CN116432068A - Method and system for predicting water content of materials of roller type de-enzyming dryer - Google Patents
Method and system for predicting water content of materials of roller type de-enzyming dryer Download PDFInfo
- Publication number
- CN116432068A CN116432068A CN202310392640.0A CN202310392640A CN116432068A CN 116432068 A CN116432068 A CN 116432068A CN 202310392640 A CN202310392640 A CN 202310392640A CN 116432068 A CN116432068 A CN 116432068A
- Authority
- CN
- China
- Prior art keywords
- water content
- samples
- sample set
- materials
- sample
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 title claims abstract description 108
- 239000000463 material Substances 0.000 title claims abstract description 103
- 238000000034 method Methods 0.000 title claims abstract description 67
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 29
- 238000003064 k means clustering Methods 0.000 claims abstract description 11
- 238000012795 verification Methods 0.000 claims description 11
- 238000012360 testing method Methods 0.000 claims description 9
- 238000011156 evaluation Methods 0.000 claims description 4
- 239000013598 vector Substances 0.000 claims description 4
- 239000011159 matrix material Substances 0.000 claims description 2
- 238000001035 drying Methods 0.000 abstract description 12
- 238000005516 engineering process Methods 0.000 abstract description 9
- 238000005259 measurement Methods 0.000 abstract description 6
- 230000033228 biological regulation Effects 0.000 abstract description 4
- 238000004519 manufacturing process Methods 0.000 abstract description 4
- 230000005611 electricity Effects 0.000 abstract description 2
- 239000000523 sample Substances 0.000 description 79
- 241001122767 Theaceae Species 0.000 description 44
- 230000008569 process Effects 0.000 description 15
- 102000004190 Enzymes Human genes 0.000 description 14
- 108090000790 Enzymes Proteins 0.000 description 14
- 230000009849 deactivation Effects 0.000 description 13
- 238000010586 diagram Methods 0.000 description 7
- 238000012545 processing Methods 0.000 description 5
- 238000010438 heat treatment Methods 0.000 description 4
- 230000015572 biosynthetic process Effects 0.000 description 3
- 239000011248 coating agent Substances 0.000 description 3
- 238000000576 coating method Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000012546 transfer Methods 0.000 description 3
- 229910000831 Steel Inorganic materials 0.000 description 2
- 238000010521 absorption reaction Methods 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 239000003205 fragrance Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 150000008442 polyphenolic compounds Chemical class 0.000 description 2
- 235000013824 polyphenols Nutrition 0.000 description 2
- 239000010959 steel Substances 0.000 description 2
- 102000030523 Catechol oxidase Human genes 0.000 description 1
- 108010031396 Catechol oxidase Proteins 0.000 description 1
- 102000004316 Oxidoreductases Human genes 0.000 description 1
- 108090000854 Oxidoreductases Proteins 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000005094 computer simulation Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000004134 energy conservation Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000002255 enzymatic effect Effects 0.000 description 1
- 230000009123 feedback regulation Effects 0.000 description 1
- 238000000855 fermentation Methods 0.000 description 1
- 230000004151 fermentation Effects 0.000 description 1
- 230000004907 flux Effects 0.000 description 1
- 239000002737 fuel gas Substances 0.000 description 1
- 239000007789 gas Substances 0.000 description 1
- 238000011065 in-situ storage Methods 0.000 description 1
- 239000002932 luster Substances 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000007500 overflow downdraw method Methods 0.000 description 1
- 230000003647 oxidation Effects 0.000 description 1
- 238000007254 oxidation reaction Methods 0.000 description 1
- 230000001590 oxidative effect Effects 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
Images
Classifications
-
- 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
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Probability & Statistics with Applications (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a method and a system for predicting the water content of materials of a drum-type de-enzyming dryer, wherein data information of a sample set is collected, and an initial clustering center and the number of clusters are determined based on the sample set; classifying samples in the sample set based on the initial cluster center and the number of clusters; and establishing a linear model of data information of samples in various types and the water content of the materials, and establishing a material water content prediction model based on the obtained various linear models. An initial clustering center algorithm and a clustering number determining method are designed based on a K-means clustering method, a drum-type water-removing dryer parameter sample set is clustered, a linear model of the water content of materials is built in various types, further a prediction model of the water content of the materials is built, online real-time measurement and effective prediction of the water-removing and drying technology of the drum-type water-removing dryer are achieved, closed-loop regulation and control of the water content of the materials are better achieved, accurate temperature control is further achieved, and accordingly production electricity consumption of the full-electric water-removing machine is reduced.
Description
Technical Field
The invention relates to the technical field of tea fixation, in particular to a method and a system for predicting the water content of materials of a roller type fixation dryer.
Background
The tea enzyme deactivation is an important procedure in tea processing, and is mainly characterized in that the oxidase activity in fresh leaves is destroyed and passivated at high temperature, the enzymatic oxidation of tea polyphenol and the like in the fresh leaves is inhibited, and partial moisture of the fresh leaves is evaporated, so that the tea is softened, the tea is convenient to knead and shape, meanwhile, the green odor is dispersed, and the formation of good aroma is promoted. The tea is rich in polyphenol oxidase, and if the tea is not deactivated, polyphenol substances are subjected to oxidative fermentation under the action of the enzyme, so that the tea can lose color and luster quickly and then be damaged. The tea enzyme deactivation process is also the process of preliminary drying of tea.
The tea enzyme deactivation technology comprises hot air enzyme deactivation, steam enzyme deactivation, frying enzyme deactivation and the like, and the enzyme deactivation machine correspondingly generates along with the continuous development of the technology. The drum-type de-enzyming and drying machine can be used for de-enzyming and drying simultaneously, and the tea is prepared by the de-enzyming process, so that the fragrance of the finished product can be greatly improved, and the drum-type de-enzyming and drying machine has the advantages of compact machine body structure, convenience in operation, high efficiency and the like and is widely used. The machine consists of a roller, a transmission device, a fan, a frame, a heating system and the like. The cylinder is rolled by steel plate, and the inner wall is welded with guide vane plate, and guide vane angle is 24 °, and its effect is helpful to advance the leaf or push out the material to throw down after taking tealeaves to certain height. The fan can be rotated in a forward and reverse direction to blow the leaves out of the barrel or suck and discharge moisture. Annular sliding rails closely connected with the cylinder body are arranged at two ends of the roller, and the sliding rails are supported on the frame by bearings. The frame is generally made of angle steel.
The fixation process plays a decisive role in the formation of tea quality, accurate monitoring of the water content of the tea material in the link is particularly important for processing quality and saving electric energy, but because the water content of the material is changed greatly and quickly, the accurate online detection and real-time feedback regulation and control are not easy in processing, the fixation technology of the conventional drum-type fixation dryer cannot measure and effectively predict the water content of the material in real time, and cannot realize closed-loop regulation and control of the water content of the material, so that the power consumption of the full-electric fixation machine is higher, and the tea quality is difficult to effectively promote. In addition, the traditional heating system also faces the problem of being heated by fuel gas, which is unfavorable for energy conservation and environmental protection.
Disclosure of Invention
In view of the above, the invention provides a method and a system for predicting the water content of materials of a roller type water-removing dryer, which aim to realize online real-time measurement and effective prediction of the water-removing technology of the roller type water-removing dryer, so that the water content of the materials is better regulated and controlled in a closed loop, and further, the accurate temperature control is realized, thereby reducing the production electricity consumption of the full-electric water-removing machine.
In order to solve the technical problems, the technical scheme of the invention is to provide a method for predicting the water content of materials of a roller type de-enzyming dryer, which comprises the following steps:
collecting data information of a sample set, and determining an initial clustering center and the number of clusters based on the sample set;
classifying samples in the sample set based on the initial cluster center and the number of clusters;
establishing a linear model of data information of samples in various types and the water content of materials, and establishing a material water content prediction model based on the obtained various linear models;
and predicting the samples in the sample set by using the material water content prediction model to obtain a water content prediction result of the samples.
As an embodiment, the collecting data of the sample includes:
and obtaining data information of the sample by measuring auxiliary variable information of the sample, wherein the auxiliary variables comprise roller temperature, roller inclination angle, roller rotating speed, fresh leaf water content and feeding speed.
As an embodiment, the method for determining an initial cluster center includes:
determining a first clustering center according to relation information among samples in the sample set;
and carrying out clustering iteration according to the first clustering center to determine an initial clustering center.
As one embodiment, the method for determining the first clustering center according to the relation information among the samples in the sample set comprises
Determining the distance between each sample in the sample set:
calculating to obtain the average distance between samples based on the distance between samples in the sample set;
a sample density is calculated based on the distance between the samples and the average spacing between the samples, and a first cluster center is determined based on the sample density.
As one embodiment, the method for determining the number of clusters includes:
determining the number of clusters based on a satisfactory clustering method, specifically,
and adopting a cluster evaluation index as modeling root mean square error, and judging the number of clusters according to the result of the root mean square error.
As one embodiment, the classifying the samples in the sample set based on the initial cluster center and the number of clusters includes:
and classifying the samples in the sample set by adopting a K-means clustering algorithm based on the initial clustering center and the clustering number.
As an implementation manner, the establishing a linear model of data information and material water content of samples in each class, and establishing a material water content prediction model based on the obtained linear models, includes:
a least square method is utilized to establish a linear model of data information of samples in various types and the water content of materials, in particular,
and (3) establishing a linear model of the material moisture content and the roller temperature, the roller rotating speed, the roller inclination angle, the feeding speed and the fresh leaf material moisture content by using a recursive least square method.
As one embodiment, the collected sample set is divided into a test sample set and a verification sample set, a material water content prediction model is built by using the test sample set, and the built material water content prediction model is verified by using the verification sample set.
In addition, the invention also provides a material water content prediction system of the roller type de-enzyming dryer, which comprises the following components:
the information acquisition module is used for acquiring data information of a sample set and determining an initial clustering center and the number of clusters based on the sample set;
the sample classification module is used for classifying samples in the sample set based on the initial cluster center and the number of clusters;
the model building module is used for building a linear model of data information of samples in various types and the water content of the materials, and building a material water content prediction model based on the obtained various linear models;
and the result prediction module is used for predicting the samples in the sample set by using the material water content prediction model to obtain a water content prediction result of the samples.
As one embodiment, the result prediction module further includes a prediction evaluation unit, and the verification unit is configured to evaluate a water content prediction result of the sample.
The invention provides a method and a system for predicting the water content of materials of a drum-type de-enzyming dryer, wherein data information of a sample set is collected, and an initial clustering center and the number of clusters are determined based on the sample set; classifying samples in the sample set based on the initial cluster center and the number of clusters; establishing a linear model of data information of samples in various types and the water content of materials, and establishing a material water content prediction model based on the obtained various linear models; and predicting the samples in the sample set by using the material water content prediction model to obtain a water content prediction result of the samples. An initial clustering center algorithm and a clustering number determining method are designed based on a K-means clustering method, a K-means algorithm is utilized to cluster a drum-type de-enzyming dryer process parameter sample set, a linear model of the water content of materials is built in each type, a model switching algorithm is designed, a prediction model of the water content of the materials is built based on the linear model, on-line real-time measurement and effective prediction of the de-enzyming technology of the drum-type de-enzyming dryer are achieved, therefore, closed-loop regulation and control of the water content of the materials are better conducted, the tea processing process is enhanced, and the tea quality is improved. And further realize accurate accuse temperature to reduce the production power consumption of full electric energy fixation machine.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the description of the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a schematic diagram of steps of a method for predicting a water content of a material in a drum-type de-enzyming dryer according to an embodiment of the present invention;
FIG. 2 is a diagram illustrating a multi-model modeling principle according to an embodiment of the present invention;
FIG. 3 is a schematic diagram showing a comparison of a predicted water content of a material and an actual water content of the material according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of model prediction error according to an embodiment of the present invention;
fig. 5 is a schematic structural diagram of a material water content prediction system of a drum-type de-enzyming dryer according to an embodiment of the present invention.
Detailed Description
In order that those skilled in the art will better understand the embodiments of the present invention, a technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in which it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The tea enzyme deactivation technology comprises hot air enzyme deactivation, steam enzyme deactivation, frying enzyme deactivation and the like, and the enzyme deactivation machine correspondingly generates along with the continuous development of the technology. The drum-type de-enzyming and drying machine can be used for de-enzyming and drying simultaneously, and the tea is prepared by the de-enzyming process, so that the fragrance of the finished product can be greatly improved, and the drum-type de-enzyming and drying machine has the advantages of compact machine body structure, convenience in operation, high efficiency and the like and is widely used. The embodiment is described with respect to tea hot working in a drum-type de-enzyming dryer.
The water-removing machine is heated by full electric energy, an electric energy heating system consists of a far infrared heat source, a high-efficiency heat-absorbing coating on the outer wall of a roller and the like, and the roller body is arranged in the heating system. The principle of fixation is that the far infrared heat source converts electric energy into far infrared rays to irradiate the high-efficiency heat absorption coating, the high-efficiency heat absorption coating heats up and heats the rotating cylinder after absorbing the far infrared energy, tea leaves are thrown and rotated in the cylinder by friction and the rotating centrifugal force of the cylinder, and the heat is absorbed to achieve the aim of fixation or drying.
Further, the drum-type de-enzyming dryer heats the outer wall of the drum through the heat source, the inner wall of the drum directly heats tea leaves, and the moisture in the heated tea leaves continuously diffuses to the surfaces of the tea leaves and forms saturated water vapor on the surfaces. Because the rotation of the roller forms the hot air flowing to the discharge end of the roller in the roller, the hot air brings the water vapor on the surface of the tea out of the roller, and finally, the purpose of proper amount of tea water is achieved. The principle calculation formula of the water content of the material is as follows:
wherein p is the water content of the material, p0 is the water content of the fresh leaves, m is the feeding amount which depends on the feeding rate of the de-enzyming dryer, t is the de-enzyming time, and k is waterAnd (5) dividing the drying rate.
The drying of the tea moisture in the roller can be regarded as a forced convection mass transfer process in a straight pipe. The mass flux density, i.e. the moisture drying rate of tea leaves, in the forced convection mass transfer process in the drum according to the principle of heat transfer theory is related to the parameters of drum temperature, air flow speed in the drum, air flow humidity, drum diameter and the like.
The de-enzyming time determines the residence time of the tea leaves in the drum. The residence time of the tea leaves is related to the length and inclination angle of the drum, the diameter of the drum, the rotational speed of the drum, the gas flow rate in the drum, and the diameter of the tea leaves.
TABLE 1 variation of the moisture content of materials
Thus, all the variables relating to the water content of the material shown in Table 1 were obtained. The diameter, length and the like of the roller belong to invariants, the temperature of the roller is a key quantity affecting the water loss rate, the rotating speed and the inclined angle of the roller are key quantities affecting the enzyme deactivation time, and the intrinsic endowment and the feeding speed of fresh tea are key factors determining the water content of the material, so the temperature of the roller, the inclined angle of the roller, the rotating speed of the roller, the water content of the fresh tea and the feeding speed are selected as auxiliary variables for predicting the water content of the material.
Fig. 1 is a schematic diagram of a method for predicting a water content of a material in a drum-type de-enzyming dryer according to an embodiment of the present invention.
S11, collecting data information of a sample set, and determining an initial clustering center and the number of clusters based on the sample set.
The data information of the samples was obtained by measuring the auxiliary variable information of the samples, including drum temperature, drum tilt angle, drum rotation speed, fresh leaf moisture content and feed rate, as shown in table 1 above.
It should be noted that, the sample set collected in this embodiment is divided into a test sample set and a verification sample set (for convenience of understanding, in this embodiment, the sample set is set to 350 sets of data, 300 sets of data are taken as the test sample set, and 50 sets of data are taken as the verification sample set), that is, algorithm experiments and model establishment are performed by using the test sample set; and verifying the obtained algorithm and model by using a verification sample set, and determining the accuracy and feasibility of the algorithm and model.
The K-means clustering algorithm is used in a large amount in the data mining algorithm due to the fact that the K-means clustering algorithm is high in convergence speed and simple in structure, the clustering effect depends on the selection and the classification number of initial clustering centers, the initial clustering center algorithm and the clustering number determining method are designed based on the K-means clustering method, and specifically data information of samples is acquired according to vectors s= (t, n, a, v, y0, y), wherein t is the temperature of a roller, n is the rotating speed of the roller, a is the sine value of the inclination angle of the roller, v is the feeding speed, y0 is the fresh leaf moisture content and y is the moisture content of materials.
The implementation steps of the selection algorithm of the initial clustering center are as follows:
(1) Calculating the distance between the samples
x ir Is the r component of the i-th sample.
(2) Calculating the average spacing between samples:
(3) Calculating the sample density den (i), den (i) being d ij(j＝1…300) Is satisfied by (1)
Taking the sample point with the maximum density as the first clustering center.
(4) Calculating the distance from the rest samples to the existing clustering center, and carrying out clustering iteration, wherein the distance from the ith sample to the existing clustering center is dis (i) =min { d (i, j) }, and j is the number of the existing clustering center point; such that formula j=den (i) +dis (i) - (den (i) -dis (i)) 2 The sample i with the maximum value is the next cluster center.
(5) Repeating the step (4) until the number of the clustering centers is the number of clusters.
According to the formula J=den (i) +dis (i) - (den (i) -dis (i)) 2 The selected cluster center ensures that the density of the new initial cluster center is larger, and the distance from the new initial cluster center to the existing initial cluster center is longer, so that the selection of the point with small density or at the edge as the initial cluster center is avoided.
Further, the number of clusters is determined. As with most other clustering algorithms, the K-means clustering algorithm also requires a pre-given number of clusters. The purpose of clustering is to convert the nonlinear model into linear models of all the subareas, and the clustering number is the number of the final linear subareas; the number of clusters has an important influence on the similarity of samples in various types and the difference of samples among the types so as to influence modeling accuracy and speed, however, the number of clusters is difficult to determine in advance under the condition that the distribution boundary of the samples is not known. The comparison method, the fusion method and the trial-and-error method all require larger calculated quantity for determining the number of clusters, and the embodiment adopts a method based on satisfactory clustering to determine the number of clusters under the condition of no priori knowledge. The adopted clustering evaluation index is modeling root mean square error:
y i for the actual output value of the ith sample, < >>
And predicting an output value for the model of the ith sample, wherein nt is the number of clustered samples. Setting the initial value of N to 2, if RMSE<0.02N is the number of clusters; otherwise, N is increased and initial cluster center clustering is determined and modeled according to a selection algorithm of the initial cluster center until the clustering precision meets RMSE<0.02. The clustering initial value algorithm can greatly improve the operation speed because the existing clustering center is not recalculated when the number of clusters is increased.
It should be noted that, unlike the comparison method for determining the number of clusters, the number of clusters determined in this embodiment is the minimum number of clusters that satisfies the modeling accuracy, which is found under the road conditions of low operation, but not the number of clusters that satisfies the highest modeling accuracy.
S12, classifying the samples in the sample set based on the initial cluster center and the number of clusters.
Based on the obtained initial clustering centers and the number of clusters, classifying samples in the sample set by adopting a K-means clustering algorithm, specifically, determining that the initial clustering centers determined according to the algorithm are all points in the samples, and setting a vector set as Z 1 (j) (j=1, 2,3,..n), N is the number of clusters, then the K-means clustering algorithm implementation process is:
(1) Calculating Euclidean distance d (v (i), Z of each sample to the clustering center I (j) I=1, 2,3,) N, j=1, 2,3,) N, N is the sample volume if: d (x (i), Z I (k))＝min{d(x(i),Z I (j) Sample x (i) then scores into the kth class.
(2) Calculating an error criterion function, wherein the formula is as follows:
n i for the i-th sample number, x () i Samples classified into class i.
(3) Judging: if S (I) -S (I-1) |<ζ then the algorithm ends, otherwise i=i+1, and according to formula
The new cluster center is recalculated.
(4) Returning to the step (1) until all samples are classified.
And S13, establishing a linear model of data information of samples in various types and the water content of the materials, and establishing a material water content prediction model based on the obtained various linear models.
In order to convert the global nonlinear model establishment into a local linear model, and then realize global modeling according to the identification of the local model and the switching of the model, the embodiment adopts a multi-model modeling mode as shown in fig. 2. The essence of multi-model modeling is the decomposition and synthesis process of the model, and is an effective scheme for solving nonlinear system modeling. According to the algorithm, classifying the sample data sets, then establishing linear models Mi of the water content of the materials and the sample data in each type, finding out the type to which x belongs according to the corresponding mode switching algorithm for new sample data x, calculating the water content of the materials when the sample data is x according to the linear models in the corresponding type, establishing each type of linear model by adopting a recursive least square method in order to facilitate real-time updating of each type of linear model, and selecting the linear model of the type where 2 clustering centers with the shortest x distance are located to predict the water content of the materials.
Specifically, the method for establishing the linear model in each class comprises the steps of dividing tea data samples into N classes according to the algorithm, and establishing the linear model of the material moisture content, the sine value of the roller temperature, the roller rotating speed and the roller inclination angle, the feeding speed and the fresh leaf moisture content in each class by using a recursive least square method: y is i ＝w i1 t+w i2 n+w i3 a+w i4 v+w i5 y 0 I.e.
Wherein: x= [ t, n, a, v, y 0 ]；W i ＝[w i1 ,w i2 ,w i3 ,w i4 ,w i5 ] T 。i＝1,2,3,...,N，w ij (j=1, 2,3,.. 5) parameters to be identified for building a linear model in class i, the recursive least squares identification procedure is:
wherein (1)>
Ni is the number of samples of the i-th class. K (K) is calculated first, then W (K) and P (K) are calculated, the initial value of K (K) is a smaller parameter, and the initial value of P (K) is a diagonal matrix with a sufficiently large diagonal value.
For tea making auxiliary variable parameter vectors x (i) = [ t (i), n (i), a (i), v (i), y set according to user requirements 0 (i)]Calculating the distance d (x (i), Z) from x (i) to each cluster center I (k) A kind of electronic device. Selecting x (i) to distanceClosest cluster center Z I (r)、Z I (s) let x (i) to Z I (r)、Z I The distances of(s) are d 1 、d 2 And d 1 ≤d 2 。
If d 1 ≤0.5d 2 Then y=w r T x, in contrast,
s14, predicting the samples in the sample set by using a material water content prediction model to obtain a water content prediction result of the samples.
After the water content prediction model is obtained, the embodiment also uses the sample in the verification sample set to verify the obtained prediction model, so that the feasibility and the accuracy of the model are verified according to the comparison between the verification result and the actual water content value.
The embodiment provides a method and a system for predicting the water content of materials of a drum-type de-enzyming dryer, which are characterized in that data information of a sample set is collected, and an initial clustering center and the number of clusters are determined based on the sample set; classifying samples in the sample set based on the initial cluster center and the number of clusters; establishing a linear model of data information of samples in various types and the water content of materials, and establishing a material water content prediction model based on the obtained various linear models; and predicting the samples in the sample set by using a material water content prediction model to obtain a water content prediction result of the samples. An initial clustering center algorithm and a clustering number determining method are designed based on a K-means clustering method, a K-means algorithm is utilized to cluster a process parameter sample set of the drum-type water-removing dryer, a linear model of the water content of materials is built in each type, a model switching algorithm is designed, and then a prediction model of the water content of the materials is built based on the linear model, so that online real-time measurement and effective prediction of the water content of the materials are realized, closed-loop regulation and control are better carried out on the water content of the materials, the tea processing technology is enhanced, and the tea quality is improved. And further realize accurate accuse temperature to reduce the production power consumption of full electric energy fixation machine.
Further, in the embodiment, the model verification is performed by adopting in-situ sampling, and a 6CST-55 numerical control infrared roller type de-enzyming dryer produced by Sichuan middle measuring instrument company is adopted, and the roller temperature, the roller rotating speed and the inclination angle of the roller feeding end can be accurately controlled by the roller type de-enzyming dryer. In order to control the feeding speed, a 6CST-10 type automatic feeder is arranged in front of a 6CST-55 type de-enzyming dryer in the experiment. The fresh leaf moisture content and the material moisture content are measured by using an IR-3000 near infrared online moisture meter of MoistTech company in the United states, the distance between a sensor and a sample is 20cm during measurement, the fresh leaf moisture content and the material moisture content are sampled and measured 5 times, the average value is used as a final measured value, and the material moisture content is measured when the material is cooled to room temperature.
The fresh tea leaves come from autumn tea in the same tea garden, 350 tests are arranged according to the picking schedule of the tea leaves, 300 groups of data are used as modeling data, and the rest 50 groups of data are used for verifying the accuracy of the model.
Based on the sample data, the method for determining the number of clusters provided by the embodiment discovers that when the number of clusters is 5, better modeling accuracy can be obtained. At this time, the predicted and measured values of the water content model of the sample material are shown in fig. 3, the root mean square error RMSE is 0.01945, and the maximum error MAXE is 0.0463, where MAXE is defined as:
the model prediction error is shown in fig. 4, and it can be seen that the measurement model has better prediction accuracy in each sample interval. Meanwhile, in fig. 3, the situation that the deviation between the predicted value and the measured value is large can be seen from the individual sample points, which is mainly caused by the fact that the accuracy of the local linear model is not high due to the fact that the modeling samples are few, and of course, the fact that the test samples are distributed at the edges of each subarea is also possible.
By comparing the predicted value and the true value of the model, the accuracy and feasibility of the second prediction method of the material water content prediction model provided by the invention are verified, and conditions are provided for realizing the control of the material water content.
Further, as shown in fig. 5, a schematic structural diagram of a material water content prediction system of a drum-type de-enzyming dryer according to an embodiment of the present invention is provided.
The utility model provides a material moisture content prediction system of drum-type dryer, includes information acquisition module, sample classification module, model establishment module, result prediction module, wherein:
the information acquisition module is used for acquiring data information of the sample set and determining an initial clustering center and the number of clusters based on the sample set;
the sample classification module is used for classifying samples in the sample set based on the initial clustering center and the clustering number;
the model building module is used for building a linear model of data information of samples in various types and the water content of the materials, and building a material water content prediction model based on the obtained various linear models;
and the result prediction module is used for predicting the samples in the sample set by using the material water content prediction model to obtain a water content prediction result of the samples.
According to the embodiment, the information acquisition module acquires the data information of the sample, the initial clustering center and the clustering number are determined according to the acquired information, so that the original tea is classified, then various linear models are built, the material water content prediction model is built based on the acquired various linear models, the material water content can be predicted based on part of the data information, the control of the material water content is realized, the fixation efficiency and fixation capacity are improved, and the fixation quality of the tea is further improved.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or system. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or system that comprises the element.
The foregoing is merely a preferred embodiment of the present invention, and it should be noted that the above-mentioned preferred embodiment should not be construed as limiting the invention, and the scope of the invention should be defined by the appended claims. It will be apparent to those skilled in the art that various modifications and adaptations can be made without departing from the spirit and scope of the invention, and such modifications and adaptations are intended to be comprehended within the scope of the invention.
Claims (10)
1. The method for predicting the water content of the materials of the roller type de-enzyming dryer is characterized by comprising the following steps of:
collecting data information of a sample set, and determining an initial clustering center and the number of clusters based on the sample set;
classifying samples in the sample set based on the initial cluster center and the number of clusters;
establishing a linear model of data information of samples in various types and the water content of materials, and establishing a material water content prediction model based on the obtained various linear models;
and predicting the samples in the sample set by using the material water content prediction model to obtain a water content prediction result of the samples.
2. The method for predicting the water content of materials in a drum-type de-enzyming dryer according to claim 1, wherein the data of the collected samples comprises:
and obtaining data information of the sample by measuring auxiliary variable information of the sample, wherein the auxiliary variables comprise roller temperature, roller inclination angle, roller rotating speed, fresh leaf material moisture content and feeding speed.
3. The method for predicting the water content of materials of a drum-type de-enzyming dryer according to claim 1, wherein the method for determining the initial clustering center comprises the following steps:
determining a first clustering center according to relation information among samples in the sample set;
and carrying out clustering iteration according to the first clustering center to determine an initial clustering center.
4. The method for predicting water content of materials in a drum-type de-enzyming dryer as claimed in claim 3, wherein said determining a first cluster center based on the relationship information among samples in the sample set includes
Determining the distance between each sample in the sample set:
calculating to obtain the average distance between samples based on the distance between samples in the sample set;
a sample density is calculated based on the distance between the samples and the average spacing between the samples, and a first cluster center is determined based on the sample density.
5. The method for predicting the water content of materials of a drum-type de-enzyming dryer of claim 4, wherein the method for determining the number of clusters comprises the following steps:
determining the number of clusters based on a satisfactory clustering method, specifically,
adopting the clustering evaluation index as modeling root mean square error
And judging the number of clusters according to the root mean square error result. Wherein y is i For the actual output value of the ith sample, < >>
And predicting an output value for the model of the ith sample, wherein nt is the number of clustered samples.
6. The method for predicting the water content of materials in a drum-type de-enzyming dryer according to claim 5, wherein classifying the samples in the sample set based on the initial cluster center and the number of clusters comprises:
based on the initial cluster center and the number of clusters, classifying the samples in the sample set by adopting a K-means clustering algorithm, specifically,
let its vector set be Z 1 (j) (j=1, 2,3, n., N is the number of the clusters,
(1) Calculating Euclidean distance d (v (i), Z of each sample to the clustering center I (j) I=1, 2,3, N, j=1, 2,3, N is the sample volume, if d (x (i), Z) is satisfied I (k))＝min{d(x(i),Z I (j) Sample x (i) is scored into the kth class).
(2) Calculating an error criterion function
n i For the i-th sample number, x () i Samples classified into class i.
(3) Judging: if S (I) -S (I-1) |<ζ then the algorithm ends, otherwise i=i+1, and as
The new cluster center is recalculated.
(4) Returning to (1).
7. The method for predicting the water content of materials of a drum-type de-enzyming dryer according to claim 2, wherein the steps of establishing a linear model of data information of samples in each type and the water content of materials, and establishing a material water content prediction model based on the obtained linear model include:
a least square method is utilized to establish a linear model of data information of samples in various types and the water content of materials, in particular,
a linear model of the material moisture content and the roller temperature, the roller rotating speed, the roller inclination angle, the feeding speed and the fresh leaf material moisture content is established by using a recursive least square method: y is i ＝w i1 t+w i2 n+w i3 a+w i4 v+w i5 y 0 I.e. y i ＝W i T x, wherein x= [ t, n, a, v, y 0 ]，W i ＝[w i1 ,w i2 ,w i3 ,w i4 ,w i5 ] T 。i＝1,2,3,...,N，w ij (j=1, 2,3,.. 5) parameters to be identified for building a linear model in class i, the recursive least squares identification procedure is:
wherein:
ni is the number of samples of class i. K (K) is calculated first, then W (K) and P (K) are calculated, the initial value of K (K) is a smaller parameter, and the initial value of P (K) is a diagonal matrix with a sufficiently large diagonal value.
8. The method for predicting the water content of materials of a drum-type de-enzyming dryer according to claim 1, further comprising:
dividing the collected sample set into a test sample set and a verification sample set, establishing a material water content prediction model by using the test sample set, and verifying the established material water content prediction model by using the verification sample set.
9. A material moisture content prediction system of drum-type de-enzyming drying-machine, characterized by comprising:
the information acquisition module is used for acquiring data information of a sample set and determining an initial clustering center and the number of clusters based on the sample set;
the sample classification module is used for classifying samples in the sample set based on the initial cluster center and the number of clusters;
the model building module is used for building a linear model of data information of samples in various types and the water content of the materials, and building a material water content prediction model based on the obtained various linear models;
and the result prediction module is used for predicting the samples in the sample set by using the material water content prediction model to obtain a water content prediction result of the samples.
10. The system for predicting the water content of materials of a drum-type de-enzyming dryer of claim 9, wherein said result predicting module further comprises a predicting and evaluating unit, and said verifying unit is used for evaluating the water content predicting result of said sample.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310392640.0A CN116432068A (en) | 2023-04-12 | 2023-04-12 | Method and system for predicting water content of materials of roller type de-enzyming dryer |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310392640.0A CN116432068A (en) | 2023-04-12 | 2023-04-12 | Method and system for predicting water content of materials of roller type de-enzyming dryer |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116432068A true CN116432068A (en) | 2023-07-14 |
Family
ID=87093977
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310392640.0A Pending CN116432068A (en) | 2023-04-12 | 2023-04-12 | Method and system for predicting water content of materials of roller type de-enzyming dryer |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116432068A (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20090106305A (en) * | 2008-04-05 | 2009-10-08 | 서희동 | A method to produce black tea using deep sea water |
| CN107066823A (en) * | 2017-04-18 | 2017-08-18 | 中山大学 | Based on plant, soil, microorganism heavy-metal contaminated soil repairing effect integrated evaluating method |
| CN113063754A (en) * | 2021-04-21 | 2021-07-02 | 中国热带农业科学院橡胶研究所 | Leaf phosphorus content detection method based on weighted environment variable clustering |
-
2023
- 2023-04-12 CN CN202310392640.0A patent/CN116432068A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20090106305A (en) * | 2008-04-05 | 2009-10-08 | 서희동 | A method to produce black tea using deep sea water |
| CN107066823A (en) * | 2017-04-18 | 2017-08-18 | 中山大学 | Based on plant, soil, microorganism heavy-metal contaminated soil repairing effect integrated evaluating method |
| CN113063754A (en) * | 2021-04-21 | 2021-07-02 | 中国热带农业科学院橡胶研究所 | Leaf phosphorus content detection method based on weighted environment variable clustering |
Non-Patent Citations (1)
| Title |
|---|
| JIANGWEN TANG等: "Research on Soft-sensing Method of Water Content of Fixed Tea Based on Improved Clustering", 《FOREST CHEMICALS REVIEW》, pages 2 - 5 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109740175B (en) | Outlier discrimination method for power curve data of wind turbine generator | |
| CN108131848B (en) | A kind of DSG slot type photo-thermal power station heat collecting field fluid properties prediction technique | |
| CN110173710B (en) | Energy-saving type climate feedback dynamic regulation and control device and method for eliminating smoke plume | |
| CN105180640B (en) | The method for drying wood of multi-parameter monitoring | |
| CN109577064A (en) | Mechanism modeling method for predicting energy consumption and evaporation capacity of drying part of toilet paper machine | |
| CN109724398B (en) | Wood drying control method and device based on artificial intelligence | |
| CN204757580U (en) | Energy -saving full -automatic postforming borduring machine | |
| CN101131344A (en) | Method for on-line measuring drying property of tobacco and kinetic analysis | |
| CN109342279B (en) | Mixed soft measurement method based on grinding mechanism and neural network | |
| CN108960492A (en) | A kind of exhaust enthalpy of turbine prediction technique based on PSO-SVR soft-sensing model | |
| CN206057142U (en) | A kind of super low concentration wet flue gas flue dust on-line measurement device | |
| CN111046532A (en) | Coherent cluster recognition method based on elbow criterion | |
| CN114115393A (en) | Method for controlling moisture and temperature at outlet of cut tobacco dryer for sheet cut tobacco making line | |
| CN105334738B (en) | A kind of method of evaluating performance suitable for tobacco processing course pid control circuit | |
| CN113157018A (en) | Method and device for controlling temperature of dryer, computer equipment and storage medium | |
| CN116432068A (en) | Method and system for predicting water content of materials of roller type de-enzyming dryer | |
| CN114642268A (en) | Cut tobacco drying intelligent control system and method based on volatile moisture content | |
| CN108446452B (en) | A kind of mixed-flow pump impeller Robust Optimal Design | |
| CN207313919U (en) | Carbon fiber starching drying device | |
| CN113280490A (en) | New wind system control method and system based on Internet of things and artificial intelligence and storage medium | |
| CN110286197B (en) | Method for representing consistency of cut tobacco processing strength in drum drying process | |
| CN111728253A (en) | Control method and system for tobacco airflow drying strength | |
| CN116746697A (en) | Environmental humidity control system and method for tobacco leaf baking process of sheet baking machine | |
| CN110826229A (en) | Cut tobacco drying head process modeling and optimal setting control method based on long-term prediction | |
| CN111062509A (en) | Wind power medium-short term prediction method and system based on improved neural network |
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 |