CN103559543A - Method and device for predicting blast furnace gas occurrence amount - Google Patents
Method and device for predicting blast furnace gas occurrence amount Download PDFInfo
- Publication number
- CN103559543A CN103559543A CN201310566822.1A CN201310566822A CN103559543A CN 103559543 A CN103559543 A CN 103559543A CN 201310566822 A CN201310566822 A CN 201310566822A CN 103559543 A CN103559543 A CN 103559543A
- Authority
- CN
- China
- Prior art keywords
- blast furnace
- furnace gas
- gas generating
- generating capacity
- neural network
- 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
Images
Landscapes
- Feedback Control In General (AREA)
Abstract
The invention discloses a method and device for predicting a blast furnace gas occurrence amount. The method includes the steps of obtaining blast furnace condition information, intensified smelting information and blast furnace gas occurrence amount data, setting up a BP neural network, selecting an input variable of the neural network according to the blast furnace condition information and the intensified smelting information, dividing the input variable into a first training set and a first test set, taking the blast furnace gas occurrence amount data as output of the neural network, dividing the blast furnace gas occurrence amount data into a second training set and a second test set, taking an input variable in the first training set and a blast furnace gas occurrence amount in the second training set as input data of the neural network and output data of the neural network respectively to conduct model training till the neural network is converged, inputting an input variable in the first test set into the converged neural network, outputting the predicted blast furnace gas occurrence amount, comparing the predicted blast furnace gas occurrence amount with a blast furnace gas occurrence amount in the second test set, and testing the prediction effect. According to the method, the blast furnace gas occurrence amount can be accurately predicted.
Description
Technical field
The present invention relates to coal gas prediction field, relate in particular a kind of Forecasting Methodology and device of blast furnace gas generating capacity.
Background technology
Iron and steel enterprise is the industry of high energy consumption, high pollution, maximum discharge, and energy-saving and cost-reducing be one of Tough questions of facing of steel industry always.The primary energy of producing consumption due to iron and steel has 40% left/right rotation to become by-product gas, wherein blast furnace gas accounts for 45%, can the by-product gas therefore, producing in steel manufacture process obtain reasonably utilizing and will directly having influence on energy consumption cost and the effects of energy saving and emission reduction of whole smelter.
Reasonably gas balance scheduling can reduce the amount of diffusing, and reduces the use amount of primary energy, reduces the production cost of iron and steel enterprise, and the generating capacity of Accurate Prediction blast furnace gas is the basis of gas balance scheduling.In actual production at present, the prediction of emergence size of smelter blast furnace gas still be take to static prediction as main, static prediction is mainly with the plan of blast furnace gas system and is planned to foundation, generating capacity to blast furnace gas in a period of time, in conjunction with during the factors such as the production schedule, turnaround plan or technological transformation project, simply predict.
Therefore the generating capacity that, the Forecasting Methodology of existing blast furnace gas generating capacity can not Accurate Prediction blast furnace gas.
Summary of the invention
In view of this, in order to solve the problem that the Forecasting Methodology of existing blast furnace gas generating capacity can not Accurate Prediction blast furnace gas generating capacity, provide a kind of Forecasting Methodology and device of blast furnace gas generating capacity, technical scheme is as follows:
A Forecasting Methodology for blast furnace gas generating capacity, comprising:
Obtain conditions of blast furnace information, strengthening smelting information, blast furnace gas generating capacity data;
Build BP neural network;
According to described conditions of blast furnace information, strengthening smelting information, choose the input variable of described BP neural network;
Described input variable is divided into the first training set and the first test set;
Output using described blast furnace gas generating capacity data as described BP neural network, is divided into the second training set and the second test set;
Inputoutput data using the blast furnace gas generating capacity in the input variable in described the first training set and described the second training set as described BP neural network carries out model training, until the convergence of described BP neural network;
Input variable in described the first test set is inputted to the BP neural network after described convergence, the blast furnace gas generating capacity of the prediction of output;
By the blast furnace gas generating capacity comparison in the blast furnace gas generating capacity of described prediction and described the second test set, check prediction effect.
Preferably, in the Forecasting Methodology of above-mentioned blast furnace gas generating capacity, also comprise:
Utilize interpolation method to ask for substituting of missing data and improper data to described blast furnace gas generating capacity data and described input variable;
Blast furnace gas generating capacity after described interpolation method is processed is carried out to Kalman filtering.
Preferably, in the Forecasting Methodology of above-mentioned blast furnace gas generating capacity, according to described conditions of blast furnace information, strengthening smelting information, choose the input variable of described BP neural network, comprising:
Described conditions of blast furnace information, strengthening smelting information are carried out to Linear correlative analysis and grey relational grade analysis successively, choose there is the strong degree of association working of a furnace factor as input variable;
According to described strengthening smelting information, choose Rich Oxygen Amount, injecting coal quantity as input variable.
Preferably, in the Forecasting Methodology of above-mentioned blast furnace gas generating capacity, described structure BP neural network, comprising:
Determine the structure of BP neural network, comprise the hidden layer number of plies, input layer, hidden layer, each node layer number of output layer;
Determine that neuron connects weights and neuron threshold value.
Preferably, in the Forecasting Methodology of above-mentioned blast furnace gas generating capacity, also comprise:
Respectively the input variable in described the first training set and the first test set is normalized;
Respectively the blast furnace gas generating capacity in described the second training set and the second test set is normalized;
Blast furnace gas generating capacity to the described prediction of output carries out renormalization processing, the blast furnace gas generating capacity after being predicted.
A prediction unit for blast furnace gas generating capacity, comprising:
Acquiring unit, for obtaining conditions of blast furnace information, strengthening smelting information, blast furnace gas generating capacity data;
Neural network construction unit, for building BP neural network;
Input variable is chosen unit, for choose the input variable of described BP neural network according to described conditions of blast furnace information, strengthening smelting information;
The first division unit, for being divided into described input variable the first training set and the first test set;
The second division unit, the output for using described blast furnace gas generating capacity data as described BP neural network, is divided into the second training set and the second test set;
Training unit, carries out model training for the inputoutput data using the blast furnace gas generating capacity in the input variable of described the first training set and described the second training set as described BP neural network, until the convergence of described BP neural network;
Processing unit, for inputting the input variable of described the first test set the BP neural network after described convergence, the blast furnace gas generating capacity of the prediction of output;
Verification unit, for by the blast furnace gas generating capacity comparison of the blast furnace gas generating capacity of described prediction and described the second test set, checks prediction effect.
Preferably, in the prediction unit of above-mentioned blast furnace gas generating capacity, also comprise:
Data processing unit, for utilizing interpolation method to ask for substituting of missing data and improper data to described blast furnace gas generating capacity data and described input variable;
Filter unit, carries out Kalman filtering for the blast furnace gas generating capacity to after described interpolation method is processed.
Preferably, in the prediction unit of above-mentioned blast furnace gas generating capacity, described input variable is chosen unit, comprising:
First chooses module, for the working of a furnace factor of described conditions of blast furnace information, strengthening smelting information being carried out to Linear correlative analysis and grey relational grade analysis successively, choose thering is the strong degree of association as input variable;
Second chooses module, for according to described strengthening smelting information, chooses Rich Oxygen Amount, injecting coal quantity as input variable.
Preferably, in the prediction unit of above-mentioned blast furnace gas generating capacity, described neural network construction unit, comprising:
The first determination module, for determining the structure of BP neural network, comprises the hidden layer number of plies, input layer, hidden layer, each node layer number of output layer;
The second determination module, for determining that neuron connects weights and neuron threshold value.
Preferably, in the prediction unit of above-mentioned blast furnace gas generating capacity, also comprise:
The first normalized module, for being normalized the input variable of described the first training set and the first test set respectively;
The second normalized module, for being normalized the blast furnace gas generating capacity of described the second training set and the second test set respectively;
Renormalization processing module, for carrying out renormalization processing, the blast furnace gas generating capacity after being predicted to the blast furnace gas generating capacity of the described prediction of output.
In technique scheme, there is following beneficial effect:
Known via above-mentioned technical scheme, compared with prior art, the blast furnace gas premeasuring method that the embodiment of the present invention provides, by building BP neural network, the input variable of choosing and blast furnace gas generating capacity data are divided into respectively to training set and test set, inputoutput data using the data of training set as BP neural network carries out model training, until BP neural network convergence, by the BP neural network after the input variable input convergence in test set, the blast furnace gas generating capacity of the prediction of output, by the blast furnace gas generating capacity of prediction and the blast furnace gas generating capacity comparison in test set, check prediction effect, therefore, can predict accurately blast furnace gas generating capacity.
Accompanying drawing explanation
In order to be illustrated more clearly in the embodiment of the present invention or technical scheme of the prior art, to the accompanying drawing of required use in embodiment or description of the Prior Art be briefly described below, apparently, accompanying drawing in the following describes is only embodiments of the invention, for those of ordinary skills, do not paying under the prerequisite of creative work, other accompanying drawing can also be provided according to the accompanying drawing providing.
A kind of schematic flow sheet of the blast furnace gas prediction of emergence size method that Fig. 1 provides for the embodiment of the present invention;
Another schematic flow sheet of the blast furnace gas prediction of emergence size method that Fig. 2 provides for the embodiment of the present invention;
Another schematic flow sheet of the blast furnace gas prediction of emergence size method that Fig. 3 provides for the embodiment of the present invention;
A kind of structural representation of the blast furnace gas prediction of emergence size device that Fig. 4 provides for the embodiment of the present invention.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is clearly and completely described, obviously, described embodiment is only the present invention's part embodiment, rather than whole embodiment.Embodiment based in the present invention, those of ordinary skills, not making the every other embodiment obtaining under creative work prerequisite, belong to the scope of protection of the invention.
Referring to Fig. 1, the embodiment of the present invention provides a kind of Forecasting Methodology of blast furnace gas generating capacity, comprising:
Step 110: obtain conditions of blast furnace information, strengthening smelting information, blast furnace gas generating capacity data.
Step 120: build BP neural network.
Step 130: the input variable of choosing BP neural network according to conditions of blast furnace information, strengthening smelting information.
Conditions of blast furnace information, strengthening smelting information are carried out to Linear correlative analysis and grey relational grade analysis successively, choose there is the strong degree of association working of a furnace factor as input variable, further, choose and there are fuel ratio, the sintering number percent of the strong degree of association, the air quantity of blowing, air-supply wind-warm syndrome, blast pressure, 6 working of a furnace factors of coke ratio as neural network input variable;
According to described strengthening smelting information, choose Rich Oxygen Amount, injecting coal quantity as input variable.
Step 140: input variable is divided into the first training set and the first test set.
Step 150: the output using blast furnace gas generating capacity data as BP neural network, is divided into the second training set and the second test set.
Step 160: the inputoutput data using the blast furnace gas generating capacity in the input variable in the first training set and the second training set as BP neural network carries out model training, until the convergence of BP neural network.
Step 170: by the BP neural network after the input variable input convergence in the first test set, the blast furnace gas generating capacity of the prediction of output.
Step 180: by the blast furnace gas generating capacity of prediction and the blast furnace gas generating capacity comparison in the second test set, check prediction effect.
The blast furnace gas premeasuring method that the embodiment of the present invention provides, by building BP neural network, the input variable of choosing and blast furnace gas generating capacity data are divided into respectively to training set and test set, inputoutput data using the data of training set as BP neural network carries out model training, until BP neural network convergence, by the BP neural network after the input variable input convergence in test set, the blast furnace gas generating capacity of the prediction of output, by the blast furnace gas generating capacity of prediction and the blast furnace gas generating capacity comparison in test set, check prediction effect, therefore, can predict accurately blast furnace gas generating capacity.
Referring to Fig. 2, the embodiment of the present invention provides a kind of Forecasting Methodology of blast furnace gas generating capacity, comprising:
Step 210: obtain conditions of blast furnace information, strengthening smelting information, blast furnace gas generating capacity data.
Step 220: build BP neural network.
Step 230: the input variable of choosing BP neural network according to conditions of blast furnace information, strengthening smelting information.
Conditions of blast furnace information, strengthening smelting information are carried out to Linear correlative analysis and grey relational grade analysis successively, choose there is the strong degree of association working of a furnace factor as input variable, further, choose and there are fuel ratio, the sintering number percent of the strong degree of association, the air quantity of blowing, air-supply wind-warm syndrome, blast pressure, 6 working of a furnace factors of coke ratio as neural network input variable;
According to described strengthening smelting information, choose Rich Oxygen Amount, injecting coal quantity as input variable.
Step 240: input variable is divided into the first training set and the first test set.
Step 250: the output using blast furnace gas generating capacity data as BP neural network, is divided into the second training set and the second test set.
Step 260: utilize interpolation method to ask for substituting of missing data and improper data to blast furnace gas generating capacity data and input variable, blast furnace gas generating capacity after interpolation method is processed is carried out to Kalman filtering, inputoutput data using the blast furnace gas generating capacity in the input variable in the first training set and the second training set as BP neural network carries out model training, until the convergence of BP neural network.
Because signals collecting is abnormal, network service failure and other reasons, may cause some shortage of data in blast furnace gas generating capacity data and input variable, or some data variation is abnormal, as excessive or too small.For such abnormal data, need on the basis of qualitative analysis, be processed.In the present embodiment, adopt interpolation method to ask for the substitution value of missing data and improper data, further, utilize interpolation method to ask for substituting of missing data and improper data to blast furnace gas generating capacity data and input variable, then the blast furnace gas generating capacity after interpolation method is processed is carried out to Kalman filtering, obtain normal data.
Utilizing interpolation method to ask for substituting of missing data and improper data to blast furnace gas generating capacity data and input variable can utilize following formula to carry out:
x
0(i)=x
0(k)+[x
0(j)-x
0(k)]·(i-k)/(j-k);
Wherein:
X
0(k), x
0(j) be known normal data, known normal data represents normal given data before and after being replaced by of missing data or abnormal data;
X
0(i) be substituting of missing data or improper data;
k<i<j;
I represents missing data or the abnormal data sequence number in data sequence;
J, k represent the sequence number of given data in data sequence, the interpolation of normal given data before and after missing data or abnormal data are replaced by.
Step 270: by the BP neural network after input variable input convergence in the first test set, the blast furnace gas generating capacity of the prediction of output.
Step 280: by the blast furnace gas generating capacity of prediction and the blast furnace gas generating capacity comparison in the second test set, check prediction effect.
The blast furnace gas premeasuring method that the embodiment of the present invention provides, by building BP neural network, the input variable of choosing and blast furnace gas generating capacity data are divided into respectively to training set and test set, inputoutput data using the data of training set as BP neural network carries out model training, until BP neural network convergence, by the BP neural network after the input variable input convergence in test set, the blast furnace gas generating capacity of the prediction of output, by the blast furnace gas generating capacity of prediction and the blast furnace gas generating capacity comparison in test set, check prediction effect, therefore, can predict accurately blast furnace gas generating capacity.
Further, in the present embodiment, adopt interpolation method to ask for the substitution value of missing data and improper data, can obtain accurate inputoutput data, effectively raise precision of prediction.
Referring to Fig. 3, the embodiment of the present invention provides a kind of Forecasting Methodology of blast furnace gas generating capacity, comprising:
Step 310: obtain conditions of blast furnace information, strengthening smelting information, blast furnace gas generating capacity data.
Step 320: build BP neural network.
Step 330: the input variable of choosing BP neural network according to conditions of blast furnace information, strengthening smelting information.
Conditions of blast furnace information, strengthening smelting information are carried out to Linear correlative analysis and grey relational grade analysis successively, choose there is the strong degree of association working of a furnace factor as input variable, further, choose and there are fuel ratio, the sintering number percent of the strong degree of association, the air quantity of blowing, air-supply wind-warm syndrome, blast pressure, 6 working of a furnace factors of coke ratio as neural network input variable;
According to described strengthening smelting information, choose Rich Oxygen Amount, injecting coal quantity as input variable.
Step 340: input variable is divided into the first training set and the first test set.
Step 350: the output using blast furnace gas generating capacity data as BP neural network, is divided into the second training set and the second test set.
Step 360: respectively the input variable in the first training set and the first test set is normalized; Respectively the blast furnace gas generating capacity in the second training set and the second test set is normalized.
Before neural metwork training, in order to prevent because of the excessive neuron output saturation that makes of clean input absolute value, and then make weights adjustment enter error curved surface flat region, need to be normalized to the data of training set and test set the data x to each training set and test set
i, i=1,2 ..., n, adopts following normalization formula:
Obtain the data after normalization, input layer and output layer data-mapping are arrived between [0,1], this nondimensionalization method, has solved dimension and the caused problem of numerical values recited difference.
Wherein, x
i(k) represent that gained after renormalization is mapped to the data of original data area;
X
i' (k) represent the data after normalization;
Step 370: the inputoutput data using the blast furnace gas generating capacity in the input variable in the first training set and the second training set as BP neural network carries out model training, until the convergence of BP neural network.
Step 380: by the BP neural network after input variable input convergence in the first test set, the blast furnace gas generating capacity of the prediction of output.
Step 390: by the blast furnace gas generating capacity of prediction and the blast furnace gas generating capacity comparison in the second test set, the blast furnace gas generating capacity of the prediction of output is carried out to renormalization processing, the blast furnace gas generating capacity check prediction effect after being predicted.
After training finishes, the blast furnace gas generating capacity of the prediction of model output is carried out to renormalization processing, can utilize following formula that model output is mapped to original data area:
x
i(k)=x
i′(k)·[max(x
i)-min(x
i)]+min(x
i);
Wherein, x
i(k) represent that gained after renormalization is mapped to the data of original data area;
X
i' (k) represent the data after normalization,
In the present embodiment, by the normalization of data and renormalization are processed, can obtain more accurate predicted data, effectively raise precision of prediction.
Referring to Fig. 4, the embodiment of the present invention provides a kind of prediction unit of blast furnace gas generating capacity, comprising:
Acquiring unit U110, for obtaining conditions of blast furnace information, strengthening smelting information, blast furnace gas generating capacity data;
Neural network construction unit U120, for building BP neural network;
Input variable is chosen unit U130, for choose the input variable of described BP neural network according to described conditions of blast furnace information, strengthening smelting information;
The first division unit U140, for being divided into described input variable the first training set and the first test set;
The second division unit U150, the output for using described blast furnace gas generating capacity data as described BP neural network, is divided into the second training set and the second test set;
Training unit U160, carries out model training for the inputoutput data using the blast furnace gas generating capacity in the input variable of described the first training set and described the second training set as described BP neural network, until the convergence of described BP neural network;
Processing unit U170, for inputting the input variable of described the first test set the BP neural network after described convergence, the blast furnace gas generating capacity of the prediction of output;
Verification unit U180, for by the blast furnace gas generating capacity comparison of the blast furnace gas generating capacity of described prediction and described the second test set, checks prediction effect.
Preferably, in other embodiments of the invention, also comprise:
Data processing unit, for utilizing interpolation method to ask for substituting of missing data and improper data to described blast furnace gas generating capacity data and described input variable;
Filter unit, carries out Kalman filtering for the blast furnace gas generating capacity to after described interpolation method is processed.
Preferably, in other embodiments of the invention, described input variable is chosen unit, comprising:
First chooses module, for the working of a furnace factor of described conditions of blast furnace information, strengthening smelting information being carried out to Linear correlative analysis and grey relational grade analysis successively, choose thering is the strong degree of association as input variable;
Second chooses module, for according to described strengthening smelting information, chooses Rich Oxygen Amount, injecting coal quantity as input variable.
Preferably, in other embodiments of the invention, described neural network construction unit, comprising:
The first determination module, for determining the structure of BP neural network, comprises the hidden layer number of plies, input layer, hidden layer, each node layer number of output layer;
The second determination module, for determining that neuron connects weights and neuron threshold value.
Preferably, in other embodiments of the invention, also comprise:
The first normalized module, for being normalized the input variable of described the first training set and the first test set respectively;
The second normalized module, for being normalized the blast furnace gas generating capacity of described the second training set and the second test set respectively;
Renormalization processing module, for carrying out renormalization processing, the blast furnace gas generating capacity after being predicted to the blast furnace gas generating capacity of the described prediction of output.
In this instructions, each embodiment adopts the mode of going forward one by one to describe, and each embodiment stresses is the difference with other embodiment, between each embodiment identical similar part mutually referring to.For the disclosed device of embodiment, because it corresponds to the method disclosed in Example, so description is fairly simple, relevant part partly illustrates referring to method.
Finally, also it should be noted that, in this article, such as first, second, third and the fourth class relational terms be only used for an entity or operation to separate with another entity or operational zone, and not necessarily require or imply and between these entities or operation, have the relation of any this reality or sequentially.And, term " comprises ", " comprising " or its any other variant are intended to contain comprising of nonexcludability, thereby the process, method, article or the equipment that make to comprise a series of key elements not only comprise those key elements, but also comprise other key elements of clearly not listing, or be also included as the intrinsic key element of this process, method, article or equipment.The in the situation that of more restrictions not, the key element being limited by statement " comprising ... ", and be not precluded within process, method, article or the equipment that comprises described key element and also have other identical element.
While for convenience of description, describing above device, with function, being divided into various unit describes respectively.Certainly, when implementing the application, the function of each unit can be realized in same or a plurality of software and/or hardware.
Above-mentioned explanation to the disclosed embodiments, makes professional and technical personnel in the field can realize or use the present invention.To the multiple modification of these embodiment, will be apparent for those skilled in the art, General Principle as defined herein can, in the situation that not departing from the spirit or scope of the present invention, realize in other embodiments.Therefore, the present invention will can not be restricted to these embodiment shown in this article, but will meet the widest scope consistent with principle disclosed herein and features of novelty.
Claims (10)
1. a Forecasting Methodology for blast furnace gas generating capacity, is characterized in that, comprising:
Obtain conditions of blast furnace information, strengthening smelting information, blast furnace gas generating capacity data;
Build BP neural network;
According to described conditions of blast furnace information, strengthening smelting information, choose the input variable of described BP neural network;
Described input variable is divided into the first training set and the first test set;
Output using described blast furnace gas generating capacity data as described BP neural network, is divided into the second training set and the second test set;
Inputoutput data using the blast furnace gas generating capacity in the input variable in described the first training set and described the second training set as described BP neural network carries out model training, until the convergence of described BP neural network;
Input variable in described the first test set is inputted to the BP neural network after described convergence, the blast furnace gas generating capacity of the prediction of output;
By the blast furnace gas generating capacity comparison in the blast furnace gas generating capacity of described prediction and described the second test set, check prediction effect.
2. method according to claim 1, is characterized in that, also comprises:
Utilize interpolation method to ask for substituting of missing data and improper data to described blast furnace gas generating capacity data and described input variable;
Blast furnace gas generating capacity after described interpolation method is processed is carried out to Kalman filtering.
3. method according to claim 1, is characterized in that, chooses the input variable of described BP neural network according to described conditions of blast furnace information, strengthening smelting information, comprising:
Described conditions of blast furnace information, strengthening smelting information are carried out to Linear correlative analysis and grey relational grade analysis successively, choose there is the strong degree of association working of a furnace factor as input variable;
According to described strengthening smelting information, choose Rich Oxygen Amount, injecting coal quantity as input variable.
4. method according to claim 1, is characterized in that, described structure BP neural network, comprising:
Determine the structure of BP neural network, comprise the hidden layer number of plies, input layer, hidden layer, each node layer number of output layer;
Determine that neuron connects weights and neuron threshold value.
5. method according to claim 1, is characterized in that, also comprises:
Respectively the input variable in described the first training set and the first test set is normalized;
Respectively the blast furnace gas generating capacity in described the second training set and the second test set is normalized;
Blast furnace gas generating capacity to the described prediction of output carries out renormalization processing, the blast furnace gas generating capacity after being predicted.
6. a prediction unit for blast furnace gas generating capacity, is characterized in that, comprising:
Acquiring unit, for obtaining conditions of blast furnace information, strengthening smelting information, blast furnace gas generating capacity data;
Neural network construction unit, for building BP neural network;
Input variable is chosen unit, for choose the input variable of described BP neural network according to described conditions of blast furnace information, strengthening smelting information;
The first division unit, for being divided into described input variable the first training set and the first test set;
The second division unit, the output for using described blast furnace gas generating capacity data as described BP neural network, is divided into the second training set and the second test set;
Training unit, carries out model training for the inputoutput data using the blast furnace gas generating capacity in the input variable of described the first training set and described the second training set as described BP neural network, until the convergence of described BP neural network;
Processing unit, for inputting the input variable of described the first test set the BP neural network after described convergence, the blast furnace gas generating capacity of the prediction of output;
Verification unit, for by the blast furnace gas generating capacity comparison of the blast furnace gas generating capacity of described prediction and described the second test set, checks prediction effect.
7. device according to claim 6, is characterized in that, also comprises:
Data processing unit, for utilizing interpolation method to ask for substituting of missing data and improper data to described blast furnace gas generating capacity data and described input variable;
Filter unit, carries out Kalman filtering for the blast furnace gas generating capacity to after described interpolation method is processed.
8. device according to claim 6, is characterized in that, described input variable is chosen unit, comprising:
First chooses module, for the working of a furnace factor of described conditions of blast furnace information, strengthening smelting information being carried out to Linear correlative analysis and grey relational grade analysis successively, choose thering is the strong degree of association as input variable;
Second chooses module, for according to described strengthening smelting information, chooses Rich Oxygen Amount, injecting coal quantity as input variable.
9. device according to claim 6, is characterized in that, described neural network construction unit, comprising:
The first determination module, for determining the structure of BP neural network, comprises the hidden layer number of plies, input layer, hidden layer, each node layer number of output layer;
The second determination module, for determining that neuron connects weights and neuron threshold value.
10. device according to claim 6, is characterized in that, also comprises:
The first normalized module, for being normalized the input variable of described the first training set and the first test set respectively;
The second normalized module, for being normalized the blast furnace gas generating capacity of described the second training set and the second test set respectively;
Renormalization processing module, for carrying out renormalization processing, the blast furnace gas generating capacity after being predicted to the blast furnace gas generating capacity of the described prediction of output.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310566822.1A CN103559543B (en) | 2013-11-13 | 2013-11-13 | The Forecasting Methodology of a kind of blast furnace gas generating capacity and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310566822.1A CN103559543B (en) | 2013-11-13 | 2013-11-13 | The Forecasting Methodology of a kind of blast furnace gas generating capacity and device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN103559543A true CN103559543A (en) | 2014-02-05 |
CN103559543B CN103559543B (en) | 2016-09-07 |
Family
ID=50013784
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201310566822.1A Active CN103559543B (en) | 2013-11-13 | 2013-11-13 | The Forecasting Methodology of a kind of blast furnace gas generating capacity and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103559543B (en) |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104615856A (en) * | 2015-01-06 | 2015-05-13 | 莱芜钢铁集团电子有限公司 | Gas consumption prediction model establishing method and device based on hot blast stove group |
CN105177199A (en) * | 2015-08-31 | 2015-12-23 | 南京南瑞继保电气有限公司 | Blast furnace gas generation amount soft measurement method |
CN106755972A (en) * | 2016-12-14 | 2017-05-31 | 中国地质大学(武汉) | A kind of method that sintering process comprehensive coke ratio is predicted based on Data Dimensionality Reduction method |
CN106777684A (en) * | 2016-12-14 | 2017-05-31 | 中国地质大学(武汉) | A kind of method for setting up comprehensive coke ratio forecast model and prediction comprehensive coke ratio |
CN106845012A (en) * | 2017-02-17 | 2017-06-13 | 大连理工大学 | A kind of blast furnace gas system model membership function based on multiple target Density Clustering determines method |
CN108520157A (en) * | 2018-04-20 | 2018-09-11 | 江苏永钢集团有限公司 | A kind of measuring method of blast furnace gas occurrence quantity |
CN109086949A (en) * | 2018-09-20 | 2018-12-25 | 鞍钢集团自动化有限公司 | Blast furnace gas occurrence quantity and its calorific value prediction technique based on the variation of coal gas composition |
CN109886396A (en) * | 2019-03-18 | 2019-06-14 | 国家电网有限公司 | A kind of transmission line galloping on-line prediction system and method |
CN110825381A (en) * | 2019-09-29 | 2020-02-21 | 南京大学 | CNN-based bug positioning method combining source code semantics and grammatical features |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1403593A (en) * | 2002-10-17 | 2003-03-19 | 浙江大学 | Blast furnace smelt controlling method with intelligent control system |
CN1594601A (en) * | 2004-06-17 | 2005-03-16 | 上海交通大学 | Method for stabilizing blast furnace operation based on equilibrium point |
CN101319925A (en) * | 2008-07-11 | 2008-12-10 | 昆明理工大学 | Coal gas measuring method by utilization of BP neural network |
CN103160626A (en) * | 2011-12-14 | 2013-06-19 | 鞍钢股份有限公司 | Method for determining cold blast furnace hearth |
-
2013
- 2013-11-13 CN CN201310566822.1A patent/CN103559543B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1403593A (en) * | 2002-10-17 | 2003-03-19 | 浙江大学 | Blast furnace smelt controlling method with intelligent control system |
CN1594601A (en) * | 2004-06-17 | 2005-03-16 | 上海交通大学 | Method for stabilizing blast furnace operation based on equilibrium point |
CN101319925A (en) * | 2008-07-11 | 2008-12-10 | 昆明理工大学 | Coal gas measuring method by utilization of BP neural network |
CN103160626A (en) * | 2011-12-14 | 2013-06-19 | 鞍钢股份有限公司 | Method for determining cold blast furnace hearth |
Non-Patent Citations (1)
Title |
---|
肖冬峰等: ""基于改进BP网络的高炉煤气发生量预测模型"", 《浙江大学学报(工学版)》, vol. 46, no. 11, 30 November 2012 (2012-11-30) * |
Cited By (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104615856B (en) * | 2015-01-06 | 2017-09-29 | 莱芜钢铁集团电子有限公司 | Gas consumption amount forecast model method for building up and device based on hot-blast stove group |
CN104615856A (en) * | 2015-01-06 | 2015-05-13 | 莱芜钢铁集团电子有限公司 | Gas consumption prediction model establishing method and device based on hot blast stove group |
CN105177199A (en) * | 2015-08-31 | 2015-12-23 | 南京南瑞继保电气有限公司 | Blast furnace gas generation amount soft measurement method |
CN106755972A (en) * | 2016-12-14 | 2017-05-31 | 中国地质大学(武汉) | A kind of method that sintering process comprehensive coke ratio is predicted based on Data Dimensionality Reduction method |
CN106777684A (en) * | 2016-12-14 | 2017-05-31 | 中国地质大学(武汉) | A kind of method for setting up comprehensive coke ratio forecast model and prediction comprehensive coke ratio |
CN106755972B (en) * | 2016-12-14 | 2018-11-27 | 中国地质大学(武汉) | A method of sintering process comprehensive coke ratio is predicted based on Data Dimensionality Reduction method |
CN106777684B (en) * | 2016-12-14 | 2019-12-10 | 中国地质大学(武汉) | Method for establishing comprehensive coke ratio prediction model and predicting comprehensive coke ratio |
CN106845012B (en) * | 2017-02-17 | 2019-09-27 | 大连理工大学 | A kind of blast furnace gas system model subordinating degree function based on multiple target Density Clustering determines method |
CN106845012A (en) * | 2017-02-17 | 2017-06-13 | 大连理工大学 | A kind of blast furnace gas system model membership function based on multiple target Density Clustering determines method |
CN108520157A (en) * | 2018-04-20 | 2018-09-11 | 江苏永钢集团有限公司 | A kind of measuring method of blast furnace gas occurrence quantity |
CN109086949A (en) * | 2018-09-20 | 2018-12-25 | 鞍钢集团自动化有限公司 | Blast furnace gas occurrence quantity and its calorific value prediction technique based on the variation of coal gas composition |
CN109086949B (en) * | 2018-09-20 | 2021-09-21 | 鞍钢集团自动化有限公司 | Blast furnace gas generation amount and heat value prediction method based on gas component change |
CN109886396A (en) * | 2019-03-18 | 2019-06-14 | 国家电网有限公司 | A kind of transmission line galloping on-line prediction system and method |
CN110825381A (en) * | 2019-09-29 | 2020-02-21 | 南京大学 | CNN-based bug positioning method combining source code semantics and grammatical features |
Also Published As
Publication number | Publication date |
---|---|
CN103559543B (en) | 2016-09-07 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103559543A (en) | Method and device for predicting blast furnace gas occurrence amount | |
Kuai et al. | An application of system dynamics for evaluating planning alternatives to guide a green industrial transformation in a resource-based city | |
WO2021189739A1 (en) | Production plan-based iron and steel plant oxygen load prediction method | |
CN111582568B (en) | Power data-based method for predicting reworking of enterprises in spring festival | |
CN109252855B (en) | Method and device for determining final cumulative yield of gas well | |
CN108764626B (en) | Energy-saving diagnosis method and device | |
Cheng | ASSESSMENT OF MINE VENTILATION SYSTEM RELIABILITY USING RANDOM SIMULATION METHOD. | |
CN111549193B (en) | Furnace changing method, furnace changing device and control equipment for multiple blast furnace hot blast stoves | |
CN102592004B (en) | System and method for on-line analysis and diagnosis of whole-process energy-consuming conditions of integrated iron and steel works | |
CN105184496A (en) | Integrated evaluation method of health index of mine ventilation system | |
Liu et al. | Construction quality risk management of projects on the basis of rough set and neural network | |
CN107679651A (en) | A kind of moon electricity demand forecasting method based on grey forecasting model Yu BP artificial nerve network models | |
CN116739147A (en) | BIM-based intelligent energy consumption management and dynamic carbon emission calculation combined method and system | |
Yu et al. | Research on the post occupancy evaluation of green public building environmental performance combined with carbon emissions accounting | |
CN114997722A (en) | Steel enterprise byproduct gas balance optimization regulation and control system and method | |
CN114759557A (en) | Coal-fired unit AGC adjustment performance prediction method | |
CN113837912A (en) | Method for analyzing influence factors of carbon emission in construction industry | |
CN112508231A (en) | Medium-and-long-term power load prediction method and system based on system dynamics | |
Jin et al. | A causal model-based scheduling approach for coke oven gas system in steel industry | |
Jain et al. | Using model calibration to develop a measurement and verification framework for managing the energy performance gap in buildings | |
Chen et al. | Research on Project Resource Allocation and Lean Procurement Strategy Based on System Dynamics’ | |
CN106875097B (en) | Rail transit product field reliability assessment method and device based on mixed population | |
CN104751233A (en) | Contract capacity optimization system and optimization method | |
CN109190858A (en) | A kind of power network construction project investment policy making period method | |
CN107480107A (en) | Combustion gas gas consumption Forecasting Methodology |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant |