CN110205427A - A kind of intelligence hot-blast stove Optimal Control System and method - Google Patents
A kind of intelligence hot-blast stove Optimal Control System and method Download PDFInfo
- Publication number
- CN110205427A CN110205427A CN201910537759.6A CN201910537759A CN110205427A CN 110205427 A CN110205427 A CN 110205427A CN 201910537759 A CN201910537759 A CN 201910537759A CN 110205427 A CN110205427 A CN 110205427A
- Authority
- CN
- China
- Prior art keywords
- time
- lstm network
- hot
- blast stove
- gas flow
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- C—CHEMISTRY; METALLURGY
- C21—METALLURGY OF IRON
- C21B—MANUFACTURE OF IRON OR STEEL
- C21B9/00—Stoves for heating the blast in blast furnaces
Landscapes
- Engineering & Computer Science (AREA)
- Chemical & Material Sciences (AREA)
- Manufacturing & Machinery (AREA)
- Materials Engineering (AREA)
- Metallurgy (AREA)
- Organic Chemistry (AREA)
- Regulation And Control Of Combustion (AREA)
Abstract
The invention belongs to metallurgy industry Combustion of Hot Air Furnace technical field more particularly to a kind of intelligent hot-blast stove Optimal Control System and methods.The system includes: remaining air supply time prediction module: for obtaining the remaining air supply time of hot-blast stove according to mixer selector valve aperture parameter and exhaust gas temperature parameter based on the first LSTM network；Afterburning time prediction module: for obtaining the afterburning time of hot-blast stove according to gas flow parameter and exhaust gas temperature parameter based on the 2nd LSTM network；Optimization module: for according to acquisition remaining air supply time and the afterburning time, dynamic adjust gas flow；Wherein, mixer selector valve aperture, the historical data of exhaust gas temperature and gas flow are chosen in advance, and the structural parameters of the first LSTM network and the 2nd LSTM network are obtained through BP algorithm training.The system is based on LSTM network and improves the production efficiency of blast furnace ironmaking so that the burning of hot-blast stove matches with air supply time.
Description
Technical field
The invention belongs to metallurgy industry Combustion of Hot Air Furnace technical field more particularly to a kind of intelligent hot-blast stove optimal control systems
System and method.
Background technique
Steel is known as the grain of industry, is the most commonly used metal material of the mankind, intensity is high, good mechanical property, resource
It is abundant, it is at low cost, it is suitable for being mass produced, suffers from and be widely applied in the every field of social production life, being can not
Or scarce strategic basic industries product, it is occupied an important position in the national economic development.
Hot-blast stove is the carrier of the high wind-warm syndrome of blast furnace, according to the work of " accumulation of heat " principle blast heating to desired temperature, is used
To improve the returns and efficiency of blast furnace.Hot-blast stove is primarily present two aspect problems: first is that cannot obtain during burning enough
Heat is to meet the requirement in air feed stage；Second is that hot-blast stove has reached institute's calorific requirement, but do not arrive change heater stage yet, to maintain temperature
Degree, causes the waste of coal gas.During operation, gas flow, pressure, the fluctuation of calorific value and air mass flow, pressure, temperature
Combustion process will be had an impact.
Stove operation is larger by the interference of extraneous factor, and the time blown every time is often different, the combustion of each hot-blast stove
The burning time is often also different, furnace must be just changed in advance when causing operator or not up to optimal combustion path, so that hot wind
Quality decline；Have reached and change furnace requirement, and the stage of blowing still without end so that needing additional coal gas to maintain temperature
Degree, causes coal gas to waste.
At present although most domestic hot-blast stove realizes basic automatization, but other than individual large blast furnaces, really may be used
Automatic control by practical Combustion of Hot Air Furnace process is also rarely found, and most of medium and small blast furnace is nearly all to be manually operated
At the Combustion System of hot-blast stove.Since automatization level is not high, the factors such as artificial experience deficiency, Combustion of Hot Air Furnace is good or bad,
The hot blast temperature for causing hot-blast stove to provide is lower, and state labile, is unfavorable for the efficient production of blast furnace ironmaking in this way.
Summary of the invention
(1) technical problems to be solved
For existing technical problem, the present invention provides a kind of intelligent hot-blast stove Optimal Control System, the system base
In LSTM network, so that the burning of hot-blast stove matches with air supply time, the production efficiency of blast furnace ironmaking is improved.
(2) technical solution
A kind of intelligent hot-blast stove Optimal Control System provided by the invention, hot-blast stove in combustion in include quickly rising
Thermophase pushes up warm stabilization sub stage and adjusting stage afterburning time, which is applied to adjusting stage afterburning time,
The system includes:
Remaining air supply time prediction module: for being based on the first LSTM network, according to mixer selector valve aperture parameter and exhaust gas temperature
Parameter is spent, the remaining air supply time of hot-blast stove is obtained；
Afterburning time prediction module: for being based on the 2nd LSTM network, according to gas flow parameter and exhaust gas temperature
Parameter obtains the afterburning time of hot-blast stove；
Optimization module: for according to acquisition remaining air supply time and the afterburning time, dynamic adjust gas flow；
Wherein, mixer selector valve aperture, the historical data of exhaust gas temperature and gas flow are chosen in advance, are obtained through BP algorithm training
The structural parameters of first LSTM network and the 2nd LSTM network.
The present invention also provides a kind of intelligent hot-blast stove optimal control methods, include the following steps:
A1, the residue of hot-blast stove is obtained according to mixer selector valve aperture parameter and exhaust gas temperature parameter based on the first LSTM network
Air supply time；
A2, it is based on the 2nd LSTM network, according to gas flow parameter and exhaust gas temperature parameter, obtains the remaining combustion of hot-blast stove
Burn the time；
A3, the remaining air supply time according to acquisition and afterburning time, dynamic adjust gas flow；
Wherein, mixer selector valve aperture, the historical data of exhaust gas temperature and gas flow are chosen in advance, are obtained through BP algorithm training
The structural parameters of first LSTM network and the 2nd LSTM network.
Further, in the step A1, period distances obtain mixer selector valve aperture parameter and exhaust gas temperature parameter, described
Interval time is 1-2s；
In the step A2, period distances obtain gas flow parameter and exhaust gas temperature parameter, the interval time
For 1-2s.
Further, in the step A3, when the absolute value of afterburning time and the difference of remaining air supply time are small
In 120s, gas flow is not adjusted；
When the absolute value of afterburning time and the difference of remaining air supply time are more than or equal to 120s, adjustment gas flow.
Further, when the absolute value of afterburning time and the difference of remaining air supply time are more than or equal to 120s, adjustment
Gas flow, comprising:
When remaining air supply time be greater than the afterburning time, reduce the 5% of current gas flow；
When remaining air supply time be less than the afterburning time, increase the 5% of current gas flow.
Further, in the step A3, the interval time that dynamic adjusts gas flow is 20-40s.
Further, the first LSTM network and the 2nd LSTM network include input layer and full articulamentum；
The mixer selector valve aperture parameter and exhaust gas temperature parameter that will acquire input the input layer of the first LSTM network, the first LSTM
The full articulamentum of network exports remaining air supply time；
The gas flow parameter and exhaust gas temperature parameter that will acquire input the input layer of the 2nd LSTM network, the 2nd LSTM net
The full articulamentum of network exports the afterburning time；
Further, the optimization of dropout is introduced to the full articulamentum of the first LSTM network and the 2nd LSTM network respectively
Method.
(3) beneficial effect
In intelligence hot-blast stove Optimal Control System provided by the invention, using LSTM network come processing sequence data, with biography
System RNN is compared, and avoids the problem of gradient disappears, and prediction effect is good, and precision is high.When according to afterburning time and remaining air-supply
Between difference, the combustion phases of dynamic adjustment in real time, burning matches with air supply time, and both the wind-warm syndrome of certifiable hot-blast stove reaches in this way
To requirement, and avoid the problem that the waste of coal gas caused by burning time is too long and lifetime of hot-air stove reduce.Simultaneously because air-fuel
The optimization of ratio and the reduction of fuel quantity, can also substantially reduce the discharge of combustion product oxynitrides etc., be conducive to environment guarantor
Shield.
Detailed description of the invention
Fig. 1 is the structural schematic diagram of LSTM in the present invention；
Fig. 2 is the flow chart of combustion process in the present invention.
Specific embodiment
In order to preferably explain the present invention, in order to understand, with reference to the accompanying drawing, by specific embodiment, to this hair
It is bright to be described in detail.
In intelligence hot-blast stove Optimal Control System provided by the invention, by the way of 3-4 hot-blast stove alternation into
Row, a hot-blast stove are blown, and remaining 2-3 hot-blast stove is in combustion phases, that is to say, that the time of a combustion phases
It is 2-3 air supply time.
The working principle of hot-blast stove includes cycle period, in a cycle period, includes two stages, respectively burns
Stage and air-supply stage.In combustion phases, coal gas is inputted to combustion chamber and air burns, thus heat storage room, accumulation of heat
Room stores heat by checker brick come accumulation of heat, is heated to certain temperature to it, pause burning goes to next stage, putting in blast.
In the air-supply stage, air blower blows cold wind enters regenerative chamber after cold air duct, and regenerative chamber keeps higher accumulation of heat horizontal at this time,
Regenerative chamber heats itself by checker brick again after cold wind entrance, by being sent into blast furnace after heat exchange.The accumulation of heat energy of regenerative chamber
Power constantly reduces, and wind-warm syndrome constantly increases.When regenerative chamber can not be by cold air heating to desired target temperature, then it is transferred to burning
State, i.e., from being on air transferred to main combustion period, such cycle operation.
As shown in Fig. 2, when hot-blast stove includes the stage of being rapidly heated, pushes up warm stabilization sub stage and afterburning in combustion
Between the adjusting stage, intelligent hot-blast stove Optimal Control System provided by the invention is applied to adjusting stage afterburning time, should
System includes:
Remaining air supply time prediction module: for being based on the first LSTM network, according to mixer selector valve aperture parameter and exhaust gas temperature
Parameter is spent, the remaining air supply time of hot-blast stove is obtained；
Afterburning time prediction module: for being based on the 2nd LSTM network, according to gas flow parameter and exhaust gas temperature
Parameter obtains the afterburning time of hot-blast stove；
Optimization module: for according to acquisition remaining air supply time and the afterburning time, dynamic adjust gas flow；
Wherein, mixer selector valve aperture, the historical data of exhaust gas temperature and gas flow are chosen in advance, are obtained through BP algorithm training
The structural parameters of first LSTM network and the 2nd LSTM network.
The present invention also provides a kind of intelligent hot-blast stove optimal control methods, include the following steps:
A1, the residue of hot-blast stove is obtained according to mixer selector valve aperture parameter and exhaust gas temperature parameter based on the first LSTM network
Air supply time.Specifically, OPC Server is connected by KingView, obtains mixer selector valve aperture and exhaust gas temperature in real time, to preparatory
The input layer input mixer selector valve aperture and exhaust gas temperature of trained first LSTM network, wherein sampling interval 1-2s, preferably
For 1s, sampling time step-length is 12-16s, preferably 15s, the remaining air supply time of full articulamentum output of the first LSTM network.For
The phenomenon that preventing over-fitting generation introduces the optimization method of dropout in full articulamentum.
A2, it is based on the 2nd LSTM network, according to gas flow parameter and exhaust gas temperature parameter, obtains the remaining combustion of hot-blast stove
Burn the time.Specifically, OPC Server is connected by KingView, gas flow and exhaust gas temperature is obtained in real time, to the 2nd LSTM
The input layer input gas flow and exhaust gas temperature of network, wherein sampling interval 1-2s, preferably 1s, sampling time step-length
Full articulamentum for 12-16s, preferably 15s, the 2nd LSTM network exports the afterburning time.Over-fitting shows in order to prevent
As occurring, the optimization method of dropout is introduced in full articulamentum.
Wherein, include: to the training process of the first LSTM network and the 2nd LSTM network
By taking No.1 furnace as an example, OPC Server is connected by KingView, required exhaust gas temperature, coal gas can be acquired in real time
The data such as flow, mixer selector valve aperture, dome temperature.By collected data according to 8:2 ratio cut partition be training set and test set.
The network structure that the first LSTM network and the 2nd LSTM network is established using Tensorflow, inputs first for training set respectively
LSTM network and the 2nd LSTM network are joined using the network that Adam optimizer calculates the first LSTM network and the 2nd LSTM network
Number.Trained first LSTM network and the 2nd LSTM network are used for test set test, test set error is calculated, adjusts first
The hyper parameter of LSTM network and the 2nd LSTM network, is trained again, until test set result is in allowable range of error.
Further, as shown in Figure 1, the first LSTM network and the 2nd LSTM network internal propagated forward calculation method such as
Under:
ft=σ (Wf·[ht-1,Xt]+bf) (1)
it=σ (Wi·[ht-1,Xt]+bi) (2)
ot=σ (Wo·[ht-1,Xt]+bo) (5)
ht=ot·tanh(ct) (6)
In formula, ftIt indicates to forget door, σ indicates sigmoid function, WfIndicate the weight matrix of forgetting door, ht-1Indicate upper one
The output data at moment, XtIndicate the input data at current time, [ht-1,Xt] indicate ht-1、XtIt is spliced into one in the horizontal direction
A long vector, bfIndicate the bias of forgetting door, itIndicate input gate, WiIndicate the weight coefficient of input gate, biIndicate input gate
Bias,Indicate location mode currently entered, WcIndicate the weight matrix of location mode currently entered, bcExpression is worked as
The bias of the location mode of preceding input, ctIndicate the location mode at current time, otIndicate out gate, WoIndicate out gate
Weight matrix, boIndicate the bias of out gate, htIndicate final output.
A3, the remaining air supply time according to acquisition and afterburning time, dynamic adjust gas flow.
It is less than 120s when afterburning time and the absolute value of the difference of remaining air supply time, does not adjust gas flow；When
The absolute value of the difference of afterburning time and remaining air supply time is more than or equal to 120s, adjusts gas flow, in which: work as residue
Air supply time is greater than the afterburning time, reduces the 5% of current gas flow；When remaining air supply time is less than afterburning
Between, increase the 5% of current gas flow.
The course of work:
Each hot-blast stove successively passes through combustion phases and air-supply stage.Include: in combustion phases
It is rapidly heated the stage: being passed through coal gas into hot-blast stove, and set maximum value Q for gas flowmax(Qmax=
160000m3/ h), when detection dome temperature reaches the 90%-97.5% of temperature preset value T, suitably reduces gas flow, protect
It demonstrate,proves dome temperature and is no more than temperature preset value T, into the top warm stabilization sub stage.Preferably, 1400 are set by temperature preset value T
℃。
Push up the warm stabilization sub stage: every 30s detects a dome temperature and exhaust gas temperature, according to dome temperature and exhaust gas temperature pair
Gas flow is suitably adjusted, and when dome temperature variation is in ± 2.5% range, gas flow is not adjusted.When detecting
When exhaust gas temperature and the temperature difference of temperature preset value T are less than 30 DEG C, into adjusting stage afterburning time.
Adjusting stage afterburning time: according to the sampling interval obtain the exhaust gas temperature of each sampling time section, gas flow,
The data such as mixer selector valve aperture are based on the first LSTM network and the 2nd LSTM network, when obtaining remaining air supply time and afterburning
Between, dynamic adjusts gas flow.Wherein, to prevent fluctuation from excessively frequently having an impact to hot-blast stove, a dant is adjusted every 30s
Throughput.It is less than 120s when afterburning time and the absolute value of the difference of remaining air supply time, does not adjust gas flow；When surplus
The absolute value of the difference of remaining burning time and remaining air supply time is more than or equal to 120s, adjusts gas flow, in which: when residue is sent
The wind time is greater than the afterburning time, reduces the 5% of current gas flow；When remaining air supply time be less than the afterburning time,
Increase the 5% of current gas flow.
When detecting 375 DEG C of exhaust gas temperature setting, reach and change furnace requirement, combustion phases terminates, into the air-supply stage.
In this way, the air-fuel ratio of optimization hot-blast stove, not only avoids burning and has reached state, but the stage of blowing does not terminate institute still
The waste of bring coal gas；In turn avoiding the air-supply stage has terminated, combustion process but remain unfulfilled and caused by hot wind quality do not reach
Target situation.
Due to extraneous a variety of factors, the reaction of the combustion phases of each hot-blast stove may be different, therefore in afterburning
Between in the adjusting stage, the first LSTM network of each hot-blast stove and the 2nd LSTM network are respectively trained.
The technical principle of the invention is described above in combination with a specific embodiment, these descriptions are intended merely to explain of the invention
Principle shall not be construed in any way as a limitation of the scope of protection of the invention.Based on explaining herein, those skilled in the art
It can associate with other specific embodiments of the invention without creative labor, these modes fall within this hair
Within bright protection scope.
Claims (8)
1. a kind of intelligence hot-blast stove Optimal Control System, hot-blast stove includes the stage of being rapidly heated, top Wen Wending in combustion
The system is applied to adjusting stage afterburning time, which is characterized in that this is by stage and adjusting stage afterburning time
System includes:
Remaining air supply time prediction module: for being based on the first LSTM network, joined according to mixer selector valve aperture parameter and exhaust gas temperature
Number, obtains the remaining air supply time of hot-blast stove；
Afterburning time prediction module: for being based on the 2nd LSTM network, according to gas flow parameter and exhaust gas temperature parameter,
Obtain the afterburning time of hot-blast stove；
Optimization module: for according to acquisition remaining air supply time and the afterburning time, dynamic adjust gas flow；
Wherein, mixer selector valve aperture, the historical data of exhaust gas temperature and gas flow are chosen in advance, obtain first through BP algorithm training
The structural parameters of LSTM network and the 2nd LSTM network.
2. a kind of intelligence hot-blast stove optimal control method, which comprises the steps of:
A1, it is based on the first LSTM network, according to mixer selector valve aperture parameter and exhaust gas temperature parameter, obtains the remaining air-supply of hot-blast stove
Time；
A2, it is based on the 2nd LSTM network, according to gas flow parameter and exhaust gas temperature parameter, when obtaining the afterburning of hot-blast stove
Between；
A3, the remaining air supply time according to acquisition and afterburning time, dynamic adjust gas flow；
Wherein, mixer selector valve aperture, the historical data of exhaust gas temperature and gas flow are chosen in advance, obtain first through BP algorithm training
The structural parameters of LSTM network and the 2nd LSTM network.
3. intelligence hot-blast stove optimal control method according to claim 2, which is characterized in that
In the step A1, period distances obtain mixer selector valve aperture parameter and exhaust gas temperature parameter, the interval time are
1-2s；
In the step A2, period distances obtain gas flow parameter and exhaust gas temperature parameter, and the interval time is 1-
2s。
4. intelligence hot-blast stove optimal control method according to claim 2, which is characterized in that
In the step A3, it is less than 120s when afterburning time and the absolute value of the difference of remaining air supply time, does not adjust
Gas flow；
When the absolute value of afterburning time and the difference of remaining air supply time are more than or equal to 120s, adjustment gas flow.
5. intelligence hot-blast stove optimal control method according to claim 4, which is characterized in that when the afterburning time and remain
The absolute value of the difference of remaining air supply time is more than or equal to 120s, adjusts gas flow, comprising:
When remaining air supply time be greater than the afterburning time, reduce the 5% of current gas flow；
When remaining air supply time be less than the afterburning time, increase the 5% of current gas flow.
6. intelligence hot-blast stove optimal control method according to claim 2, which is characterized in that in the step A3, move
The interval time that state adjusts gas flow is 20-40s.
7. intelligence hot-blast stove optimal control method according to claim 2, which is characterized in that
The first LSTM network and the 2nd LSTM network include input layer and full articulamentum；
The mixer selector valve aperture parameter and exhaust gas temperature parameter that will acquire input the input layer of the first LSTM network, the first LSTM network
Full articulamentum export remaining air supply time；
The gas flow parameter and exhaust gas temperature parameter that will acquire input the input layer of the 2nd LSTM network, the 2nd LSTM network
Full articulamentum exports the afterburning time；
8. intelligence hot-blast stove optimal control method according to claim 7, which is characterized in that respectively to the first LSTM network
The optimization method of dropout is introduced with the full articulamentum of the 2nd LSTM network.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910537759.6A CN110205427B (en) | 2019-06-20 | 2019-06-20 | Intelligent hot blast stove optimization control system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910537759.6A CN110205427B (en) | 2019-06-20 | 2019-06-20 | Intelligent hot blast stove optimization control system and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110205427A true CN110205427A (en) | 2019-09-06 |
| CN110205427B CN110205427B (en) | 2020-06-16 |
Family
ID=67793720
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910537759.6A Active CN110205427B (en) | 2019-06-20 | 2019-06-20 | Intelligent hot blast stove optimization control system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110205427B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110717628A (en) * | 2019-10-09 | 2020-01-21 | 浪潮软件股份有限公司 | Goods source optimal distribution model construction method, optimal distribution model and optimal distribution method |
| CN111676365A (en) * | 2020-06-09 | 2020-09-18 | 邯郸钢铁集团有限责任公司 | LSTM-based method for predicting annealing temperature set value of transition steel coil |
| CN113420843A (en) * | 2021-08-23 | 2021-09-21 | 湖南省计量检测研究院 | Heat testing method, device, equipment and medium based on micro fluidized bed |
| CN114279494A (en) * | 2021-12-16 | 2022-04-05 | 金陵科技学院 | Fiber grating blast furnace shell temperature compensation detection method based on improved LSTM |
| CN114610093A (en) * | 2022-03-09 | 2022-06-10 | 科大智能物联技术股份有限公司 | Burning furnace air supply control method based on variable period prediction of hot blast stove |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1766130A (en) * | 2005-12-09 | 2006-05-03 | 河北理工大学 | Harmonization control method for blast furnace hot blast stove system |
| CN101892338A (en) * | 2010-07-22 | 2010-11-24 | 首钢总公司 | Constant wind temperature control system for hot blast stove |
| CN108251591A (en) * | 2018-01-15 | 2018-07-06 | 上海大学 | Utilize the top bottom blowing converter producing process control method of LSTM systems |
-
2019
- 2019-06-20 CN CN201910537759.6A patent/CN110205427B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1766130A (en) * | 2005-12-09 | 2006-05-03 | 河北理工大学 | Harmonization control method for blast furnace hot blast stove system |
| CN101892338A (en) * | 2010-07-22 | 2010-11-24 | 首钢总公司 | Constant wind temperature control system for hot blast stove |
| CN108251591A (en) * | 2018-01-15 | 2018-07-06 | 上海大学 | Utilize the top bottom blowing converter producing process control method of LSTM systems |
Non-Patent Citations (2)
| Title |
|---|
| 高志强等: "《深度学习 从入门到实践》", 30 June 2018, 中国铁道出版社 * |
| 龙红明: "《冶金过程数学模型与人工智能应用》", 30 June 2010, 龙红明 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110717628A (en) * | 2019-10-09 | 2020-01-21 | 浪潮软件股份有限公司 | Goods source optimal distribution model construction method, optimal distribution model and optimal distribution method |
| CN110717628B (en) * | 2019-10-09 | 2023-05-23 | 浪潮软件股份有限公司 | Goods source optimizing model construction method, optimizing model and optimizing method |
| CN111676365A (en) * | 2020-06-09 | 2020-09-18 | 邯郸钢铁集团有限责任公司 | LSTM-based method for predicting annealing temperature set value of transition steel coil |
| CN113420843A (en) * | 2021-08-23 | 2021-09-21 | 湖南省计量检测研究院 | Heat testing method, device, equipment and medium based on micro fluidized bed |
| CN114279494A (en) * | 2021-12-16 | 2022-04-05 | 金陵科技学院 | Fiber grating blast furnace shell temperature compensation detection method based on improved LSTM |
| CN114279494B (en) * | 2021-12-16 | 2023-05-26 | 金陵科技学院 | Fiber bragg grating blast furnace shell temperature compensation detection method based on improved LSTM |
| CN114610093A (en) * | 2022-03-09 | 2022-06-10 | 科大智能物联技术股份有限公司 | Burning furnace air supply control method based on variable period prediction of hot blast stove |
| CN114610093B (en) * | 2022-03-09 | 2023-03-03 | 科大智能物联技术股份有限公司 | Burning furnace air supply control method based on variable period prediction of hot blast stove |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110205427B (en) | 2020-06-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110205427A (en) | A kind of intelligence hot-blast stove Optimal Control System and method | |
| CN109248928A (en) | A kind of hot-rolling heating furnace dynamic method for controlling furnace temperature | |
| CN112795716B (en) | Efficient and practical hot blast stove burning control method | |
| CN107245570A (en) | A kind of cold rolled annealed stove heat combination control method | |
| CN106011352A (en) | Optimized control system for coal gas flow of blast-furnace hot blast stove | |
| CN102121786A (en) | Hot air system of grate-rotary kiln-circular cooler | |
| CN206430540U (en) | A kind of drying furnace system of temperature automatic adjustment control | |
| CN106362648A (en) | Heat accumulating type quick pyrolysis furnace temperature control method capable of controlling air-fuel ratio | |
| CN116499272B (en) | Intelligent kiln control method for tracking multi-energy combustion carbon emission | |
| CN205874470U (en) | Blast funnace hot blast stove optimal control system | |
| CN106127351A (en) | A kind of blast funnace hot blast stove initially burns producer gas flow optimization system | |
| CN105404143B (en) | Gas furnace kiln furnace pressure computer intelligence fuzzy control power-economizing method | |
| CN112666834B (en) | Heating furnace temperature control method adaptive to severe fluctuation of fuel gas heat value | |
| CN101892366A (en) | Low oxygen content and low nitrogen consumption control system and method for radiant tube-heated non-oxidation roller hearth furnace | |
| CN111998689B (en) | Method and system for controlling carbon anode roasting process for aluminum | |
| CN112695193B (en) | Coordination control method for ensuring efficient operation of hot rolling heating furnace burner | |
| KR100361246B1 (en) | Method for controlling equalized back pressure of blast furnace hot stove | |
| CN108728629A (en) | A kind of energy-saving straight fire heating apparatus for continuous heat treatment | |
| CN103525959A (en) | Method for controlling air temperature of blast furnace | |
| CN109579044A (en) | A kind of walking beam furnace air-fuel ratio dynamic Feedforward method for controlling combustion | |
| JPS63226524A (en) | Combustion control in hot blast furnace | |
| CN109141051A (en) | A kind of optimum design method of heat accumulating type industrial heating furnace controling of the pressure of the oven | |
| CN104531978A (en) | Non-oriented silicon steel heating method and modeling control method of non-oriented silicon steel heating method | |
| TWI817819B (en) | Method and computer program product for controlling hot blast stove | |
| CN115537486B (en) | Combustion control method of hot blast stove |
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 |