CN116165987A - Intelligent production line control system of electronic-grade potassium hydroxide - Google Patents

Intelligent production line control system of electronic-grade potassium hydroxide Download PDF

Info

Publication number
CN116165987A
CN116165987A CN202310307002.4A CN202310307002A CN116165987A CN 116165987 A CN116165987 A CN 116165987A CN 202310307002 A CN202310307002 A CN 202310307002A CN 116165987 A CN116165987 A CN 116165987A
Authority
CN
China
Prior art keywords
feature
ultrasonic
frequency domain
temperature
vector
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.)
Withdrawn
Application number
CN202310307002.4A
Other languages
Chinese (zh)
Inventor
林德荣
蓝伟雄
丘建洲
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.)
Fujian Deer Technology Corp
Original Assignee
Fujian Deer Technology Corp
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 Fujian Deer Technology Corp filed Critical Fujian Deer Technology Corp
Priority to CN202310307002.4A priority Critical patent/CN116165987A/en
Publication of CN116165987A publication Critical patent/CN116165987A/en
Withdrawn legal-status Critical Current

Links

Images

Classifications

    • GPHYSICS
    • G05CONTROLLING; REGULATING
    • G05BCONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B19/00Programme-control systems
    • G05B19/02Programme-control systems electric
    • G05B19/418Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM]
    • G05B19/41885Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM] characterised by modeling, simulation of the manufacturing system
    • GPHYSICS
    • G05CONTROLLING; REGULATING
    • G05BCONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B2219/00Program-control systems
    • G05B2219/30Nc systems
    • G05B2219/32Operator till task planning
    • G05B2219/32339Object oriented modeling, design, analysis, implementation, simulation language
    • YGENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02PCLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/02Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]

Landscapes

  • Engineering & Computer Science (AREA)
  • Manufacturing & Machinery (AREA)
  • General Engineering & Computer Science (AREA)
  • Quality & Reliability (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Investigating Or Analyzing Materials By The Use Of Ultrasonic Waves (AREA)

Abstract

An intelligent production line control system for electronic grade potassium hydroxide is disclosed. Firstly, carrying out frequency domain analysis based on Fourier transform on an ultrasonic signal in a preset time period generated by an ultrasonic generating device to obtain a plurality of ultrasonic frequency domain eigenvalues, then, passing the ultrasonic frequency domain eigenvalues through a multi-scale eigenvector to obtain ultrasonic frequency domain eigenvectors, then, passing temperature values at a plurality of preset time points in the preset time period through the multi-scale eigenvector to obtain a temperature time sequence eigenvector, then, carrying out associated coding on the ultrasonic frequency domain eigenvector and the temperature time sequence eigenvector to obtain a cooperative eigenvector, and finally, passing the cooperative eigenvector through a classifier to obtain a classification result for indicating that the frequency value of the ultrasonic signal at the current time point should be increased or decreased. Thus, the crystallization efficiency of potassium hydroxide can be improved.

