CN116432088A - Intelligent monitoring method and system for layer thickness of composite optical film - Google Patents

Intelligent monitoring method and system for layer thickness of composite optical film Download PDF

Info

Publication number
CN116432088A
CN116432088A CN202310488375.6A CN202310488375A CN116432088A CN 116432088 A CN116432088 A CN 116432088A CN 202310488375 A CN202310488375 A CN 202310488375A CN 116432088 A CN116432088 A CN 116432088A
Authority
CN
China
Prior art keywords
layer thickness
data
thickness data
optical film
composite optical
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202310488375.6A
Other languages
Chinese (zh)
Other versions
CN116432088B (en
Inventor
金正植
赵成权
李在成
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.)
Bao Xin New Material Suzhou Co ltd
Original Assignee
Bao Xin New Material Suzhou Co ltd
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 Bao Xin New Material Suzhou Co ltd filed Critical Bao Xin New Material Suzhou Co ltd
Priority to CN202310488375.6A priority Critical patent/CN116432088B/en
Publication of CN116432088A publication Critical patent/CN116432088A/en
Application granted granted Critical
Publication of CN116432088B publication Critical patent/CN116432088B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
    • G01B21/00—Measuring arrangements or details thereof, where the measuring technique is not covered by the other groups of this subclass, unspecified or not relevant
    • G01B21/02—Measuring arrangements or details thereof, where the measuring technique is not covered by the other groups of this subclass, unspecified or not relevant for measuring length, width, or thickness
    • G01B21/08—Measuring arrangements or details thereof, where the measuring technique is not covered by the other groups of this subclass, unspecified or not relevant for measuring length, width, or thickness for measuring thickness
    • 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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/004—Artificial life, i.e. computing arrangements simulating life
    • G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]

Landscapes

  • Engineering & Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Evolutionary Computation (AREA)
  • Artificial Intelligence (AREA)
  • General Engineering & Computer Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Biology (AREA)
  • Biomedical Technology (AREA)
  • Health & Medical Sciences (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Length Measuring Devices By Optical Means (AREA)

Abstract

The invention relates to the field of film layer thickness monitoring, in particular to an intelligent monitoring method and system for the layer thickness of a composite optical film, which are used for acquiring layer thickness data of each position of the composite optical film and acquiring a layer thickness monitoring matrix; calculating a distance factor between each layer thickness data and the layer thickness data in a local range; obtaining local variation indexes of each layer thickness data according to the distribution relation between each layer thickness data and the layer thickness data contained in the local range; selecting an initial core point by combining the local variation index and updating a new core point; constructing a clustering parameter optimized objective function, and dividing each layer thickness data according to optimal clustering parameters obtained by combining the firefly optimization algorithm with the objective function; extracting noise data and filtering to obtain the category of each noise data filtering value; and monitoring the thickness of the composite optical film layer according to the classification result. The invention intelligently monitors the thickness of the composite optical film layer, avoids the influence of noise points and improves the monitoring precision.

Description

Intelligent monitoring method and system for layer thickness of composite optical film
Technical Field
The application relates to the field of film layer thickness monitoring, in particular to an intelligent monitoring method and system for a composite optical film layer thickness.
Background
The modern society is rapidly developed, and living conditions of people are continuously improved, so that demands of people for various products are continuously increased, and the development of the packaging industry is promoted. The packaging is defined in the mind of everyone, and the packaging is visible everywhere, in daily life, people can contact the packaging every day or the packaging, people know that the outer packaging film of food can prevent the food from being polluted by the outside, so that the quality guarantee period of the food is prolonged, the demand of the film is also very large, the film is very thin as the name implies, people cannot distinguish the quality by naked eyes and feel, and therefore under the related thickness standard requirements, the quality of the film is monitored by a thickness gauge and a thickness measuring sensor, the quality of the film is mainly the thickness of the film, if the thickness of the film is uneven, the sealing effect is poor, the service life is short, the protection of the product is weak, the working efficiency and the product quality are reduced, and the monitoring of the thickness of the film is very important.
In summary, the invention provides an intelligent monitoring method and system for the layer thickness of a composite optical film, which are characterized in that a layer thickness monitoring matrix is constructed by collecting layer thickness data of each position of the composite optical film to be tested through a thickness measuring sensor, classification is carried out on the layer thickness data through self-adaptive clustering parameters, the higher data classification precision is achieved, noise data in the layer thickness data are extracted, the influence of the noise data on the layer thickness monitoring of the composite optical film to be tested is avoided, the intelligent monitoring of the layer thickness of the composite optical film to be tested is realized according to the classification result of the layer thickness data, early warning prompt is carried out on the uneven layer thickness of the composite optical film to be tested, the secondary processing is facilitated, and the layer thickness quality of the composite optical film is ensured.
Disclosure of Invention
In order to solve the technical problems, the invention provides an intelligent monitoring method and system for the thickness of a composite optical film layer, which are used for solving the existing problems.
The invention discloses an intelligent monitoring method and system for the layer thickness of a composite optical film, which adopts the following technical scheme:
the embodiment of the invention provides a method for intelligently monitoring the layer thickness of a composite optical film, which comprises the following steps:
acquiring layer thickness data of each position of the composite optical film to be detected through a thickness measuring sensor, and acquiring a layer thickness monitoring matrix;
obtaining local variation indexes of each layer thickness data according to the distribution relation between each layer thickness data and the layer thickness data contained in the local range; obtaining an initial core point of a DBSCAN clustering algorithm and updating a new core point according to local variation indexes of each layer thickness data;
taking the clustering radius and the minimum points in the class in the DBSCAN clustering algorithm as optimization parameters, and performing iterative optimization on the optimization parameters through a firefly optimization algorithm; obtaining an objective function of the optimization process according to the distribution condition of the layer thickness data in each class in the iterative optimization process and the spatial relationship between center points of each class;
according to a firefly optimization algorithm, combining an objective function to obtain an optimal cluster radius and the minimum number in the optimal class; obtaining a layer thickness data category classification result according to the initial core points, the updating of the new core points, the optimal clustering radius and the minimum points in the optimal class; obtaining noise data according to the layer thickness data classification result;
obtaining each noise filtering value according to the average value of the non-noise data contained in each noise local window; obtaining the category of each noise data filtering value according to the data difference between each noise data filtering value and each center point;
and performing intelligent monitoring on the layer thickness of the composite optical film to be detected according to classification results of all layer thickness data in the layer thickness monitoring matrix.
Preferably, the local variation index of each layer thickness data is obtained according to the distribution relation between each layer thickness data and the layer thickness data contained in the local range, and the expression is as follows:
Figure SMS_1
in the method, in the process of the invention,
Figure SMS_18
for layer thickness data
Figure SMS_21
Is used for determining the local variation index of the (c),
Figure SMS_23
for layer thickness data
Figure SMS_2
Variance of the distance factor from the layer thickness data in the local range,
Figure SMS_6
for layer thickness data
Figure SMS_10
Layer thickness data sets contained within a local range of (c),
Figure SMS_14
respectively to-be-measured composite optical film
Figure SMS_4
A position of,
Figure SMS_12
Layer thickness data at the location(s),
Figure SMS_16
as a logarithmic function based on a natural constant e,
Figure SMS_13
for layer thickness data
Figure SMS_17
Layer thickness data in a local range of (2)
Figure SMS_20
The number of occurrences in the local layer thickness data set,
Figure SMS_19
local layer thickness data set for all layer thickness data contained in a local rangeThe sum of the number of occurrences of the combination,
Figure SMS_22
as an exponential function based on a natural constant e,
Figure SMS_24
for the multiplication operation, Σ is the sum operation,
Figure SMS_25
is a distance threshold; wherein,,
Figure SMS_7
for layer thickness data
Figure SMS_11
And layer thickness data
Figure SMS_15
Layer thickness data in local range
Figure SMS_5
The distance factor between the two is expressed as follows:
Figure SMS_26
in the method, in the process of the invention,
Figure SMS_27
for layer thickness data
Figure SMS_28
The variance of all layer thickness data in a local range of (a),
Figure SMS_29
to take absolute value.
Preferably, the updating of the initial core point and the new core point of the DBSCAN clustering algorithm is obtained according to the local variation index of each layer thickness data, specifically: taking layer thickness data with the smallest local variation index in the layer thickness monitoring matrix as an initial core point; and after the class clustering corresponding to the initial core point is completed, the layer thickness data with the minimum local variation index in the layer thickness data except the initial core point and the layer thickness data in the corresponding class is used as a new core point, and the core points are updated in sequence.
Preferably, the objective function of the optimization process is obtained according to the distribution condition of the layer thickness data in each class and the spatial relationship between the center points of each class in the iterative optimization process, and the expression is as follows:
Figure SMS_30
wherein F is an objective function, Z is the class number of the layer thickness data cluster,
Figure SMS_40
for the variance of the layer thickness data in class i, max () is the sign of the maximum value,
Figure SMS_34
is the spatial average distance of each layer thickness data in class i to the center point of the class, wherein,
Figure SMS_36
，
Figure SMS_45
for the row and column numbers of layer thickness data x in the layer thickness monitoring matrix in class i,
Figure SMS_48
is the center point of category i
Figure SMS_46
The number of rows and columns in the layer thickness monitoring matrix,
Figure SMS_49
for the total number of layer thickness data in category i,
Figure SMS_39
is the spatial average distance of each layer thickness data in category s to the center point of the category, wherein,
Figure SMS_43
，
Figure SMS_31
for the row and column numbers of layer thickness data y in the layer thickness monitoring matrix in category s,
Figure SMS_35
is the center point of category s
Figure SMS_32
The number of rows and columns in the layer thickness monitoring matrix,
Figure SMS_37
for the total number of layer thickness data in category s,
Figure SMS_41
is the center point of category i
Figure SMS_50
Center point with category sA spatial distance between the two, wherein,
Figure SMS_44
，
Figure SMS_47
is the center point of category i
Figure SMS_51
Center point with category s
Figure SMS_33
Corresponding to the absolute value of the difference in layer thickness data,
Figure SMS_38
for the multiplication operation, Σ is the sum operation.
Preferably, the noise data is obtained according to the layer thickness data classification result, specifically: and taking all the categories containing only one data as noise categories, and recording the data in the noise categories as noise data.
Preferably, the obtaining the category of each noise data filtering value according to the data difference between each noise data filtering value and each center point specifically includes: and calculating the absolute value of the difference value between the noise data filtering value and each class of center points, and taking the class with the smallest absolute value as the class to which the noise data filtering value belongs.
Preferably, the intelligent monitoring of the layer thickness of the composite optical film to be tested is performed according to the classification result of all layer thickness data in the layer thickness monitoring matrix, specifically: counting the classified category number of all layer thickness data in the layer thickness monitoring matrix, and when the classified category number is equal to 1, uniformly measuring the layer thickness of the composite optical film; when the number of the classified categories is larger than 1, the thickness of the composite optical film layer to be measured is uneven.
In a second aspect, the present invention provides an intelligent monitoring system for a layer thickness of a composite optical film, including a processor and a memory, where the processor is configured to process instructions stored in the memory, so as to implement the above-mentioned intelligent monitoring method for a layer thickness of a composite optical film.
The invention has at least the following beneficial effects:
according to the invention, the category of each layer thickness data is obtained by analyzing the layer thickness data at each position of the composite optical film, the noise data in the layer thickness monitoring matrix is extracted, and the noise data is filtered to obtain each noise filtering value, so that the influence of the noise data in the intelligent monitoring process of the layer thickness of the composite optical film is avoided, and the intelligent monitoring precision of the layer thickness of the composite optical film is improved;
according to the method, the initial core points in the layer thickness data category division process are selected through the local variation indexes of each layer thickness data, and the core points are iteratively updated, so that the problems of low clustering speed, low clustering precision and the like caused by poor selection of the initial core points are solved, meanwhile, the clustering parameters in the layer thickness data category division process are optimized by combining a firefly algorithm and constructing an objective function, the randomness of manually selecting the clustering parameters is solved, the layer thickness data category division precision is improved, and the problem of misclassification of the layer thickness data is avoided.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions and advantages of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a method and system for intelligently monitoring the thickness of a composite optical film layer.
Detailed Description
In order to further describe the technical means and effects adopted by the present invention to achieve the preset purpose, the following detailed description refers to specific implementation, structure, characteristics and effects of a method and a system for intelligently monitoring the thickness of a composite optical film layer according to the present invention, which are provided by the present invention, with reference to the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The invention provides a method and a system for intelligently monitoring the thickness of a composite optical film layer, which are concretely described below with reference to the accompanying drawings.
The embodiment of the invention provides a method and a system for intelligently monitoring the thickness of a composite optical film layer.
Specifically, the method and system for intelligently monitoring the layer thickness of a composite optical film according to the present embodiment provide a method for intelligently monitoring the layer thickness of a composite optical film as follows, referring to fig. 1, the method includes the following steps:
and S001, collecting layer thickness data of each position of the composite optical film to be detected through a thickness measuring sensor, and obtaining a layer thickness monitoring matrix.
Firstly, the thickness data of each position of the composite optical film to be measured is collected by the thickness measuring sensor, and the specific thickness measuring sensor model and the thickness data collecting process are known techniques, which are not described in the embodiment. In order to facilitate analysis and monitoring of the layer thickness of the composite optical film, in this embodiment, the dimensions of the composite optical film to be tested obtained each time are set to be regular square, so as to construct a layer thickness monitoring matrix according to the collected layer thickness data of each position of the composite optical film to be tested, where the layer thickness monitoring matrix of the composite optical film to be tested specifically includes:
Figure SMS_52
in the method, in the process of the invention,
Figure SMS_54
is a composite optical film to be tested
Figure SMS_56
Layer thickness data at the location(s),
Figure SMS_58
and K is a layer thickness monitoring matrix corresponding to the composite optical film to be tested.
So far, the thickness data of each position of the composite optical film to be detected can be acquired through the thickness sensor, and the thickness monitoring matrix of the composite optical film to be detected can be obtained.
Step S002, optimizing the clustering parameters by combining with a firefly algorithm, classifying each layer thickness data in the layer thickness monitoring matrix according to the optimized clustering algorithm, extracting noise data, and obtaining the final belonging category of each layer thickness data.
When the thickness intelligent monitoring of the thickness of the composite optical film to be detected is carried out based on the thickness data acquired by the thickness measuring sensor, the thickness measuring sensor is extremely easy to be influenced by external environment noise in the data acquisition process, so that a large number of noise points exist in the acquired monitoring data; meanwhile, noise is generated by components in the sensor, and the acquisition of the layer thickness data is influenced to some extent, so that the noise data exist in the layer thickness monitoring matrix, and therefore, for the layer thickness monitoring matrix, the layer thickness data are classified to extract the noise data. In this embodiment, in consideration of the fact that the DBSCAN (Density-Based Spatial Clustering of Applications with Noise), that is, the Density-based clustering algorithm, can identify isolated noise points in data in the clustering process, the DBSCAN clustering algorithm is adopted to classify each layer thickness data in the layer thickness monitoring matrix so as to extract the noise point types.
However, in this embodiment, considering that the initial core points, the clustering radius and the intra-class minimum points need to be selected in the clustering process of the DBSCAN algorithm, the selection of the initial core points, the clustering radius and the intra-class minimum points of the conventional DBSCAN algorithm are mostly randomly selected by people, the process will affect the final clustering effect, and in order to improve the classification accuracy of the layer thickness data, the noise data is accurately identified, and in this embodiment, the initial core points are selected in a self-adaptive manner:
for each layer thickness data in the layer thickness monitoring matrix, the embodiment uses the layer thickness data
Figure SMS_59
For the sake of example, the acquisition of layer thickness data
Figure SMS_60
Layer thickness data in the local w×w range, the value of W can be set by the practitioner, and in this embodiment, w=5, according to the layer thickness data
Figure SMS_61
Data difference and space difference of each layer thickness data in local range to obtain layer thickness data
Figure SMS_62
And layer thickness data
Figure SMS_63
The distance factor between each layer thickness data in the local range is expressed as follows:
Figure SMS_64
wherein e is a natural constant,
Figure SMS_66
for layer thickness data
Figure SMS_70
And layer thickness data
Figure SMS_73
Layer thickness data in local range
Figure SMS_67
A factor of the distance between them,
Figure SMS_69
respectively to-be-measured composite optical film
Figure SMS_72
A position of,
Figure SMS_75
Layer thickness data at the location(s),
Figure SMS_65
for layer thickness data
Figure SMS_68
The variance of the layer thickness data contained within the local range of (c),
Figure SMS_71
in order to take the absolute value of the operation,
Figure SMS_74
is a multiplication operation;
repeating the method to obtain layer thickness data
Figure SMS_76
Distance factor from each layer thickness data in local range and according to layer thickness data
Figure SMS_77
And the layer thickness contained in the local areaObtaining layer thickness data from distribution relation among data
Figure SMS_78
Is expressed as layer thickness data
Figure SMS_79
The local variation index expression of (2) is:
Figure SMS_80
in the method, in the process of the invention,
Figure SMS_88
for layer thickness data
Figure SMS_81
Is used for determining the local variation index of the (c),
Figure SMS_85
for layer thickness data
Figure SMS_92
Variance of the distance factor from the layer thickness data in the local range,
Figure SMS_96
for layer thickness data
Figure SMS_99
Layer thickness data sets contained within a local range of (c),
Figure SMS_101
respectively to-be-measured composite optical film
Figure SMS_93
A position of,
Figure SMS_84
Layer thickness data at the location(s),
Figure SMS_89
as a logarithmic function based on a natural constant e,
Figure SMS_87
for layer thickness data
Figure SMS_91
Layer thickness data in a local range of (2)
Figure SMS_95
The number of occurrences in the local layer thickness data set,
Figure SMS_90
for the sum of the number of occurrences of all layer thickness data contained in the local range in the local layer thickness data set,
Figure SMS_94
as an exponential function based on a natural constant e,
Figure SMS_98
for the multiplication operation, Σ is the sum operation,
Figure SMS_100
for the distance threshold, the distance threshold value implementer can set itself, and this embodiment sets it to 3. Wherein, the local variation index
Figure SMS_82
The larger the layer thickness data
Figure SMS_86
The more uneven the distribution of the layer thickness data in the local range is, the more obvious the density change of the local range is, when the layer thickness data is selected as an initial core point, the layer thickness data in the local range of the layer thickness data is scattered and belongs to different categories, and the layer thickness data in the local range is not easy to cluster rapidly in the initial clustering process, so that the clustering speed and the clustering effect are influenced;
repeating the method to obtain the local variation index of each layer thickness data;
according to the local variation index of each layer thickness data in the layer thickness data matrix, the initial core point of the DBSCAN clustering algorithm and the update of the core point are obtained according to the local variation index of the layer thickness data, and the layer thickness data with the minimum local variation index is used as the initial core point in the embodiment; considering that the DBSCAN clustering process is an iterative clustering process, after the category clustering corresponding to the initial core point is completed, the layer thickness data with the minimum local variation index in the layer thickness data except the initial core point and the layer thickness data in the corresponding category is taken as a new core point, and the core points are sequentially updated, so that the selection of the initial core point and the iterative updating of the core points in the DBSCAN clustering algorithm are realized.
Thus, the initial core point and the updating process of the core point in the classification process of the layer thickness data can be obtained according to the method.
Further, in consideration of the fact that in the clustering process of the DBSCAN clustering algorithm, the clustering radius in the clustering parameters and the setting of the minimum points in the class can have a large influence on the clustering result, and the setting of the minimum points in the class can influence the effect and the accuracy of data class division; meanwhile, when the layer thickness data category is classified according to the DBSCAN clustering algorithm, if the set clustering radius is too small, a large part of data cannot be aggregated because the number of points for creating a dense area is not met, so that a large amount of normal data is mistakenly regarded as isolated abnormal data; if the cluster radius is too large, the data of different categories will be combined, so that most of the data of different categories are in the same category, and therefore, the cluster parameters are optimized to obtain the optimal cluster parameters, and the randomness of manual setting is prevented.
In the embodiment, the optimal cluster radius and the minimum point number in the optimal class are obtained through a firefly algorithm. Firstly, taking the cluster radius and the minimum points in the class in a DBSCAN cluster algorithm as optimization parameters of a firefly optimization algorithm, and in order to avoid the problem that the search efficiency is low or a local optimal solution is sunk due to the fact that the search range of the optimization parameters is too large and too small in the optimization process, presetting the search range of the cluster radius and the minimum points in the class, wherein a specific search range setting implementer can set the search range of the cluster radius and the minimum points in the class as follows:
Figure SMS_102
，
Figure SMS_103
. It should be noted that, the brightness value, the attraction degree, the position update calculation formula and the specific optimization process of firefly are known techniques, and are not described in detail in the present embodiment.
Then, in order to ensure the precision in the classification process of the layer thickness data and improve the optimization effect of the clustering parameters, according to the variance of the layer thickness data in each class, the space average distance from the layer thickness data in each class to the class center point, the space distance between the class center points and the data difference between the class center points in the iterative optimization process, an objective function of the optimization process is obtained, wherein the expression of the objective function is as follows:
Figure SMS_104
wherein F is an objective function, Z is the class number of the layer thickness data cluster,
Figure SMS_114
for the variance of the layer thickness data in class i, max () is the sign of the maximum value,
Figure SMS_106
is the spatial average distance of each layer thickness data in class i to the center point of the class, wherein,
Figure SMS_110
，
Figure SMS_118
for the row and column numbers of layer thickness data x in the layer thickness monitoring matrix in class i,
Figure SMS_122
is the center point of category i
Figure SMS_120
The number of rows and columns in the layer thickness monitoring matrix,
Figure SMS_124
for the total number of layer thickness data in category i,
Figure SMS_117
is the spatial average distance of each layer thickness data in category s to the center point of the category, wherein,
Figure SMS_121
，
Figure SMS_105
for the row and column numbers of layer thickness data y in the layer thickness monitoring matrix in category s,
Figure SMS_112
is the center point of category s
Figure SMS_108
The number of rows and columns in the layer thickness monitoring matrix,
Figure SMS_109
for the total number of layer thickness data in category s,
Figure SMS_113
is the center point of category i
Figure SMS_116
Center point with category s
Figure SMS_115
A spatial distance between the two, wherein,
Figure SMS_119
，
Figure SMS_123
is the center point of category i
Figure SMS_125
Center point with category s
Figure SMS_107
Corresponding to the absolute value of the difference in layer thickness data,
Figure SMS_111
for the multiplication operation, Σ is the sum operation.
In the objective function expression, the objective function expression is
Figure SMS_126
The smaller the layer thickness data fluctuation degree in each category is, the smaller the layer thickness data fluctuation degree is;
Figure SMS_128
the smaller the spatial distance between the layer thickness data in each class is, the larger the spatial distance between each class and the layer thickness data difference is; therefore, the smaller the objective function, the better the layer thickness data class classification effect.
Finally, setting the optimization iteration times, and realizing the iterative optimization of the optimization parameters by combining an objective function with a firefly optimization algorithm, wherein an operator of the iteration times can set the iteration times to 500 times, and the optimization parameters corresponding to the minimum objective function are used as the optimal clustering radius in the layer thickness data category dividing process
Figure SMS_129
And minimum points in the optimal class
Figure SMS_130
The classification of the layer thickness data in the layer thickness monitoring matrix can be realized.
And obtaining each parameter of the DBSCAN clustering algorithm according to the selection update of the initial core points, the optimal clustering radius and the minimum points in the optimal class, and further obtaining the class of each layer thickness data in the layer thickness monitoring matrix. Further, in this embodiment, noise data in the layer thickness monitoring matrix is extracted, and noise categories are identified according to the data size in the categories in consideration of the isolation of the noise, and in this embodiment, the category containing only one data is taken as the noise category, and the data in the noise category is recorded as the noise data;
repeating the method to extract all noise data.
In order to avoid the influence of the noise data on the thickness monitoring of the composite optical film layer, filtering processing is sequentially carried out on each noise data, a w partial window is obtained by taking the noise data as a center, a value implementation person of w can set the value implementation person of w to w=5, and the filtering value of each noise data is obtained according to the average value of the non-noise data contained in each noise data partial window.
Further classifying the data filtering values of the noise points, and obtaining the class of the data filtering values of the noise points according to the data difference between the data filtering values of the noise points and the center points of the noise points in the embodiment. And calculating the absolute value of the difference value between the noise data filtering value and each class of center points, and taking the class with the smallest absolute value as the class to which the noise data filtering value belongs.
Repeating the method to obtain the category of each noise data filtering value and classifying all layer thickness data in the layer thickness monitoring matrix.
Therefore, the clustering parameters in the layer thickness data category division process are optimized according to the method, category division of the layer thickness data in the layer thickness monitoring matrix is achieved, noise data are extracted and filtered, influence of the noise data on the layer thickness monitoring of the composite optical thin film to be detected is prevented, and the clustering precision is high.
And step S003, intelligently monitoring the layer thickness condition of the composite optical film to be tested based on the classification result of the layer thickness data.
Performing intelligent monitoring on the layer thickness of the composite optical film to be detected according to the classification result of all layer thickness data in the layer thickness monitoring matrix, obtaining the classified classification number of all layer thickness data in the layer thickness monitoring matrix, and when the classified classification number is equal to 1, uniformly coating the layer thickness of the composite optical film to be detected; when the number of categories after category division is greater than 1, the thickness of the composite optical film to be measured is uneven, and the composite optical film needs to be processed again to meet the use requirement, so that the problems of poor protection effect, low sealing performance and the like of products when the composite optical film is used for product packaging due to uneven thickness of the composite optical film are avoided.
Furthermore, the embodiment also provides an intelligent monitoring system for the layer thickness of the composite optical film, which comprises a processor, a memory and a computer program stored on the memory and capable of running on the processor, wherein the computer program realizes the steps of the intelligent monitoring method for the layer thickness of the composite optical film when being executed by the processor.
In summary, the embodiment of the invention analyzes the layer thickness data at each position of the composite optical film to obtain the category of each layer thickness data, extracts the noise data in the layer thickness monitoring matrix, and filters the noise data to obtain the filtering value of each noise data, thereby avoiding the influence of the noise data in the intelligent monitoring process of the layer thickness of the composite optical film and improving the intelligent monitoring precision of the layer thickness of the composite optical film;
according to the embodiment of the invention, the initial core points in the layer thickness data category division process are selected through the local variation indexes of each layer thickness data, and the core points are iteratively updated, so that the problems of low clustering speed, low clustering precision and the like caused by poor selection of the initial core points are solved, meanwhile, the clustering parameters in the layer thickness data category division process are optimized by combining a firefly algorithm and constructing an objective function, the randomness of manually selecting the clustering parameters is solved, the layer thickness data category division precision is improved, and the problem of misclassification of the layer thickness data is avoided.
It should be noted that: the sequence of the embodiments of the present invention is only for description, and does not represent the advantages and disadvantages of the embodiments. And the foregoing description has been directed to specific embodiments of this specification. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
In this specification, each embodiment is described in a progressive manner, and the same or similar parts of each embodiment are referred to each other, and each embodiment mainly describes differences from other embodiments.
The above embodiments are only for illustrating the technical solution of the present application, and are not limiting; the technical solutions described in the foregoing embodiments are modified or some of the technical features are replaced equivalently, so that the essence of the corresponding technical solutions does not deviate from the scope of the technical solutions of the embodiments of the present application, and all the technical solutions are included in the protection scope of the present application.

Claims (8)

1. The intelligent monitoring method for the thickness of the composite optical film layer is characterized by comprising the following steps:
acquiring layer thickness data of each position of the composite optical film to be detected through a thickness measuring sensor, and acquiring a layer thickness monitoring matrix;
obtaining local variation indexes of each layer thickness data according to the distribution relation between each layer thickness data and the layer thickness data contained in the local range; obtaining an initial core point of a DBSCAN clustering algorithm and updating a new core point according to local variation indexes of each layer thickness data;
taking the clustering radius and the minimum points in the class in the DBSCAN clustering algorithm as optimization parameters, and performing iterative optimization on the optimization parameters through a firefly optimization algorithm; obtaining an objective function of the optimization process according to the distribution condition of the layer thickness data in each class in the iterative optimization process and the spatial relationship between center points of each class;
according to a firefly optimization algorithm, combining an objective function to obtain an optimal cluster radius and the minimum number in the optimal class; obtaining a layer thickness data category classification result according to the initial core points, the updating of the new core points, the optimal clustering radius and the minimum points in the optimal class; obtaining noise data according to the layer thickness data classification result;
obtaining each noise filtering value according to the average value of the non-noise data contained in each noise local window; obtaining the category of each noise data filtering value according to the data difference between each noise data filtering value and each center point;
and performing intelligent monitoring on the layer thickness of the composite optical film to be detected according to classification results of all layer thickness data in the layer thickness monitoring matrix.
2. The method for intelligently monitoring the layer thickness of a composite optical film according to claim 1, wherein the local variation index of each layer thickness data is obtained according to the distribution relation between each layer thickness data and the layer thickness data contained in the local range, and the expression is:
Figure QLYQS_1
in the method, in the process of the invention,
Figure QLYQS_20
for layer thickness data->
Figure QLYQS_22
Local variation index of->
Figure QLYQS_24
For layer thickness data->
Figure QLYQS_4
Variance of distance factor from layer thickness data in local range, +.>
Figure QLYQS_9
For layer thickness data->
Figure QLYQS_12
Layer thickness data set contained in a local range, < ->
Figure QLYQS_16
Respectively is a composite optical film to be measured->
Figure QLYQS_6
Position, & gt>
Figure QLYQS_15
Layer thickness data at the location,/>
Figure QLYQS_19
As a logarithmic function based on a natural constant e, < ->
Figure QLYQS_21
For layer thickness data->
Figure QLYQS_23
Layer thickness data in the local region +.>
Figure QLYQS_25
The number of occurrences in the local layer thickness dataset, +.>
Figure QLYQS_5
Is the sum of the number of occurrences of all layer thickness data contained in the local range in the local layer thickness data set,/for>Is an exponential function based on a natural constant e, < ->
Figure QLYQS_13
For multiplication operation, Σ is a sum operation, +.>
Figure QLYQS_17
Is a distance threshold; wherein (1)>
Figure QLYQS_7
For layer thickness data->
Figure QLYQS_10
Data of layer thickness->
Figure QLYQS_14
Layer thickness data in the local region +.>
Figure QLYQS_2
The distance factor between the two is expressed as follows:
Figure QLYQS_26
in the method, in the process of the invention,
Figure QLYQS_27
for layer thickness data->
Figure QLYQS_28
Variance of all layer thickness data in a local range, +.>
Figure QLYQS_29
To take absolute value.
3. The method for intelligently monitoring the layer thickness of the composite optical film according to claim 1, wherein the updating of the initial core point and the new core point of the DBSCAN clustering algorithm is obtained according to the local variation index of each layer thickness data, specifically: taking layer thickness data with the smallest local variation index in the layer thickness monitoring matrix as an initial core point; and after the class clustering corresponding to the initial core point is completed, the layer thickness data with the minimum local variation index in the layer thickness data except the initial core point and the layer thickness data in the corresponding class is used as a new core point, and the core points are updated in sequence.
4. The intelligent monitoring method of composite optical film layer thickness according to claim 1, wherein the objective function of the optimization process is obtained according to the distribution condition of the layer thickness data in each class and the spatial relationship between the center points of each class in the iterative optimization process, and the expression is:
Figure QLYQS_30
wherein F is an objective function, Z is the class number of the layer thickness data cluster,
Figure QLYQS_39
for the variance of the layer thickness data in class i, max () is the sign of maximum value, ++>
Figure QLYQS_32
Is the spatial average distance of each layer thickness data in class i to the center point of the class, wherein,
Figure QLYQS_35
，/>
Figure QLYQS_43
for the row, column number, and/or the +/of the layer thickness monitoring matrix of the layer thickness data x in class i>
Figure QLYQS_47
Center point for category i->
Figure QLYQS_49
Row, column number, < >>
Figure QLYQS_51
For the total number of layer thickness data in category i, +.>
Figure QLYQS_40
Is the spatial average distance of each layer thickness data in category s to the center point of the category, wherein,
Figure QLYQS_44
，/>
Figure QLYQS_31
for the row, column number, and/or the +/of the layer thickness data y in the layer thickness monitoring matrix in category s>
Figure QLYQS_37
Is the center point of category s->
Figure QLYQS_34
Row, column number, < >>
Figure QLYQS_36
For layer thicknesses in category sTotal number of data->
Figure QLYQS_41
Center point for category i->
Figure QLYQS_45
Center point with category s->
Figure QLYQS_42
A spatial distance between the two, wherein,
Figure QLYQS_46
，/>
Figure QLYQS_48
center point for category i->
Figure QLYQS_50
Center point with category s->
Figure QLYQS_33
Absolute value of the difference of the corresponding layer thickness data, +.>
Figure QLYQS_38
For the multiplication operation, Σ is the sum operation.
5. The method for intelligently monitoring the layer thickness of the composite optical film according to claim 1, wherein noise data is obtained according to the classification result of layer thickness data, specifically: and taking all the categories containing only one data as noise categories, and recording the data in the noise categories as noise data.
6. The method for intelligently monitoring the layer thickness of a composite optical film according to claim 1, wherein the category to which each noise data filtering value belongs is obtained according to the data difference between each noise data filtering value and each center point, specifically: and calculating the absolute value of the difference value between the noise data filtering value and each class of center points, and taking the class with the smallest absolute value as the class to which the noise data filtering value belongs.
7. The method for intelligently monitoring the layer thickness of the composite optical film according to claim 1, wherein the intelligent monitoring of the layer thickness of the composite optical film to be tested is performed according to classification results of all layer thickness data in a layer thickness monitoring matrix, specifically: counting the classified category number of all layer thickness data in the layer thickness monitoring matrix, and when the classified category number is equal to 1, uniformly measuring the layer thickness of the composite optical film; when the number of the classified categories is larger than 1, the thickness of the composite optical film layer to be measured is uneven.
8. A composite optical film layer thickness intelligent monitoring system, comprising a processor and a memory, wherein the processor is configured to process instructions stored in the memory to implement a composite optical film layer thickness intelligent monitoring method as claimed in any one of claims 1-7.
CN202310488375.6A 2023-05-04 2023-05-04 Intelligent monitoring method and system for layer thickness of composite optical film Active CN116432088B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310488375.6A CN116432088B (en) 2023-05-04 2023-05-04 Intelligent monitoring method and system for layer thickness of composite optical film

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310488375.6A CN116432088B (en) 2023-05-04 2023-05-04 Intelligent monitoring method and system for layer thickness of composite optical film

Publications (2)

Publication Number Publication Date
CN116432088A true CN116432088A (en) 2023-07-14
CN116432088B CN116432088B (en) 2023-11-07

Family

ID=87087281

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310488375.6A Active CN116432088B (en) 2023-05-04 2023-05-04 Intelligent monitoring method and system for layer thickness of composite optical film

Country Status (1)

Country Link
CN (1) CN116432088B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117540281A (en) * 2024-01-09 2024-02-09 深圳市宇辉光学科技有限公司 Data optimization analysis system and method applied to optical film

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20170371886A1 (en) * 2016-06-22 2017-12-28 Agency For Science, Technology And Research Methods for identifying clusters in a dataset, methods of analyzing cytometry data with the aid of a computer and methods of detecting cell sub-populations in a plurality of cells
CN110083665A (en) * 2019-05-05 2019-08-02 贵州师范大学 Data classification method based on the detection of improved local outlier factor
CN113344019A (en) * 2021-01-20 2021-09-03 昆明理工大学 K-means algorithm for improving decision value selection initial clustering center
CN115358349A (en) * 2022-10-19 2022-11-18 江苏益捷思信息科技有限公司 Data optimization clustering method
CN115730967A (en) * 2022-11-28 2023-03-03 天翼数字生活科技有限公司 Sales condition prediction method and device, storage medium and computer equipment
CN115797299A (en) * 2022-12-05 2023-03-14 常宝新材料(苏州)有限公司 Defect detection method of optical composite film
CN115935203A (en) * 2022-12-06 2023-04-07 华南理工大学 Distributed clustering method, device and medium on wireless sensor network
CN115984213A (en) * 2022-12-29 2023-04-18 西安电子科技大学广州研究院 Industrial product appearance defect detection method based on deep clustering

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20170371886A1 (en) * 2016-06-22 2017-12-28 Agency For Science, Technology And Research Methods for identifying clusters in a dataset, methods of analyzing cytometry data with the aid of a computer and methods of detecting cell sub-populations in a plurality of cells
CN110083665A (en) * 2019-05-05 2019-08-02 贵州师范大学 Data classification method based on the detection of improved local outlier factor
CN113344019A (en) * 2021-01-20 2021-09-03 昆明理工大学 K-means algorithm for improving decision value selection initial clustering center
CN115358349A (en) * 2022-10-19 2022-11-18 江苏益捷思信息科技有限公司 Data optimization clustering method
CN115730967A (en) * 2022-11-28 2023-03-03 天翼数字生活科技有限公司 Sales condition prediction method and device, storage medium and computer equipment
CN115797299A (en) * 2022-12-05 2023-03-14 常宝新材料(苏州)有限公司 Defect detection method of optical composite film
CN115935203A (en) * 2022-12-06 2023-04-07 华南理工大学 Distributed clustering method, device and medium on wireless sensor network
CN115984213A (en) * 2022-12-29 2023-04-18 西安电子科技大学广州研究院 Industrial product appearance defect detection method based on deep clustering

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117540281A (en) * 2024-01-09 2024-02-09 深圳市宇辉光学科技有限公司 Data optimization analysis system and method applied to optical film
CN117540281B (en) * 2024-01-09 2024-03-22 深圳市宇辉光学科技有限公司 Data optimization analysis system and method applied to optical film

Also Published As

Publication number Publication date
CN116432088B (en) 2023-11-07

Similar Documents

Publication Publication Date Title
US20190228527A1 (en) System and method of classification of biological particles
CN115276006A (en) Load prediction method and system for power integration system
CN111161814A (en) DRGs automatic grouping method based on convolutional neural network
CN116432088B (en) Intelligent monitoring method and system for layer thickness of composite optical film
CN111079620A (en) Leukocyte image detection and identification model construction method based on transfer learning and application
CN116109195B (en) Performance evaluation method and system based on graph convolution neural network
CN108734216A (en) Classification of power customers method, apparatus and storage medium based on load curve form
CN112270596A (en) Risk control system and method based on user portrait construction
CN112529638B (en) Service demand dynamic prediction method and system based on user classification and deep learning
CN104239722A (en) Forecasting method based on recognition of correlational relationship between factors
CN111477328B (en) Non-contact psychological state prediction method
CN116564409A (en) Machine learning-based identification method for sequencing data of transcriptome of metastatic breast cancer
CN112330095A (en) Quality management method based on decision tree algorithm
CN117494013A (en) Multi-scale weight sharing convolutional neural network and electroencephalogram emotion recognition method thereof
CN117435937A (en) Smart electric meter abnormal data identification method, device, equipment and storage medium
CN111816298B (en) Event prediction method and device, storage medium, terminal and cloud service system
CN117034153A (en) Milling cutter abrasion recognition method based on three-dimensional variation modal decomposition feature fusion
CN116229330A (en) Method, system, electronic equipment and storage medium for determining video effective frames
CN116504314A (en) Gene regulation network construction method based on cell dynamic differentiation
CN110265151B (en) Learning method based on heterogeneous temporal data in EHR
CN109840479B (en) Health state matching method and device
CN118395219B (en) Structural performance detection method and system for tunnel lining construction
CN104200383A (en) Application of multivariate regression analysis to tax decision
Yuhang et al. Research on data cleaning in text clustering
CN117877736B (en) Intelligent ring abnormal health data early warning method based on machine learning

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant