CN114936746B - Quality analysis method for wear-resistant, illumination-resistant, tensile and shrinkage-free warning tape corona composite film - Google Patents
Quality analysis method for wear-resistant, illumination-resistant, tensile and shrinkage-free warning tape corona composite film Download PDFInfo
- Publication number
- CN114936746B CN114936746B CN202210385057.2A CN202210385057A CN114936746B CN 114936746 B CN114936746 B CN 114936746B CN 202210385057 A CN202210385057 A CN 202210385057A CN 114936746 B CN114936746 B CN 114936746B
- Authority
- CN
- China
- Prior art keywords
- film
- corona
- quality
- neural network
- resistant
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06395—Quality analysis or management
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/40—Analysis of texture
- G06T7/41—Analysis of texture based on statistical description of texture
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Development Economics (AREA)
- Educational Administration (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Quality & Reliability (AREA)
- General Business, Economics & Management (AREA)
- Operations Research (AREA)
- Marketing (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Game Theory and Decision Science (AREA)
- Tourism & Hospitality (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Probability & Statistics with Applications (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Treatments Of Macromolecular Shaped Articles (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The invention relates to a quality analysis method for a wear-resistant and illumination-resistant tensile non-shrinkage warning tape corona composite film, which comprises the steps of obtaining environmental humidity data of each layer of film before the composite film is not compounded, and calculating the humidity change degree of all the films; acquiring image information of each layer of film of the composite film during corona treatment, and determining the abnormal degree of the corona of the film; obtaining a film quality index based on the humidity change degree and the film corona abnormal degree; performing phase space reconstruction on the film quality index to obtain a tracking index of a phase space; determining the standard deviation of the tracking index to obtain a span value of a reference quality interval; and if the reference mass interval span value updated for multiple times is continuously increased for more than two times, new quality degradation occurs in the composite film, and the environmental humidity data and the corona texture information of each layer of film are used as the input of a neural network model to identify and analyze the film quality of the film corona treatment. Namely, the scheme of the invention can analyze the fluctuation of good quality or bad quality in the production process of the composite membrane.
Description
Technical Field
The invention relates to the field of thin film materials, in particular to a quality analysis method for a wear-resistant and illumination-resistant tensile non-shrinkage warning tape corona composite film.
Background
The wear-resistant, illumination-resistant, stretchable and non-shrinkage warning tape is formed by simultaneously compounding polyethylene terephthalate (PET), polyvinyl chloride (PVC) and UV films; when the three films are compounded, the defects of the corona treatment can be relieved by the three films together, but the process quality still needs to be controlled, and if the process is not controlled, the durable quality of the adhesive tape can be deteriorated.
Disclosure of Invention
In order to solve the technical problems, the invention aims to provide a quality analysis method for a wear-resistant and illumination-resistant tensile non-shrinkage warning tape corona composite film, which adopts the following technical scheme:
the invention provides a quality analysis method for a wear-resistant, illumination-resistant, tensile and shrinkage-free warning tape corona composite film, which comprises the following steps:
acquiring environmental humidity data of each layer of film before the composite film is not compounded, and calculating the humidity change degree of all the films; the composite film is formed by compounding three layers of films;
acquiring image information of each layer of film of the composite film during corona treatment, acquiring corona texture information of the corresponding film, and determining the abnormal degree of corona of the film;
obtaining a film quality index based on the humidity change degree and the film corona abnormal degree;
performing phase space reconstruction on the film quality index to obtain a tracking index of a phase space; determining a standard deviation of the tracking index by using the tracking index, and obtaining a span value of a reference quality interval based on the standard deviation;
and if the reference mass interval span value updated for multiple times is continuously increased for more than two times in the phase space reconstruction process, determining that new quality degradation occurs in the composite film corresponding to the current moment, taking the environmental humidity data and the corona texture information of each layer of film as the input of a neural network model, and identifying and analyzing the film quality of the corona treatment of the film based on the neural network model.
Preferably, calculating the texture variance of the corona texture information to obtain the abnormal degree of the film, and further amplifying the abnormal degrees of all the films to obtain the abnormal degree of the corona of the film.
Preferably, the film quality index is the product of the humidity change degree and the corona abnormity degree of the film.
Preferably, the method for acquiring the span value of the reference mass interval includes:
acquiring tracking indexes at different moments, and calculating the average value P of all the tracking indexes T And standard deviation σ P ；
Determining a reference mass interval as [ P ] T -3σ P ,P T +3σ P ]；
Determining a span value 6 sigma of the reference quality interval according to the up-and-down floating value of the reference quality interval P 。
Preferably, the training process of the neural network model is as follows:
respectively collecting environmental humidity data and film corona abnormal degrees corresponding to the three layers of films to form 6-dimensional input vectors as network input of the neural network model, marking the 6-dimensional input vectors in the data set by using labels, marking the abnormal labels as 0 and marking the normal labels as 1, outputting the labels as the network, and performing iterative training on the neural network model by using a loss function to obtain the trained neural network model.
Preferably, the neural network model is a multi-layer feedforward artificial neural network of an error back propagation algorithm.
The invention has the beneficial effects that:
the scheme of the invention can construct the phase space by analyzing the minimum unit comprehensive texture and humidity environment in the compounding process of three films, solves the problem that the traditional texture analysis cannot be used in the phase space, simultaneously solves the phase space tracking problem that the environmental humidity influences the corona effect, and simultaneously can integrate the tracking index and the span value of the reference quality interval, thereby analyzing the quality fluctuation in the production process of the composite film.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions and advantages of the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
FIG. 1 is a composite schematic view of a tape composite film;
FIG. 2 is a flow chart of the steps of a method for analyzing the quality of a corona composite film of a wear-resistant, light-resistant, stretchable and non-shrinkable warning tape according to the present invention.
Detailed Description
To further explain the technical means and effects of the present invention adopted to achieve the predetermined objects, the embodiments, structures, features and effects thereof according to the present invention will be described in detail below with reference to the accompanying drawings and preferred embodiments. In the following description, the different references to "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, the particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.
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 scene of the invention is directed at the quality analysis of the corona composite film of the wear-resistant and illumination-resistant stretching non-shrinkage warning tape, wherein the composite film is formed by simultaneously compounding PET, PVC and UV films, as shown in figure 1, the composite film is formed by compounding films after corona treatment, and then the quality of the composite film is analyzed.
Specifically, referring to fig. 2, the method for analyzing the quality of the corona composite film of the wear-resistant, light-resistant, stretchable and non-shrinkable warning tape according to the present invention comprises the following steps:
step 1, acquiring environmental humidity data of each layer of film before the composite film is not compounded, and calculating the humidity change degree of all the films; the composite film is formed by compounding three layers of films.
In this embodiment, the environmental humidity data of each layer of film before the composite film is not compounded is collected as the representation of the state of the film before the composite film is compounded, because the corona treatment process of each film is in 3 different environments, the corona condition and the overall condition of each environment are analyzed, so as to identify whether the current treatment quality is better or not; specifically, because the film need avoid moisture too big factor that influences quality such as dewfall when leading to the corona, need do the drying process of big amount of wind to it, every membrane material need have independent processing production line, and when the humidity difference that the production line was located is great, can influence the surface quality of membrane material to the quality that the parameter when making the skew corona debugging of membrane material quality corresponds, thereby influence the corona treatment quality, consequently, need acquire the humidity data of different films.
In this embodiment, the drying humidity environment of each film drying production line is similar, and when the difference between the drying humidity of the three films is large, the drying effect of one film is consideredIf an abnormality occurs, the working environment of the corona equipment is different, so that the humidity value of each minimum unit (set distance) of the film needs to be recorded, and H = { H = } 1 ，H 2 ，H 3 }。
In the above embodiment, the composite film of the adhesive tape is segmented and analyzed according to a set distance, and specifically, the set distance may be 5cm.
Further, according to the acquired environmental humidity data of the three films, the humidity change degrees of the three films are determined, so as to represent different degrees between the environments of the three films, it should be noted that in the embodiment, when the humidity change data is calculated, it is required to determine that the environmental humidity data of one of the films is standard data, that is, the environmental humidity data is used as reference data to represent whether the environmental humidity data of the other films is abnormal.
And 2, acquiring image information of each layer of film of the composite film during corona treatment, acquiring corona texture information of the corresponding film, and determining the corona abnormal degree of the film.
In this embodiment, the PVC layer film of the warning tape is pure yellow, and the rest of the film is black. It should be noted that, the acquisition of the image of the thin film after corona is a well-known technology, and details thereof are not described herein.
In this embodiment, the film image is also subjected to graying processing.
For a film with an arbitrarily set distance, the surface property of the film is changed in the corona process, so that the pixel statistical value of the texture at each stage is collected; if the corona level is too high, it can cause the film surface to age and become brittle, creating brittle cracks, and also causing corona breakdown, creating continuous breakdown pores that appear as continuous, nearly black pixels. When the corona degree is insufficient, the film surface also has the phenomenon of precipitation of internal additives, which is represented by the appearance of single pixels close to white on the image.
Meanwhile, the abnormal degree of corona of each film can be influenced by the environment and the working state of the machine, and the statistical characteristics in the gray scale images of the films are different.
Therefore, based on the above analysis, it is found that the variance of the pixel values in the image corresponding to the film at the set distance (5 cm) is calculated, and the degree of film abnormality E of the current film can be represented by the variance i Then the three films are abnormal to the extent of
E＝{E 1 ,E 2 ,E 3 }。
And further, performing three-stage amplification treatment on the film abnormal degrees of the three films to obtain the film corona abnormal degree:
E＝∏(1+E i )
at the moment, E represents the abnormal degree of the three films with the corona abnormal textures, and the larger E is, the more abnormal corona areas appear in the current process of any film is, and the quality of the produced adhesive tape is poorer; whereas the corona quality is better.
And 3, obtaining a film quality index based on the humidity change degree and the corona abnormal degree of the film.
Wherein, the film quality indexes are as follows:
W＝Range(H)*E
when the humidity difference in the film production environment is large, the discharge abnormality occurs during the corona treatment of one film, so that the texture variance of the film is large, and therefore the film quality index W of the film can be sensed very sensitively by combining the environmental humidity H for drying the three films and the texture variance E of the film.
W in the above can also represent a chaotic state of corona treatment in which there is a state transition of the state phase space.
In the above-described acquisition of the film quality index, it is considered that the conventional corona equipment often uses a monolithic long electrode, and if the corona treatment is performed for a long period of time, there is a problem that crystalline foreign matter adheres to the film, so that it is necessary to determine the quality of the film to be pretreated and the crystallization problem in the corona process, and further analysis of the film quality abnormality is performed by integrating the quality. Meanwhile, it should be noted that, if the humidity in the air is relatively high, the efficiency of the corona equipment is also reduced when various films are subjected to corona treatment.
Step 4, carrying out phase space reconstruction on the film quality index to obtain a tracking index of a phase space; and determining the standard deviation of the tracking index by using the tracking index, and obtaining a span value of the reference quality interval based on the standard deviation.
In this embodiment, since the determination of the film quality index W in step 3 cannot reflect whether the quality index is abnormal in the production process, the multi-time spatial scale of the film quality index is analyzed, specifically, the process of performing phase-space reconstruction on the film quality index is as follows:
1) When the production line normally produces, the time t =0 of the reference analysis is set, and N film quality indexes { W } are recorded through the corona analysis result 0(1) ，W 0(2) ，...，W 0(N) And N is the recording distance of the corona analysis result, a mutual information method is used for selecting a delay time parameter tau, a false adjacent point method is used for selecting an embedded dimension parameter m, and the phase space reconstruction mode is as follows:
W 0(1) ＝[w 0(1) ，w 0(1+τ) ，...，w 0(1+(m-1)×τ) ]
……
W 0(N) ＝[w 0(N) ，w 0(N+τ) ，...，w 0(N+(m-1)×τ) ]
……
W 0(N-(m-1)×τ) ＝[w 0(N-(m-1)×τ) ，w 0(N-(m-2)×τ) ，...，w 0(N) ]
at this point, the phase space of the change of the film quality index in N production processes is reconstructed and taken as the reference phase space W 0 。
2) Updating data points w in a form of continuous sliding windows, and recording N film quality indexes { w when the moving length is t x 5cm t(1) ，w t(2) ，...，w t(N) Using a space W with a reference phase 0 Reconstructing the phase spaces of the N film quality indexes at the t moment by the same delay time tau and the embedding dimension m, and simultaneously reconstructing the current phase space in the same way as the above in the subsequent production process:
W(1)＝[w(1)，w(1+τ)，...，w(1+(m-1)×τ)]
……
W (N) ＝[w (N) ，w (N+τ) ，...，w (N+(m-1)×τ) ]
……
W (N-(m-1)×τ) ＝[w (N-(m-1)×τ) ，w (N-(m-2)×τ) ，...，w (N) ]。
in this way, a real-time varying phase space W is obtained.
3) For a certain vector W in the phase space at time t t (n) finding the p (p > 2 (tau x m)) nearest vectors { W) in the reference phase space (k) ，W (k+1) ，...，W (k+p-1) K = {1,. Multidot., N- (m-1) τ }.
4) Based on the above processing method, the data point w is updated in the form of a continuous sliding window, and a reference mass interval J within the production distance T is obtained by taking a production distance T (which corresponds to a length of the film) as an observation distance:
for the phase space, there are:
A k ＝{W 0(k+τ) ...W 0(k+p+τ-1) }
B k ＝{{W′ 0(k+τ-1) ，...，W′ 0(k) }...W′ 0(k+p+τ-2) ，...W′ 0(k+p-1) }
W t (n) the tracking function is:
calculating W 0 And vector W 0(k) ，W 0(k+1) ，....，W 0(k+p-1) The farthest distance is Dr n Design of phase space weights
Let N self-increment by 1, continue to calculate P (N) until N = N- (m-1) τ.
Then, all tracking functions corresponding to all vectors in the phase space at the time t are utilized, and a tracking index of the phase space at the time t is calculated:
wherein q (n) is a weight function,
m is the correlation dimension of the phase space at the time t; dr n Is a reference phase space W 0 And vector W 0(k) ，W 0(k+1) ，....，W 0(k+p-1) The farthest distance, P t(n) Is a tracking function of the phase space at time t.
To this end, P is calculated 1 、P 2 ...P T Average value P of T tracking indexes T And standard deviation sigma P 。
The above-mentioned tracking index reveals each minimum unit P of the film (n) An indication of the phase space state of the process over a production distance T. When the state index changes greatly in a period of time, the environment which depends on the corona treatment quality in the production process changes obviously, and generally:
when the humidity of the drying environment of different films changes slowly, the quality indexes of the three films fluctuate, P t The change is not excessive; when the electrode of the corona equipment is aged, the electrode can be slightly changed under the impact condition of high-voltage and high-frequency electric sparks, but the distribution of the textures of the film cannot be suddenly changed, so that P t Nor is it excessively varied.
In summary, the standard deviation σ p Will be smaller so that the 3sigma criterion can be used to estimate the change in quality of the film as it is produced.
The reference quality interval span value obtained in this embodiment is:
acquiring tracking indexes at different moments, and calculating the average value P of all tracking indexes T And standard deviation σ P ；
Determining the reference mass interval J as [ P ] T -3σ P ，P T +3σ P ]；
Determining a span value J of the reference quality interval according to the upper and lower floating values of the reference quality interval range =6 σ P; when J is range When the sample distribution is larger, the sample distribution is considered to be more discrete, namely, the state transition degree of the film quality of the three films after corona in the phase space is larger, and the quality in the production distance T is more unstable.