Description

Intelligent production line control system of electronic-grade potassium hydroxide
Technical Field
The present application relates to the field of intelligent control, and more particularly, to a control system for an intelligent production line of electronic grade potassium hydroxide.
Background
With the development of global high-new electronic circuit boards, the global demand for high-purity electronic-grade potassium hydroxide is continuously rising, and industrial-grade potassium hydroxide is not satisfied. Therefore, it is necessary to purify industrial grade potassium hydroxide to improve quality.
In view of the above problems, chinese patent CN113860336a discloses a method for preparing electronic grade potassium hydroxide, which comprises stirring, dissolving industrial grade potassium hydroxide solution and potassium hydroxide flake base, cooling, crystallizing, and centrifuging to obtain high purity electronic grade potassium hydroxide solution. However, in the actual production process, crystallization efficiency is found to be low, impurities of other components are present in the crystal, and it is difficult to obtain an electronic grade potassium hydroxide solution with high purity, which is because: the crystallization of potassium hydroxide solution is realized by only slowly bringing the temperature to a predetermined temperature during crystallization, which results in the formation of byproducts, and the temperature is difficult to control accurately during crystallization, resulting in lower crystallization efficiency of potassium hydroxide.
Thus, an optimized intelligent production line control system for electronic grade potassium hydroxide is desired.
Disclosure of Invention
The present application has been made in order to solve the above technical problems. The embodiment of the application provides a control system of an intelligent production line of electronic-grade potassium hydroxide. Firstly, carrying out frequency domain analysis based on Fourier transform on an ultrasonic signal in a preset time period generated by an ultrasonic generating device to obtain a plurality of ultrasonic frequency domain eigenvalues, then, passing the ultrasonic frequency domain eigenvalues through a multi-scale eigenvector to obtain ultrasonic frequency domain eigenvectors, then, passing temperature values at a plurality of preset time points in the preset time period through the multi-scale eigenvector to obtain a temperature time sequence eigenvector, then, carrying out associated coding on the ultrasonic frequency domain eigenvector and the temperature time sequence eigenvector to obtain a cooperative eigenvector, and finally, passing the cooperative eigenvector through a classifier to obtain a classification result for indicating that the frequency value of the ultrasonic signal at the current time point should be increased or decreased. Thus, the crystallization efficiency of potassium hydroxide can be improved.
According to one aspect of the present application, there is provided a control system for an intelligent production line of electronic grade potassium hydroxide, comprising: the data acquisition module is used for acquiring ultrasonic signals of a preset time period generated by the ultrasonic generating device and temperature values of a plurality of preset time points in the preset time period; the frequency domain analysis module is used for carrying out frequency domain analysis based on Fourier transform on the ultrasonic signals so as to obtain a plurality of ultrasonic frequency domain characteristic values; the ultrasonic wave feature extraction module is used for enabling the plurality of ultrasonic wave frequency domain feature values to pass through a multi-scale feature susceptor comprising a first convolution layer and a second convolution layer so as to obtain ultrasonic wave frequency domain feature vectors; the temperature characteristic extraction module is used for passing the temperature values of a plurality of preset time points in the preset time period through the multi-scale characteristic receptor comprising the first convolution layer and the second convolution layer to obtain a temperature time sequence characteristic vector; the characteristic association coding module is used for carrying out association coding on the ultrasonic frequency domain characteristic vector and the temperature time sequence characteristic vector so as to obtain a cooperative characteristic matrix; and the ultrasonic frequency control module is used for enabling the collaborative feature matrix to pass through a classifier to obtain a classification result, wherein the classification result is used for indicating that the frequency value of an ultrasonic signal at the current time point should be increased or decreased.
In the control system of the intelligent production line of electronic grade potassium hydroxide, the multi-scale feature susceptor comprises a first convolution layer and a second convolution layer which are parallel, and a multi-scale feature fusion layer connected with the first convolution layer and the second convolution layer, wherein the first convolution layer and the second convolution layer use one-dimensional convolution kernels with different scales.
In the control system of the intelligent production line of electronic grade potassium hydroxide, the ultrasonic characteristic extraction module comprises: an ultrasonic input vector arrangement unit for arranging the plurality of ultrasonic frequency domain feature values into ultrasonic input vectors; a first scale ultrasonic feature extraction unit, configured to perform one-dimensional convolution encoding on the ultrasonic input vector by using a first convolution layer of the multi-scale feature sensor to obtain a first scale ultrasonic feature vector, where the first convolution layer has a first one-dimensional convolution kernel with a first length; a second-scale ultrasonic feature extraction unit, configured to perform one-dimensional convolution encoding on the ultrasonic input vector using a second convolution layer of the multi-scale feature susceptor to obtain a second-scale ultrasonic feature vector, where the second convolution layer has a second one-dimensional convolution kernel with a second length, and the first length is different from the second length; and the ultrasonic multi-scale fusion unit is used for cascading the first-scale ultrasonic feature vector and the second-scale ultrasonic feature vector by using the multi-scale feature fusion layer of the multi-scale feature receptor so as to obtain the ultrasonic frequency domain feature vector.
In the control system of the intelligent production line of electronic grade potassium hydroxide, the temperature characteristic extraction module comprises: a temperature input vector arrangement unit configured to arrange temperature values at a plurality of predetermined time points within the predetermined period of time as a temperature input vector; a first scale temperature feature extraction unit, configured to perform one-dimensional convolutional encoding on the temperature input vector by using a first convolutional layer of the multi-scale feature susceptor to obtain a first scale temperature feature vector, where the first convolutional layer has a first one-dimensional convolution kernel with a first length; a second scale temperature feature extraction unit, configured to perform one-dimensional convolution encoding on the temperature input vector using a second convolution layer of the multi-scale feature susceptor to obtain a second scale temperature feature vector, where the second convolution layer has a second one-dimensional convolution kernel with a second length, and the first length is different from the second length; and the temperature multi-scale fusion unit is used for cascading the first-scale temperature characteristic vector and the second-scale temperature characteristic vector by using the multi-scale characteristic fusion layer of the multi-scale characteristic receptor so as to obtain the temperature frequency domain characteristic vector.
In the control system of the intelligent production line of electronic grade potassium hydroxide, the feature association coding module comprises: the optimization factor calculation unit is used for calculating the Helmholtz class free energy factors of the ultrasonic frequency domain feature vector and the temperature time sequence feature vector to obtain a first Helmholtz class free energy factor and a second Helmholtz class free energy factor; the weighting optimization unit is used for weighting the ultrasonic frequency domain feature vector and the temperature time sequence feature vector by taking the first Helmholtz class free energy factor and the second Helmholtz class free energy factor as weighting weights so as to obtain an optimized ultrasonic frequency domain feature vector and an optimized temperature time sequence feature vector; and the optimized association characteristic fusion unit is used for carrying out association coding on the optimized ultrasonic frequency domain characteristic vector and the optimized temperature time sequence characteristic vector so as to obtain the collaborative characteristic matrix.
In the above control system for an intelligent production line of electronic grade potassium hydroxide, the optimization factor calculating subunit is configured to: calculating Helmholtz free energy factors of the ultrasonic frequency domain feature vector and the temperature time sequence feature vector according to the following optimization formula to obtain a first Helmholtz free energy factor and a second Helmholtz free energy factor; wherein, the optimization formula is:
Figure SMS_2
Figure SMS_8
Wherein (1)>
Figure SMS_11
Characteristic values representing respective positions in the ultrasonic frequency domain characteristic vector,/or->
Figure SMS_3
Characteristic values representing respective positions in the temperature timing characteristic vector, +.>
Figure SMS_5
And->
Figure SMS_7
Classification probability values respectively representing the ultrasonic frequency domain feature vector and the temperature time series feature vector, and +.>
Figure SMS_10
Is the length of the feature vector, +.>
Figure SMS_1
Represents a logarithmic function with base 2, +.>
Figure SMS_4
Representing an exponential operation, ++>
Figure SMS_6
And->
Figure SMS_9
Representing the first helmholtz-like free energy factor and the second helmholtz-like free energy factor, respectively.
In the above control system for an intelligent production line of electronic grade potassium hydroxide, the optimization association feature fusion unit is configured to: performing association coding on the optimized ultrasonic frequency domain feature vector and the optimized temperature time sequence feature vector by using the following association coding formula to obtain the collaborative feature matrix; wherein, the association coding formula is:
Figure SMS_12
wherein (1)>
Figure SMS_13
Representing the optimized ultrasonic frequency domain feature vector, < >>
Figure SMS_14
A transpose vector representing the optimized ultrasonic frequency domain feature vector,>
Figure SMS_15
representing the optimized temperature time sequence feature vector, < >>
Figure SMS_16
Representing the synergistic feature matrix,>
Figure SMS_17
representing vector multiplication.
In the control system of the intelligent production line of electronic grade potassium hydroxide, the ultrasonic frequency control module comprises: the matrix unfolding unit is used for unfolding the collaborative feature matrix into a classification feature vector according to a row vector or a column vector; the full-connection coding unit is used for carrying out full-connection coding on the classification characteristic vectors by using a plurality of full-connection layers of the classifier so as to obtain coded classification characteristic vectors; and the classification unit is used for passing the coding classification feature vector through a Softmax classification function of the classifier to obtain the classification result.
Compared with the prior art, the control system of the intelligent production line of the electronic-grade potassium hydroxide firstly carries out frequency domain analysis based on Fourier transform on ultrasonic signals of a preset time period generated by an ultrasonic generating device to obtain a plurality of ultrasonic frequency domain feature values, then the ultrasonic frequency domain feature values pass through a multi-scale feature sensor to obtain ultrasonic frequency domain feature vectors, then temperature values of a plurality of preset time points in the preset time period pass through the multi-scale feature sensor to obtain temperature time sequence feature vectors, then the ultrasonic frequency domain feature vectors and the temperature time sequence feature vectors are subjected to associated coding to obtain a collaborative feature matrix, and finally the collaborative feature matrix passes through a classifier to obtain a classification result for indicating that the frequency value of the ultrasonic signals of the current time point should be increased or decreased. Thus, the crystallization efficiency of potassium hydroxide can be improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the description of the embodiments will be briefly introduced below, it being obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art. The following drawings are not intended to be drawn to scale, with emphasis instead being placed upon illustrating the principles of the present application.
Fig. 1 is an application scenario diagram of a control system of an intelligent production line of electronic grade potassium hydroxide according to an embodiment of the present application.
FIG. 2 is a block diagram of a control system for an intelligent production line of electronic grade potassium hydroxide according to an embodiment of the present application.
Fig. 3 is a block diagram schematic diagram of the ultrasonic feature extraction module in the control system of the intelligent production line of electronic grade potassium hydroxide according to an embodiment of the application.
Fig. 4 is a block diagram of the temperature feature extraction module in the control system of the intelligent production line of electronic grade potassium hydroxide according to the embodiment of the application.
Fig. 5 is a block diagram of the feature-related encoding module in the control system of the intelligent production line of electronic grade potassium hydroxide according to the embodiment of the application.
Fig. 6 is a block diagram of the ultrasonic frequency control module in the control system of the intelligent production line of electronic grade potassium hydroxide according to the embodiment of the application.
FIG. 7 is a flow chart of a method for controlling an intelligent production line of electronic grade potassium hydroxide according to an embodiment of the present application.
FIG. 8 is a schematic diagram of a system architecture of a method for intelligent production line control of electronic grade potassium hydroxide according to an embodiment of the present application.
Detailed Description
The following description of the embodiments of the present application will be made clearly and fully with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some, but not all embodiments of the present application. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments of the present application without making any inventive effort, are also within the scope of the present application.
As used in this application and in the claims, the terms "a," "an," "the," and/or "the" are not specific to the singular, but may include the plural, unless the context clearly dictates otherwise. In general, the terms "comprises" and "comprising" merely indicate that the steps and elements are explicitly identified, and they do not constitute an exclusive list, as other steps or elements may be included in a method or apparatus.
Although the present application makes various references to certain modules in a system according to embodiments of the present application, any number of different modules may be used and run on a user terminal and/or server. The modules are merely illustrative, and different aspects of the systems and methods may use different modules.
Flowcharts are used in this application to describe the operations performed by systems according to embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed in order precisely. Rather, the various steps may be processed in reverse order or simultaneously, as desired. Also, other operations may be added to or removed from these processes.
Hereinafter, example embodiments according to the present application will be described in detail with reference to the accompanying drawings. It should be apparent that the described embodiments are only some of the embodiments of the present application and not all of the embodiments of the present application, and it should be understood that the present application is not limited by the example embodiments described herein.
As described above, chinese patent CN113860336a discloses a preparation method of electronic grade potassium hydroxide, in which crystallization efficiency is found to be low in the actual preparation process, impurities of other components may exist in the crystal, and it is difficult to obtain electronic grade potassium hydroxide solution with high purity, which is the following reasons: the crystallization of potassium hydroxide solution is realized by only slowly bringing the temperature to a predetermined temperature during crystallization, which results in the formation of byproducts, and the temperature is difficult to control accurately during crystallization, resulting in lower crystallization efficiency of potassium hydroxide. Thus, an optimized intelligent production line control system for electronic grade potassium hydroxide is desired.
It should be understood that ultrasonic crystallization is the use of ultrasonic energy to control the crystallization process. The nucleation and growth process can be controlled by ultrasonic waves, thereby optimizing the crystallization process. Therefore, ultrasonic crystallization can be used to replace the conventional cooling crystallization, so that the supersaturation degree of the solution can be effectively controlled by controlling ultrasonic waves, and the growth rate of crystals in the solution can be changed, thereby inducing potassium hydroxide crystallization. In the actual preparation process, the ultrasonic treatment can obviously shorten the crystallization starting time of the potassium hydroxide solution and refine crystal grains of the potassium hydroxide solution. In a low supersaturation potassium hydroxide solution environment where nucleation is difficult, the use of ultrasonic waves can effectively promote potassium hydroxide crystallization nucleation. However, as the ultrasonic frequency increases, the ultrasonic treatment time increases, the cooling rate of the saturated potassium hydroxide solution decreases, and the rate of decrease in the solution temperature decreases, resulting in a decrease in the crystallization rate of potassium hydroxide.
Based on this, in the technical solution of the present application, it is considered that in the process of inducing crystallization of potassium hydroxide by actually performing ultrasonic treatment, the crystallization efficiency is affected due to both the variation of the cooling temperature of the solution and the variation of the frequency of the ultrasonic wave. Therefore, it is desirable to further optimize the crystallization efficiency by cooperative control of the ultrasonic signal and the cooling temperature value. However, considering that since the ultrasonic signal is liable to have a large amount of noise signals in the process of acquisition, this may cause interference to cooperative control, and the correlation characteristic between the frequency domain time-series variation characteristic of the ultrasonic signal and the time-series dynamic variation characteristic of the temperature value is difficult to capture and extract, resulting in lower accuracy of judgment of the frequency value of the ultrasonic signal for optimal induced crystallization.
In recent years, deep learning and neural networks have been widely used in the fields of computer vision, natural language processing, text signal processing, and the like. The development of deep learning and neural networks provides new solutions and schemes for mining the correlation feature distribution information between the frequency domain time sequence variation features of the ultrasonic signals and the time sequence dynamic variation features of the temperature values. Those of ordinary skill in the art will appreciate that a deep learning based deep neural network model may adjust parameters of the deep neural network model by appropriate training strategies, such as by a gradient descent back-propagation algorithm, to enable modeling of complex nonlinear correlations between things, which is obviously suitable for modeling and establishing complex mappings between frequency domain time-varying features of the ultrasound signal and time-varying dynamic features of the temperature values.
Specifically, in the technical solution of the present application, first, an ultrasonic signal of a predetermined period of time generated by an ultrasonic wave generating device and temperature values at a plurality of predetermined time points within the predetermined period of time are acquired. Next, considering that for the ultrasonic signal of the predetermined period, since the ultrasonic signal is a time domain signal, the time domain signal is more intuitive to the dominance of the features in the time correlation, but because the ultrasonic signal is weaker, it is interfered by external noise, resulting in lower feature extraction accuracy for the ultrasonic signal, thereby affecting the detection judgment of the mapping relationship between the frequency domain time sequence variation feature and the time sequence variation of the temperature of the ultrasonic signal. The characteristics of the frequency domain signals are different from those of the time domain signals, the ultrasonic signals are converted into the frequency domain, and the frequency time sequence dynamic change condition of the ultrasonic signals can be determined through the implicit characteristic distribution information of the ultrasonic signals in the frequency domain. Therefore, in the technical solution of the present application, the frequency domain analysis based on fourier transform is performed on the ultrasonic signal to obtain a plurality of ultrasonic frequency domain feature values.
Then, it is considered that there is an association relationship between a plurality of ultrasonic frequency-domain feature values due to the ultrasonic signal, and there are different association characteristics under different class spans of the plurality of ultrasonic frequency-domain feature values. Therefore, in order to enable sufficient expression of the correlation features of the plurality of ultrasonic frequency domain feature values, the frequency domain time series correlation features of the ultrasonic signal are extracted, and in the technical solution of the present application, the plurality of ultrasonic frequency domain feature values are further passed through a multi-scale feature susceptor including a first convolution layer and a second convolution layer to obtain ultrasonic frequency domain feature vectors. In particular, here, the first convolution layer and the second convolution layer perform feature extraction on feature vectors of the plurality of ultrasonic frequency domain feature value arrangements by using one-dimensional convolution kernels of different scales, so as to extract multi-scale associated feature information under different frequency domain feature value spans among the plurality of ultrasonic frequency domain feature values, namely frequency domain time sequence associated feature information of the ultrasonic signal.
Further, the temperature value has a time-series dynamic change law in a time dimension, and has different mode state change characteristics at different time period spans within the predetermined time period. Therefore, in the technical scheme of the application, after the temperature values of a plurality of preset time points in the preset time period are arranged as the temperature input vector, feature mining is performed in the multi-scale feature sensor comprising the first convolution layer and the second convolution layer, so that multi-scale neighborhood associated feature information of the temperature values in different time spans is extracted, and therefore the temperature time sequence feature vector is obtained.
And then, further carrying out association coding on the ultrasonic frequency domain feature vector and the temperature time sequence feature vector so as to integrate association feature distribution information between the multi-scale association feature of the ultrasonic frequency domain feature value and the time sequence dynamic multi-scale change feature of the temperature, namely, the cooperative association feature information between the frequency time sequence change feature of the ultrasonic and the time sequence change feature of the temperature, thereby obtaining a cooperative feature matrix. Then, the collaborative feature matrix is used as a classification feature matrix to pass through a classifier to obtain a classification result, wherein the classification result is used for representing that the frequency value of the ultrasonic signal at the current time point is increased or decreased.
That is, in the technical solution of the present application, the label of the classifier includes that the frequency value of the ultrasonic signal at the current time point should be increased (first label) and that the frequency value of the ultrasonic signal at the current time point should be decreased (second label), wherein the classifier determines to which classification label the classification feature matrix belongs by a soft maximum function. It should be noted that the first tag p1 and the second tag p2 do not include the concept of artificial setting, and in fact, during the training process, the computer model does not have the concept of "the frequency value of the ultrasonic signal at the current time point should be increased or decreased", which is only two kinds of classification tags, and the probability that the output feature is under the two kinds of classification tags, that is, the sum of p1 and p2 is one. Therefore, the classification result that the frequency value of the ultrasonic signal should be increased or decreased is actually a class probability distribution converted into a class classification conforming to the natural law by classifying the tag, and the physical meaning of the natural probability distribution of the tag is essentially used instead of the language text meaning that the frequency value of the ultrasonic signal at the current time point should be increased or decreased. It should be understood that, in the technical solution of the present application, the classification label of the classifier is a control policy label that the frequency value of the ultrasonic signal at the current time point should be increased or decreased, so after the classification result is obtained, the frequency value of the ultrasonic signal at the current time point may be adaptively adjusted based on the classification result, so as to further optimize the crystallization efficiency through cooperative control of the frequency value of the ultrasonic signal and the cooling temperature value.
In particular, in the technical solution of the present application, considering the data heterogeneity between the ultrasonic frequency domain feature vector and the temperature value expressed by the temperature time sequence feature vector, and the difference in feature distribution direction between the ultrasonic frequency domain feature correlation and the temperature value time sequence correlation, for the ultrasonic frequency domain feature vector and the temperature time sequence feature vector, there may be a weak class correlation distribution instance with respect to the class label of the classifier in their respective overall feature distributions, that is, the compatibility of the ultrasonic frequency domain feature vector and the temperature time sequence feature vector with respect to the overall feature distribution of the class probability of the classifier is low, which may affect the correlation fusion effect of the cooperative feature matrix with respect to the ultrasonic frequency domain feature vector and the temperature time sequence feature vector, thereby affecting the accuracy of the classification result of the cooperative feature matrix.
Based on this, the helmholtz-like free energy factor of the ultrasonic frequency domain feature vector and the temperature timing feature vector is preferably calculated, specifically:
Figure SMS_18
Figure SMS_19
,/>
Figure SMS_20
and->
Figure SMS_21
Respectively represent the ultrasonic frequency domain feature vector +. >
Figure SMS_22
And the temperature timing feature vector +.>
Figure SMS_23
Is a classification probability value of>
Figure SMS_24
Is the length of the feature vector.
Here, the ultrasonic frequency domain feature vector may be based on the helmholtz free energy formula
Figure SMS_27
And the temperature timing feature vector +.>
Figure SMS_29
The respective feature value sets describe the energy values of the predetermined category as the free energy of classification of the feature vector as a whole by using them to +.>
Figure SMS_30
And the temperature timing feature vector +.>
Figure SMS_26
Weighting is performed to obtain the ultrasonic frequency domain feature vector +.>
Figure SMS_28
And the temperature timing feature vector +.>
Figure SMS_31
Focusing on the class-dependent prototype instance (prototype instance) distribution of features overlapping with the true instance (groundtruth instance) distribution in the classification target domain so as to be in the ultrasonic frequency domain feature vector +>
Figure SMS_32
And the temperature timing feature vector
Figure SMS_25
There is a weak correlation-like within the overall feature distribution of (a)Under the condition of the distribution instance, incremental learning is realized by carrying out ambiguity labeling on the distribution instance, so that the compatibility of class probability of the overall feature distribution relative to a classifier is improved, and the association fusion effect of the cooperative feature matrix on the ultrasonic frequency domain feature vector and the temperature time sequence feature vector is improved, so that the accuracy of the classification result of the cooperative feature matrix is improved. Therefore, the frequency value of the ultrasonic signal can be adaptively adjusted in real time and accurately based on the actual temperature change condition, so that the crystallization efficiency and the purity of the prepared electronic-grade potassium hydroxide are optimized.
Fig. 1 is an application scenario diagram of a control system of an intelligent production line of electronic grade potassium hydroxide according to an embodiment of the present application. As shown in fig. 1, in this application scenario, first, an ultrasonic signal (e.g., D1 shown in fig. 1) of a predetermined period of time generated by an ultrasonic generating device (e.g., M shown in fig. 1) and temperature values (e.g., D2 shown in fig. 1) of a plurality of predetermined time points in the predetermined period of time are acquired, and then the ultrasonic signal and the temperature values of the plurality of predetermined time points in the predetermined period of time are input to a server (e.g., S shown in fig. 1) of a control algorithm of a smart product line in which electronic grade potassium hydroxide is deployed, wherein the server is capable of processing the ultrasonic signal and the temperature values of the plurality of predetermined time points in the predetermined period of time using the control algorithm of the smart product line of electronic grade potassium hydroxide to obtain a classification result indicating that the frequency value of the ultrasonic signal of the current time point should be increased or decreased.
Having described the basic principles of the present application, various non-limiting embodiments of the present application will now be described in detail with reference to the accompanying drawings.
FIG. 2 is a block diagram of a control system for an intelligent production line of electronic grade potassium hydroxide according to an embodiment of the present application. As shown in fig. 2, the control system 100 of the intelligent production line of electronic grade potassium hydroxide according to the embodiment of the present application includes: a data acquisition module 110 for acquiring an ultrasonic signal of a predetermined period of time generated by an ultrasonic wave generating device and temperature values at a plurality of predetermined time points within the predetermined period of time; the frequency domain analysis module 120 is configured to perform fourier transform-based frequency domain analysis on the ultrasonic signal to obtain a plurality of ultrasonic frequency domain feature values; an ultrasonic feature extraction module 130, configured to pass the plurality of ultrasonic frequency domain feature values through a multi-scale feature susceptor that includes a first convolution layer and a second convolution layer to obtain an ultrasonic frequency domain feature vector; a temperature feature extraction module 140, configured to pass temperature values at a plurality of predetermined time points within the predetermined time period through the multi-scale feature susceptor including the first convolution layer and the second convolution layer to obtain a temperature time sequence feature vector; the feature association encoding module 150 is configured to perform association encoding on the ultrasonic frequency domain feature vector and the temperature time sequence feature vector to obtain a collaborative feature matrix; and an ultrasonic frequency control module 160 for passing the collaborative feature matrix through a classifier to obtain a classification result, wherein the classification result is used for indicating that the frequency value of the ultrasonic signal at the current time point should be increased or decreased.
More specifically, in the embodiment of the present application, the data acquisition module 110 is configured to acquire an ultrasonic signal generated by the ultrasonic generating device for a predetermined period of time and temperature values at a plurality of predetermined time points within the predetermined period of time. In the actual process of conducting ultrasonic treatment to induce crystallization of potassium hydroxide, both changes in the cooling temperature of the solution and changes in the frequency of the ultrasonic waves affect the efficiency of crystallization. Thus, the present application further optimizes crystallization efficiency by cooperative control of the ultrasonic signal and the cooling temperature value.
More specifically, in the embodiment of the present application, the frequency domain analysis module 120 is configured to perform fourier transform-based frequency domain analysis on the ultrasonic signal to obtain a plurality of ultrasonic frequency domain feature values. For the ultrasonic signal in the predetermined time period, since the ultrasonic signal is a time domain signal, the time domain signal is more intuitive to the dominance of the feature in time correlation, but because the ultrasonic signal is weaker, the ultrasonic signal is interfered by external noise, so that the feature extraction accuracy of the ultrasonic signal is lower, and the detection and judgment of the mapping relationship between the frequency domain time sequence change feature and the time sequence change of the temperature of the ultrasonic signal are further affected. The characteristics of the frequency domain signals are different from those of the time domain signals, the ultrasonic signals are converted into the frequency domain, and the frequency time sequence dynamic change condition of the ultrasonic signals can be determined through the implicit characteristic distribution information of the ultrasonic signals in the frequency domain. Therefore, in the technical solution of the present application, the frequency domain analysis based on fourier transform is performed on the ultrasonic signal to obtain a plurality of ultrasonic frequency domain feature values.
More specifically, in the embodiment of the present application, the ultrasonic feature extraction module 130 is configured to pass the plurality of ultrasonic frequency domain feature values through a multi-scale feature susceptor that includes a first convolution layer and a second convolution layer to obtain an ultrasonic frequency domain feature vector. Since there is an association relationship between the plurality of ultrasonic frequency domain feature values of the ultrasonic signal, and there are different association characteristics under different class spans of the plurality of ultrasonic frequency domain feature values. Therefore, in order to enable sufficient expression of the correlation features of the plurality of ultrasonic frequency domain feature values, the frequency domain time series correlation features of the ultrasonic signal are extracted, and in the technical solution of the present application, the plurality of ultrasonic frequency domain feature values are further passed through a multi-scale feature susceptor including a first convolution layer and a second convolution layer to obtain ultrasonic frequency domain feature vectors. In particular, here, the first convolution layer and the second convolution layer perform feature extraction on feature vectors of the plurality of ultrasonic frequency domain feature value arrangements by using one-dimensional convolution kernels of different scales, so as to extract multi-scale associated feature information under different frequency domain feature value spans among the plurality of ultrasonic frequency domain feature values, namely frequency domain time sequence associated feature information of the ultrasonic signal.
More specifically, in an embodiment of the present application, the multi-scale feature susceptor includes first and second convolution layers in parallel, and a multi-scale feature fusion layer connected to the first and second convolution layers, wherein the first and second convolution layers use one-dimensional convolution kernels having different scales.
More specifically, in the embodiment of the present application, as shown in fig. 3, the ultrasonic feature extraction module 130 includes: an ultrasonic input vector arrangement unit 131 for arranging the plurality of ultrasonic frequency domain feature values into ultrasonic input vectors; a first scale ultrasonic feature extraction unit 132 configured to perform one-dimensional convolution encoding on the ultrasonic input vector using a first convolution layer of the multi-scale feature susceptor to obtain a first scale ultrasonic feature vector, where the first convolution layer has a first one-dimensional convolution kernel of a first length; a second-scale ultrasonic feature extraction unit 133 configured to perform one-dimensional convolution encoding on the ultrasonic input vector using a second convolution layer of the multi-scale feature susceptor to obtain a second-scale ultrasonic feature vector, where the second convolution layer has a second one-dimensional convolution kernel of a second length, and the first length is different from the second length; and an ultrasonic multi-scale fusion unit 134, configured to concatenate the first-scale ultrasonic feature vector and the second-scale ultrasonic feature vector using the multi-scale feature fusion layer of the multi-scale feature susceptor to obtain the ultrasonic frequency domain feature vector.
More specifically, in the embodiment of the present application, the temperature feature extraction module 140 is configured to pass the temperature values at a plurality of predetermined time points within the predetermined time period through the multi-scale feature susceptor including the first convolution layer and the second convolution layer to obtain a temperature time sequence feature vector. The temperature value has a time sequence dynamic change rule in a time dimension, and has different mode state change characteristics in different time period spans in the preset time period. Therefore, in the technical scheme of the application, after the temperature values of a plurality of preset time points in the preset time period are arranged as the temperature input vector, feature mining is performed in the multi-scale feature sensor comprising the first convolution layer and the second convolution layer, so that multi-scale neighborhood associated feature information of the temperature values in different time spans is extracted, and therefore the temperature time sequence feature vector is obtained.
More specifically, in the embodiment of the present application, as shown in fig. 4, the temperature feature extraction module 140 includes: a temperature input vector arrangement unit 141 for arranging temperature values at a plurality of predetermined time points within the predetermined period as a temperature input vector; a first scale temperature feature extraction unit 142, configured to perform one-dimensional convolution encoding on the temperature input vector using a first convolution layer of the multi-scale feature susceptor to obtain a first scale temperature feature vector, where the first convolution layer has a first one-dimensional convolution kernel with a first length; a second scale temperature feature extraction unit 143, configured to perform one-dimensional convolution encoding on the temperature input vector using a second convolution layer of the multi-scale feature susceptor to obtain a second scale temperature feature vector, where the second convolution layer has a second one-dimensional convolution kernel with a second length, and the first length is different from the second length; and a temperature multi-scale fusion unit 144, configured to concatenate the first-scale temperature feature vector and the second-scale temperature feature vector using the multi-scale feature fusion layer of the multi-scale feature susceptor to obtain the temperature frequency domain feature vector.
It should be appreciated that convolutional neural network (Convolutional Neural Network, CNN) is an artificial neural network and has wide application in the fields of image recognition and the like. The convolutional neural network may include an input layer, a hidden layer, and an output layer, where the hidden layer may include a convolutional layer, a pooling layer, an activation layer, a full connection layer, etc., where the previous layer performs a corresponding operation according to input data, outputs an operation result to the next layer, and obtains a final result after the input initial data is subjected to a multi-layer operation. The convolutional neural network model has excellent performance in the aspect of image local feature extraction by taking a convolutional kernel as a feature filtering factor, and has stronger feature extraction generalization capability and fitting capability compared with the traditional image feature extraction algorithm based on statistics or feature engineering.
More specifically, in the embodiment of the present application, the feature correlation encoding module 150 is configured to perform correlation encoding on the ultrasonic frequency domain feature vector and the temperature time sequence feature vector to obtain a collaborative feature matrix. And integrating the correlation characteristic distribution information between the multiscale correlation characteristic of the ultrasonic frequency domain characteristic value and the time sequence dynamic multiscale variation characteristic of the temperature, namely the cooperative correlation characteristic information between the frequency time sequence variation characteristic of the ultrasonic and the time sequence variation characteristic of the temperature, thereby obtaining a cooperative characteristic matrix.
More specifically, in the embodiment of the present application, as shown in fig. 5, the feature association encoding module 150 includes: an optimization factor calculation unit 151, configured to calculate a helmholtz type free energy factor of the ultrasonic frequency domain feature vector and the temperature timing feature vector to obtain a first helmholtz type free energy factor and a second helmholtz type free energy factor; a weighting optimization unit 152, configured to weight the ultrasonic frequency domain feature vector and the temperature time sequence feature vector with the first helmholtz class free energy factor and the second helmholtz class free energy factor as weighting weights, so as to obtain an optimized ultrasonic frequency domain feature vector and an optimized temperature time sequence feature vector; and an optimized correlation feature fusion unit 153, configured to perform correlation encoding on the optimized ultrasonic frequency domain feature vector and the optimized temperature time sequence feature vector to obtain the collaborative feature matrix.
In particular, in the technical solution of the present application, considering the data heterogeneity between the ultrasonic frequency domain feature vector and the temperature value expressed by the temperature time sequence feature vector, and the difference in feature distribution direction between the ultrasonic frequency domain feature correlation and the temperature value time sequence correlation, for the ultrasonic frequency domain feature vector and the temperature time sequence feature vector, there may be a weak class correlation distribution instance with respect to the class label of the classifier in their respective overall feature distributions, that is, the compatibility of the ultrasonic frequency domain feature vector and the temperature time sequence feature vector with respect to the overall feature distribution of the class probability of the classifier is low, which may affect the correlation fusion effect of the cooperative feature matrix with respect to the ultrasonic frequency domain feature vector and the temperature time sequence feature vector, thereby affecting the accuracy of the classification result of the cooperative feature matrix. Based on this, the helmholtz-like free energy factor of the ultrasonic frequency domain feature vector and the temperature timing feature vector is preferably calculated.
More specifically, in the embodiment of the present application, the optimization factor calculating subunit 151 is configured to: calculating Helmholtz free energy factors of the ultrasonic frequency domain feature vector and the temperature time sequence feature vector according to the following optimization formula to obtain a first Helmholtz free energy factor and a second Helmholtz free energy factor; wherein, the optimization formula is:
Figure SMS_34
Figure SMS_36
wherein (1)>
Figure SMS_38
Characteristic values representing respective positions in the ultrasonic frequency domain characteristic vector,/or->
Figure SMS_35
Characteristic values representing respective positions in the temperature timing characteristic vector, +.>
Figure SMS_40
And->
Figure SMS_42
Classification probability values respectively representing the ultrasonic frequency domain feature vector and the temperature time series feature vector, and +.>
Figure SMS_43
Is the length of the feature vector, +.>
Figure SMS_33
Represents a logarithmic function with base 2, +.>
Figure SMS_37
Representing an exponential operation, ++>
Figure SMS_39
And->
Figure SMS_41
Representing the first helmholtz-like free energy factor and the second helmholtz-like free energy factor, respectively.
Here, the ultrasonic frequency domain feature vector may be based on the helmholtz free energy formula
Figure SMS_45
And the temperature timing feature vector +.>
Figure SMS_48
The respective feature value sets describe the energy values of the predetermined category as the free energy of classification of the feature vector as a whole by using them to +. >
Figure SMS_49
And the temperature timing feature vector +.>
Figure SMS_46
Weighting is performed to obtain the ultrasonic frequency domain feature vector +.>
Figure SMS_47
And the temperature timing feature vector +.>
Figure SMS_50
Focusing on the class-dependent prototype instance (prototype instance) distribution of features overlapping with the true instance (groundtruth instance) distribution in the classification target domain so as to be in the ultrasonic frequency domain feature vector +>
Figure SMS_51
And the temperature timing feature vector
Figure SMS_44
Incremental learning is achieved by fuzziness labeling of the global feature distribution with instances of weakly correlated class distributions within it, thereby improving the compatibility of the global feature distribution with respect to class probabilities of the classifier, and enhancing the collaborative feature matrix with respect to the ultrasound wavesAnd the correlation fusion effect of the frequency domain feature vector and the temperature time sequence feature vector is improved, so that the accuracy of the classification result of the cooperative feature matrix is improved. Therefore, the frequency value of the ultrasonic signal can be adaptively adjusted in real time and accurately based on the actual temperature change condition, so that the crystallization efficiency and the purity of the prepared electronic-grade potassium hydroxide are optimized.
More specifically, in the embodiment of the present application, the optimization association feature fusion unit 153 is configured to: performing association coding on the optimized ultrasonic frequency domain feature vector and the optimized temperature time sequence feature vector by using the following association coding formula to obtain the collaborative feature matrix; wherein, the association coding formula is:
Figure SMS_52
Wherein (1)>
Figure SMS_53
Representing the optimized ultrasonic frequency domain feature vector, < >>
Figure SMS_54
A transpose vector representing the optimized ultrasonic frequency domain feature vector,>
Figure SMS_55
representing the optimized temperature time sequence feature vector, < >>
Figure SMS_56
Representing the synergistic feature matrix,>
Figure SMS_57
representing vector multiplication.
More specifically, in the embodiment of the present application, the ultrasonic frequency control module 160 is configured to pass the collaborative feature matrix through a classifier to obtain a classification result, where the frequency value of the ultrasonic signal used to represent the current time point should be increased or should be decreased. That is, in the technical solution of the present application, the label of the classifier includes that the frequency value of the ultrasonic signal at the current time point should be increased (first label) and that the frequency value of the ultrasonic signal at the current time point should be decreased (second label), wherein the classifier determines to which classification label the classification feature matrix belongs by a soft maximum function.
It should be appreciated that the role of the classifier is to learn the classification rules and classifier using a given class, known training data, and then classify (or predict) the unknown data. Logistic regression (logistics), SVM, etc. are commonly used to solve the classification problem, and for multi-classification problems (multi-class classification), logistic regression or SVM can be used as well, but multiple bi-classifications are required to compose multiple classifications, but this is error-prone and inefficient, and the commonly used multi-classification method is the Softmax classification function.
More specifically, in the embodiment of the present application, as shown in fig. 6, the ultrasonic frequency control module 160 includes: a matrix developing unit 161, configured to develop the collaborative feature matrix into a classification feature vector according to a row vector or a column vector; a full-connection encoding unit 162, configured to perform full-connection encoding on the classification feature vector by using multiple full-connection layers of the classifier to obtain an encoded classification feature vector; and a classification unit 163, configured to pass the encoded classification feature vector through a Softmax classification function of the classifier to obtain the classification result.
To sum up, the control system 100 of the intelligent production line of electronic grade potassium hydroxide according to the embodiment of the present application is illustrated, which firstly performs frequency domain analysis based on fourier transform on an ultrasonic signal generated by an ultrasonic generating device for a predetermined period of time to obtain a plurality of ultrasonic frequency domain feature values, then passes the plurality of ultrasonic frequency domain feature values through a multi-scale feature susceptor to obtain an ultrasonic frequency domain feature vector, then passes a temperature value of a plurality of predetermined time points within the predetermined period of time through the multi-scale feature susceptor to obtain a temperature time sequence feature vector, then performs association coding on the ultrasonic frequency domain feature vector and the temperature time sequence feature vector to obtain a cooperative feature matrix, and finally passes the cooperative feature matrix through a classifier to obtain a classification result for indicating that the frequency value of the ultrasonic signal at the current time point should be increased or decreased. Thus, the crystallization efficiency of potassium hydroxide can be improved.
As described above, the control system 100 of the intelligent production line of electronic grade potassium hydroxide according to the embodiment of the present application may be implemented in various terminal devices, for example, a server or the like having a control algorithm of the intelligent production line of electronic grade potassium hydroxide according to the embodiment of the present application. In one example, the control system 100 of the intelligent production line based on electronic grade potassium hydroxide of the embodiments of the present application may be integrated into the terminal device as a software module and/or hardware module. For example, the control system 100 of the intelligent production line of electronic grade potassium hydroxide according to the embodiments of the present application may be a software module in the operating system of the terminal device, or may be an application program developed for the terminal device; of course, the control system 100 of the intelligent production line of electronic grade potassium hydroxide according to the embodiment of the present application can also be one of numerous hardware modules of the terminal device.
Alternatively, in another example, the control system 100 of the intelligent production line of electronic grade potassium hydroxide and the terminal device according to the embodiment of the present application may be separate devices, and the control system 100 of the intelligent production line of electronic grade potassium hydroxide may be connected to the terminal device through a wired and/or wireless network, and transmit the interactive information according to the agreed data format.
FIG. 7 is a flow chart of a method for controlling an intelligent production line of electronic grade potassium hydroxide according to an embodiment of the present application. As shown in fig. 7, a method for controlling an intelligent production line of electronic grade potassium hydroxide according to an embodiment of the present application includes: s110, acquiring ultrasonic signals of a preset time period generated by an ultrasonic generating device and temperature values of a plurality of preset time points in the preset time period; s120, carrying out frequency domain analysis based on Fourier transform on the ultrasonic signals to obtain a plurality of ultrasonic frequency domain characteristic values; s130, passing the plurality of ultrasonic frequency domain feature values through a multi-scale feature susceptor comprising a first convolution layer and a second convolution layer to obtain ultrasonic frequency domain feature vectors; s140, passing the temperature values of a plurality of preset time points in the preset time period through the multiscale feature receptors comprising the first convolution layer and the second convolution layer to obtain temperature time sequence feature vectors; s150, performing association coding on the ultrasonic frequency domain feature vector and the temperature time sequence feature vector to obtain a cooperative feature matrix; and S160, passing the collaborative feature matrix through a classifier to obtain a classification result, wherein the classification result is used for indicating that the frequency value of the ultrasonic signal at the current time point should be increased or decreased.
FIG. 8 is a schematic diagram of a system architecture of a method for intelligent production line control of electronic grade potassium hydroxide according to an embodiment of the present application. As shown in fig. 8, in the system architecture of the intelligent production line control method of electronic grade potassium hydroxide, firstly, an ultrasonic signal of a predetermined time period generated by an ultrasonic generating device and temperature values of a plurality of predetermined time points within the predetermined time period are acquired; then, carrying out frequency domain analysis based on Fourier transform on the ultrasonic signals to obtain a plurality of ultrasonic frequency domain characteristic values; then, the plurality of ultrasonic frequency domain feature values pass through a multi-scale feature susceptor comprising a first convolution layer and a second convolution layer to obtain ultrasonic frequency domain feature vectors; then, passing the temperature values at a plurality of preset time points in the preset time period through the multiscale feature receptors comprising the first convolution layer and the second convolution layer to obtain temperature time sequence feature vectors; then, performing association coding on the ultrasonic frequency domain feature vector and the temperature time sequence feature vector to obtain a cooperative feature matrix; finally, the collaborative feature matrix is passed through a classifier to obtain a classification result, wherein the classification result is used for representing that the frequency value of the ultrasonic signal at the current time point should be increased or decreased.
In a specific example, in the control method of the intelligent production line of electronic grade potassium hydroxide, the multi-scale feature susceptor comprises a first convolution layer and a second convolution layer which are parallel, and a multi-scale feature fusion layer connected with the first convolution layer and the second convolution layer, wherein the first convolution layer and the second convolution layer use one-dimensional convolution kernels with different scales.
In a specific example, in the control method of the intelligent production line of electronic grade potassium hydroxide, the step of passing the plurality of ultrasonic frequency domain eigenvalues through a multi-scale eigenvector comprising a first convolution layer and a second convolution layer to obtain ultrasonic frequency domain eigenvectors comprises: arranging the plurality of ultrasonic frequency domain feature values into ultrasonic input vectors; performing one-dimensional convolution encoding on the ultrasonic input vector by using a first convolution layer of the multi-scale feature susceptor to obtain a first-scale ultrasonic feature vector, wherein the first convolution layer has a first one-dimensional convolution kernel with a first length; performing one-dimensional convolution encoding on the ultrasonic input vector by using a second convolution layer of the multi-scale feature susceptor to obtain a second-scale ultrasonic feature vector, wherein the second convolution layer has a second one-dimensional convolution kernel with a second length, and the first length is different from the second length; and cascading the first-scale ultrasonic feature vector and the second-scale ultrasonic feature vector by using a multi-scale feature fusion layer of the multi-scale feature susceptor to obtain the ultrasonic frequency domain feature vector.
In a specific example, in the control method of the intelligent production line of electronic grade potassium hydroxide, passing the temperature values of a plurality of predetermined time points within the predetermined time period through the multi-scale feature susceptor comprising the first convolution layer and the second convolution layer to obtain the temperature time sequence feature vector comprises: arranging temperature values at a plurality of preset time points in the preset time period into a temperature input vector; performing one-dimensional convolution encoding on the temperature input vector by using a first convolution layer of the multi-scale feature susceptor to obtain a first-scale temperature feature vector, wherein the first convolution layer has a first one-dimensional convolution kernel with a first length; performing one-dimensional convolution encoding on the temperature input vector by using a second convolution layer of the multi-scale feature susceptor to obtain a second-scale temperature feature vector, wherein the second convolution layer has a second one-dimensional convolution kernel with a second length, and the first length is different from the second length; and cascading the first-scale temperature feature vector and the second-scale temperature feature vector by using a multi-scale feature fusion layer of the multi-scale feature susceptor to obtain the temperature frequency domain feature vector.
In a specific example, in the control method of the intelligent production line of electronic grade potassium hydroxide, performing association coding on the ultrasonic frequency domain feature vector and the temperature time sequence feature vector to obtain a cooperative feature matrix, including: calculating Helmholtz free energy factors of the ultrasonic frequency domain feature vector and the temperature time sequence feature vector to obtain a first Helmholtz free energy factor and a second Helmholtz free energy factor; the ultrasonic frequency domain feature vector and the temperature time sequence feature vector are weighted by taking the first Helmholtz class free energy factor and the second Helmholtz class free energy factor as weighting weights so as to obtain an optimized ultrasonic frequency domain feature vector and an optimized temperature time sequence feature vector; and performing association coding on the optimized ultrasonic frequency domain feature vector and the optimized temperature time sequence feature vector to obtain the collaborative feature matrix.
In a specific example, in the control method of the intelligent production line of electronic grade potassium hydroxide, calculating the helmholtz free energy factor of the ultrasonic frequency domain feature vector and the temperature time series feature vector to obtain a first helmholtz free energy factor and a second helmholtz free energy factor includes: calculating Helmholtz free energy factors of the ultrasonic frequency domain feature vector and the temperature time sequence feature vector according to the following optimization formula to obtain a first Helmholtz free energy factor and a second Helmholtz free energy factor; wherein, the optimization formula is:
Figure SMS_59
Figure SMS_62
Wherein (1)>
Figure SMS_66
Characteristic values representing respective positions in the ultrasonic frequency domain characteristic vector,/or->
Figure SMS_60
Characteristic values representing respective positions in the temperature timing characteristic vector, +.>
Figure SMS_63
And->
Figure SMS_65
Classification probability values respectively representing the ultrasonic frequency domain feature vector and the temperature time series feature vector, and +.>
Figure SMS_68
Is the length of the feature vector and,
Figure SMS_58
represents a logarithmic function with base 2, +.>
Figure SMS_61
Representing an exponential operation, ++>
Figure SMS_64
And->
Figure SMS_67
Representing the first helmholtz-like free energy factor and the second helmholtz-like free energy factor, respectively.
In a specific example, in the control method of the intelligent production line of electronic grade potassium hydroxide, performing association coding on the optimized ultrasonic frequency domain feature vector and the optimized temperature time sequence feature vector to obtain the collaborative feature matrix, the method includes: performing association coding on the optimized ultrasonic frequency domain feature vector and the optimized temperature time sequence feature vector by using the following association coding formula to obtain the collaborative feature matrix; wherein, the association coding formula is:
Figure SMS_69
wherein (1)>
Figure SMS_70
Representing the optimized ultrasonic frequency domain feature vector, < >>
Figure SMS_71
A transpose vector representing the optimized ultrasonic frequency domain feature vector, >
Figure SMS_72
Representing the optimized temperature time sequence feature vector, < >>
Figure SMS_73
Representing the synergistic feature matrix,>
Figure SMS_74
representing vector multiplication.
In a specific example, in the control method of the intelligent production line of electronic grade potassium hydroxide, the method for passing the collaborative feature matrix through a classifier to obtain a classification result, wherein the classification result is used for indicating that the frequency value of the ultrasonic signal at the current time point should be increased or decreased, and includes: expanding the collaborative feature matrix into classification feature vectors according to row vectors or column vectors; performing full-connection coding on the classification feature vectors by using a plurality of full-connection layers of the classifier to obtain coded classification feature vectors; and passing the coding classification feature vector through a Softmax classification function of the classifier to obtain the classification result.
Here, it will be understood by those skilled in the art that the specific operations of the respective steps in the above-described control method of the intelligent production line of electronic grade potassium hydroxide have been described in detail in the above description of the control system 100 of the intelligent production line of electronic grade potassium hydroxide with reference to fig. 1 to 6, and thus, repetitive descriptions thereof will be omitted.
According to another aspect of the present application, there is also provided a non-volatile computer-readable storage medium having stored thereon computer-readable instructions which, when executed by a computer, can perform a method as described above.
Program portions of the technology may be considered to be "products" or "articles of manufacture" in the form of executable code and/or associated data, embodied or carried out by a computer readable medium. A tangible, persistent storage medium may include any memory or storage used by a computer, processor, or similar device or related module. Such as various semiconductor memories, tape drives, disk drives, or the like, capable of providing storage functionality for software.
All or a portion of the software may sometimes communicate over a network, such as the internet or other communication network. Such communication may load software from one computer device or processor to another. For example: a hardware platform loaded from a server or host computer of the video object detection device to a computer environment, or other computer environment implementing the system, or similar functioning system related to providing information needed for object detection. Thus, another medium capable of carrying software elements may also be used as a physical connection between local devices, such as optical, electrical, electromagnetic, etc., propagating through cable, optical cable, air, etc. Physical media used for carrier waves, such as electrical, wireless, or optical, may also be considered to be software-bearing media. Unless limited to a tangible "storage" medium, other terms used herein to refer to a computer or machine "readable medium" mean any medium that participates in the execution of any instructions by a processor.
This application uses specific words to describe embodiments of the application. Reference to "a first/second embodiment," "an embodiment," and/or "some embodiments" means that a particular feature, structure, or characteristic is associated with at least one embodiment of the present application. Thus, it should be emphasized and should be appreciated that two or more references to "an embodiment" or "one embodiment" or "an alternative embodiment" in various positions in this specification are not necessarily referring to the same embodiment. Furthermore, certain features, structures, or characteristics of one or more embodiments of the present application may be combined as suitable.
Furthermore, those skilled in the art will appreciate that the various aspects of the invention are illustrated and described in the context of a number of patentable categories or circumstances, including any novel and useful procedures, machines, products, or materials, or any novel and useful modifications thereof. Accordingly, aspects of the present application may be performed entirely by hardware, entirely by software (including firmware, resident software, micro-code, etc.) or by a combination of hardware and software. The above hardware or software may be referred to as a "data block," module, "" engine, "" unit, "" component, "or" system. Furthermore, aspects of the present application may take the form of a computer product, comprising computer-readable program code, embodied in one or more computer-readable media.
Unless otherwise defined, all terms (including 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. It will be further understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the relevant art and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein.
The foregoing is illustrative of the present invention and is not to be construed as limiting thereof. Although a few exemplary embodiments of this invention have been described, those skilled in the art will readily appreciate that many modifications are possible in the exemplary embodiments without materially departing from the novel teachings and advantages of this invention. Accordingly, all such modifications are intended to be included within the scope of this invention as defined in the following claims. It is to be understood that the foregoing is illustrative of the present invention and is not to be construed as limited to the specific embodiments disclosed, and that modifications to the disclosed embodiments, as well as other embodiments, are intended to be included within the scope of the appended claims. The invention is defined by the claims and their equivalents.

Claims (8)

1. An intelligent production line control system for electronic grade potassium hydroxide, which is characterized by comprising: the data acquisition module is used for acquiring ultrasonic signals of a preset time period generated by the ultrasonic generating device and temperature values of a plurality of preset time points in the preset time period; the frequency domain analysis module is used for carrying out frequency domain analysis based on Fourier transform on the ultrasonic signals so as to obtain a plurality of ultrasonic frequency domain characteristic values; the ultrasonic wave feature extraction module is used for enabling the plurality of ultrasonic wave frequency domain feature values to pass through a multi-scale feature susceptor comprising a first convolution layer and a second convolution layer so as to obtain ultrasonic wave frequency domain feature vectors; the temperature characteristic extraction module is used for passing the temperature values of a plurality of preset time points in the preset time period through the multi-scale characteristic receptor comprising the first convolution layer and the second convolution layer to obtain a temperature time sequence characteristic vector; the characteristic association coding module is used for carrying out association coding on the ultrasonic frequency domain characteristic vector and the temperature time sequence characteristic vector so as to obtain a cooperative characteristic matrix; and the ultrasonic frequency control module is used for enabling the collaborative feature matrix to pass through a classifier to obtain a classification result, wherein the classification result is used for indicating that the frequency value of an ultrasonic signal at the current time point should be increased or decreased.
2. The intelligent production line control system of electronic grade potassium hydroxide according to claim 1, wherein the multi-scale feature susceptor comprises a first convolution layer and a second convolution layer in parallel, and a multi-scale feature fusion layer coupled to the first convolution layer and the second convolution layer, wherein the first convolution layer and the second convolution layer use one-dimensional convolution kernels having different scales.
3. The intelligent production line control system of electronic grade potassium hydroxide according to claim 2, wherein the ultrasonic feature extraction module comprises: an ultrasonic input vector arrangement unit for arranging the plurality of ultrasonic frequency domain feature values into ultrasonic input vectors; a first scale ultrasonic feature extraction unit, configured to perform one-dimensional convolution encoding on the ultrasonic input vector by using a first convolution layer of the multi-scale feature sensor to obtain a first scale ultrasonic feature vector, where the first convolution layer has a first one-dimensional convolution kernel with a first length; a second-scale ultrasonic feature extraction unit, configured to perform one-dimensional convolution encoding on the ultrasonic input vector using a second convolution layer of the multi-scale feature susceptor to obtain a second-scale ultrasonic feature vector, where the second convolution layer has a second one-dimensional convolution kernel with a second length, and the first length is different from the second length; and the ultrasonic multi-scale fusion unit is used for cascading the first-scale ultrasonic feature vector and the second-scale ultrasonic feature vector by using the multi-scale feature fusion layer of the multi-scale feature receptor so as to obtain the ultrasonic frequency domain feature vector.
4. A control system for an intelligent production line of electronic grade potassium hydroxide according to claim 3, wherein the temperature characteristic extraction module comprises: a temperature input vector arrangement unit configured to arrange temperature values at a plurality of predetermined time points within the predetermined period of time as a temperature input vector; a first scale temperature feature extraction unit, configured to perform one-dimensional convolutional encoding on the temperature input vector by using a first convolutional layer of the multi-scale feature susceptor to obtain a first scale temperature feature vector, where the first convolutional layer has a first one-dimensional convolution kernel with a first length; a second scale temperature feature extraction unit, configured to perform one-dimensional convolution encoding on the temperature input vector using a second convolution layer of the multi-scale feature susceptor to obtain a second scale temperature feature vector, where the second convolution layer has a second one-dimensional convolution kernel with a second length, and the first length is different from the second length; and the temperature multi-scale fusion unit is used for cascading the first-scale temperature characteristic vector and the second-scale temperature characteristic vector by using the multi-scale characteristic fusion layer of the multi-scale characteristic receptor so as to obtain the temperature frequency domain characteristic vector.
5. The intelligent production line control system of electronic grade potassium hydroxide according to claim 4, wherein the feature-related encoding module comprises: the optimization factor calculation unit is used for calculating the Helmholtz class free energy factors of the ultrasonic frequency domain feature vector and the temperature time sequence feature vector to obtain a first Helmholtz class free energy factor and a second Helmholtz class free energy factor; the weighting optimization unit is used for weighting the ultrasonic frequency domain feature vector and the temperature time sequence feature vector by taking the first Helmholtz class free energy factor and the second Helmholtz class free energy factor as weighting weights so as to obtain an optimized ultrasonic frequency domain feature vector and an optimized temperature time sequence feature vector; and the optimized association characteristic fusion unit is used for carrying out association coding on the optimized ultrasonic frequency domain characteristic vector and the optimized temperature time sequence characteristic vector so as to obtain the collaborative characteristic matrix.
6. The intelligent production line control system of electronic grade potassium hydroxide according to claim 5, wherein the optimization factor calculation subunit is configured to: calculating Helmholtz free energy factors of the ultrasonic frequency domain feature vector and the temperature time sequence feature vector according to the following optimization formula to obtain a first Helmholtz free energy factor and a second Helmholtz free energy factor; wherein, the optimization formula is:
Figure QLYQS_2
Figure QLYQS_6
Wherein (1)>
Figure QLYQS_9
Characteristic values representing respective positions in the ultrasonic frequency domain characteristic vector,/or->
Figure QLYQS_3
Characteristic values representing respective positions in the temperature timing characteristic vector, +.>
Figure QLYQS_4
And->
Figure QLYQS_8
Respectively represent the ultrasonic frequency domain feature vector sumClassification probability value of the temperature timing feature vector, and +.>
Figure QLYQS_11
Is the length of the feature vector, +.>
Figure QLYQS_1
Represents a logarithmic function with base 2, +.>
Figure QLYQS_5
Representing an exponential operation, ++>
Figure QLYQS_7
And->
Figure QLYQS_10
Representing the first helmholtz-like free energy factor and the second helmholtz-like free energy factor, respectively.
7. The intelligent production line control system of electronic grade potassium hydroxide according to claim 6, wherein the optimization association feature fusion unit is configured to: performing association coding on the optimized ultrasonic frequency domain feature vector and the optimized temperature time sequence feature vector by using the following association coding formula to obtain the collaborative feature matrix; wherein, the association coding formula is:
Figure QLYQS_12
wherein (1)>
Figure QLYQS_13
Representing the optimized ultrasonic frequency domain feature vector, < >>
Figure QLYQS_14
A transpose vector representing the optimized ultrasonic frequency domain feature vector,>
Figure QLYQS_15
representing the optimized temperatureTiming feature vector,/->
Figure QLYQS_16
Representing the synergistic feature matrix, >
Figure QLYQS_17
Representing vector multiplication.
8. The intelligent production line control system of electronic grade potassium hydroxide according to claim 7, wherein the ultrasonic frequency control module comprises: the matrix unfolding unit is used for unfolding the collaborative feature matrix into a classification feature vector according to a row vector or a column vector; the full-connection coding unit is used for carrying out full-connection coding on the classification characteristic vectors by using a plurality of full-connection layers of the classifier so as to obtain coded classification characteristic vectors; and the classification unit is used for passing the coding classification feature vector through a Softmax classification function of the classifier to obtain the classification result.
CN202310307002.4A 2023-03-27 2023-03-27 Intelligent production line control system of electronic-grade potassium hydroxide Withdrawn CN116165987A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310307002.4A CN116165987A (en) 2023-03-27 2023-03-27 Intelligent production line control system of electronic-grade potassium hydroxide

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310307002.4A CN116165987A (en) 2023-03-27 2023-03-27 Intelligent production line control system of electronic-grade potassium hydroxide

Publications (1)

Publication Number Publication Date
CN116165987A true CN116165987A (en) 2023-05-26

Family

ID=86416507

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310307002.4A Withdrawn CN116165987A (en) 2023-03-27 2023-03-27 Intelligent production line control system of electronic-grade potassium hydroxide

Country Status (1)

Country Link
CN (1) CN116165987A (en)

Similar Documents

Publication Publication Date Title
Hatami et al. Classification of time-series images using deep convolutional neural networks
Yuan et al. Tensor ring decomposition with rank minimization on latent space: An efficient approach for tensor completion
CN116036639B (en) Control system and method for ultrasonic wave induced hexafluorophosphate crystallization
Du et al. Recognition of concurrent control chart patterns using wavelet transform decomposition and multiclass support vector machines
Zhang et al. Imbalanced data classification based on scaling kernel-based support vector machine
CN115845428B (en) Ultrasonic device for inducing crystallization of hexafluorophosphoric acid and control method thereof
WO2020010301A1 (en) Delimitation in unsupervised classification of gestures
CN116164497A (en) Rectification control system and method in liquid oxygen preparation process
US20230083437A1 (en) Hyperdimensional learning using variational autoencoder
CN110991500A (en) Small sample multi-classification method based on nested integrated depth support vector machine
CN117308077A (en) Water supply control system for evaporator of nuclear power unit of reactor
CN116597198A (en) Preparation method of energy-saving environment-friendly ceramic marmite
CN116385945A (en) Video interaction action detection method and system based on random frame complement and attention
EP4328867A1 (en) Percentile-based pseudo-label selection for multi-label semi-supervised classification
CN116165987A (en) Intelligent production line control system of electronic-grade potassium hydroxide
CN116059681A (en) Ultrasonic control method and system for controlling crystallization nucleation
CN116166072A (en) Electronic grade hexafluoropropylene cracking temperature control system
CN116177858A (en) Preparation method and system of high-purity quartz crucible
Hu et al. Siamese network object tracking algorithm combining attention mechanism and correlation filter theory
CN117999560A (en) Hardware-aware progressive training of machine learning models
CN116232699A (en) Training method of fine-grained network intrusion detection model and network intrusion detection method
Deshpande et al. Detection of Plant Leaf Disease by Generative Adversarial and Deep Convolutional Neural Network
US20220044113A1 (en) Asynchronous Neural Network Systems
Wu et al. Time Series Classification Based on Adaptive Feature Adjustment and Multi-scale AGRes2Net
Trentin et al. Unsupervised nonparametric density estimation: A neural network approach

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
WW01 Invention patent application withdrawn after publication
WW01 Invention patent application withdrawn after publication

Application publication date: 20230526