temperatures T of the control regions corresponding to the sets of input dataspi. In the present embodiment, the number of control regions is 4, and thus TspiIncluding the set temperature T of the control area 1sp1 Control region 2 set temperature Tsp2And the set temperature T of the control area 3sp3The set temperature T of the control area 4sp4. The division of the different control areas is shown in fig. 1.
And B, training the BP neural network by using the training data in the step A to obtain 4 BP neural network models.
Step C, determining a stepping period corresponding to the target yield, and simultaneously adjusting the input value of the neural network model obtained after training, wherein the adjustment of the input value of the neural network model obtained after training comprises the key adjustment of the thermal uniformity degree value of the discharged steel billet, so that the heating precision value of the steel billet is in the range of-5 ℃; and D, taking the stepping period, the hot uniformity degree value of the discharged steel billet and the steel billet heating precision value in the step C as input data of the 4 BP neural network models in the step B to obtain the set temperature value of each control area corresponding to the target yield.
In the step A, the billet is heated with the precision delta TdisAnd the thermal uniformity degree delta T of the discharged steel billetmaxAnd solving a plurality of groups of training data of the walking beam type billet heating furnace under different yields by using a Hooke-Jeeves direct search algorithm.
Specifically, the yield varies from 65ton/hr to 130ton/hr, starting from the minimum yield, and every 1.75ton/hr increase in yield is calculated by the Hooke-Jeeves direct search algorithmThe set temperature at the corresponding yield is obtained. Since the productivity of the heating furnace increases in inverse proportion to the step period of the heating furnace, the main program of the heating furnace changes the productivity of the heating furnace by changing the step period of the heating furnace. The variation range of the corresponding heating furnace stepping period is 180s to 360s, and the set temperature under the corresponding stepping period is calculated every 5 seconds of the stepping period. Therefore, there are 37 sets of set temperature data at different step periods, including Δ T at the timedisAnd Δ Tmax. Of these, 10 sets of training data are shown in fig. 2.
The method selects a typical three-layer BP neural network to predict the set temperature of the heating furnace. In the step B, the BP neural network is a three-layer BP neural network; as shown in fig. 3, the input layer of the BP neural network has 3 neurons, and the 3 neurons of the input layer represent the step period WR and the billet heating accuracy (deviation between the billet tapping temperature and the set temperature) Δ TdisAnd the degree of thermal uniformity (maximum temperature difference on the discharged steel billet) of the discharged steel billetmax(ii) a The hidden layer of the BP neural network has 8 neurons; the output layer of the BP neural network is provided with 1 neuron, and the neuron of the output layer correspondingly represents the set temperature T of a control areaspi. Therefore, the invention constructs 4 BP neural network models with the same structure to respectively predict Tsp1、Tsp2、Tsp3、Tsp4。
The 4 BP neural network models obtained by training are:
fig. 4 is a comparison graph of the predicted values and the actual values of the trained neural networks, and it can be seen from fig. 4 that the predicted values of the four trained neural networks well match the actual values.
In the step C, the value range of WR is 180-360 s, and delta T is used for ensuring the best tapping temperature of the tapped billetdisIs 0. In the formula,. DELTA.TmaxThe value of (D) is determined by the calculated Delta T of the heating furnace programdisBy manually setting Δ T in different WR rangesmaxValue of Δ T calculated by the programdisThe range is. + -. 5 ℃. Δ T in different WR rangesmaxThe values are shown in figure 5.
In order to verify the reasonability of the invention, the steel billet tapping temperature under different stepping periods is simulated and compared based on the actual working data of the heating furnace, and the simulation result is analyzed. As the furnace step cycle changes, the set temperature at the new step cycle can be calculated in a matter of hundreds of milliseconds. As can be seen from fig. 6, comparing the furnace installation temperature data obtained by the method of the present invention with the actual industrial data, the method of the present invention can greatly improve the heating accuracy of the furnace.
While the present invention has been described with reference to the embodiments shown in the drawings, the present invention is not limited to the embodiments, which are illustrative and not restrictive, and it will be apparent to those skilled in the art that various changes and modifications can be made therein without departing from the spirit and scope of the invention as defined in the appended claims.
Claims (3)
1. A method for rapidly acquiring set temperatures of a walking beam type billet heating furnace under different yields is characterized by comprising the following steps:
a, acquiring a plurality of groups of training data of the walking beam type billet heating furnace under different yields, wherein the training data comprises input data and output data; the input data comprises a stepping period WR and billet heating precision Delta TdisDelta T of degree of thermal uniformity of discharged steel billetsmax(ii) a The output data includes set temperatures T of the control regions corresponding to the sets of input dataspi(ii) a The heating precision delta T of the steel billetdisThe deviation of the tapping temperature of the steel billet and the set temperature is the delta T of the thermal uniformity degree of the tapped steel billetmaxThe maximum temperature difference on the discharged steel billet is obtained;
b, training the BP neural network by using the training data in the step A to obtain i BP neural network models, wherein i is the number of the control areas;
step C, determining a stepping period corresponding to the target yield, and simultaneously adjusting the input value of the neural network model obtained after training, wherein the adjustment of the input value of the neural network model obtained after training comprises the adjustment of the thermal uniformity degree value of the discharged steel billet, so that the heating precision value of the steel billet is in the range of-5 ℃; and D, taking the stepping period, the hot uniformity degree value of the discharged steel billet and the steel billet heating precision value in the step C as input data of the i BP neural network models in the step B to obtain the set temperature value of each control area corresponding to the target yield.
2. The method for rapidly obtaining the set temperatures at different productivities of the walking beam type billet heating furnace according to claim 1, wherein the billet heating accuracy Δ T is used in the step AdisDelta T of degree of thermal uniformity of discharged steel billetsmaxAnd solving a plurality of groups of training data of the walking beam type billet heating furnace under different yields by using a Hooke-Jeeves direct search algorithm.
3. The method for rapidly acquiring the set temperatures of the walking beam type billet heating furnace with different production rates as claimed in claim 1, wherein in the step B, the BP neural network is a three-layer BP neural network; the input layer of the BP neural network has 3 neurons, and the 3 neurons of the input layer respectively represent a stepping period WR and a billet heating precision Delta TdisDelta T of degree of thermal uniformity of discharged steel billetsmax(ii) a The hidden layer of the BP neural network has 8 neurons; the output layer of the BP neural network is provided with 1 neuron, and the neuron of the output layer correspondingly represents the set temperature T of a control areaspi(ii) a The i BP neural network models obtained by training are as follows:
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810528045.4A CN108694288B (en) | 2018-05-29 | 2018-05-29 | Method for rapidly acquiring set temperatures of walking beam type billet heating furnace under different yields |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810528045.4A CN108694288B (en) | 2018-05-29 | 2018-05-29 | Method for rapidly acquiring set temperatures of walking beam type billet heating furnace under different yields |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108694288A CN108694288A (en) | 2018-10-23 |
CN108694288B true CN108694288B (en) | 2021-04-30 |
Family
ID=63847944
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810528045.4A Expired - Fee Related CN108694288B (en) | 2018-05-29 | 2018-05-29 | Method for rapidly acquiring set temperatures of walking beam type billet heating furnace under different yields |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108694288B (en) |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO1997028669A1 (en) * | 1996-01-31 | 1997-08-07 | Asm America, Inc. | Model-based predictive control of thermal processing |
CN1644257A (en) * | 2004-12-15 | 2005-07-27 | 大连理工大学 | Heating furnace comprehensive optimizing controlling system designing and controlling method for hot rolling process |
CN1757263A (en) * | 2003-03-06 | 2006-04-05 | 莱克公司 | Analytical furnace with predictive temperature control |
CN103019097A (en) * | 2012-11-29 | 2013-04-03 | 北京和隆优化控制技术有限公司 | Optimal control system for steel rolling heating furnace |
CN103146906A (en) * | 2013-02-28 | 2013-06-12 | 首钢总公司 | Parameter adjustment and control method for two-stage control model of walking beam heating furnace |
CN105045949A (en) * | 2015-05-26 | 2015-11-11 | 浙江中控研究院有限公司 | Walking beam furnace steel billet temperature modeling and on-line correcting method |
-
2018
- 2018-05-29 CN CN201810528045.4A patent/CN108694288B/en not_active Expired - Fee Related
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO1997028669A1 (en) * | 1996-01-31 | 1997-08-07 | Asm America, Inc. | Model-based predictive control of thermal processing |
CN1757263A (en) * | 2003-03-06 | 2006-04-05 | 莱克公司 | Analytical furnace with predictive temperature control |
CN1644257A (en) * | 2004-12-15 | 2005-07-27 | 大连理工大学 | Heating furnace comprehensive optimizing controlling system designing and controlling method for hot rolling process |
CN103019097A (en) * | 2012-11-29 | 2013-04-03 | 北京和隆优化控制技术有限公司 | Optimal control system for steel rolling heating furnace |
CN103146906A (en) * | 2013-02-28 | 2013-06-12 | 首钢总公司 | Parameter adjustment and control method for two-stage control model of walking beam heating furnace |
CN105045949A (en) * | 2015-05-26 | 2015-11-11 | 浙江中控研究院有限公司 | Walking beam furnace steel billet temperature modeling and on-line correcting method |
Non-Patent Citations (7)
Title |
---|
Application of neural network to the supervisory control of a reheating furnace in the steel industry;Young ll Kim等;《Control Engineering Practice》;19980409;第6卷(第8期);1009-1014页 * |
Development of a first-principles hybrid model for large-scale reheating furnaces;Yukun Hu等;《Applied Energy》;20160420;第173卷;555-566页 * |
Model-based multi-objective optimisation of reheating furnace operations using genetic algorithm;Yukun Hu等;《Energy Procedia》;20171231;第142卷;2143-2151页 * |
Modeling of Reheating-Furnace Dynamics Using Neural Network Based on Improved Sequential-Learning Algorithm;Yingxin Liao等;《2006 IEEE International Conference on Control Applications》;20061006;3175-3181页 * |
Nonlinear dynamic simulation and control of large-scale reheating furnace operations using a zone method based model;Yukun Hu等;《Applied Thermal Engineering》;20180214;第135卷;41-53页 * |
步进式加热炉智能控制方法研究;李丹;《中国优秀硕士学位论文全文数据库 工程科技Ⅰ辑》;20160315(第3期);B023-37页 * |
热轧过程加热炉综合优化控制策略及控制方法研究;金迪;《中国优秀博硕士学位论文全文数据库(硕士) 工程科技Ⅰ辑》;20050715(第3期);B022-66页 * |
Also Published As
Publication number | Publication date |
---|---|
CN108694288A (en) | 2018-10-23 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106636610B (en) | A kind of double dimension walking beam furnace heating curve optimal setting methods based on time and furnace superintendent | |
CN102176221B (en) | Coke furnace temperature predicting method based on dynamic working conditions in coke furnace heating and burning process | |
CN109248928A (en) | A kind of hot-rolling heating furnace dynamic method for controlling furnace temperature | |
CN102880046B (en) | Chemical multi-variable process decoupling prediction function control method | |
CN104101105B (en) | The composite control method of a kind of heating crude oil furnace temperature and device thereof | |
CN109858780B (en) | Steelmaking-continuous casting production scheduling optimization method | |
CN111522229B (en) | Parameter self-tuning MIMO different factor offset format model-free control method | |
KR20180073434A (en) | System for control temperature pattern of strip in continuous annealing line and the method of the same | |
CN104894362A (en) | Method for setting temperature of heating furnace in cold and hot steel billet mixed loading | |
CN104615856A (en) | Gas consumption prediction model establishing method and device based on hot blast stove group | |
Zanoli et al. | Two‐Layer Linear MPC Approach Aimed at Walking Beam Billets Reheating Furnace Optimization | |
Cao et al. | A dynamic subspace model for predicting burn-through point in iron sintering process | |
CN102520617A (en) | Prediction control method for unminimized partial decoupling model in oil refining industrial process | |
CN116127760A (en) | Heating furnace temperature optimization method based on hybrid modeling and improved sand dune cat swarm algorithm | |
CN114410956B (en) | Online control system and method for intermittent aluminum coil annealing furnace | |
CN108694288B (en) | Method for rapidly acquiring set temperatures of walking beam type billet heating furnace under different yields | |
CN103146905A (en) | Heating furnace temperature decision-making method based on billet optimizing heating curve | |
CN111411215A (en) | Furnace temperature comprehensive decision-making method for multiple steel billet objects | |
CN106191411B (en) | A kind of time inside furnace control method for steel plate heat treatment | |
CN108398017B (en) | A kind of reactor heating furnace beacon flint control method | |
CN113900379A (en) | Neural network-based predictive control algorithm | |
CN102799723A (en) | Muffle furnace heating capacity calculation, prediction and simulation method | |
CN108895855B (en) | Temperature optimization method is arranged in walking beam furnace | |
CN106773699B (en) | The multi-model predictive function control method of the genetic optimization of electric furnace | |
JPH09296229A (en) | Method for controlling combustion of continuous heating furnace |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20210430 |