CN110763830A - Method for predicting content of free calcium oxide in cement clinker - Google Patents
Method for predicting content of free calcium oxide in cement clinker Download PDFInfo
- Publication number
- CN110763830A CN110763830A CN201911232888.0A CN201911232888A CN110763830A CN 110763830 A CN110763830 A CN 110763830A CN 201911232888 A CN201911232888 A CN 201911232888A CN 110763830 A CN110763830 A CN 110763830A
- Authority
- CN
- China
- Prior art keywords
- calcium oxide
- free calcium
- content
- cement clinker
- hidden layer
- 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
- BRPQOXSCLDDYGP-UHFFFAOYSA-N calcium oxide Chemical compound [O-2].[Ca+2] BRPQOXSCLDDYGP-UHFFFAOYSA-N 0.000 title claims abstract description 60
- ODINCKMPIJJUCX-UHFFFAOYSA-N calcium oxide Inorganic materials [Ca]=O ODINCKMPIJJUCX-UHFFFAOYSA-N 0.000 title claims abstract description 59
- 239000000292 calcium oxide Substances 0.000 title claims abstract description 59
- 239000004568 cement Substances 0.000 title claims abstract description 50
- 238000000034 method Methods 0.000 title claims abstract description 36
- 238000000605 extraction Methods 0.000 claims abstract description 19
- 238000012549 training Methods 0.000 claims abstract description 15
- 238000000354 decomposition reaction Methods 0.000 claims abstract description 13
- 238000013528 artificial neural network Methods 0.000 claims abstract description 9
- 239000011159 matrix material Substances 0.000 claims description 36
- 210000002569 neuron Anatomy 0.000 claims description 12
- 238000009827 uniform distribution Methods 0.000 claims description 12
- 210000002364 input neuron Anatomy 0.000 claims description 6
- 210000004205 output neuron Anatomy 0.000 claims description 6
- 238000005070 sampling Methods 0.000 claims description 6
- 238000012360 testing method Methods 0.000 claims description 4
- 238000001514 detection method Methods 0.000 claims description 3
- 230000006870 function Effects 0.000 claims description 3
- 238000005259 measurement Methods 0.000 abstract description 4
- 239000000428 dust Substances 0.000 abstract description 3
- 238000000691 measurement method Methods 0.000 abstract description 3
- 239000000779 smoke Substances 0.000 abstract description 3
- 238000011545 laboratory measurement Methods 0.000 abstract description 2
- 238000001354 calcination Methods 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- LYCAIKOWRPUZTN-UHFFFAOYSA-N Ethylene glycol Chemical compound OCCO LYCAIKOWRPUZTN-UHFFFAOYSA-N 0.000 description 3
- PEDCQBHIVMGVHV-UHFFFAOYSA-N Glycerine Chemical compound OCC(O)CO PEDCQBHIVMGVHV-UHFFFAOYSA-N 0.000 description 3
- UQSXHKLRYXJYBZ-UHFFFAOYSA-N Iron oxide Chemical compound [Fe]=O UQSXHKLRYXJYBZ-UHFFFAOYSA-N 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- OYPRJOBELJOOCE-UHFFFAOYSA-N Calcium Chemical compound [Ca] OYPRJOBELJOOCE-UHFFFAOYSA-N 0.000 description 1
- LFQSCWFLJHTTHZ-UHFFFAOYSA-N Ethanol Chemical compound CCO LFQSCWFLJHTTHZ-UHFFFAOYSA-N 0.000 description 1
- UUGLJVMIFJNVFH-UHFFFAOYSA-N Hexyl benzoate Chemical compound CCCCCCOC(=O)C1=CC=CC=C1 UUGLJVMIFJNVFH-UHFFFAOYSA-N 0.000 description 1
- VYPSYNLAJGMNEJ-UHFFFAOYSA-N Silicium dioxide Chemical compound O=[Si]=O VYPSYNLAJGMNEJ-UHFFFAOYSA-N 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 229910052791 calcium Inorganic materials 0.000 description 1
- 239000011575 calcium Substances 0.000 description 1
- 238000013098 chemical test method Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- TWNQGVIAIRXVLR-UHFFFAOYSA-N oxo(oxoalumanyloxy)alumane Chemical compound O=[Al]O[Al]=O TWNQGVIAIRXVLR-UHFFFAOYSA-N 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 239000002994 raw material Substances 0.000 description 1
- 229910052814 silicon oxide Inorganic materials 0.000 description 1
- 239000000243 solution Substances 0.000 description 1
- 239000002904 solvent Substances 0.000 description 1
- 239000012086 standard solution Substances 0.000 description 1
Images
Classifications
-
- 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
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
- G01N33/38—Concrete; Lime; Mortar; Gypsum; Bricks; Ceramics; Glass
- G01N33/383—Concrete or cement
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
-
- 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
- G06N3/06—Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons
- G06N3/061—Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons using biological neurons, e.g. biological neurons connected to an integrated circuit
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Theoretical Computer Science (AREA)
- Mathematical Physics (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Data Mining & Analysis (AREA)
- Software Systems (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Mathematical Analysis (AREA)
- Neurology (AREA)
- Artificial Intelligence (AREA)
- Pure & Applied Mathematics (AREA)
- Chemical & Material Sciences (AREA)
- Computational Linguistics (AREA)
- Mathematical Optimization (AREA)
- Evolutionary Computation (AREA)
- Computational Mathematics (AREA)
- Food Science & Technology (AREA)
- Pathology (AREA)
- Microelectronics & Electronic Packaging (AREA)
- Immunology (AREA)
- Biochemistry (AREA)
- Analytical Chemistry (AREA)
- Medicinal Chemistry (AREA)
- Ceramic Engineering (AREA)
- Algebra (AREA)
- Databases & Information Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a method for predicting the content of free calcium oxide in cement clinker, which comprises the following steps: s1: collecting a cement clinker sample; s2: constructing a time sequence of free calcium oxide content of cement clinker; s3: performing time sequence decomposition on the time sequence of the content of free calcium oxide in the mud clinker based on an empirical mode decomposition method; s4: constructing input and output sample pairs required by a training feature extraction module; s5: training the feature extraction submodule, namely a modularized echo state neural network, and learning the multi-time scale features of the time sequence: s6: training the prediction module-echo state neural network: s7: and predicting the content of free calcium oxide in the clinker. The method predicts the content of free calcium oxide at the next moment based on off-line experimental data, and solves the problem of lag of laboratory measurement results; compared with the measurement method of an on-line analyzer, the method has low cost, and the measurement accuracy is less influenced by on-site smoke dust and actual working conditions.
Description
Technical Field
The invention relates to a method for predicting the content of free calcium oxide in cement clinker, and belongs to the technical field of cement production control.
Background
The content of clinker free calcium oxide in the cement calcination process is an important standard for measuring the cement quality, and the content of clinker free calcium oxide represents the residual degree after calcium oxide is combined with silicon oxide, aluminum oxide and iron oxide in the calcination of raw materials, and the strength of the cement clinker is directly influenced by the content of the calcium oxide. Free calcium oxide of cement clinker is one of important indexes for reflecting the quality of the cement clinker, and the content of the free calcium oxide in the clinker during the calcination process of the cement clinker needs to be controlled within a reasonable range. The real-time prediction of the content of free calcium oxide in clinker is crucial to the guarantee of the calcination quality of cement clinker and the optimization of the calcination process.
At present, most laboratories of cement factories manually measure the content of free calcium oxide in cement clinker, adopt a glycerol-mango alcohol system as an extraction solvent, generate alkaline glycerol with the free calcium, and titrate with a hexanol benzoate standard solution. The on-line analyzer measuring method utilizes the fact that after ethylene glycol reacts with free calcium oxide in cement clinker, the conductivity of a solution and the content of the free calcium oxide form a certain proportional relation, and the content of the free calcium oxide is indirectly obtained through conductivity measurement.
The chemical test method has large delay and cannot truly reflect the production condition in the rotary kiln, so that the quality of the clinker is difficult to be effectively controlled. The online analyzer measurement method has the advantages of high equipment cost and high maintenance cost, and the measurement accuracy is easily influenced by site smoke dust and actual working conditions.
Disclosure of Invention
Aiming at the defects of the method, the invention provides a method for predicting the content of free calcium oxide in cement clinker, which can accurately predict the free calcium oxide in cement clinker in real time.
The technical scheme adopted for solving the technical problems is as follows:
the embodiment of the invention provides a method for predicting the content of free calcium oxide in cement clinker, which comprises the following steps:
s1: collecting a cement clinker sample;
s2: constructing a time sequence of free calcium oxide content of cement clinker;
s3: performing time sequence decomposition on the time sequence of the content of free calcium oxide in the mud clinker based on an empirical mode decomposition method:
s4: constructing input and output sample pairs required by a training feature extraction module;
s5: training the feature extraction submodule, namely a modularized echo state neural network, and learning the multi-time scale features of the time sequence:
s6: training the prediction module-echo state neural network:
s7: and predicting the content of free calcium oxide in the clinker.
As a possible implementation manner of this embodiment, in step S1, the sampling period is 1 hour, n times of continuous sampling are performed to obtain cement clinker samples online, and laboratory offline detection is performed manually, and the content u of free calcium oxide in cement clinker is recorded1,u2,……,un。
As a possible implementation manner of this embodiment, in step S3, the process of performing time-series decomposition on the time series of the free calcium oxide content of the clinker includes the following steps:
s3.1: setting an initial characteristic sequence number p to be 0; let h0(t) U (t), U (t) is cement clinker free calcium oxide content time sequence, U (t) ut,t=1,2,……,n;
S3.2: setting the initial cycle number q to be 0;
s3.3: updating the cycle number q to q + 1;
s3.4: finding out all extreme points of the time sequence U (t) of the content of free calcium oxide of the cement clinker;
s3.5: forming a lower envelope e for the minimum points by cubic spline interpolationmin(t) forming an upper envelope e for the maximamax(t);
S3.6: calculating the mean of the extreme points:
s3.7: details of extraction:
dq(t)=hp(t)-m(t); (2)
s3.8: for detail signal dq(t) repeating steps S3.3 to S3.7 until a detail signal dq(t) has a mean value of 0;
s3.9: updating the characteristic sequence number p to p + 1;
s3.10: recording a feature sequence IMFp(t);
S3.11: calculating a residual sequence:
hp(t)=hp-1(t)-IMFp(t); (3)
s3.12: to hp(t) repeating S3.2 to S3.11 until hp(t) is a monotonic function;
s3.13: let IMFp(t)=hp(t)。
As a possible implementation manner of this embodiment, in step S4, the input/output sample pairs required for constructing the training feature extraction module are:
Ωi=(U(t),IMFi(t)) (4)
wherein i is 1,2, …, p + 1; t is 1,2, …, n.
As a possible implementation manner of this embodiment, the step S5 specifically includes the following steps:
s5.1: setting an initial module number k to be 0;
s5.2: updating the number k of the modules to k + 1;
s5.3: setting module parameters: number of input neurons 1, number of output neurons 1, number of hidden layer neurons nkSpectral radius ρkDegree of sparsity spk;
S5.4: randomly generating networks with uniform distribution over the interval (-0.1,0.1)Input weight matrix
S5.5: randomly generating sparsity sp with uniform distribution over the interval (-1,1)kOf (2) matrix
S5.7: calculating hidden layer weight Wk:
S5.8: computing hidden layer neuron states:
where t is 1,2, L L, n, sk(0)=0;
S5.9: constructing a hidden layer state matrix:
s5.10: calculating a network output weight matrix:
s5.11: and (3) calculating network output:
s5.12: repeating steps S5.2 to S5.11 until k ═ p + 1;
s5.13: collecting submodule output:
as a possible implementation manner of this embodiment, the step S6 includes the following steps:
S6.2: setting module parameters: number of input neurons p +1, number of output neurons 1, number of hidden layer neurons nRSpectral radius ρRDegree of sparsity spR;
S6.3: randomly generating a network input weight matrix with uniform distribution over an interval (-0.1,0.1)
S6.4: randomly generating sparsity sp with uniform distribution over the interval (-1,1)ROf (2) matrix
S6.6: calculating hidden layer weight WR
S6.7: computing hidden layer neuron states sR(t):
Wherein s isR(0)=0;
S6.8: constructing a hidden layer state matrix HR:
S6.9: computing network output weight matrix
As a possible implementation manner of this embodiment, the step S7 specifically includes: inputting the test sample into the trained feature extraction module, and outputting by the prediction module to obtain the free calcium oxide content of the cement clinker at the next moment.
The technical scheme of the embodiment of the invention has the following beneficial effects:
aiming at the problem of online prediction of the free calcium oxide content of the cement clinker, the invention designs an intelligent prediction method of the free calcium oxide content of the cement clinker based on time sequence decomposition and a neural network, and realizes accurate prediction of the free calcium oxide content of the cement clinker; the empirical mode decomposition and the modularized echo state are combined, a time series multi-time scale feature learning method is provided, the automatic extraction of the time series multi-time scale features is realized, and the free calcium oxide of the cement clinker can be accurately predicted in real time.
The method predicts the content of free calcium oxide at the next moment based on off-line experimental data, and solves the problem of lag of laboratory measurement results; compared with the measurement method of an on-line analyzer, the method has low cost, and the measurement accuracy is less influenced by on-site smoke dust and actual working conditions.
Description of the drawings:
FIG. 1 is a flow chart illustrating a method for predicting free calcium oxide content of cement clinker in accordance with an exemplary embodiment;
FIG. 2 is a diagram of a predictive model topology;
FIG. 3 is a comparison graph of predicted values and measured values;
fig. 4 is a prediction error map.
Detailed Description
The invention is further illustrated by the following examples in conjunction with the accompanying drawings:
in order to clearly explain the technical features of the present invention, the following detailed description of the present invention is provided with reference to the accompanying drawings. The following disclosure provides many different embodiments, or examples, for implementing different features of the invention. To simplify the disclosure of the present invention, the components and arrangements of specific examples are described below. Furthermore, the present invention may repeat reference numerals and/or letters in the various examples. This repetition is for the purpose of simplicity and clarity and does not in itself dictate a relationship between the various embodiments and/or configurations discussed. It should be noted that the components illustrated in the figures are not necessarily drawn to scale. Descriptions of well-known components and processing techniques and procedures are omitted so as to not unnecessarily limit the invention.
FIG. 1 is a flow chart illustrating a method for predicting free calcium oxide content of cement clinker according to an exemplary embodiment. As shown in fig. 1, a method for predicting the content of free calcium oxide in cement clinker provided by the embodiment of the present invention includes the following steps:
s1: collecting a cement clinker sample;
s2: constructing a time sequence of free calcium oxide content of cement clinker;
s3: performing time sequence decomposition on the time sequence of the content of free calcium oxide in the mud clinker based on an empirical mode decomposition method:
s4: constructing input and output sample pairs required by a training feature extraction module;
s5: training the feature extraction submodule, namely a modularized echo state neural network, and learning the multi-time scale features of the time sequence:
s6: training the prediction module-echo state neural network:
s7: and predicting the content of free calcium oxide in the clinker.
As a possible implementation manner of this embodiment, in step S1, the sampling period is 1 hour, n times of continuous sampling are performed to obtain cement clinker samples online, and laboratory offline detection is performed manually, and the content u of free calcium oxide in cement clinker is recorded1,u2,……,un。
As a possible implementation manner of this embodiment, in step S3, the process of performing time-series decomposition on the time series of the free calcium oxide content of the clinker includes the following steps:
s3.1: setting an initial characteristic sequence number p to be 0; let h0(t) U (t), U (t) is cement clinker free calcium oxide content time sequence, U (t) ut,t=1,2,……,n;
S3.2: setting the initial cycle number q to be 0;
s3.3: updating the cycle number q to q + 1;
s3.4: finding out all extreme points of the time sequence U (t) of the content of free calcium oxide of the cement clinker;
s3.5: forming a lower envelope e for the minimum points by cubic spline interpolationmin(t) forming an upper envelope e for the maximamax(t);
S3.6: calculating the mean of the extreme points:
s3.7: details of extraction:
dq(t)=hp(t)-m(t); (2)
s3.8: for detail signal dq(t) repeating steps S3.3 to S3.7 until a detail signal dq(t) has a mean value of 0;
s3.9: updating the characteristic sequence number p to p + 1;
s3.10: recording a feature sequence IMFp(t);
S3.11: calculating a residual sequence:
hp(t)=hp-1(t)-IMFp(t); (3)
s3.12: to hp(t) repeating S3.2 to S3.11 until hp(t) is a monotonic function;
s3.13: let IMFp+1(t)=hp(t)。
As a possible implementation manner of this embodiment, in step S4, the input/output sample pairs required for constructing the training feature extraction module are:
Ωi=(U(t),IMFi(t)) (4)
wherein i is 1,2, …, p + 1; t is 1,2, …, n.
As a possible implementation manner of this embodiment, the step S5 specifically includes the following steps:
s5.1: setting an initial module number k to be 0;
s5.2: updating the number k of the modules to k + 1;
s5.3: setting module parameters: number of input neurons 1, number of output neurons 1, number of hidden layer neurons nkSpectral radius ρkDegree of sparsity spk;
S5.4: randomly generating network input weight moments in uniform distribution over an interval (-0.1,0.1)Matrix of
S5.5: randomly generating sparsity sp with uniform distribution over the interval (-1,1)kOf (2) matrix
S5.7: calculating hidden layer weight Wk:
S5.8: computing hidden layer neuron states:
wherein t is 1,2, … …, n; s, sk(0)=0;
S5.9: constructing a hidden layer state matrix:
s5.10: calculating a network output weight matrix:
s5.11: and (3) calculating network output:
s5.12: repeating steps S5.2 to S5.11 until k ═ p + 1;
s5.13: collecting submodule output:
as a possible implementation manner of this embodiment, the step S6 includes the following steps:
S6.2: setting module parameters: number of input neurons p +1, number of output neurons 1, number of hidden layer neurons nRSpectral radius ρRDegree of sparsity spR;
S6.3: randomly generating a network input weight matrix with uniform distribution over an interval (-0.1,0.1)
S6.4: randomly generating sparsity sp with uniform distribution over the interval (-1,1)ROf (2) matrix
S6.6: calculating hidden layer weight WR
S6.7: computing hidden layer neuron states sR(t):
Wherein s isR(0)=0;
S6.8: constructing a hidden layer state matrix HR:
As a possible implementation manner of this embodiment, the step S7 specifically includes: as shown in fig. 2, the test sample is input into the trained feature extraction module, and the output of the prediction module is the free calcium oxide content of the cement clinker at the next moment.
The test result is shown in figure 3, the X axis is sample (number/hour), the Y axis is free calcium oxide content (%), ○ is prediction result, is actually measured free calcium oxide content, the error between the model prediction result and the actually measured free calcium oxide content is shown in figure 4, the X axis is sample (number/hour), the Y axis is the difference between the model prediction result and the actually measured free calcium oxide content, and the result proves the effectiveness of the method.
The foregoing is only a preferred embodiment of the present invention, and it will be apparent to those skilled in the art that various modifications and improvements can be made without departing from the principle of the present invention, and these modifications and improvements are also considered to be within the scope of the present invention.
Claims (7)
1. A method for predicting the content of free calcium oxide in cement clinker is characterized by comprising the following steps:
s1: collecting a cement clinker sample;
s2: constructing a time sequence of free calcium oxide content of cement clinker;
s3: performing time sequence decomposition on the time sequence of the content of free calcium oxide in the mud clinker based on an empirical mode decomposition method:
s4: constructing input and output sample pairs required by a training feature extraction module;
s5: training the feature extraction submodule, namely a modularized echo state neural network, and learning the multi-time scale features of the time sequence:
s6: training the prediction module-echo state neural network:
s7: and predicting the content of free calcium oxide in the clinker.
2. The method as claimed in claim 1, wherein in step S1, the sampling period is 1 hour, the cement clinker sample is obtained on-line n times by continuous sampling, the laboratory off-line detection is performed manually, and the cement clinker free calcium oxide content u is recorded1,u2,……,un。
3. The method as claimed in claim 2, wherein the step of performing time-series decomposition of the free calcium oxide content of the clinker in step S3 comprises the steps of:
s3.1: setting an initial characteristic sequence number p to be 0; let h0(t) U (t), U (t) is cement clinker free calcium oxide content time sequence, U (t) ut,t=1,2,……,n;
S3.2: setting the initial cycle number q to be 0;
s3.3: updating the cycle number q to q + 1;
s3.4: finding out all extreme points of the time sequence U (t) of the content of free calcium oxide in the cement clinker;
s3.5: forming a lower envelope e for the minimum points by cubic spline interpolationmin(t) forming an upper envelope e for the maximamax(t);
S3.6: calculating the mean of the extreme points:
s3.7: details of extraction:
dq(t)=hp(t)-m(t); (2)
s3.8: for detail signal dq(t) repeating steps S3.3 to S3.7 until a detail signal dq(t) has a mean value of 0;
s3.9: updating the characteristic sequence number p to p + 1;
s3.10: recording a feature sequence IMFp(t);
S3.11: calculating a residual sequence:
hp(t)=hp-1(t)-IMFp(t); (3)
s3.12: to hp(t) repeating S3.2 to S3.11 until hp(t) is a monotonic function;
s3.13: let IMFp(t)=hp(t)。
4. The method as claimed in claim 3, wherein in step S4, the input and output sample pairs required for constructing the training feature extraction module are:
Ωi=(U(t),IMFi(t)) (4)
wherein i is 1,2, …, p + 1; t is 1,2, …, n.
5. The method as claimed in claim 4, wherein said step S5 includes the following steps:
s5.1: setting an initial module number k to be 0;
s5.2: updating the number k of the modules to k + 1;
s5.3: setting module parameters: number of input neurons 1, number of output neurons 1, number of hidden layer neurons nkSpectral radius ρkDegree of sparsity spk;
S5.4: randomly generating a network input weight matrix with uniform distribution over an interval (-0.1,0.1)
S5.5: randomly generating sparsity sp with uniform distribution over the interval (-1,1)kOf (2) matrix
S5.7: calculating hidden layer weight Wk:
S5.8: computing hidden layer neuron states:
where t is 1,2, … …, n, sk(0)=0;
S5.9: constructing a hidden layer state matrix:
s5.10: calculating a network output weight matrix:
s5.11: and (3) calculating network output:
s5.12: repeating steps S5.2 to S5.11 until k ═ p + 1;
s5.13: collecting submodule output:
6. the method as claimed in claim 5, wherein said step S6 includes the steps of:
S6.2: setting module parameters: number of input neurons p +1, number of output neurons 1, number of hidden layer neurons nRSpectral radius ρRDegree of sparsity spR;
S6.3: randomly generating a network input weight matrix with uniform distribution over an interval (-0.1,0.1)
S6.4: randomly generating sparsity sp with uniform distribution over the interval (-1,1)ROf (2) matrix
S6.6: calculating hidden layer weight WR
S6.7: computing hidden layer neuron states sR(t):
Wherein s isR(0)=0;
S6.8: constructing a hidden layer state matrix HR:
7. The method for predicting the content of free calcium oxide in cement clinker according to any one of claims 1 to 6, wherein the step S7 is specifically as follows: inputting the test sample into the trained feature extraction module, and outputting by the prediction module to obtain the free calcium oxide content of the cement clinker at the next moment.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911232888.0A CN110763830B (en) | 2019-12-04 | 2019-12-04 | Method for predicting content of free calcium oxide in cement clinker |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911232888.0A CN110763830B (en) | 2019-12-04 | 2019-12-04 | Method for predicting content of free calcium oxide in cement clinker |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110763830A true CN110763830A (en) | 2020-02-07 |
CN110763830B CN110763830B (en) | 2022-04-05 |
Family
ID=69340973
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201911232888.0A Active CN110763830B (en) | 2019-12-04 | 2019-12-04 | Method for predicting content of free calcium oxide in cement clinker |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110763830B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112712861A (en) * | 2021-01-07 | 2021-04-27 | 北京明略软件系统有限公司 | Model construction method, device, equipment and computer readable medium |
CN114236104A (en) * | 2021-10-28 | 2022-03-25 | 阿里云计算有限公司 | Method, device, equipment, medium and product for measuring free calcium oxide |
Citations (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102147273A (en) * | 2010-01-29 | 2011-08-10 | 大连理工大学 | Data-based blast-furnace gas dynamic predication method for metallurgical enterprises |
CN104537444A (en) * | 2015-01-13 | 2015-04-22 | 安徽理工大学 | Gas outburst predicting method based on EMD and ELM |
CN104899656A (en) * | 2015-06-05 | 2015-09-09 | 三峡大学 | Wind power combined predication method based on ensemble average empirical mode decomposition and improved Elman neural network |
CN105373856A (en) * | 2015-11-23 | 2016-03-02 | 江苏省电力公司南通供电公司 | Wind electricity power short-term combined prediction method considering run detection method reconstruction |
CN105930916A (en) * | 2016-04-07 | 2016-09-07 | 大连理工大学 | Parallel modular neural network-based byproduct gas real-time prediction method |
CN106126896A (en) * | 2016-06-20 | 2016-11-16 | 中国地质大学(武汉) | The mixed model wind speed forecasting method learnt based on empirical mode decomposition and the degree of depth and system |
CN106202946A (en) * | 2016-07-18 | 2016-12-07 | 燕山大学 | Clinker free calcium levels Forecasting Methodology based on degree of depth belief network model |
CN106295798A (en) * | 2016-08-29 | 2017-01-04 | 江苏省电力试验研究院有限公司 | Empirical mode decomposition and Elman neural network ensemble wind-powered electricity generation Forecasting Methodology |
CN107016453A (en) * | 2016-12-08 | 2017-08-04 | 中国农业大学 | A kind of aquaculture dissolved oxygen prediction method and device |
CN108804800A (en) * | 2018-06-04 | 2018-11-13 | 桂林电子科技大学 | Lithium ion battery SOC on-line prediction methods based on echo state network |
CN109147878A (en) * | 2018-10-08 | 2019-01-04 | 燕山大学 | A kind of clinker free calcium flexible measurement method |
CN109165798A (en) * | 2018-10-19 | 2019-01-08 | 燕山大学 | A kind of Free Calcium Oxide Contents in Cement Clinker on-line prediction method and system |
CN109583566A (en) * | 2018-11-20 | 2019-04-05 | 东北农业大学 | Ammonia concentration prediction technique in a kind of cold ground pig house based on empirical mode decomposition and Elman neural network |
CN110163433A (en) * | 2019-05-21 | 2019-08-23 | 上海海事大学 | A kind of ship method for predicting |
CN110309979A (en) * | 2019-07-09 | 2019-10-08 | 南方电网科学研究院有限责任公司 | Power load prediction method, device and equipment based on echo state network |
US20190325862A1 (en) * | 2018-04-23 | 2019-10-24 | Eta Compute, Inc. | Neural network for continuous speech segmentation and recognition |
CN110517488A (en) * | 2019-08-19 | 2019-11-29 | 南京理工大学 | The Short-time Traffic Flow Forecasting Methods with Recognition with Recurrent Neural Network are decomposed based on timing |
-
2019
- 2019-12-04 CN CN201911232888.0A patent/CN110763830B/en active Active
Patent Citations (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102147273A (en) * | 2010-01-29 | 2011-08-10 | 大连理工大学 | Data-based blast-furnace gas dynamic predication method for metallurgical enterprises |
CN104537444A (en) * | 2015-01-13 | 2015-04-22 | 安徽理工大学 | Gas outburst predicting method based on EMD and ELM |
CN104899656A (en) * | 2015-06-05 | 2015-09-09 | 三峡大学 | Wind power combined predication method based on ensemble average empirical mode decomposition and improved Elman neural network |
CN105373856A (en) * | 2015-11-23 | 2016-03-02 | 江苏省电力公司南通供电公司 | Wind electricity power short-term combined prediction method considering run detection method reconstruction |
CN105930916A (en) * | 2016-04-07 | 2016-09-07 | 大连理工大学 | Parallel modular neural network-based byproduct gas real-time prediction method |
CN106126896A (en) * | 2016-06-20 | 2016-11-16 | 中国地质大学(武汉) | The mixed model wind speed forecasting method learnt based on empirical mode decomposition and the degree of depth and system |
CN106202946A (en) * | 2016-07-18 | 2016-12-07 | 燕山大学 | Clinker free calcium levels Forecasting Methodology based on degree of depth belief network model |
CN106295798A (en) * | 2016-08-29 | 2017-01-04 | 江苏省电力试验研究院有限公司 | Empirical mode decomposition and Elman neural network ensemble wind-powered electricity generation Forecasting Methodology |
CN107016453A (en) * | 2016-12-08 | 2017-08-04 | 中国农业大学 | A kind of aquaculture dissolved oxygen prediction method and device |
US20190325862A1 (en) * | 2018-04-23 | 2019-10-24 | Eta Compute, Inc. | Neural network for continuous speech segmentation and recognition |
CN108804800A (en) * | 2018-06-04 | 2018-11-13 | 桂林电子科技大学 | Lithium ion battery SOC on-line prediction methods based on echo state network |
CN109147878A (en) * | 2018-10-08 | 2019-01-04 | 燕山大学 | A kind of clinker free calcium flexible measurement method |
CN109165798A (en) * | 2018-10-19 | 2019-01-08 | 燕山大学 | A kind of Free Calcium Oxide Contents in Cement Clinker on-line prediction method and system |
CN109583566A (en) * | 2018-11-20 | 2019-04-05 | 东北农业大学 | Ammonia concentration prediction technique in a kind of cold ground pig house based on empirical mode decomposition and Elman neural network |
CN110163433A (en) * | 2019-05-21 | 2019-08-23 | 上海海事大学 | A kind of ship method for predicting |
CN110309979A (en) * | 2019-07-09 | 2019-10-08 | 南方电网科学研究院有限责任公司 | Power load prediction method, device and equipment based on echo state network |
CN110517488A (en) * | 2019-08-19 | 2019-11-29 | 南京理工大学 | The Short-time Traffic Flow Forecasting Methods with Recognition with Recurrent Neural Network are decomposed based on timing |
Non-Patent Citations (7)
Title |
---|
JUN ZHAO ET AL.: "A Two-Stage Online Prediction Method for a Blast", 《IEEE TRANSACTIONS ON CONTROL SYSTEMS TECHNOLOGY》 * |
JUNFEI QIAO ET AL.: "Growing Echo-State Network With", 《IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS》 * |
LIHUA SHEN ET AL.: "A novel echo state network for multivariate and nonlinear time seriesprediction", 《APPLIED SOFT COMPUTING》 * |
宋菁华等: "改进型EMD-Elman神经网络在铁水硅含量预测中的应用", 《化工学报》 * |
王磊: "基于灵敏度分析的模块化回声状态网络修剪算法", 《自动化学报》 * |
肖勇: "基于模块化回声状态网络的实时电力负荷预测", 《电网技术》 * |
赵朋程等: "用于水泥熟料 fCaO 预测的多核最小二乘支持向量机模型", 《化工学报》 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112712861A (en) * | 2021-01-07 | 2021-04-27 | 北京明略软件系统有限公司 | Model construction method, device, equipment and computer readable medium |
CN114236104A (en) * | 2021-10-28 | 2022-03-25 | 阿里云计算有限公司 | Method, device, equipment, medium and product for measuring free calcium oxide |
Also Published As
Publication number | Publication date |
---|---|
CN110763830B (en) | 2022-04-05 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111804146B (en) | Intelligent ammonia injection control method and intelligent ammonia injection control device | |
CN110807554B (en) | Generation method and system based on wind power/photovoltaic classical scene set | |
CN110763830B (en) | Method for predicting content of free calcium oxide in cement clinker | |
CN110378044B (en) | Multi-time scale convolution neural network soft measurement method based on attention mechanism | |
CN103336107B (en) | Soft measurement method for f-CaO content of cement clinker | |
CN108152235B (en) | Heavy metal content inversion method combining soil indoor and outdoor spectra | |
CN107766986A (en) | Leak integral form echo state network on-line study photovoltaic power Forecasting Methodology | |
CN102693451A (en) | Method for predicting ammonia process flue gas desulphurization efficiency based on multiple parameters | |
CN110701796B (en) | Energy-saving control system of hot water system based on cloud prediction algorithm | |
WO2022121932A1 (en) | Adaptive deep learning-based intelligent forecasting method, apparatus and device for complex industrial system, and storage medium | |
CN112949894B (en) | Output water BOD prediction method based on simplified long-short-term memory neural network | |
CN114239397A (en) | Soft measurement modeling method based on dynamic feature extraction and local weighted deep learning | |
CN114239421A (en) | Method and system for predicting coal powder combustion characteristic parameters based on BP neural network | |
CN111580570A (en) | Container liquid level monitoring method and system | |
CN114330647A (en) | Model training method and device and silicon rod weight prediction method | |
CN103279030B (en) | Dynamic soft measuring modeling method and device based on Bayesian frame | |
CN117350897A (en) | Method and system for predicting conversion power by considering air humidity | |
CN106529680B (en) | A kind of multiple dimensioned extreme learning machine integrated modelling approach based on empirical mode decomposition | |
CN110909492A (en) | Sewage treatment process soft measurement method based on extreme gradient lifting algorithm | |
Zhou et al. | Anode effect prediction of aluminum electrolysis using GRNN | |
CN113707240B (en) | Component parameter robust soft measurement method based on semi-supervised nonlinear variation Bayesian hybrid model | |
CN115577856A (en) | Method and system for predicting construction cost and controlling balance of power transformation project | |
CN113377075A (en) | Method and device for optimizing rare earth extraction process in real time and computer readable storage medium | |
CN117970978B (en) | Automatic regulation and control system of constant temperature and humidity experiment box suitable for experiments | |
CN114580798B (en) | Device point location prediction method and system based on transformer |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |