CN112275439A - Cement raw material vertical mill differential pressure soft measurement modeling method, storage medium and system - Google Patents

Cement raw material vertical mill differential pressure soft measurement modeling method, storage medium and system Download PDF

Info

Publication number
CN112275439A
CN112275439A CN202011097514.5A CN202011097514A CN112275439A CN 112275439 A CN112275439 A CN 112275439A CN 202011097514 A CN202011097514 A CN 202011097514A CN 112275439 A CN112275439 A CN 112275439A
Authority
CN
China
Prior art keywords
cement raw
training
output
neuron
differential pressure
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
Application number
CN202011097514.5A
Other languages
Chinese (zh)
Inventor
袁铸钢
季玉玺
王孝红
张强
孟庆金
刘钊
于洪亮
景绍洪
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
University of Jinan
Original Assignee
University of Jinan
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by University of Jinan filed Critical University of Jinan
Priority to CN202011097514.5A priority Critical patent/CN112275439A/en
Publication of CN112275439A publication Critical patent/CN112275439A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B02—CRUSHING, PULVERISING, OR DISINTEGRATING; PREPARATORY TREATMENT OF GRAIN FOR MILLING
    • B02C—CRUSHING, PULVERISING, OR DISINTEGRATING IN GENERAL; MILLING GRAIN
    • B02C25/00—Control arrangements specially adapted for crushing or disintegrating
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • G06F18/232—Non-hierarchical techniques
    • G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
    • G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G06N3/088—Non-supervised learning, e.g. competitive learning

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Evolutionary Computation (AREA)
  • Data Mining & Analysis (AREA)
  • Artificial Intelligence (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Software Systems (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Mathematical Physics (AREA)
  • Molecular Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Computing Systems (AREA)
  • Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • General Health & Medical Sciences (AREA)
  • Evolutionary Biology (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Probability & Statistics with Applications (AREA)
  • Food Science & Technology (AREA)
  • Medical Informatics (AREA)
  • Computer Hardware Design (AREA)
  • Geometry (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention provides a cement raw material vertical mill differential pressure soft measurement modeling method, a storage medium and a system, wherein the method comprises the following steps: step 1: determining parameters required by the model according to the comparison of historical production data of a cement plant and the trend of the pressure difference of a vertical mill for historical cement raw materials; step 2: establishing a model of the parameters through training of a radial basis function neural network; and step 3: and obtaining the pressure difference of the vertical mill for the cement raw materials during actual production according to the actual production data of the cement plant and the model. According to the scheme, the neural network training is utilized to obtain the model, the pressure difference of the cement raw material vertical mill is obtained through high-precision prediction of the model, and the subjectivity and the hysteresis in the production process of raw material grinding can be reduced.

Description

Cement raw material vertical mill differential pressure soft measurement modeling method, storage medium and system
Technical Field
The invention relates to the technical field of intelligent cement production, in particular to a cement raw material vertical mill differential pressure soft measurement modeling method, a storage medium and a system.
Background
The grinding process of raw materials of the vertical mill is a key part of the novel dry cement production. The quality of the raw meal and the stability of the powder process are key factors for measuring the quality of the raw meal grinding process. In the field grinding process, the proper pressure difference in the mill is an important index for keeping the mill to work stably, and the stable production can be ensured only by ensuring the stability of the load of the vertical mill.
The cement raw material vertical mill production process has the characteristics of multivariable, nonlinearity, strong coupling and the like, an accurate mathematical model is difficult to establish, key parameters cannot be measured on line, and the setting of the parameters is mainly manually adjusted by the experience of an operator, so that the raw material grinding production process has subjectivity and hysteresis.
Disclosure of Invention
In view of the above, the present invention provides a modeling method, a storage medium and a system for soft measurement of differential pressure of a cement raw mill, so as to overcome the defects of the prior art.
One part of the embodiment of the invention provides a cement raw meal vertical mill differential pressure soft measurement modeling method, which comprises the following steps:
step 1: determining parameters required by the model according to the comparison of historical production data of a cement plant and the trend of the pressure difference of a vertical mill for historical cement raw materials;
step 2: establishing a model of the parameters through training of a radial basis function neural network;
and step 3: and obtaining the pressure difference of the vertical mill for the cement raw materials during actual production according to the actual production data of the cement plant and the model.
Some embodiments of the present invention provide a storage medium, in which program information is stored, and a computer reads the program information and executes the above-mentioned soft measurement modeling method for differential pressure of a cement raw mill.
In some embodiments of the present invention, a soft measurement modeling system for differential pressure of a cement raw mill is provided, which includes: the device comprises at least one processor and at least one memory, wherein program information is stored in the at least one memory, and the at least one processor reads the program information and then executes the cement raw meal vertical mill differential pressure soft measurement modeling method.
Compared with the prior art, the technical scheme provided by the invention at least has the following effects: the invention provides a design method of a raw cement material vertical mill differential pressure measuring device based on a radial basis function neural network, which can reduce subjectivity and hysteresis in the production process of raw material grinding through high-precision prediction.
Drawings
FIG. 1 is a flow chart of a modeling method for differential pressure soft measurement of a cement raw mill according to an embodiment of the present invention;
FIG. 2 is a block diagram of a standard RBF neural network architecture according to one embodiment of the present invention;
FIG. 3 is a radial basis function neural network structure based on Gaussian kernel according to an embodiment of the present invention.
FIG. 4 is a block diagram of a soft measurement modeling apparatus for differential pressure of a cement raw mill according to an embodiment of the present invention;
fig. 5 is a block diagram of a hardware structure of a cement raw meal vertical mill differential pressure soft measurement modeling system according to an embodiment of the invention.
Detailed Description
The embodiments of the present invention will be further described with reference to the accompanying drawings. In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicate orientations or positional relationships based on the orientations or positional relationships shown in the drawings, and are only for convenience of description of the present invention, and do not indicate or imply that the device or assembly referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus, should not be construed as limiting the present invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance. Wherein the terms "first position" and "second position" are two different positions.
In some embodiments of the present invention, a soft measurement modeling method for differential pressure of a cement raw mill is provided, as shown in fig. 1, including the following steps:
step 1: determining parameters required by the model according to the comparison of historical production data of a cement plant and the trend of the pressure difference of a vertical mill for historical cement raw materials; the parameters include feed volume, mill outlet temperature, mill vibration, mill current, and bed thickness. Which comprises the following steps:
step 11: eliminating abnormal values in the parameter samples by adopting an Edarad criterion; removing abnormal values in the parameter sample by the following steps:
step 111: calculating the mean value, xiRepresents the ith parameter sample, the value of i is in the range of 1 to n, mu represents the mean value of n parameter samples:
Figure BDA0002724235290000031
step 112: calculating the deviation e of each parameter samplei：ei＝xi-μ，eiRepresenting parametric samples xiA deviation of (a);
step 113: calculating an experimental standard deviation sigma:
Figure BDA0002724235290000032
step 114: when the sample xiIs satisfied with | eiIf | > 3 σ, x is determinediRemoving abnormal values;
step 12: the noise in the remaining parameter samples is processed by sliding filtering to obtain the parameters. Noise in the parametric samples is processed by sliding filtering as follows:
Figure BDA0002724235290000033
where M denotes the time window length of the mean filtering, xi' denotes the parameter samples after the sliding filtering process.
Step 2: and establishing a model for the parameters through training of a radial basis function neural network. The method comprises the following steps:
step 21: as shown in fig. 2, selecting a radial basis function neural network center includes: determining the center of the radial basis function neural network by adopting a direct calculation method or a self-organizing learning method; the grinding machine mechanism is complex, the coupling is strong, the self-organizing learning is generally selected, the unsupervised learning method of the center can be selected through a K-clustering algorithm, the process of self-organizing selecting the center is as follows, and according to the standard radial basis function neural network structure, the radial basis function neural network learning of the self-organizing selecting center is divided into a self-organizing learning stage and a supervised learning stage of unsupervised learning:
step 211: in the self-organizing learning stage of unsupervised learning, the weight C between the input layer and the hidden layer is solvedjThe method comprises the following steps:
step 2111: network initialization, randomly selecting h training parameter samples as clustering centers ci。
Step 2112: grouping input training parameter samples according to a training sample nearest rule, and grouping input training parameter samples according to a training parameter sample xpAnd cluster center ciThe Euclidean distance between xpRespective cluster sets theta assigned to input training parameter samplespWherein the cluster set θpP training parameter samples are included.
Step 2113: readjusting the clustering center to calculate the average value of the training parameter samples in each clustering set until the average value is unchanged, and solving the variance deltai：
Figure BDA0002724235290000041
cmaxThe maximum value of the training parameter sample average values in each cluster set is used.
Step 212: in the stage of supervised learning, the calculation of weights from the hidden layer to the output layer includes:
step 2121: solving the partial derivative for w for the loss function:
Figure BDA0002724235290000042
step 2122: cluster center c such that the partial derivative result in step 2121 is 0iAs a cluster set thetapOf the center of (c).
Preferably, the steps of expressing and calculating the parameters of the network training in step 22 are as follows:
step 221: as shown in FIG. 3, according to the radial basis function neural network topology based on the Gaussian kernel, an input vector X and an output vector Y and a width vector D of a radial basis function are prepared before trainingj；
The input to the output has hysteresis, and the input quantity is correspondingly matched with the sampling time according to the expert experience. For example, field experience defines a material-induced lag time of 15 min. Then the corresponding table of the input quantity and time matching in the grinding process of the cement raw materials is as follows:
TABLE 1 Cement raw meal differential pressure model input and sampling time
Figure BDA0002724235290000043
Optionally, the radial basis network transfer function is a real-valued function whose value depends only on the distance from the origin. Or the distance from any point to the point c, where c is the center point:
Φ(x)＝Φ(||x||)；
Φ(x,c)＝Φ(||x-c||)；
wherein x and c are kernel function centers.
In the above, the radial basis network transfer function may take various forms. The following are common:
1. gaussian function
Figure BDA0002724235290000051
2. Abnormal S-shaped function
Figure BDA0002724235290000052
3. Inverse distortion correction function
Figure BDA0002724235290000053
Step 222: when training is performed by inputting the test set for the ith time, firstly, an input vector X is defined: x ═ X1,x2,...,xn]T(ii) a Wherein n is the number of input layer number units; explicit output vector Y and desired output vector O: y ═ Y1,y2,...,yn]T，O＝[o1,o2,...,on]TWhere q is the number of output layer units;
step 223: initializing a connection weight from the hidden layer to the output layer; wk＝[wk1,wk2,...,wkp]T(h ═ 1,2,. ·, q); wherein p is the number of hidden layer units and q is the number of output layer units;
initializing weights from a hidden layer to an output layer; wk＝[wk1,wk2,...,wkp]T(h ═ 1,2,. ·, q); wherein mink is the minimum value expected to be output by the kth output neuron of the training set, and maxk is the maximum value;
initializing the Central parameter C of hidden layer neuronsjiWhere i is 1,2, …, n. The initial value of the central parameter of the radial basis function neural network is as follows:
Figure BDA0002724235290000054
initializing a width vector: (ii) a The width vector is:
Figure BDA0002724235290000061
xkiis the i-th input parameter in the k-th neuron, dfA width modulation factor of less than 1 may facilitate the local responsiveness of the neural network.
Step 224: computing hidden layer neuron output values zj：
Figure BDA0002724235290000062
Step 225: calculating output Y of the output neuron:
Y＝[y1,y2,…,yq]T；
Figure BDA0002724235290000063
wkjis the weight between the kth neuron of the output layer and the jth neuron of the hidden layer;
the iterative selection gradient descent method of the weight parameters comprises the following steps:
Figure BDA0002724235290000064
Figure BDA0002724235290000065
Figure BDA0002724235290000066
wherein Wkj(t) is the tuning weight at the time of the t-th iterative computation between the kth output neuron and the jth hidden layer neuron; c. Ckj(t) is the center of the jth hidden layer neuron for the ith input neuron at the time of the tth iterative computationA component; the width of the jth hidden layer neuron corresponding to the center of the ith input neuron in the tth iterative computation is defined, wherein eta is a learning factor, and E is a neural network evaluation function:
Figure BDA0002724235290000067
ylkfor the net output value of the kth output neuron at the l input sample, OlkThe expected output value of the kth output neuron at the ith input sample.
Step 22: and carrying out network training on the radial basis function neural network by using the parameters.
And step 3: and obtaining the pressure difference of the vertical mill for the cement raw materials during actual production according to the actual production data of the cement plant and the model. The pressure difference of the vertical mill for the cement raw meal can be obtained by directly substituting actual production data into the model.
In addition, the present invention also provides a soft measurement device for differential pressure of a cement raw material vertical mill, as shown in fig. 4, comprising:
a selection input unit 401 configured with various parameters for modeling; selecting a proper radial basis function; and establishing a neural network by the selected parameters.
And the radial basis function unit 402 is configured to serve as a basis of the hidden unit to build a hidden layer space, and the input variable is mapped to the high-dimensional space through the transformation of the hidden layer on the input variable.
The center point unit 403 of the radial basis function is configured to determine the input vector mapping relationship.
And a linear mapping unit 404 configured to output the implicit layer data to obtain the soft measurement model.
In one embodiment, the radial basis function center point comprises: the acquisition unit of the radial basis function central point is configured for directly selecting a given training book set or selecting the given training book set in a clustering mode;
the self-organizing learning unit 405 is configured to enable the centers to better reflect the mutual information of the training sets when the centers of the radial basis functions are not clustered. In one embodiment, the self-organizing learning unit 405 includes: the configuration is used for neural network resource redistribution, so that the center of the neuron of the hidden layer of the radial basis function is positioned in a key part of an input space.
The invention also provides a storage medium which can be read and written by a computer, wherein the storage medium is stored with program instructions, and the computer reads the program information and then executes the cement raw meal vertical mill differential pressure soft measurement modeling method in any scheme.
Some embodiments of the present invention further provide a system for modeling differential pressure soft measurement of a cement raw mill, as shown in fig. 5, including: at least one processor 501 and at least one memory 502, wherein at least one memory 502 stores program information, and at least one processor 501 reads the program information and executes any one of the above soft measurement modeling methods for cement raw mill differential pressure. The apparatus may further comprise: an input device 503 and an output device 504. The processor 501, the memory 502, the input device 503 and the output device 504 may be connected by a bus or other means, and fig. 5 illustrates the connection by a bus as an example.
The technical scheme provided by the embodiment of the invention determines the parameters required by the model based on the actual production data of the cement plant; establishing a model for the determined parameters through training of a radial basis function neural network; the pressure difference of the vertical mill for the cement raw material is measured by the model. The invention provides a design method of a raw cement material vertical mill differential pressure measuring device based on a radial basis function neural network, which can reduce subjectivity and hysteresis in the production process of raw material grinding through high-precision prediction.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present invention.

Claims (10)

1. A cement raw material vertical mill differential pressure soft measurement modeling method is characterized by comprising the following steps:
step 1: determining parameters required by the model according to the comparison of historical production data of a cement plant and the trend of the pressure difference of a vertical mill for historical cement raw materials;
step 2: establishing a model of the parameters through training of a radial basis function neural network;
and step 3: and obtaining the pressure difference of the vertical mill for the cement raw materials during actual production according to the actual production data of the cement plant and the model.
2. The modeling method for differential pressure soft measurement of cement raw meal vertical mill according to claim 1, characterized in that in step 1:
the parameters include feed volume, mill outlet temperature, mill vibration, mill current, and bed thickness.
3. The modeling method for differential pressure soft measurement of cement raw meal vertical mill according to claim 2, wherein the step 1 comprises:
step 11: eliminating abnormal values in the parameter samples by adopting an Edarad criterion;
step 12: the noise in the remaining parameter samples is processed by sliding filtering to obtain the parameters.
4. The modeling method for differential pressure soft measurement of cement raw meal vertical mill according to claim 3, characterized in that:
in step 11, abnormal values in the parameter samples are removed through the following steps:
step 111: calculating the mean value, xiRepresents the ith parameter sample, the value of i is in the range of 1 to n, mu represents the mean value of n parameter samples:
Figure FDA0002724235280000011
step 112: for obtaining samples of each parameterDeviation ei：ei＝xi-μ，eiRepresenting parametric samples xiA deviation of (a);
step 113: calculating an experimental standard deviation sigma:
Figure FDA0002724235280000012
step 114: when the sample xiIs satisfied with | eiIf | > 3 σ, x is determinediRemoving abnormal values;
the noise in the parameter samples is processed by sliding filtering in step 12 as follows:
Figure FDA0002724235280000021
where M denotes the time window length of the mean filtering, xi' denotes the parameter samples after the sliding filtering process.
5. The modeling method for differential pressure soft measurement of cement raw meal vertical mill according to claim 4, wherein the step 2 comprises:
step 21: selecting a radial basis function neural network center, comprising: determining the center of the radial basis function neural network by adopting a direct calculation method or a self-organizing learning method;
step 22: and carrying out network training on the radial basis function neural network by using the parameters.
6. The modeling method for differential pressure soft measurement of a cement raw mill as claimed in claim 5, wherein the step of determining the center of the radial basis function neural network by the self-organizing learning method in step 21 selects the center by the K-clustering algorithm, comprising:
step 211: in the self-organizing learning stage of unsupervised learning, the weight C between the input layer and the hidden layer is solvedjThe method comprises the following steps:
step 2111: network initialization, randomly selecting h training parameter samples as clustering centers ci；
Step 2112: grouping input training parameter samples according to a training sample nearest rule, and grouping input training parameter samples according to a training parameter sample xpAnd cluster center ciThe Euclidean distance between xpRespective cluster sets theta assigned to input training parameter samplespWherein the cluster set θpP training parameter samples are included;
step 2113: readjusting the clustering center to calculate the average value of the training parameter samples in each clustering set until the average value is unchanged, and solving the variance deltai：
Figure FDA0002724235280000022
cmaxThe maximum value of the training parameter sample average values in each cluster set is taken as the maximum value;
step 212: in the stage of supervised learning, the calculation of weights from the hidden layer to the output layer includes:
step 2121: solving the partial derivative for w for the loss function:
Figure FDA0002724235280000031
step 2122: cluster center c such that the partial derivative result in step 2121 is 0iAs a cluster set thetapOf the center of (c).
7. The modeling method for differential pressure soft measurement of cement raw mill as claimed in claim 6, wherein each parameter expression and calculation step of network training in step 22 is as follows:
step 221: preparing an input vector X and an output vector Y and a width vector D of a radial basis function before training according to a radial basis function neural network topological structure based on a Gaussian kernelj；
Figure FDA0002724235280000032
xkiIs the ith input parameter in the kth neuron;
step 222: when training is performed by inputting the test set for the ith time, firstly, an input vector X is defined: x ═ X1,x2,...,xn]T(ii) a Wherein n is the number of input layer number units; explicit output vector Y and desired output vector O: y ═ Y1,y2,...,yn]T，O＝[o1,o2,...,on]TWhere q is the number of output layer units;
step 223: initializing a connection weight from the hidden layer to the output layer; initializing weights from a hidden layer to an output layer; initializing a central parameter of a hidden layer neuron; initializing a width vector;
step 224: computing hidden layer neuron output values zj：
Figure FDA0002724235280000033
Step 225: calculating output Y of the output neuron:
Y＝[y1,y2,…,yq]T；
Figure FDA0002724235280000034
wkjis the weight between the kth neuron of the output layer and the jth neuron of the hidden layer;
the iterative selection gradient descent method of the weight parameters comprises the following steps:
Figure FDA0002724235280000041
Figure FDA0002724235280000042
Figure FDA0002724235280000043
wherein Wkj(t) is the tuning weight at the time of the t-th iterative computation between the kth output neuron and the jth hidden layer neuron; c. Ckj(t) is the central component of the jth hidden layer neuron for the ith input neuron at the time of the tth iterative computation; the width of the jth hidden layer neuron corresponding to the center of the ith input neuron in the tth iterative computation is defined, wherein eta is a learning factor, and E is a neural network evaluation function:
Figure FDA0002724235280000044
ylkfor the net output value of the kth output neuron at the l input sample, OlkThe expected output value of the kth output neuron at the ith input sample.
8. The modeling method for differential pressure soft measurement of cement raw mill as claimed in claim 7, wherein the self-organizing learning algorithm in step 21 comprises:
after parameter initialization, given eta and alpha and iteration precision epsilon, the root mean square error RMS of the network output is calculated:
Figure FDA0002724235280000045
if RMS is less than or equal to epsilon, finishing training; otherwise, the weight iteration is continuously calculated, and the weight parameters, the radial basis function center and the width of the radial basis function are adjusted to carry out the iterative calculation.
9. A storage medium having stored therein program information, wherein a computer reads the program information and executes the method for modeling differential pressure soft measurement of cement raw meal according to any one of claims 1 to 8.
10. The utility model provides a cement raw grinds pressure differential soft measurement modeling system immediately which characterized in that includes:
at least one processor and at least one memory, wherein the at least one memory stores program information, and the at least one processor reads the program information and executes the method for modeling differential pressure soft measurements of cement raw meal according to any one of claims 1 to 8.
CN202011097514.5A 2020-10-14 2020-10-14 Cement raw material vertical mill differential pressure soft measurement modeling method, storage medium and system Pending CN112275439A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011097514.5A CN112275439A (en) 2020-10-14 2020-10-14 Cement raw material vertical mill differential pressure soft measurement modeling method, storage medium and system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011097514.5A CN112275439A (en) 2020-10-14 2020-10-14 Cement raw material vertical mill differential pressure soft measurement modeling method, storage medium and system

Publications (1)

Publication Number Publication Date
CN112275439A true CN112275439A (en) 2021-01-29

Family

ID=74496250

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011097514.5A Pending CN112275439A (en) 2020-10-14 2020-10-14 Cement raw material vertical mill differential pressure soft measurement modeling method, storage medium and system

Country Status (1)

Country Link
CN (1) CN112275439A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114733640A (en) * 2022-03-03 2022-07-12 江苏丰尚智能科技有限公司 Method and device for adjusting processing parameters of pulverizer and computer equipment

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110751202A (en) * 2019-10-16 2020-02-04 济南大学 Method for establishing load soft measurement model of cement combined grinding ball mill and application
CN110766234A (en) * 2019-10-29 2020-02-07 济南大学 Cement cooling process grate pressure prediction method based on information fusion
US20200074280A1 (en) * 2018-08-28 2020-03-05 Apple Inc. Semi-supervised learning using clustering as an additional constraint
CN111639820A (en) * 2020-06-09 2020-09-08 国电南瑞科技股份有限公司 Energy consumption parameter optimization method and system for cement kiln production
CN111701697A (en) * 2020-05-26 2020-09-25 上海万澄环保科技有限公司 Cement raw material grinding system and automatic optimization control method thereof

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20200074280A1 (en) * 2018-08-28 2020-03-05 Apple Inc. Semi-supervised learning using clustering as an additional constraint
CN110751202A (en) * 2019-10-16 2020-02-04 济南大学 Method for establishing load soft measurement model of cement combined grinding ball mill and application
CN110766234A (en) * 2019-10-29 2020-02-07 济南大学 Cement cooling process grate pressure prediction method based on information fusion
CN111701697A (en) * 2020-05-26 2020-09-25 上海万澄环保科技有限公司 Cement raw material grinding system and automatic optimization control method thereof
CN111639820A (en) * 2020-06-09 2020-09-08 国电南瑞科技股份有限公司 Energy consumption parameter optimization method and system for cement kiln production

Non-Patent Citations (5)

* Cited by examiner, † Cited by third party
Title
GSCIENTY: "基于K-均值聚类方法求取基函数中心c"https://blog.csdn.net/gscienty/article/details/53366974", 《CSDN技术交流论坛》 *
肖明: "《大学计算机基础》", 30 July 2019 *
蔡学云: "水泥生产生料立磨智能控制系统研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 *
风翼冰舟: "人工神经网络——径向基函数(RBF)神经网络", 《CSDN技术交流论坛》 *
黄建华等: "《单片机原理及应用》", 30 August 2016 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114733640A (en) * 2022-03-03 2022-07-12 江苏丰尚智能科技有限公司 Method and device for adjusting processing parameters of pulverizer and computer equipment
CN114733640B (en) * 2022-03-03 2023-09-12 江苏丰尚智能科技有限公司 Method and device for adjusting processing parameters of pulverizer and computer equipment

Similar Documents

Publication Publication Date Title
CN108920863B (en) Method for establishing energy consumption estimation model of robot servo system
CN103927412A (en) Real-time learning debutanizer soft measurement modeling method on basis of Gaussian mixture models
CN114997528B (en) Production parameter adjusting method and system based on injection molding product quality analysis
CN109409425B (en) Fault type identification method based on neighbor component analysis
CN110059824A (en) A kind of neural net prediction method based on principal component analysis
CN107403196B (en) Method for predicting concentration of butane at bottom of debutanizer by instant learning modeling based on spectral clustering analysis
CN114398049A (en) Self-adaptive dynamic updating method for digital twin model of discrete manufacturing workshop
US20230140696A1 (en) Method and system for optimizing parameter intervals of manufacturing processes based on prediction intervals
CN112275439A (en) Cement raw material vertical mill differential pressure soft measurement modeling method, storage medium and system
CN114583767B (en) Data-driven wind power plant frequency modulation response characteristic modeling method and system
CN112329142A (en) Method for identifying uncertain dynamic loads of wing structure based on support vector regression
CN108490782A (en) A kind of method and system being suitable for complex industrial process product quality indicator missing data completion based on selective double layer integrated study
CN115169809A (en) Smart city evaluation method and device
CN107766887A (en) A kind of local weighted deficiency of data mixes clustering method
CN116976209A (en) Data-driven unmanned aerial vehicle system frequency domain identification method and system based on joint decision
CN114800036B (en) Equipment health state assessment method
CN115526276A (en) Wind tunnel balance calibration load prediction method with robustness
CN111160464B (en) Industrial high-order dynamic process soft measurement method based on multi-hidden-layer weighted dynamic model
CN113191614B (en) Polypropylene product quality real-time monitoring method based on slow characteristic regression analysis
CN108875961A (en) A kind of online weighting extreme learning machine method based on pre- boundary's mechanism
CN110673470B (en) Industrial non-stationary process soft measurement modeling method based on local weighting factor model
CN114115150A (en) Data-based heat pump system online modeling method and device
CN118016218B (en) Intelligent analysis method for high-temperature-resistant rotary joint material performance data
CN106127326B (en) Chemical material processing melt index forecasting method
CN109902762A (en) The data preprocessing method deviateed based on 1/2 similarity

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
RJ01 Rejection of invention patent application after publication

Application publication date: 20210129

RJ01 Rejection of invention patent application after publication