It should be noted that the phase space reconstruction of the present invention is to perform characterization processing on the texture index of the film, and then perform multi-scale time domain analysis on the moving process of the film, so as to realize an improved spatial warping (Wrapping) effect, so that the film texture index of the wear-resistant, light-resistant, stretch-free, shrink-free warning tape is used as a space for all possible states of the quality of the composite film after the corona treatment of the three films, thereby realizing the analysis on the quality change of the composite film.
The minimum analysis interval t in the above embodiment is related to the rotation angle of the roller and the moving distance of the film; the corona treatment is generally carried out at a constant speed, so the t equivalent time is actually the minimum analysis unit with a set distance of 5cm.
And 5, if the reference mass interval span value updated for multiple times is continuously increased for more than two times in the phase space reconstruction process, determining that new quality degradation occurs to the composite film corresponding to the current moment, taking the environmental humidity data and the corona texture information of each layer of film as the input of a neural network model, and identifying and analyzing the quality of the film subjected to corona treatment on the basis of the neural network model.
In this embodiment, J is calculated each time w is updated range J before the minimum unit move based on the last time range(t-2) J of the last moment minimum unit movement range(t-1) And J moved at the current time range(t) Making a comparison when J range(t-1) ＜J range(t) And J range(t-2) ＜J range(t-1) When the new quality deterioration of the composite film is considered to occur at the present momentAt this point, analysis of w is performed:
when updated w enters the phase space, if J is caused range(t) Than the previous time J range(t-1) If the value is large, it is determined that the three film samples corresponding to w cannot well belong to the tracking function of the phase space, that is, an abnormal condition occurs.
At this time, in order to overcome the delay of tracking in the phase space reconstruction process, the abnormal condition needs to be further identified and analyzed; specifically, in this embodiment, a neural network model is adopted, and the input of the neural network model is the environmental humidity data H = { H) corresponding to the three thin films corresponding to the current composite film 1 ，H 2 ，H 3 And corresponding film anomaly degree E = { E = } 1 ，E 2 ，E 3 6-dimensional input vector X = { H) } 1 ，H 2 ，H 3 ，E 1 ，E 2 ，E 3 And outputting the abnormal recognition result of the composite membrane.
Based on the quality identification of the neural network model to the composite membrane, the tracking error condition caused by tracking delay in the phase space reconstruction process can be compensated, and the accuracy of the quality prediction analysis of the composite membrane is improved.
Note that J is range(t-1) ≥J range(t) In the process, the quality analysis of the film can be directly carried out according to the film quality index w obtained by updating the phase space of the composite film.
In the above embodiment, each time the film quality index w is updated means that the quality of three films are considered at the same time, and when the corona treatment abnormality occurs on any one film, the film quality index w is changed.
In the above embodiment, the training process of the neural network model is as follows:
label data: acquiring a training data set, wherein the data set is used for acquiring environmental humidity data and the corona abnormal degree of the films corresponding to the three layers of films and forming a 6-dimensional input vector, and X = { H) = is input to the 6-dimensional input vector in the data set 1 ，H 2 ，H 3 ，E 1 ，E 2 ，E 3 Mark, namely mark the abnormity as 0, mark the normal ratio as 1, and mark the abnormal ratio as 0The recorded results are output as a neural network model, and the input is a 6-dimensional input vector in the data set.
The neural network model provided by the invention is a multilayer feedforward artificial neural network adopting an Error Back-propagation Algorithm (Error Back-propagation Algorithm), and the Algorithm is widely applied to fitting and predicting of a nonlinear system due to good nonlinear approximation capability, generalization capability and adaptability of use. The BP neural network model comprises an input layer, a hidden layer and an output layer:
the neural network model structure is 6-15-1, namely the number of input layer nodes is 6, and the neural network model structure is used for inputting 6-dimensional input vectors X = { H } obtained in real time 1 ，H 2 ，H 3 ，E 1 ，E 2 ，E 3 And 10 hidden layer nodes are provided, and an implementer can adjust the hidden layer nodes according to the training fitting condition. The number of output nodes is 2, and cross entropy Loss is used for supervision and training.
In this embodiment, the neural network model is used to identify the quality of the composite film, because the phase space reconstruction cannot instantly know whether the current corona discharge processing state of the film is abnormal, the BP neural network is trained to perform instantaneous analysis, so as to mark the minimum unit with the abnormality in real time, and facilitate the control of the quality of the production in the later stage.
Meanwhile, the roller of the corona equipment in the film production is considered to be in periodic operation, the working characteristics of the corona equipment are slowly changed, and the chaos state of the corona treatment can be subjected to abnormal analysis by tracking the change in a phase space; the traditional phase space is simple to analyze time-varying signals, the space signals during film production are mapped to a time domain, the system change of corona abnormal textures and humidity difference occurring in the minimum unit is tracked, so that the tracking index is measured and calculated, a BP deep neural network is trained on the basis of the fluctuation condition of the 3sigma standard interval span of the tracking index, and the generalization capability and the real-time performance of system abnormal analysis are improved.
The above-mentioned embodiments are only used for illustrating the technical solutions of the present application, and not for limiting the same; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; such modifications and substitutions do not substantially depart from the spirit and scope of the embodiments of the present application and are intended to be included within the scope of the present application.
Claims (1)
1. A method for analyzing the quality of a corona composite film of a wear-resistant, illumination-resistant, tensile and non-shrinkage warning tape is characterized by comprising the following steps of:
acquiring environmental humidity data of each layer of film before the composite film is not compounded, and calculating the humidity change degree of all the films; the composite film is formed by compounding three layers of films;
acquiring image information of each layer of film of the composite film during corona treatment, acquiring corona texture information of the corresponding film, and determining the abnormal degree of corona of the film;
obtaining a film quality index based on the humidity change degree and the film corona abnormal degree;
performing phase space reconstruction on the film quality index to obtain a tracking index of a phase space; determining a standard deviation of the tracking index by using the tracking index, and obtaining a span value of a reference quality interval based on the standard deviation;
if the reference mass interval span value updated for multiple times is continuously increased for more than two times in the phase space reconstruction process, determining that new quality degradation occurs to the composite film corresponding to the current moment, taking the environmental humidity data and the corona texture information of each layer of film as the input of a neural network model, and identifying and analyzing the film quality of the corona treatment of the film based on the neural network model;
calculating the texture variance of the corona texture information to obtain the abnormal degree of the film, and further amplifying the abnormal degree of the film of all the films to obtain the abnormal degree of the corona of the film;
the quality index of the film is the product of the humidity change degree and the corona abnormal degree of the film;
the method for acquiring the span value of the reference quality interval comprises the following steps:
acquiring tracking indexes at different moments, and calculating the average value of all tracking indexes
And standard deviation>
；
Determining a reference mass interval of
；
Determining a span value of the reference quality interval according to the upper and lower floating values of the reference quality interval
；
The training process of the neural network model comprises the following steps:
respectively acquiring environmental humidity data and film corona abnormal degrees corresponding to three layers of films to form 6-dimensional input vectors as network input of a neural network model, marking the 6-dimensional input vectors in a data set by using a label, marking the abnormal vectors as 0 and the normal vectors as 1, outputting the label as a network, and performing iterative training on the neural network model by using a loss function to obtain a trained neural network model;
the neural network model is a multilayer feedforward artificial neural network of an error back propagation algorithm.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210385057.2A CN114936746B (en) | 2022-04-13 | 2022-04-13 | Quality analysis method for wear-resistant, illumination-resistant, tensile and shrinkage-free warning tape corona composite film |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210385057.2A CN114936746B (en) | 2022-04-13 | 2022-04-13 | Quality analysis method for wear-resistant, illumination-resistant, tensile and shrinkage-free warning tape corona composite film |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114936746A CN114936746A (en) | 2022-08-23 |
| CN114936746B true CN114936746B (en) | 2023-04-07 |
Family
ID=82862441
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210385057.2A Active CN114936746B (en) | 2022-04-13 | 2022-04-13 | Quality analysis method for wear-resistant, illumination-resistant, tensile and shrinkage-free warning tape corona composite film |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114936746B (en) |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111875826A (en) * | 2020-07-22 | 2020-11-03 | 上海大学 | Coarse grained molecular dynamics/Monte Carlo simulation method and proton exchange membrane material |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104915550B (en) * | 2015-05-25 | 2016-06-29 | 东南大学 | Bearing life Forecasting Methodology based on Multiple Time Scales modeling |
| EP3726318B1 (en) * | 2019-04-17 | 2022-07-13 | ABB Schweiz AG | Computer-implemented determination of a quality indicator of a production batch-run that is ongoing |
| CN111222762A (en) * | 2019-12-27 | 2020-06-02 | 江苏南高智能装备创新中心有限公司 | Solar cell panel coating process state monitoring and quality control system |
| CN111679949A (en) * | 2020-04-23 | 2020-09-18 | 平安科技（深圳）有限公司 | Anomaly detection method based on equipment index data and related equipment |
| CN113764048A (en) * | 2020-06-05 | 2021-12-07 | 中国石油天然气股份有限公司 | Polypropylene production quality on-line measuring system |
| CN113989279B (en) * | 2021-12-24 | 2022-03-22 | 武汉华康龙兴工贸有限公司 | Plastic film quality detection method based on artificial intelligence and image processing |
-
2022
- 2022-04-13 CN CN202210385057.2A patent/CN114936746B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111875826A (en) * | 2020-07-22 | 2020-11-03 | 上海大学 | Coarse grained molecular dynamics/Monte Carlo simulation method and proton exchange membrane material |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114936746A (en) | 2022-08-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2021174771A1 (en) | Human-machine collaborative video anomaly detection method | |
| CN110890102A (en) | Engine defect detection algorithm based on RNN voiceprint recognition | |
| CN113780412B (en) | Fault diagnosis model training method and system and fault diagnosis method and system | |
| CN109919032A (en) | A kind of video anomaly detection method based on action prediction | |
| CN109583355B (en) | People flow counting device and method based on boundary selection | |
| CN110958467B (en) | Video quality prediction method and device and electronic equipment | |
| CN110927535A (en) | Power equipment partial discharge severity evaluation method based on extreme learning machine | |
| CN115937091A (en) | Transformer substation equipment defect image detection method based on changeable patch | |
| Chatterjee et al. | Sensing wettability condition of insulation surface employing convolutional neural network | |
| CN111881970A (en) | Intelligent outer broken image identification method based on deep learning | |
| CN111291818B (en) | Non-uniform class sample equalization method for cloud mask | |
| Huang et al. | Study on hydrophobicity detection of composite insulators of transmission lines by image analysis | |
| Do et al. | Dynamic background modeling based on radial basis function neural networks for moving object detection | |
| CN111222798B (en) | Complex industrial process key index soft measurement method | |
| CN114936746B (en) | Quality analysis method for wear-resistant, illumination-resistant, tensile and shrinkage-free warning tape corona composite film | |
| CN115953630A (en) | Cross-domain small sample image classification method based on global-local knowledge distillation | |
| CN114781458A (en) | Power distribution network initial fault identification method based on time-frequency memory neural network of adaptive wavelet and attention mechanism | |
| CN113283334B (en) | Classroom concentration analysis method, device and storage medium | |
| CN114048546A (en) | Graph convolution network and unsupervised domain self-adaptive prediction method for residual service life of aircraft engine | |
| CN115761647A (en) | Computer vision-based method and system for detecting motion trail of object in transformer substation | |
| CN113177563B (en) | Post-chip anomaly detection method integrating CMA-ES algorithm and sequential extreme learning machine | |
| CN113537240B (en) | Deformation zone intelligent extraction method and system based on radar sequence image | |
| Sizyakin et al. | Defect detection on videos using neural network | |
| CN114596433A (en) | Insulator identification method | |
| Shi et al. | Human visual perception based image quality assessment for video prediction |
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 |