CN116434859A - Blast furnace gas production consumption prediction method and device, electronic equipment and storage medium - Google Patents
Blast furnace gas production consumption prediction method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN116434859A CN116434859A CN202310402188.1A CN202310402188A CN116434859A CN 116434859 A CN116434859 A CN 116434859A CN 202310402188 A CN202310402188 A CN 202310402188A CN 116434859 A CN116434859 A CN 116434859A
- Authority
- CN
- China
- Prior art keywords
- consumption
- gas production
- prediction
- blast furnace
- historical data
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 276
- 238000004519 manufacturing process Methods 0.000 title claims abstract description 180
- 238000003860 storage Methods 0.000 title claims abstract description 19
- 230000008569 process Effects 0.000 claims abstract description 156
- 238000012549 training Methods 0.000 claims abstract description 27
- 238000012216 screening Methods 0.000 claims abstract description 10
- 238000004422 calculation algorithm Methods 0.000 claims description 46
- 238000011156 evaluation Methods 0.000 claims description 25
- 238000005457 optimization Methods 0.000 claims description 20
- 230000007246 mechanism Effects 0.000 claims description 14
- 230000015654 memory Effects 0.000 claims description 12
- 238000004590 computer program Methods 0.000 claims description 10
- 238000003062 neural network model Methods 0.000 claims description 10
- 239000007789 gas Substances 0.000 description 162
- 229910000831 Steel Inorganic materials 0.000 description 25
- 239000010959 steel Substances 0.000 description 25
- 238000010586 diagram Methods 0.000 description 10
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical compound [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 description 8
- 238000004364 calculation method Methods 0.000 description 8
- 239000003034 coal gas Substances 0.000 description 5
- 238000004891 communication Methods 0.000 description 5
- 238000010219 correlation analysis Methods 0.000 description 5
- 238000010438 heat treatment Methods 0.000 description 5
- 238000013528 artificial neural network Methods 0.000 description 4
- 230000008859 change Effects 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 229910052742 iron Inorganic materials 0.000 description 4
- 239000002245 particle Substances 0.000 description 4
- 238000012360 testing method Methods 0.000 description 4
- 238000012545 processing Methods 0.000 description 3
- 238000011084 recovery Methods 0.000 description 3
- IJGRMHOSHXDMSA-UHFFFAOYSA-N Atomic nitrogen Chemical compound N#N IJGRMHOSHXDMSA-UHFFFAOYSA-N 0.000 description 2
- CURLTUGMZLYLDI-UHFFFAOYSA-N Carbon dioxide Chemical compound O=C=O CURLTUGMZLYLDI-UHFFFAOYSA-N 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000010354 integration Effects 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000005096 rolling process Methods 0.000 description 2
- UGFAIRIUMAVXCW-UHFFFAOYSA-N Carbon monoxide Chemical compound [O+]#[C-] UGFAIRIUMAVXCW-UHFFFAOYSA-N 0.000 description 1
- 238000003723 Smelting Methods 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000006227 byproduct Substances 0.000 description 1
- 229910002092 carbon dioxide Inorganic materials 0.000 description 1
- 239000001569 carbon dioxide Substances 0.000 description 1
- 229910002091 carbon monoxide Inorganic materials 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000001276 controlling effect Effects 0.000 description 1
- 238000009792 diffusion process Methods 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000002068 genetic effect Effects 0.000 description 1
- 239000001257 hydrogen Substances 0.000 description 1
- 229910052739 hydrogen Inorganic materials 0.000 description 1
- 125000004435 hydrogen atom Chemical class [H]* 0.000 description 1
- YHXISWVBGDMDLQ-UHFFFAOYSA-N moclobemide Chemical compound C1=CC(Cl)=CC=C1C(=O)NCCN1CCOCC1 YHXISWVBGDMDLQ-UHFFFAOYSA-N 0.000 description 1
- 229910052757 nitrogen Inorganic materials 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 239000000047 product Substances 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 230000033764 rhythmic process Effects 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- 238000002791 soaking Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2462—Approximate or statistical queries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2474—Sequence data queries, e.g. querying versioned data
-
- 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
- G06N3/044—Recurrent networks, e.g. Hopfield networks
- G06N3/0442—Recurrent networks, e.g. Hopfield networks characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
-
- 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
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/04—Manufacturing
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Strategic Management (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Probability & Statistics with Applications (AREA)
- Economics (AREA)
- Fuzzy Systems (AREA)
- Computing Systems (AREA)
- General Business, Economics & Management (AREA)
- Molecular Biology (AREA)
- Tourism & Hospitality (AREA)
- Life Sciences & Earth Sciences (AREA)
- Databases & Information Systems (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Marketing (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Operations Research (AREA)
- Primary Health Care (AREA)
- Manufacturing & Machinery (AREA)
- Quality & Reliability (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a method, a device, electronic equipment and a storage medium for predicting the production and consumption of blast furnace gas, wherein the method comprises the steps of obtaining historical data, wherein the historical data comprises gas production and consumption historical data and process parameter historical data, screening the process parameter historical data based on the correlation between the gas production and consumption and the process parameter to obtain target process parameter historical data, generating a data sample set based on the gas production and consumption historical data and the target process parameter historical data, training different types of initial models by using the data sample set to obtain a plurality of prediction models, integrating the plurality of prediction models to obtain a comprehensive prediction model, and predicting the production and consumption of the blast furnace gas based on the comprehensive prediction model; the comprehensive prediction model integrated by different types of prediction models predicts the production and consumption of the blast furnace gas, can be suitable for various complex working conditions with different production and consumption characteristics of the blast furnace gas, improves the stability of the production and consumption prediction of the blast furnace gas, and ensures the accuracy of the production and consumption prediction of the blast furnace gas.
Description
Technical Field
The present disclosure relates to the field of gas prediction technologies, and in particular, to a method and an apparatus for predicting production and consumption of blast furnace gas, an electronic device, and a storage medium.
Background
In the iron and steel smelting industry, about 40% of primary energy is converted into auxiliary energy, and the amount of byproduct blast furnace gas in iron and steel enterprises is huge, and the generation amount of blast furnace gas per hour often exceeds 100 ten thousand m 3 Is the most important secondary energy source in iron and steel enterprises. In the regulation and control process of blast furnace gas, the problems of mismatching of energy regulation and control and production rhythm and untimely regulation and control lag often exist, so that the problems of overlarge pressure fluctuation and even diffusion of a blast furnace gas system can be caused, and potential safety hazards and energy waste are caused. The prediction of the production and consumption of the blast furnace gas can help regulatory personnel to predict the balance condition of the production and consumption of the blast furnace gas for a period of time in advance, thereby realizing advanced reaction and active regulation, effectively improving the stability of a gas system, ensuring safe and safe supply, reducing energy source emission and realizing the aims of safety, economy and environmental protection.
Most of the existing blast furnace gas production and consumption prediction methods adopt a single prediction model, are difficult to adapt to complicated and changeable actual procedures of a blast furnace, have large fluctuation of accuracy for predicting the blast furnace gas production and consumption, and are unstable.
Disclosure of Invention
In view of the above drawbacks of the prior art, the present application provides a method, an apparatus, an electronic device, and a storage medium for predicting the production and consumption of blast furnace gas, so as to solve the technical problem that the accuracy of the existing single prediction model for predicting the production and consumption of blast furnace gas fluctuates greatly and is not stable enough.
The method for predicting the production and consumption of the blast furnace gas provided by the application comprises the following steps: acquiring historical data, wherein the historical data comprises gas production and consumption historical data and process parameter historical data; screening the process parameter historical data based on the correlation between the gas production and consumption and the process parameter to obtain target process parameter historical data, and generating a data sample set based on the gas production and consumption historical data and the target process parameter historical data; training different types of initial models by using the data sample set to obtain a plurality of prediction models, and integrating the plurality of prediction models to obtain a comprehensive prediction model; and predicting the blast furnace gas production and consumption based on the comprehensive prediction model.
In an embodiment of the present application, training different types of initial models by using the data sample set to obtain a plurality of prediction models, and integrating the plurality of prediction models to obtain a comprehensive prediction model, including: and training different types of initial models by using data samples of each working procedure, integrating a plurality of prediction models obtained by training to obtain a plurality of comprehensive prediction models so as to predict the blast furnace gas production consumption of different working procedures, wherein the comprehensive prediction models are in one-to-one correspondence with the working procedures, and the data sample sets comprise data samples of different working procedures.
In one embodiment of the present application, after obtaining a plurality of comprehensive prediction models, the method includes: configuring weights for all the prediction models in the comprehensive prediction models corresponding to each procedure, and carrying out weight optimization based on an optimization algorithm to determine the optimal weight combination of each comprehensive prediction model; and predicting the blast furnace gas production and consumption of different working procedures based on the comprehensive prediction models corresponding to the different working procedures and the optimized weight combination of the comprehensive prediction models.
In one embodiment of the present application, the weight optimizing based on the optimizing algorithm to determine the preferred weight combination of each comprehensive prediction model includes: determining the gas production and consumption characteristics of each process according to the gas production and consumption historical data of each process, wherein the historical data comprise the gas production and consumption historical data of different processes; for the comprehensive prediction model corresponding to each process, adjusting the weight value of each prediction model based on the gas production and consumption characteristics of the process to obtain a plurality of weight combinations of the comprehensive prediction model, and determining an evaluation algorithm corresponding to the process based on the gas production and consumption characteristics of the process to calculate the prediction error when the comprehensive prediction model calls each weight combination according to the evaluation algorithm corresponding to the process; if the prediction error is smaller than or equal to a preset threshold value, determining a weight combination corresponding to the prediction error as a preferred weight combination of the comprehensive prediction model; and if the prediction error is larger than the preset threshold, readjusting the weight value of each prediction model based on the gas production and consumption characteristics of the working procedure until the prediction error is smaller than or equal to the preset threshold.
In an embodiment of the present application, determining an evaluation algorithm corresponding to the process based on the gas production and consumption characteristics of the process includes: if the gas production and consumption characteristics of the working procedure are suddenly fluctuated, taking the mean square error as an evaluation algorithm corresponding to the working procedure; and if the gas production and consumption characteristics of the working procedure are stable fluctuation, the relative error is used as an evaluation algorithm corresponding to the working procedure.
In an embodiment of the present application, predicting blast furnace gas production and consumption of different processes based on comprehensive prediction models corresponding to the different processes and preferred weight combinations of the comprehensive prediction models includes: acquiring current process parameter data of different procedures; and inputting the current technological parameter data of each working procedure into a comprehensive prediction model corresponding to the working procedure, and calling the optimal weight combination of the comprehensive prediction model to obtain the blast furnace gas production and consumption prediction results of different working procedures.
In an embodiment of the present application, the initial model includes at least one of a process mechanism model, a differential integrated moving average autoregressive model, a back propagation neural network model, and a long and short memory neural network model.
In an embodiment of the present application, there is also provided a blast furnace gas production and consumption prediction apparatus, including: the acquisition module is used for acquiring historical data, wherein the historical data comprises gas production consumption historical data and process parameter historical data; the sample generation module is used for screening the process parameter historical data based on the correlation between the gas production consumption and the process parameter to obtain target process parameter historical data, and generating a data sample set based on the gas production consumption historical data and the target process parameter historical data; the training module is used for training different types of initial models by utilizing the data sample set to obtain a plurality of prediction models, and integrating the plurality of prediction models to obtain a comprehensive prediction model; and the prediction module is used for predicting the blast furnace gas production and consumption based on the comprehensive prediction model.
In an embodiment of the present application, there is also provided an electronic device including: one or more processors; and a storage device for storing one or more programs which, when executed by the one or more processors, cause the electronic equipment to implement the blast furnace gas production and consumption prediction method as described above.
In an embodiment of the present application, there is also provided a computer-readable storage medium having stored thereon a computer program which, when executed by a processor of a computer, causes the computer to perform the blast furnace gas production and consumption prediction method as described above.
The invention has the beneficial effects that: the invention provides a method, a device, electronic equipment and a storage medium for predicting the production and consumption of blast furnace gas, wherein the method for predicting the production and consumption of blast furnace gas generates a data sample set based on gas production and consumption historical data and process parameter historical data, trains different types of initial models, integrates a plurality of prediction models obtained by training into a comprehensive prediction model, predicts the production and consumption of blast furnace gas through the comprehensive prediction model formed by integrating a plurality of different types of prediction models, can be suitable for various complex working conditions with different production and consumption characteristics of blast furnace gas, improves the stability of the production and consumption prediction of blast furnace gas, and ensures the accuracy of the production and consumption prediction of blast furnace gas.
Drawings
FIG. 1 is a flow chart illustrating a method of blast furnace gas production and consumption prediction according to an exemplary embodiment of the present application;
FIG. 2 is a schematic diagram illustrating a gas consumption correlation analysis according to an embodiment of the present application;
FIG. 3 is a wave diagram of blast furnace gas generation shown in an embodiment of the present application;
FIG. 4 is a graph showing the fluctuation of blast furnace stove consumption according to an embodiment of the present application;
FIG. 5 is a graph showing the fluctuation of small rod gas consumption in accordance with an embodiment of the present application;
FIG. 6 is a schematic diagram illustrating weight optimization using a particle swarm algorithm according to an embodiment of the present application;
FIG. 7 is a simplified flow diagram illustrating integrated predictive model creation in accordance with an embodiment of the present application;
fig. 8 is a block diagram of a blast furnace gas production and consumption prediction apparatus according to an exemplary embodiment of the present application.
Detailed Description
Other advantages and effects of the present application will become apparent to those skilled in the art from the present disclosure, when the following description of the embodiments is taken in conjunction with the accompanying drawings. The present application may be embodied or carried out in other specific embodiments, and the details of the present application may be modified or changed from various points of view and applications without departing from the spirit of the present application. It should be noted that the following embodiments and features in the embodiments may be combined with each other without conflict.
It should be noted that, the illustrations provided in the following embodiments merely illustrate the basic concepts of the application by way of illustration, and only the components related to the application are shown in the drawings and are not drawn according to the number, shape and size of the components in actual implementation, and the form, number and proportion of the components in actual implementation may be arbitrarily changed, and the layout of the components may be more complex.
It should be noted that, in this application, "first", "second", and the like are merely distinguishing between similar objects, and are not limited to the order or precedence of similar objects. The description of variations such as "comprising," "having," etc., means that the subject of the word is not exclusive, except for the examples shown by the word.
It should be understood that the various numbers, step numbers, etc. described in this application are for ease of description and are not intended to limit the scope of this application. The size of the reference numerals in this application does not mean the order of execution, and the order of execution of the processes should be determined by their functions and inherent logic.
In the following description, numerous details are set forth to provide a more thorough explanation of embodiments of the present application, however, it will be apparent to one skilled in the art that embodiments of the present application may be practiced without these specific details, in other embodiments, well-known structures and devices are shown in block diagram form, rather than in detail, in order to avoid obscuring the embodiments of the present application.
It should be noted that most of the existing blast furnace gas production and consumption prediction methods adopt a single prediction model, are not tightly combined with the characteristics of gas production and consumption in different working procedures, and result in insufficient consideration of influence factors of the blast furnace production and consumption prediction, so that a certain difference exists between a model prediction result and actual conditions, the method is difficult to adapt to complex and changeable actual working procedures of the blast furnace, and the accuracy of the blast furnace gas production and consumption prediction is high, low, large in fluctuation and unstable. In addition, the existing blast furnace gas production and consumption prediction method also needs to carry out screening and testing of algorithms according to the blast furnace gas production and consumption characteristics of different process units, so that a great deal of time and effort are consumed, and the selection of high-low algorithm of the gas production and consumption prediction accuracy has great influence.
To solve these problems, embodiments of the present application respectively propose a blast furnace gas production and consumption prediction method, a blast furnace gas production and consumption prediction apparatus, an electronic device, a computer-readable storage medium, and a computer program product, and these embodiments will be described in detail below.
Referring to fig. 1, fig. 1 is a flowchart illustrating a method for predicting production and consumption of blast furnace gas according to an exemplary embodiment of the present application. As shown in fig. 1, in an exemplary embodiment, the blast furnace gas production and consumption prediction method at least includes steps S110 to S140, which are described in detail as follows:
Step S110, obtaining history data.
In one embodiment of the present application, the historical data includes gas production consumption historical data of at least one process and process parameter historical data of a corresponding process, wherein the gas production consumption historical data includes at least one of blast furnace gas historical occurrence, small rod heating furnace gas historical consumption, blast furnace hot blast stove historical consumption and the like, the process parameter historical data includes at least production related pressure, temperature, flow, steel billet and the like, such as steam recovery pressure, furnace steel billet temperature, steel billet specification, furnace steel billet quantity, furnace steel billet weight and the like, and the process parameter historical data of different processes are different. The initial historical data needs to be preprocessed before the historical data is acquired, the preprocessing of the basic data comprises necessary time alignment, error data elimination and the like, and the preprocessed initial historical data is used as the historical data.
Step S120, screening the process parameter historical data based on the correlation between the gas production and consumption and the process parameter to obtain target process parameter historical data, and generating a data sample set based on the gas production and consumption historical data and the target process parameter historical data.
In one embodiment of the application, the gas production and consumption characteristics of different working procedures are different, and the correlation analysis can be performed on the gas production and consumption and the technological parameters in advance, wherein the correlation analysis is mainly used for finding out the technological parameters with larger influence on the gas production and consumption from a large number of technological parameters, and introducing the technological parameters with higher correlation into the comprehensive prediction model as the technological parameters with higher correlation can effectively improve the accuracy of the comprehensive prediction model. The process parameters most relevant to the gas production consumption can be found out through calculation methods such as covariance, correlation coefficients and the like, and the process parameters with higher correlation are more helpful to improving the prediction accuracy.
Taking a steel rolling heating furnace to burn blast furnace gas to heat a steel billet as an example, referring to fig. 2, fig. 2 is a schematic diagram showing a gas consumption correlation analysis according to an embodiment of the present application. As shown in fig. 2, in the production process of heating steel billets by combusting blast furnace gas in the steel rolling heating furnace, the process parameters include steam recovery amount, steam recovery pressure, number of steel billets to be charged, steel billet specification, number of steel billets to be charged, temperature of steel billets to be charged, weight of steel billets to be charged, finished product specification, steel billet residence time, soaking section temperature, adding three sections of temperature, adding two sections of temperature and adding one section of temperature, wherein the process parameters such as number of steel billets to be charged, temperature of steel billets to be charged, weight of steel billets to be charged and the like have great influence on gas consumption, so that the process parameters such as number of steel billets to be charged, temperature of steel billets to be charged, weight of steel billets to be charged and the like can be introduced into the comprehensive prediction model as process parameters with higher correlation.
For different working procedures, the process parameter historical data of a certain working procedure is screened based on the correlation between the gas production consumption and the process parameter of the working procedure, and the process parameter historical data with higher correlation of the working procedure is obtained and is used as the target process parameter historical data of the working procedure. And generating a data sample set based on the gas production and consumption historical data and the target process parameter historical data so as to train the model. The data sample set can be divided into a training data sample set and a test data sample set, the model is trained by using the training data sample set, the trained model is tested by using the test data sample set, and the accuracy of model prediction is ensured.
In one embodiment of the present application, generating a data sample set based on gas production and consumption history data and target process parameter history data includes: taking the gas production and consumption historical data of each process and the target process parameter historical data of the process as data samples of the process to obtain data samples of different processes, and integrating the data samples of different processes to obtain a data sample set.
And step S130, training different types of initial models by using the data sample set to obtain a plurality of prediction models, and integrating the plurality of prediction models to obtain a comprehensive prediction model.
In one embodiment of the application, different types of initial models are built based on different prediction algorithms to form an algorithm pool, each initial model is trained by utilizing a data sample set, and the trained initial model is used as a prediction model to integrate a plurality of prediction models into a comprehensive prediction model. Compared with a single prediction model, the comprehensive prediction model formed by integrating a plurality of prediction models has better adaptability, has various complex working conditions with different blast furnace gas production and consumption characteristics, and has higher stability for accurately predicting the blast furnace gas than the single prediction model.
In one embodiment of the present application, the initial model includes at least one of a process mechanism model, a differentially integrated moving average autoregressive model, a back propagation neural network model, and a long and short memory neural network model.
In this embodiment, taking the blast furnace gas generation amount as an example, the training process of different algorithms is as follows:
1) Technological mechanism model prediction algorithm
According to the technological mechanism (conservation of materials and conservation of elements), a calculation formula can be obtained, the generation amount of blast furnace gas is a calculation function of the carbon monoxide, hydrogen, carbon dioxide, nitrogen components and blast furnace blast amount of the blast furnace gas at the top of a furnace, and a comprehensive prediction model of the generation amount of the blast furnace gas under the technological mechanism can be obtained through time dislocation and coefficient fitting, wherein the calculation formula of the technological mechanism is as follows:
Q Gas (gas) ＝f 1 (CO,H2,CO2,N2,Q Blowing air ) (1)
Wherein Q is Gas (gas) Is the generation amount of blast furnace gas, CO is the content of carbon monoxide, H2 is the content of hydrogen, CO2 is the content of carbon dioxide, N2 is the content of nitrogen, Q Blowing air Is blast furnace blast volume.
2) Time series model prediction algorithm
The time sequence model comprises MA (moving average mode, moving average model), ARIMA (Autoregressive Integrated Moving Average model, differential integration moving average autoregressive model) and the like, and the main mechanism is to predict according to the gas production and consumption historical data, and the formula is as follows:
wherein Q is Gas (gas) Is the generation amount of the blast furnace gas,
for the fitting coefficient at the time p, the fitting coefficient at the time p represents the influence factors at different times, Q t-p For the gas flow at the moment t-p epsilon t To fit the error term, t is the current time, t-p is the time p time units before the current time, p=0, 1,2,3, ….
3) BP (back propagation) neural network model prediction algorithm
The BP neural network model is the same as the time sequence model, the prediction is mainly carried out through gas production and consumption historical data, the training process mainly comprises the steps of deriving parameters in the BP neural network by utilizing errors, and continuously reducing training errors, wherein the formula is as follows:
Q Gas (gas) ＝f 3 (Q) A (3)
Wherein Q is Gas (gas) The generation amount of the blast furnace gas is Q, and the generation amount of the blast furnace gas is historical data.
4) LSTM neural network model prediction algorithm
The LSTM (Long Short-Term Memory) neural network can better grasp the characteristics of trend, period and the like of data change, and simultaneously put a plurality of groups of process parameter historical data into the neural network. The principle is similar to that of a BP neural network, but a large amount of process parameter historical data is introduced, the purpose of prediction is finally realized, and the formula can be written as follows:
Q gas (gas) ＝f 4 (Q,F Process parameter 1 ,F Process parameter 2 ,F Process parameter 3 ,…,F Process parameter n ) (4)
Wherein Q is Gas (gas) Is the blast furnace gas generation amount, Q is the blast furnace gas generation amount historical data, F Process parameter n N=1, 2,3 …, which is a history of the process parameter n.
The four predictive models described above are only examples of the present application and are not limiting. In addition, other prediction mechanism models can be introduced, and the more comprehensive prediction models are introduced and the different calculation mechanisms are, the better the combined use effect is.
In one embodiment of the present application, step S130 includes: and training different types of initial models by using the data samples of each process, integrating a plurality of prediction models obtained by training to obtain a plurality of comprehensive prediction models so as to predict the blast furnace gas production consumption of different processes, wherein the comprehensive prediction models correspond to the processes one by one, and the data sample set comprises the data samples of different processes.
In this embodiment, since the gas production and consumption characteristics of different processes are different, model training is required for different processes, specifically, for a certain process, a data sample of the process is selected from a data sample set, a plurality of initial models of different types are trained by using the data sample of the process, a plurality of prediction models corresponding to the process are obtained, a plurality of prediction models corresponding to the process are integrated, and a comprehensive prediction model corresponding to the process is obtained, so that the blast furnace gas production and consumption of the process is predicted by the comprehensive prediction model corresponding to the process. And by analogy, obtaining comprehensive prediction models of different procedures, wherein each comprehensive prediction model corresponds to each procedure one by one.
In one embodiment of the present application, after obtaining a plurality of integrated predictive models, the method includes: configuring weights for all the prediction models in the comprehensive prediction models corresponding to each procedure, and carrying out weight optimization based on an optimization algorithm to determine the optimal weight combination of each comprehensive prediction model; and predicting the blast furnace gas production and consumption of different working procedures based on the comprehensive prediction models corresponding to the different working procedures and the optimized weight combination of the comprehensive prediction models.
In this embodiment, for a certain process, the predictive algorithm models (predictive models) of different mechanisms in the algorithm pool are combined by means of weights to obtain a predictive model combination (comprehensive predictive model) of the process, and the combination form is as follows:
wherein f Combination of two or more kinds of materials F is a comprehensive prediction model n For predictive model n, w n For the weight of the prediction model n, n=1, 2,3 ….
Weight optimization refers to constantly changing the value of each weight w within a certain rule and range until a weight combination is found that minimizes the prediction error. And (3) carrying out weight optimization on the comprehensive prediction model by utilizing an optimization algorithm, finding out a weight combination with high prediction accuracy and good stability, and taking the weight combination as a preferable weight combination of the comprehensive prediction model so that the comprehensive prediction model predicts the blast furnace gas production consumption of the process by calling the preferable weight combination. The optimizing algorithm comprises, but is not limited to, global optimizing, particle swarm optimizing, genetic algorithm and the like. And similarly, determining the optimal weight combination of the comprehensive prediction models of different working procedures so as to predict the blast furnace gas production consumption of different working procedures.
In one embodiment of the present application, weight optimization is performed based on an optimization algorithm to determine a preferred weight combination for each comprehensive prediction model, comprising: determining the gas production and consumption characteristics of each process according to the gas production and consumption historical data of each process, wherein the historical data comprise the gas production and consumption historical data of different processes; for the comprehensive prediction model corresponding to each process, adjusting the weight value of each prediction model based on the gas production and consumption characteristics of the process to obtain a plurality of weight combinations of the comprehensive prediction model, and determining an evaluation algorithm corresponding to the process based on the gas production and consumption characteristics of the process so as to calculate the prediction error when the comprehensive prediction model calls each weight combination according to the evaluation algorithm corresponding to the process; if the prediction error is smaller than or equal to a preset threshold value, determining a weight combination corresponding to the prediction error as a preferable weight combination of the comprehensive prediction model; and if the prediction error is greater than the preset threshold, readjusting the weight value of each prediction model based on the gas production and consumption characteristics of the working procedure until the prediction error is less than or equal to the preset threshold.
In this embodiment, the process of weight optimization for the comprehensive prediction model includes: the weight value of each prediction model is adjusted to obtain a plurality of weight combinations, an adjustment condition can be preset before adjustment, for example, the sum of all weights is 1, and the weight change value cannot exceed a preset weight change threshold value each time, wherein the preset weight change threshold value can be 0.1, or can be set to other values according to actual needs. And (3) controlling the comprehensive prediction model to call a weight combination, inputting the test data sample set into the comprehensive prediction model to obtain a gas production and consumption prediction result, and calculating a prediction error when the comprehensive prediction model calls the weight combination by using an evaluation method through the gas production and consumption prediction result and a gas production and consumption actual result (gas production and consumption historical data). And similarly, obtaining the prediction error when the comprehensive prediction model calls each weight combination. The methods of evaluation include, but are not limited to, absolute error, relative error, variance, standard deviation, mean square error, and the like.
And comparing the prediction error corresponding to each weight combination with a preset threshold value, and taking the weight combination as a preferable weight combination if the prediction error corresponding to one weight combination is smaller than or equal to the preset threshold value. And if the prediction errors corresponding to all the weight combinations are larger than the preset threshold, readjusting the weight values of the prediction models in the comprehensive prediction model until the prediction errors are smaller than or equal to the preset threshold. The process of readjusting the weight values can be to screen out a weight combination with smaller half of the prediction errors from a plurality of weight combinations as a basic weight combination, and adjust each weight value in the basic weight combination to iterate according to a preset adjustment condition to obtain a plurality of different new weight combinations so as to calculate the prediction errors when the comprehensive prediction model calls different new weight combinations until the new weight combination with the prediction errors smaller than or equal to a preset threshold value appears. The preset threshold value may be 10% or may be set to other values according to actual needs.
In addition, the number of times of iteration of each weight value in the basic weight combination can be statistically adjusted, if the iteration is repeated for a plurality of times, but no new weight combination with the prediction error smaller than or equal to the preset threshold value still exists, the number of times of iteration is compared with the preset iteration threshold value, and when the number of times of iteration reaches the preset iteration threshold value, the new weight combination with the minimum prediction error in a plurality of new weight combinations obtained by the current iteration is used as the optimal weight combination. The preset iteration threshold may be 100, or may be set to other values according to actual needs.
Because the production characteristics of each process unit (process for short) of the iron and steel enterprise are different, the coal gas production and consumption data characteristics (coal gas production and consumption characteristics) of the process units are also greatly different, so that the coal gas production and consumption data characteristics of the different process units need to be subjected to weight optimization so as to obtain the optimal weight combination of the comprehensive prediction model for the different process units.
Firstly, acquiring gas production and consumption historical data of different working procedures, and further determining gas production and consumption characteristics of different working procedures. Referring to fig. 3, fig. 3 is a wave diagram showing the amount of blast furnace gas generation according to an embodiment of the present application. As shown in fig. 3, the horizontal coordinate is a time axis, the vertical coordinate is BFG (blast furnace gas ) generation amount, and the blast furnace gas generation amount fluctuates up and down at different times, but the fluctuation is relatively stable, so the gas production consumption of the process is characterized by stable fluctuation. Referring to fig. 4, fig. 4 is a graph showing the fluctuation of the blast furnace stove consumption according to an embodiment of the present application. As shown in fig. 4, the horizontal coordinate is a time axis, the vertical coordinate is BFG flow, and the fluctuation of BFG flow at different moments in the blast furnace hot blast stove process unit tends to be stable and changes rapidly, but the fluctuation changes periodically, so the gas production and consumption characteristics of the process are periodically fluctuated. Referring to fig. 5, fig. 5 is a graph showing the fluctuation of the gas consumption of the stick according to an embodiment of the present application. As shown in fig. 5, the horizontal coordinate is a time axis, the vertical coordinate is a BFG flow rate, and the fluctuation of the BFG flow rate at different times is random in a small rod (small rod heating furnace) process unit, so that the gas production and consumption characteristics of the process are rapid fluctuation.
Different characteristics (gas production and consumption characteristics)) The blast furnace gas production and consumption prediction is carried out by carrying out different weight optimization. For different working procedures, the weight value of each prediction model is adjusted based on the gas production and consumption characteristics of the working procedures, a plurality of different weight combinations of the comprehensive prediction model are obtained, and an evaluation algorithm corresponding to the working procedures is determined based on the gas production and consumption characteristics of the working procedures so as to perform weight optimization. Taking a particle swarm algorithm as an example: referring to fig. 6, fig. 6 is a schematic diagram illustrating weight optimization using a particle swarm algorithm according to an embodiment of the present application. As shown in FIG. 6, first, the first generation individuals are initialized to form coverage ownership weight values, m individuals in total, and the weight value of each individual is w 1 、w 2 、w 3 、……、w n Calculating a corresponding prediction error under each individual weight according to an evaluation algorithm corresponding to the procedure; then, eliminating half of individuals with the largest errors, and changing the weights in half of individuals with smaller errors to form new generation individuals; iterating the steps until the prediction error is smaller than or equal to a preset threshold value or the iteration number reaches an upper limit value, and obtaining an optimal weight combination of the comprehensive prediction model of the procedure; the preset adjustment conditions described in the above embodiments are satisfied when the weights are changed. According to the weight optimizing step, the optimal weight combination of the comprehensive prediction model in different procedures can be obtained.
In one embodiment of the present application, an evaluation algorithm corresponding to a process is determined based on a gas production and consumption characteristic of the process, including: if the gas production and consumption characteristics of the working procedure are suddenly fluctuated, taking the mean square error as an evaluation algorithm corresponding to the working procedure; if the gas production and consumption characteristics of the working procedure are stable fluctuation, the relative error is used as an evaluation algorithm corresponding to the working procedure.
In this embodiment, different accuracy calculation methods (evaluation methods) are selected according to actual use requirements, for example, data (gas production and consumption historical data) with large fluctuation (rapid fluctuation) are required to be predicted accurately at each time point, that is, mean square error can be selected, and the error between a predicted value (gas production and consumption prediction result) and an actual value (and gas production and consumption actual result) at each time point is evaluated. For data with smoother fluctuation, the total amount in the whole time period of the prediction period needs to be accurately predicted, and the absolute error percentage, namely the relative error, can be selected.
And step S140, predicting the production and consumption of the blast furnace gas based on the comprehensive prediction model.
In one embodiment of the application, the current process parameter data of a certain process is acquired, the current process parameter data is input into the comprehensive prediction model of the process, so that the comprehensive prediction model predicts the production and consumption of the blast furnace gas of the process, and the output blast furnace gas production and consumption prediction result has higher stability than that of a single prediction model, can adapt to the blast furnace gas production and consumption prediction with different characteristics, and ensures the accuracy of the prediction.
In one embodiment of the present application, predicting blast furnace gas production and consumption of different processes based on the comprehensive prediction model and the preferred weight combination of the comprehensive prediction model corresponding to the different processes includes: acquiring current process parameter data of different procedures; and inputting the current technological parameter data of each working procedure into a comprehensive prediction model corresponding to the working procedure, and calling the optimal weight combination of the comprehensive prediction model to obtain the blast furnace gas production and consumption prediction results of different working procedures.
According to the technical scheme, as many prediction algorithm models (prediction models) with different mechanisms as possible are adopted in the algorithm pool, and according to different coal gas production and consumption data characteristics, the optimizing process can increase the weight of the prediction model which is suitable for the coal gas production and consumption data characteristics in the algorithm pool as much as possible so as to reduce prediction errors; the applicability of the comprehensive prediction model is greatly improved. In the optimization calculation method preset in the weight optimization process, a computer system can directly run without manually screening and comparing algorithms, so that a great deal of manpower and manpower time are saved. In general, a great deal of training is carried out on each initial model according to the gas production and consumption historical data and the process parameter historical data of each working procedure, and the prediction models of different mechanisms are combined in a weight optimizing mode, so that the accuracy of prediction is higher than that of a single comprehensive prediction model.
Referring to fig. 7, fig. 7 is a simplified flow chart illustrating integrated predictive model creation in accordance with an embodiment of the present application. As shown in fig. 7, the flow of the integrated prediction model establishment is as follows:
1) Historical data reading: reading gas production consumption historical data and process production parameter historical data (process parameter historical data);
2) Correlation analysis and data screening: pre-analyzing the correlation between the gas production and consumption and the technological parameters, and screening the technological parameter historical data based on the correlation to obtain technological parameter historical data with larger influence on the gas production and consumption, namely target technological parameter historical data;
3) And (3) training a prediction model: generating a data sample set based on the gas production and consumption historical data and the target process parameter historical data to train a prediction model (referred to as an initial model) of a plurality of different algorithms, wherein the prediction model comprises, but is not limited to, a process mechanism model, a differential integration moving average autoregressive model, a back propagation neural network model and a long and short memory neural network model;
4) Prediction model combination: after model training is completed, configuring weights for the prediction models of different algorithms, so as to combine all the prediction models to obtain a comprehensive prediction model, wherein the combination mode is shown in a formula (5);
5) Determining model prediction accuracy evaluation parameters (evaluation methods): determining gas production and consumption characteristics of different working procedures according to the gas production and consumption historical data of the different working procedures, and selecting corresponding evaluation calculation methods (evaluation methods) of the different working procedures according to the gas production and consumption characteristics of the different working procedures to calculate prediction errors when the comprehensive prediction model calls each weight set;
6) Weight optimizing: according to the gas production and consumption characteristics of different working procedures and corresponding evaluation methods, carrying out weight optimization on comprehensive prediction models of blast furnace gas production and consumption units (short for different working procedures) of different working procedures, and finding weight combinations applicable to the gas production and consumption characteristics of different working procedures;
7) And outputting the blast furnace gas production and consumption prediction models of different working procedures, namely outputting the weight combination of the comprehensive prediction models applicable to different working procedures as the preferable weight combination of the comprehensive prediction models of different working procedures. The output comprehensive prediction model is actually a weight combination with the smallest prediction error for different procedures when all the prediction models in the algorithm pool are combined, and after the weight combination is output, the weight combination is actually called each time the comprehensive prediction model is called, and the prediction is performed after the prediction models in the algorithm pool are integrated according to the weight combination.
For the detailed building process of the comprehensive prediction model, please refer to the descriptions in the foregoing embodiments, and no further description is given here.
Referring to fig. 8, fig. 8 is a block diagram illustrating a blast furnace gas production and consumption prediction apparatus according to an exemplary embodiment of the present application. As shown in fig. 8, the exemplary blast furnace gas production and consumption prediction apparatus includes:
an acquisition module 810 for acquiring historical data, wherein the historical data comprises gas production consumption historical data and process parameter historical data; the sample generation module 820 is configured to screen the process parameter history data based on the correlation between the gas production and consumption and the process parameter to obtain target process parameter history data, and generate a data sample set based on the gas production and consumption history data and the target process parameter history data; the training module 830 is configured to train different types of initial models by using the data sample set to obtain a plurality of prediction models, and integrate the plurality of prediction models to obtain a comprehensive prediction model; and the prediction module 840 is used for predicting the blast furnace gas production and consumption based on the comprehensive prediction model.
It should be noted that, the blast furnace gas production and consumption prediction apparatus provided in the foregoing embodiments and the blast furnace gas production and consumption prediction method provided in the foregoing embodiments belong to the same concept, and specific manners in which each module and unit perform operations have been described in detail in the method embodiments, which are not repeated herein. In practical application, the blast furnace gas production and consumption prediction device provided in the above embodiment may distribute the functions to different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above, which is not limited herein.
The embodiment also provides an electronic device, including: one or more processors; and the storage device is used for storing one or more programs, and when the one or more programs are executed by the one or more processors, the electronic equipment realizes the blast furnace gas production and consumption prediction method provided in each embodiment.
The present embodiment also provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor of a computer, causes the computer to perform the blast furnace gas production and consumption prediction method as described above. The computer-readable storage medium may be included in the electronic device described in the above embodiment or may exist alone without being incorporated in the electronic device.
The present embodiments also provide a computer program product or computer program comprising computer instructions stored in a computer readable storage medium. The processor of the computer apparatus reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer apparatus performs the blast furnace gas production and consumption prediction method provided in the above embodiments.
The electronic device provided in this embodiment includes a processor, a memory, a transceiver, and a communication interface, where the memory and the communication interface are connected to the processor and the transceiver and perform communication therebetween, the memory is used to store a computer program, the communication interface is used to perform communication, and the processor and the transceiver are used to run the computer program, so that the electronic device performs each step of the above method.
In this embodiment, the memory may include a random access memory (Random Access Memory, abbreviated as RAM), and may further include a non-volatile memory (non-volatile memory), such as at least one magnetic disk memory.
The processor may be a general-purpose processor, including a central processing unit (Central Processing Unit, CPU for short), a network processor (Network Processor, NP for short), etc.; but also digital signal processors (Digital Signal Processing, DSP for short), application specific integrated circuits (Application Specific Integrated Circuit, ASIC for short), field-programmable gate arrays (Field-Programmable Gate Array, FPGA for short) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.
The computer readable storage medium in this embodiment, as will be appreciated by those of ordinary skill in the art: all or part of the steps for implementing the method embodiments described above may be performed by computer program related hardware. The aforementioned computer program may be stored in a computer readable storage medium. The program, when executed, performs steps including the method embodiments described above; and the aforementioned storage medium includes: various media capable of storing program codes, such as ROM (read only memory), RAM (random access memory), magnetic disk or optical disk.
The above embodiments are merely illustrative of the principles of the present application and its effectiveness and are not intended to limit the present application. Modifications and variations may be made to the above-described embodiments by those of ordinary skill in the art without departing from the spirit and scope of the present application. It is therefore contemplated that the appended claims will cover all such equivalent modifications and changes as fall within the true spirit and scope of the disclosure.
Claims (10)
1. A method for predicting blast furnace gas production and consumption, the method comprising:
acquiring historical data, wherein the historical data comprises gas production and consumption historical data and process parameter historical data;
Screening the process parameter historical data based on the correlation between the gas production and consumption and the process parameter to obtain target process parameter historical data, and generating a data sample set based on the gas production and consumption historical data and the target process parameter historical data;
training different types of initial models by using the data sample set to obtain a plurality of prediction models, and integrating the plurality of prediction models to obtain a comprehensive prediction model;
and predicting the blast furnace gas production and consumption based on the comprehensive prediction model.
2. The method for predicting the production and consumption of blast furnace gas according to claim 1, wherein training different types of initial models by using the data sample set to obtain a plurality of prediction models, integrating the plurality of prediction models to obtain a comprehensive prediction model, comprises:
and training different types of initial models by using data samples of each working procedure, integrating a plurality of prediction models obtained by training to obtain a plurality of comprehensive prediction models so as to predict the blast furnace gas production consumption of different working procedures, wherein the comprehensive prediction models are in one-to-one correspondence with the working procedures, and the data sample sets comprise data samples of different working procedures.
3. The method for predicting the production and consumption of blast furnace gas according to claim 2, wherein after obtaining a plurality of comprehensive prediction models, the method comprises:
configuring weights for all the prediction models in the comprehensive prediction models corresponding to each procedure, and carrying out weight optimization based on an optimization algorithm to determine the optimal weight combination of each comprehensive prediction model;
and predicting the blast furnace gas production and consumption of different working procedures based on the comprehensive prediction models corresponding to the different working procedures and the optimized weight combination of the comprehensive prediction models.
4. A method of predicting blast furnace gas production and consumption according to claim 3, wherein the weight optimizing based on the optimizing algorithm to determine the preferred weight combination for each comprehensive prediction model comprises:
determining the gas production and consumption characteristics of each process according to the gas production and consumption historical data of each process, wherein the historical data comprise the gas production and consumption historical data of different processes;
for the comprehensive prediction model corresponding to each process, adjusting the weight value of each prediction model based on the gas production and consumption characteristics of the process to obtain a plurality of weight combinations of the comprehensive prediction model, and determining an evaluation algorithm corresponding to the process based on the gas production and consumption characteristics of the process to calculate the prediction error when the comprehensive prediction model calls each weight combination according to the evaluation algorithm corresponding to the process;
If the prediction error is smaller than or equal to a preset threshold value, determining a weight combination corresponding to the prediction error as a preferred weight combination of the comprehensive prediction model;
and if the prediction error is larger than the preset threshold, readjusting the weight value of each prediction model based on the gas production and consumption characteristics of the working procedure until the prediction error is smaller than or equal to the preset threshold.
5. The blast furnace gas production and consumption prediction method according to claim 4, wherein determining an evaluation algorithm corresponding to the process based on the gas production and consumption characteristics of the process comprises:
if the gas production and consumption characteristics of the working procedure are suddenly fluctuated, taking the mean square error as an evaluation algorithm corresponding to the working procedure;
and if the gas production and consumption characteristics of the working procedure are stable fluctuation, the relative error is used as an evaluation algorithm corresponding to the working procedure.
6. The blast furnace gas production and consumption prediction method according to any one of claims 3 to 5, wherein predicting blast furnace gas production and consumption of different processes based on the comprehensive prediction model corresponding to different processes and the preferable weight combination of the comprehensive prediction model comprises:
acquiring current process parameter data of different procedures;
And inputting the current technological parameter data of each working procedure into a comprehensive prediction model corresponding to the working procedure, and calling the optimal weight combination of the comprehensive prediction model to obtain the blast furnace gas production and consumption prediction results of different working procedures.
7. The blast furnace gas production and consumption prediction method according to any one of claims 1 to 5, wherein the initial model comprises at least one of a process mechanism model, a differentially integrated moving average autoregressive model, a counter-propagating neural network model, and a long and short memory neural network model.
8. A blast furnace gas production and consumption prediction device, the device comprising:
the acquisition module is used for acquiring historical data, wherein the historical data comprises gas production consumption historical data and process parameter historical data;
the sample generation module is used for screening the process parameter historical data based on the correlation between the gas production consumption and the process parameter to obtain target process parameter historical data, and generating a data sample set based on the gas production consumption historical data and the target process parameter historical data;
the training module is used for training different types of initial models by utilizing the data sample set to obtain a plurality of prediction models, and integrating the plurality of prediction models to obtain a comprehensive prediction model;
And the prediction module is used for predicting the blast furnace gas production and consumption based on the comprehensive prediction model.
9. An electronic device, the electronic device comprising:
one or more processors;
storage means for storing one or more programs that, when executed by the one or more processors, cause the electronic device to implement the blast furnace gas production and consumption prediction method of any one of claims 1 to 7.
10. A computer-readable storage medium, having stored thereon a computer program which, when executed by a processor of a computer, causes the computer to perform the blast furnace gas production and consumption prediction method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310402188.1A CN116434859A (en) | 2023-04-13 | 2023-04-13 | Blast furnace gas production consumption prediction method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310402188.1A CN116434859A (en) | 2023-04-13 | 2023-04-13 | Blast furnace gas production consumption prediction method and device, electronic equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116434859A true CN116434859A (en) | 2023-07-14 |
Family
ID=87079367
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310402188.1A Pending CN116434859A (en) | 2023-04-13 | 2023-04-13 | Blast furnace gas production consumption prediction method and device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116434859A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113177369A (en) * | 2021-06-15 | 2021-07-27 | 中冶赛迪技术研究中心有限公司 | Energy scheduling evaluation method and system |
| CN116841269A (en) * | 2023-07-26 | 2023-10-03 | 枣庄杰富意振兴化工有限公司 | Process adjustment method, system and storage medium based on coal tar production flow |
| CN117556968A (en) * | 2024-01-11 | 2024-02-13 | 天津瑞远粉末涂料有限公司 | Powder coating preparation method and device, electronic equipment and storage medium |
-
2023
- 2023-04-13 CN CN202310402188.1A patent/CN116434859A/en active Pending
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113177369A (en) * | 2021-06-15 | 2021-07-27 | 中冶赛迪技术研究中心有限公司 | Energy scheduling evaluation method and system |
| CN113177369B (en) * | 2021-06-15 | 2024-03-01 | 中冶赛迪技术研究中心有限公司 | Energy scheduling evaluation method and system |
| CN116841269A (en) * | 2023-07-26 | 2023-10-03 | 枣庄杰富意振兴化工有限公司 | Process adjustment method, system and storage medium based on coal tar production flow |
| CN116841269B (en) * | 2023-07-26 | 2024-01-23 | 枣庄杰富意振兴化工有限公司 | Process adjustment method, system and storage medium based on coal tar production flow |
| CN117556968A (en) * | 2024-01-11 | 2024-02-13 | 天津瑞远粉末涂料有限公司 | Powder coating preparation method and device, electronic equipment and storage medium |
| CN117556968B (en) * | 2024-01-11 | 2024-04-09 | 天津瑞远粉末涂料有限公司 | Powder coating preparation method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116434859A (en) | Blast furnace gas production consumption prediction method and device, electronic equipment and storage medium | |
| CN109062053B (en) | Denitration ammonia injection control method based on multivariate correction | |
| WO2021189739A1 (en) | Production plan-based iron and steel plant oxygen load prediction method | |
| CN111652355B (en) | Method and device for predicting silicon content of blast furnace molten iron based on LSTM and DNN | |
| CN104270281A (en) | Web Service QoS (Quality of Service) combined forecasting method | |
| CN104915518A (en) | Establishing method and application of two-dimensional prediction model of silicon content in hot metal in blast furnace | |
| CN104866923A (en) | Steel enterprise blast furnace by-product gas emergence size prediction method | |
| CN113011660A (en) | Air quality prediction method, system and storage medium | |
| CN111523730A (en) | Method for predicting coal consumption of cogeneration unit | |
| CN108154266B (en) | Economic dispatching method for wind power-containing power system | |
| CN110849149A (en) | Energy perception-based sintering batching scheme cascade optimization obtaining method and device | |
| CN110222910A (en) | A kind of active power distribution network Tendency Prediction method and forecasting system | |
| CN107545321A (en) | A kind of ARMA RBF by-product gas generating capacity combination forecasting methods based on wavelet transformation | |
| CN103983332A (en) | Method for error compensation of sensor based on HGSA-BP algorithm | |
| CN117052457A (en) | Tunnel construction dust treatment control method, system, equipment and medium | |
| Chen et al. | Research on prediction methods of energy consumption data | |
| CN116384254A (en) | Error correction method and system for predicting molten steel temperature of LF refining furnace | |
| CN112553390A (en) | Coal injection optimization control method based on Gaussian process regression | |
| CN109506248B (en) | Case reasoning boiler combustion optimization method based on online optimization | |
| CN111680720A (en) | Blast furnace molten iron silicon content prediction method based on improved CS-SVR model | |
| Jin et al. | A causal model-based scheduling approach for coke oven gas system in steel industry | |
| CN116127345B (en) | Converter steelmaking process mode design method based on deep clustering generation countermeasure network | |
| CN111860923A (en) | Boiler flue gas NO based on multi-model clustering integrationxEmission prediction algorithm | |
| CN112347695B (en) | Method and system for predicting top layer oil temperature of transformer | |
| CN113806996B (en) | Grate cooler optimization method based on time memory modeling |
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 |