CN115259089B - Production management control system for preparing electronic grade hydrofluoric acid and control method thereof - Google Patents
Production management control system for preparing electronic grade hydrofluoric acid and control method thereof Download PDFInfo
- Publication number
- CN115259089B CN115259089B CN202210767053.0A CN202210767053A CN115259089B CN 115259089 B CN115259089 B CN 115259089B CN 202210767053 A CN202210767053 A CN 202210767053A CN 115259089 B CN115259089 B CN 115259089B
- Authority
- CN
- China
- Prior art keywords
- tower
- rectifying tower
- feature vector
- parameter
- rectifying
- 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
- KRHYYFGTRYWZRS-UHFFFAOYSA-N Fluorane Chemical compound F KRHYYFGTRYWZRS-UHFFFAOYSA-N 0.000 title claims abstract description 277
- 238000004519 manufacturing process Methods 0.000 title claims abstract description 118
- 238000000034 method Methods 0.000 title claims abstract description 53
- 238000002360 preparation method Methods 0.000 claims abstract description 31
- 239000013598 vector Substances 0.000 claims description 275
- 238000010521 absorption reaction Methods 0.000 claims description 179
- 239000007788 liquid Substances 0.000 claims description 105
- 239000011159 matrix material Substances 0.000 claims description 82
- 238000013527 convolutional neural network Methods 0.000 claims description 64
- 238000012545 processing Methods 0.000 claims description 54
- 229910000040 hydrogen fluoride Inorganic materials 0.000 claims description 53
- 238000001914 filtration Methods 0.000 claims description 45
- 230000004043 responsiveness Effects 0.000 claims description 34
- 238000010992 reflux Methods 0.000 claims description 32
- 238000011176 pooling Methods 0.000 claims description 24
- 238000013528 artificial neural network Methods 0.000 claims description 23
- 230000004913 activation Effects 0.000 claims description 14
- 230000004927 fusion Effects 0.000 claims description 13
- 230000002195 synergetic effect Effects 0.000 claims description 13
- 230000036760 body temperature Effects 0.000 claims description 11
- 238000010276 construction Methods 0.000 claims description 4
- 238000012546 transfer Methods 0.000 claims description 2
- 238000004821 distillation Methods 0.000 claims 1
- 238000012544 monitoring process Methods 0.000 abstract description 14
- 239000012535 impurity Substances 0.000 description 13
- 230000008569 process Effects 0.000 description 13
- 229910052785 arsenic Inorganic materials 0.000 description 9
- RQNWIZPPADIBDY-UHFFFAOYSA-N arsenic atom Chemical compound [As] RQNWIZPPADIBDY-UHFFFAOYSA-N 0.000 description 9
- 230000008859 change Effects 0.000 description 9
- 238000010586 diagram Methods 0.000 description 9
- 239000007789 gas Substances 0.000 description 9
- 238000003860 storage Methods 0.000 description 8
- 230000005540 biological transmission Effects 0.000 description 6
- 238000004590 computer program Methods 0.000 description 6
- 238000000605 extraction Methods 0.000 description 6
- 239000007800 oxidant agent Substances 0.000 description 6
- 230000001590 oxidative effect Effects 0.000 description 6
- 230000004044 response Effects 0.000 description 6
- MHAJPDPJQMAIIY-UHFFFAOYSA-N Hydrogen peroxide Chemical compound OO MHAJPDPJQMAIIY-UHFFFAOYSA-N 0.000 description 5
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 5
- 150000001495 arsenic compounds Chemical class 0.000 description 4
- 238000013473 artificial intelligence Methods 0.000 description 4
- 238000009835 boiling Methods 0.000 description 4
- 238000009826 distribution Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 229940093920 gynecological arsenic compound Drugs 0.000 description 4
- 238000005065 mining Methods 0.000 description 3
- 238000003062 neural network model Methods 0.000 description 3
- 239000002245 particle Substances 0.000 description 3
- 239000003518 caustics Substances 0.000 description 2
- 239000012459 cleaning agent Substances 0.000 description 2
- 238000012937 correction Methods 0.000 description 2
- 238000011049 filling Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- JRKICGRDRMAZLK-UHFFFAOYSA-L peroxydisulfate Chemical compound [O-]S(=O)(=O)OOS([O-])(=O)=O JRKICGRDRMAZLK-UHFFFAOYSA-L 0.000 description 2
- 239000012286 potassium permanganate Substances 0.000 description 2
- 230000001737 promoting effect Effects 0.000 description 2
- 238000000746 purification Methods 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 239000000126 substance Substances 0.000 description 2
- 102100033620 Calponin-1 Human genes 0.000 description 1
- 102100033591 Calponin-2 Human genes 0.000 description 1
- 102100033592 Calponin-3 Human genes 0.000 description 1
- 101000945318 Homo sapiens Calponin-1 Proteins 0.000 description 1
- 101000945403 Homo sapiens Calponin-2 Proteins 0.000 description 1
- 101000945410 Homo sapiens Calponin-3 Proteins 0.000 description 1
- XUIMIQQOPSSXEZ-UHFFFAOYSA-N Silicon Chemical compound [Si] XUIMIQQOPSSXEZ-UHFFFAOYSA-N 0.000 description 1
- 239000002253 acid Substances 0.000 description 1
- 230000002378 acidificating effect Effects 0.000 description 1
- 238000007792 addition Methods 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000009924 canning Methods 0.000 description 1
- HOPSCVCBEOCPJZ-UHFFFAOYSA-N carboxymethyl(trimethyl)azanium;chloride Chemical compound [Cl-].C[N+](C)(C)CC(O)=O HOPSCVCBEOCPJZ-UHFFFAOYSA-N 0.000 description 1
- 238000009833 condensation Methods 0.000 description 1
- 230000005494 condensation Effects 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 239000007791 liquid phase Substances 0.000 description 1
- 230000000116 mitigating effect Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 230000003647 oxidation Effects 0.000 description 1
- 238000007254 oxidation reaction Methods 0.000 description 1
- 230000006798 recombination Effects 0.000 description 1
- 238000005215 recombination Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 229910052710 silicon Inorganic materials 0.000 description 1
- 239000010703 silicon Substances 0.000 description 1
Images
Classifications
-
- C—CHEMISTRY; METALLURGY
- C01—INORGANIC CHEMISTRY
- C01B—NON-METALLIC ELEMENTS; COMPOUNDS THEREOF; METALLOIDS OR COMPOUNDS THEREOF NOT COVERED BY SUBCLASS C01C
- C01B7/00—Halogens; Halogen acids
- C01B7/19—Fluorine; Hydrogen fluoride
- C01B7/191—Hydrogen fluoride
-
- C—CHEMISTRY; METALLURGY
- C01—INORGANIC CHEMISTRY
- C01B—NON-METALLIC ELEMENTS; COMPOUNDS THEREOF; METALLOIDS OR COMPOUNDS THEREOF NOT COVERED BY SUBCLASS C01C
- C01B7/00—Halogens; Halogen acids
- C01B7/19—Fluorine; Hydrogen fluoride
- C01B7/191—Hydrogen fluoride
- C01B7/195—Separation; Purification
-
- C—CHEMISTRY; METALLURGY
- C01—INORGANIC CHEMISTRY
- C01B—NON-METALLIC ELEMENTS; COMPOUNDS THEREOF; METALLOIDS OR COMPOUNDS THEREOF NOT COVERED BY SUBCLASS C01C
- C01B7/00—Halogens; Halogen acids
- C01B7/19—Fluorine; Hydrogen fluoride
- C01B7/191—Hydrogen fluoride
- C01B7/195—Separation; Purification
- C01B7/196—Separation; Purification by distillation
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B19/00—Programme-control systems
- G05B19/02—Programme-control systems electric
- G05B19/418—Total 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]
- G05B19/41885—Total 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Landscapes
- Chemical & Material Sciences (AREA)
- Organic Chemistry (AREA)
- Inorganic Chemistry (AREA)
- 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)
- Organic Low-Molecular-Weight Compounds And Preparation Thereof (AREA)
- Feedback Control In General (AREA)
Abstract
The application relates to the field of intelligent monitoring of equipment, and particularly discloses a production management control system for electronic-grade hydrofluoric acid preparation and a control method thereof.
Description
Technical Field
The invention relates to the field of intelligent monitoring of equipment, in particular to a production management control system for electronic-grade hydrofluoric acid preparation and a control method thereof.
Background
The electronic grade hydrofluoric acid is a strong acid cleaning agent and a corrosive agent and is mainly used for producing the super-large scale integrated circuit. At present, the main production method of electronic grade hydrofluoric acid comprises the following steps: firstly, chemical pretreatment is carried out on industrial anhydrous hydrofluoric acid, then rectification is carried out, the obtained hydrogen fluoride gas is cooled and absorbed by pure water, and finally filtration and filling are carried out. Because the existence of arsenic as an impurity has a serious influence on the performance of electronic devices, the removal of arsenic is a key problem in the purification process of hydrofluoric acid, and a commonly used method is to convert trivalent arsenic impurities into pentavalent arsenic compounds with high boiling points by using an oxidant, wherein the commonly used oxidant is KMnO4, crO3, persulfate and the like.
Patent 103991847 discloses a method for preparing electronic-grade hydrofluoric acid, which uses hydrogen peroxide as an oxidant to convert trivalent arsenic impurities into pentavalent arsenic compounds with high boiling points, and does not introduce additional impurities, thereby preparing electronic-grade hydrofluoric acid products with high yield, good quality and low cost, and recovering hydrogen fluoride gas in tail gas.
However, in the actual preparation process, it is found that although the technical solution disclosed in patent 103991847 can produce electronic-grade hydrofluoric acid products with higher purity, since the production line for producing electronic-grade hydrofluoric acid includes a plurality of equipments, once a certain equipment fails or the performance of a certain equipment is reduced, the production yield of the final electronic-grade hydrofluoric acid product is affected.
Therefore, a plant production management control system for electronic grade hydrofluoric acid production is desired to perform on-line monitoring of the performance of each plant in the production line.
Disclosure of Invention
The present application is proposed to solve the above-mentioned technical problems. The embodiment of the application provides a production management control system for electronic-grade hydrofluoric acid preparation and a control method thereof, which adopt a monitoring technology based on artificial intelligence, ensure the preparation yield of final electronic-grade hydrofluoric acid products by monitoring various parameters of a rectifying tower and an absorption tower in the electronic-grade hydrofluoric acid preparation process, and in the process, use the dynamic change characteristics of the products to perform responsiveness estimation so as to more accurately monitor whether the equipment performance of the rectifying tower and the absorption tower is normal or not.
According to one aspect of the present application, there is provided a production management control system for electronic grade hydrofluoric acid production comprising: the device comprises an equipment parameter acquisition module, a data processing module and a data processing module, wherein the equipment parameter acquisition module is used for acquiring a plurality of first parameters of a rectifying tower and a plurality of second parameters of an absorption tower at a plurality of preset time points, the plurality of first parameters comprise the tower kettle temperature of the rectifying tower, the tower body temperature of the rectifying tower, the tower top temperature of the rectifying tower, the tower kettle pressure of the rectifying tower, the tower top pressure of the rectifying tower, the tower in-tower reflux temperature and reflux ratio of the rectifying tower, and the plurality of second parameters comprise the temperature of the absorption tower and the pressure of the absorption tower; the product state acquisition module is used for acquiring a plurality of liquid chromatograms of the hydrogen fluoride liquid at the plurality of preset time points before first filtration through a liquid chromatograph; a product tracking encoding module, configured to pass a plurality of liquid chromatograms of the hydrogen fluoride liquid at the plurality of predetermined time points through a first convolution neural network using a three-dimensional convolution kernel to obtain a product tracking feature vector; the rectifying tower parameter coding module is used for arranging a plurality of items of first parameters of the rectifying tower at a plurality of preset time points into a rectifying tower parameter input matrix according to time dimension and parameter sample dimension and then obtaining a rectifying tower characteristic vector through a second convolutional neural network serving as a filter; the absorption tower parameter coding module is used for arranging a plurality of second parameters of the absorption towers at the plurality of preset time points into an absorption tower parameter input matrix according to the time dimension and the parameter sample dimension and then obtaining an absorption tower characteristic vector through a third convolutional neural network serving as a filter; the parameter characteristic fusion module is used for fusing the rectifying tower characteristic vector and the absorption tower characteristic vector to obtain a cooperative characteristic vector; the responsiveness estimation module is used for calculating the responsiveness estimation of the product tracking feature vector relative to the cooperative feature vector to obtain a classification feature vector; and the equipment management result generation module is used for enabling the classification characteristic vectors to pass through a classifier to obtain a classification result, and the classification result is used for indicating whether the equipment performance of the rectifying tower and the absorption tower is normal or not.
In the above production management control system for electronic grade hydrofluoric acid preparation, the product tracking coding module comprises: a feature dynamic capturing unit, configured to perform convolution processing, pooling processing and nonlinear activation processing based on the three-dimensional convolution kernel on input data respectively during forward transmission of layers by using each layer of the first convolutional neural network to output a tracking feature map from a last layer of the first convolutional neural network, where an input of the first convolutional neural network is a plurality of liquid chromatogram of the hydrogen fluoride liquid before first filtering at the plurality of predetermined time points; and the global pooling unit is used for performing global mean pooling based on a feature matrix on the tracking feature map to obtain the tracking feature vector.
In the production management control system for preparing electronic grade hydrofluoric acid, the rectifying tower parameter coding module comprises: the first matrix construction unit is used for arranging a plurality of items of first parameters of the rectifying tower at a plurality of preset time points into a plurality of items of first parameter row vectors corresponding to the rectifying tower according to the time dimension, and arranging the plurality of items of first parameter row vectors into a parameter input matrix of the rectifying tower according to the parameter sample dimension; a first filter unit for performing convolution processing, pooling processing along a feature matrix, and activation processing on input data in forward pass of layers using layers of the second convolutional neural network to generate the rectifying tower feature vector from a last layer of the second convolutional neural network, wherein an input of a first layer of the second convolutional neural network is the rectifying tower parameter input matrix.
In the above production management control system for electronic-grade hydrofluoric acid preparation, the absorption tower parameter coding module comprises: the second matrix construction unit is used for arranging a plurality of items of second parameters of the absorption towers at the plurality of preset time points into a plurality of items of second parameter row vectors corresponding to the absorption towers according to the time dimension, and arranging the plurality of items of second parameter row vectors into the parameter input matrix of the absorption towers according to the parameter sample dimension; a second filter unit for performing convolution processing, pooling processing along feature matrices, and activation processing on input data in forward pass of layers using layers of the third convolutional neural network to generate the absorption tower feature vector from a last layer of the third convolutional neural network, wherein an input of a first layer of the third convolutional neural network is the absorption tower parameter input matrix.
In the production management control system for preparing the electronic-grade hydrofluoric acid, the parameter characteristic fusion module is configured to calculate a position-weighted sum of the rectification tower characteristic vector and the absorption tower characteristic vector to obtain the synergistic characteristic vector.
In the production management control system for preparing the electronic grade hydrofluoric acid, the weighting weight of the characteristic vector of the rectifying tower is calculated by the following formula:
wherein,characteristic values representing respective positions of the rectifying column characteristic vector,and the weighting coefficient represents the characteristic vector of the rectifying tower when the characteristic vector of the rectifying tower and the characteristic vector of the absorption tower are fused.
In the production management control system for electronic grade hydrofluoric acid preparation, the weighted weight of the absorption tower eigenvector is calculated by the following formula:
wherein,an eigenvalue representing each position of the absorption tower eigenvector,and the weighting coefficient represents the characteristic vector of the absorption tower when the characteristic vector of the rectification tower and the characteristic vector of the absorption tower are fused.
In the production management control system for electronic-grade hydrofluoric acid preparation, the responsiveness estimation module is further configured to calculate a responsiveness estimation of the product tracking feature vector relative to the collaborative feature vector by the following formula to obtain the classification feature vector; wherein the formula is:
whereinRepresenting the product tracking feature vector and the product tracking feature vector,representing the co-occurrence feature vector in the image,representing the classified feature vector in a manner that the classified feature vector,representing a dot-product of a vector,the representation takes the inverse of the value for each position of the vector.
In the production management control system for preparing electronic grade hydrofluoric acid, the equipment management result generating module is further configured to: processing the classification feature vector using the classifier to obtain the classification result with a formula:whereintoIn order to be a weight matrix, the weight matrix,toIn order to be a vector of the offset,is the classification feature vector.
According to another aspect of the present application, a method of controlling a production management control system for electronic grade hydrofluoric acid production comprises: acquiring a plurality of first parameters of a rectifying tower and a plurality of second parameters of an absorption tower at a plurality of preset time points, wherein the plurality of first parameters comprise the tower kettle temperature of the rectifying tower, the body temperature of the rectifying tower, the tower top temperature of the rectifying tower, the tower kettle pressure of the rectifying tower, the tower top pressure of the rectifying tower, the tower internal reflux temperature and reflux ratio of the rectifying tower, and the plurality of second parameters comprise the temperature of the absorption tower and the pressure of the absorption tower; obtaining a plurality of liquid chromatograms of the hydrogen fluoride liquid before the first filtration at the plurality of predetermined time points by a liquid chromatograph; passing a plurality of liquid chromatograms of the hydrogen fluoride liquid at the plurality of predetermined time points before first filtering through a first convolution neural network using a three-dimensional convolution kernel to obtain a product tracking feature vector; arranging a plurality of items of first parameters of the rectifying tower at a plurality of preset time points into a rectifying tower parameter input matrix according to a time dimension and a parameter sample dimension, and then passing through a second convolution neural network serving as a filter to obtain a rectifying tower characteristic vector; a plurality of second parameters of the absorption tower at the plurality of preset time points are arrayed into an absorption tower parameter input matrix according to time dimensions and parameter sample dimensions, and then a third convolution neural network serving as a filter is used for obtaining an absorption tower characteristic vector; fusing the rectifying tower characteristic vector and the absorption tower characteristic vector to obtain a synergistic characteristic vector; calculating a responsiveness estimate of the product tracking feature vector relative to the collaborative feature vector to obtain a classification feature vector; and passing the classification characteristic vector through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the equipment performance of the rectifying tower and the absorption tower is normal or not.
In the above control method of the production management control system for electronic grade hydrofluoric acid preparation, acquiring a plurality of liquid chromatograms of the hydrogen fluoride liquid before first filtration at the plurality of predetermined time points by a liquid chromatograph includes: performing convolution processing, pooling processing and nonlinear activation processing based on the three-dimensional convolution kernel on input data respectively in the forward transmission process of layers by using each layer of the first convolution neural network so as to output a tracking characteristic map from the last layer of the first convolution neural network, wherein the input of the first convolution neural network is a plurality of liquid chromatogram maps of the hydrogen fluoride liquid before first filtration at a plurality of preset time points; and performing global mean pooling based on a feature matrix on the tracking feature map to obtain the tracking feature vector.
In the above control method for a production management control system for electronic-grade hydrofluoric acid preparation, after arranging a plurality of first parameters of a rectifying tower at a plurality of predetermined time points into a rectifying tower parameter input matrix according to a time dimension and a parameter sample dimension, obtaining a rectifying tower feature vector through a second convolutional neural network serving as a filter, the method includes: arranging a plurality of items of first parameters of the rectifying tower at a plurality of preset time points into a plurality of items of first parameter row vectors corresponding to the rectifying tower according to a time dimension, and arranging the plurality of items of first parameter row vectors into a rectifying tower parameter input matrix according to a parameter sample dimension; a first filter unit for performing convolution processing, pooling processing along a feature matrix, and activation processing on input data in forward pass of layers using layers of the second convolutional neural network to generate the rectifying column feature vector from a last layer of the second convolutional neural network, wherein an input of a first layer of the second convolutional neural network is the rectifying column parameter input matrix.
In the above control method for a production management control system for electronic-grade hydrofluoric acid preparation, after arranging a plurality of second parameters of the absorption tower at a plurality of predetermined time points into an absorption tower parameter input matrix according to a time dimension and a parameter sample dimension, obtaining an absorption tower feature vector through a third convolutional neural network serving as a filter, the method includes: arranging a plurality of items of second parameters of the absorption towers at the preset time points into a plurality of items of second parameter row vectors corresponding to the absorption towers according to time dimension, and arranging the plurality of items of second parameter row vectors into the parameter input matrix of the absorption towers according to parameter sample dimension; performing convolution processing, pooling processing along a feature matrix, and activation processing on input data in forward pass of layers using layers of the third convolutional neural network to generate the absorption tower feature vector from a last layer of the third convolutional neural network, wherein an input of a first layer of the third convolutional neural network is the absorption tower parameter input matrix.
In the above control method of the production management control system for preparing electronic grade hydrofluoric acid, fusing the rectification tower feature vector and the absorption tower feature vector to obtain a synergistic feature vector, the method includes: and calculating the position-weighted sum of the rectifying tower characteristic vector and the absorption tower characteristic vector to obtain the synergic characteristic vector.
In the above control method for the production management control system for electronic grade hydrofluoric acid preparation, the weighting weight of the rectifying tower eigenvector is calculated by the following formula:
wherein,characteristic values representing respective positions of the rectifying tower characteristic vector,and the weighting coefficient represents the characteristic vector of the rectifying tower when the characteristic vector of the rectifying tower and the characteristic vector of the absorption tower are fused.
In the control method of the production management control system for preparing the electronic grade hydrofluoric acid, the weighted weight of the characteristic vector of the absorption tower is calculated by the following formula:
wherein,an eigenvalue representing each position of the absorption tower eigenvector,and the weighting coefficient represents the characteristic vector of the absorption tower when the characteristic vector of the rectification tower and the characteristic vector of the absorption tower are fused.
In the above control method for a production management control system for electronic-grade hydrofluoric acid production, calculating a responsiveness estimate of the product tracking feature vector with respect to the synergistic feature vector to obtain a classification feature vector includes: calculating a responsiveness estimate of the product tracking feature vector relative to the collaborative feature vector to obtain the classification feature vector; wherein the formula is:
whereinRepresenting the product tracking feature vector and the product tracking feature vector,representing the co-occurrence feature vector in the image,a feature vector representing the classification of the feature vector,representing a dot-product of a vector,the representation takes the inverse of the value for each position of the vector.
Production tube for preparing electronic grade hydrofluoric acidIn the control method of the physical control system, the step of passing the classification feature vector through a classifier to obtain a classification result comprises the following steps: processing the classification feature vector using the classifier to obtain the classification result with the following formula:wherein, in the process,toIn the form of a matrix of weights,toIn order to be a vector of the offset,the classified feature vector is obtained.
According to yet another aspect of the present application, there is provided a computer readable medium having stored thereon computer program instructions which, when executed by a processor, cause the processor to perform the control method of the production management control system for electronic grade hydrofluoric acid production as described above.
Compared with the prior art, the production management control system for preparing the electronic-grade hydrofluoric acid and the control method thereof adopt an artificial intelligence-based monitoring technology, ensure the preparation yield of the final electronic-grade hydrofluoric acid product by monitoring various parameters of the rectifying tower and the absorption tower in the preparation process of the electronic-grade hydrofluoric acid, and also use the dynamic change characteristics of the product to perform responsiveness estimation so as to more accurately monitor whether the equipment performances of the rectifying tower and the absorption tower are normal or not.
Drawings
The above and other objects, features and advantages of the present application will become more apparent by describing in more detail embodiments of the present application with reference to the attached drawings. The accompanying drawings are included to provide a further understanding of the embodiments of the application and are incorporated in and constitute a part of this specification, illustrate embodiments of the application and together with the description serve to explain the principles of the application. In the drawings, like reference numbers generally represent like parts or steps.
Fig. 1 is a view of an application scenario of a production management control system for electronic-grade hydrofluoric acid production according to an embodiment of the present application.
FIG. 2 is a block diagram of a production management control system for electronic grade hydrofluoric acid production according to an embodiment of the present application.
FIG. 3 is a block diagram of a product tracking code module in a production management control system for electronic grade hydrofluoric acid production according to an embodiment of the present application.
FIG. 4 is a block diagram of a rectifier parameter encoding module in a production management control system for electronic grade hydrofluoric acid production according to an embodiment of the present application.
FIG. 5 is a flow chart of a method of controlling a production management control system for electronic grade hydrofluoric acid production according to an embodiment of the present application.
Fig. 6 is a schematic diagram of a control method of a production management control system for electronic grade hydrofluoric acid production according to an embodiment of the present application.
Detailed Description
Hereinafter, example embodiments according to the present application will be described in detail with reference to the accompanying drawings. It should be understood that the described embodiments are only some embodiments of the present application and not all embodiments of the present application, and that the present application is not limited by the example embodiments described herein.
Overview of a scene
As mentioned above, electronic-grade hydrofluoric acid is a strong acidic cleaning agent and corrosive agent, and is mainly used for producing very large scale integrated circuits. At present, the main production method of electronic grade hydrofluoric acid comprises the following steps: firstly, chemical pretreatment is carried out on industrial anhydrous hydrofluoric acid, then rectification is carried out, the obtained hydrogen fluoride gas is cooled and absorbed by pure water, and finally filtration and filling are carried out. Because the existence of arsenic as an impurity has a serious influence on the performance of electronic devices, the removal of arsenic is a key problem in the purification process of hydrofluoric acid, and a commonly used method is to convert trivalent arsenic impurities into pentavalent arsenic compounds with high boiling points by using an oxidant, wherein the commonly used oxidant is KMnO4, crO3, persulfate and the like.
Patent 103991847 discloses a method for preparing electronic grade hydrofluoric acid, which uses hydrogen peroxide as oxidant to convert trivalent arsenic impurities into pentavalent arsenic compounds with high boiling point, and no extra impurities are introduced, so as to prepare electronic grade hydrofluoric acid product with high yield, good quality and low cost, and recover hydrogen fluoride gas in tail gas.
However, in the actual preparation process, it is found that although the technical solution disclosed in patent 103991847 can produce the electronic-grade hydrofluoric acid product with higher purity, since the production line for producing the electronic-grade hydrofluoric acid comprises a plurality of devices, once a certain device fails or the performance of the certain device is reduced, the production yield of the final electronic-grade hydrofluoric acid product is affected.
Therefore, a plant production management control system for electronic-grade hydrofluoric acid production is desired to perform online monitoring of the performance of each plant in the production line.
Correspondingly, the preparation method of the electronic-grade hydrofluoric acid comprises the following steps:
(1) Introducing industrial anhydrous hydrogen fluoride liquid and pure water into a rectifying tower to form hydrofluoric acid with a first concentration; adding hydrogen peroxide solution into hydrofluoric acid with a first concentration, and carrying out oxidation treatment to oxidize arsenic and silicon impurities in the hydrofluoric acid; then rectifying in a rectifying tower, keeping impurities in a tower kettle of the rectifying tower, and obtaining purified hydrogen fluoride gas at the tower top of the rectifying tower;
(2) Introducing the purified hydrogen fluoride gas into an absorption tower, and carrying out condensation treatment to obtain hydrogen fluoride liquid;
(3) Performing first filtration on the hydrogen fluoride liquid obtained in the step (2) to remove large-particle impurities;
(4) Introducing the hydrogen fluoride liquid from which the large-particle impurities are removed after the primary filtration into a finished product intermediate tank from the bottom of the finished product intermediate tank, and absorbing the hydrogen fluoride liquid in the finished product intermediate tank by using pure water to obtain hydrofluoric acid with a second concentration;
(5) Filtering the hydrofluoric acid with the second concentration for the second time to remove small-particle impurities, and then carrying out dust-free canning to obtain an electronic grade hydrofluoric acid product;
(6) Absorbing tail gas generated in the process of preparing electronic-grade hydrofluoric acid by pure water to prepare industrial-grade hydrofluoric acid.
Based on this, the present inventors consider that if the production yield of the final electronic-grade hydrofluoric acid product is ensured, the performance of each device for producing electronic-grade hydrofluoric acid on a production line needs to be monitored on line, and in the method for producing electronic-grade hydrofluoric acid, the present inventors found that the performance of each device is mainly monitored by a rectifying tower and an absorption tower, and when the rectifying tower and the absorption tower are monitored, the liquid chromatogram of the hydrogen fluoride liquid before the first filtration is used as the responsiveness reference of the product, so that the accuracy of the device performance judgment of the rectifying tower and the absorption tower can be further improved.
Specifically, in the technical solution of the present application, first, a plurality of first parameters of the rectifying column and a plurality of second parameters of the absorption column at a plurality of predetermined time points are obtained by respective sensors disposed in the rectifying column and the absorption column. Here, the portions of the equipment whose main need to be monitored are the temperature of the bottom of the rectifying tower, the temperature of the body of the tower, the temperature of the top of the tower, the pressure of the bottom of the tower, the pressure of the top of the tower, the reflux temperature and reflux ratio in the rectifying tower, and the temperature of the absorbing tower and the pressure of the absorbing tower, and therefore, the plurality of first parameters include the bottom temperature of the rectifying tower, the temperature of the body of the rectifying tower, the temperature of the top of the rectifying tower, the pressure of the bottom of the rectifying tower, the reflux temperature and reflux ratio in the rectifying tower, and the plurality of second parameters include the temperature of the absorbing tower and the pressure of the absorbing tower.
Meanwhile, a plurality of liquid chromatograms of the hydrogen fluoride liquid before the first filtration at the plurality of predetermined time points are obtained by a liquid chromatograph as the dynamic change responsiveness characteristics of the product. Then, a convolution neural network model with excellent performance in local implicit feature extraction of the image is used for processing the liquid chromatogram of the hydrogen fluoride liquid before the first filtration. It should be understood that, since the liquid chromatogram of the hydrogen fluoride liquid before the first filtration has dynamic feature information in the time dimension when feature mining is performed on the liquid chromatogram, in order to extract such dynamic change feature information, the plurality of liquid chromatograms of the hydrogen fluoride liquid before the first filtration at the plurality of predetermined time points are further processed using a first convolution neural network of a three-dimensional convolution kernel to obtain a product tracking feature vector.
And for the plurality of first parameters and the plurality of second parameters, the plurality of first parameters and the plurality of second parameters are respectively arranged into a rectifying tower parameter input matrix and an absorbing tower parameter input matrix according to a time dimension and a parameter sample dimension, and then a convolutional neural network serving as a filter is used for obtaining a rectifying tower characteristic vector and an absorbing tower characteristic vector, so that high-dimensional implicit associated characteristic information of the plurality of first parameters and the plurality of second parameters on a time sequence in the rectifying tower parameter matrix and the absorbing tower parameter matrix integrating the time dimension and the sample dimension can be respectively mined.
It will be appreciated that the data matrix in the time and data sample dimensions is feature extracted using a convolutional neural network as a filter to obtain the rectification column feature vectorAnd the characteristic vector of the absorption towerIn the process, the filter of the convolutional neural network is used for extracting the features aiming at local feature dimensions, but when the feature vectors are fused, the filter is based on the feature vectors of the rectifying towerAnd the characteristic vector of the absorption towerIs fused, thus for the rectification column eigenvectorsAnd the characteristic vector of the absorption towerThe coefficients were calculated at the time of fusion as:
andare respectively the characteristic vector of the rectifying towerAnd the characteristic vector of the absorption towerThe characteristic value of each position.
The coefficient explicitly generalizes semantic reasoning information by explicitly generalizing semantic concepts corresponding to the feature values from bottom to top, and accordingly obtains information plasticity of space complexity of high-dimensional manifold corresponding to the features in a high-dimensional semantic space by informatively reasoning from bottom to top of the feature semantics, thereby promoting feature vector of the rectifying towerAnd the characteristic vector of the absorption towerAnd the feature fusion based on the global feature distribution is used for obtaining the cooperative feature vector, so that the accuracy of subsequent classification can be improved.
Considering that the dynamic change characteristics of the products along with time in the process of preparing the electronic grade hydrofluoric acid are response characteristics related to the control of the rectifying tower and the absorption tower, in order to more accurately monitor and control the rectifying tower and the absorption tower, a response estimation of the product tracking feature vector relative to the cooperative feature vector is calculated to obtain a classification feature vector. And classifying the classified characteristic vectors through a classifier to obtain a classification result for indicating whether the equipment performance of the rectifying tower and the absorption tower is normal or not.
In view of this, the present application proposes a production management control system for electronic grade hydrofluoric acid production, comprising: the device comprises an equipment parameter acquisition module, a data processing module and a data processing module, wherein the equipment parameter acquisition module is used for acquiring a plurality of first parameters of a rectifying tower and a plurality of second parameters of an absorption tower at a plurality of preset time points, the plurality of first parameters comprise the tower kettle temperature of the rectifying tower, the tower body temperature of the rectifying tower, the tower top temperature of the rectifying tower, the tower kettle pressure of the rectifying tower, the tower top pressure of the rectifying tower, the tower in-tower reflux temperature and reflux ratio of the rectifying tower, and the plurality of second parameters comprise the temperature of the absorption tower and the pressure of the absorption tower; the product state acquisition module is used for acquiring a plurality of liquid chromatogram maps of the hydrogen fluoride liquid before the first filtration at the plurality of preset time points through a liquid chromatograph; a product tracking encoding module, configured to pass a plurality of liquid chromatograms of the hydrogen fluoride liquid at the plurality of predetermined time points through a first convolution neural network using a three-dimensional convolution kernel to obtain a product tracking feature vector; the rectifying tower parameter coding module is used for arranging a plurality of items of first parameters of the rectifying tower at a plurality of preset time points into a rectifying tower parameter input matrix according to time dimensions and parameter sample dimensions and then obtaining a rectifying tower characteristic vector through a second convolution neural network serving as a filter; the absorption tower parameter coding module is used for arranging a plurality of second parameters of the absorption towers at the plurality of preset time points into an absorption tower parameter input matrix according to the time dimension and the parameter sample dimension and then obtaining an absorption tower characteristic vector through a third convolutional neural network serving as a filter; the parameter characteristic fusion module is used for fusing the rectifying tower characteristic vector and the absorption tower characteristic vector to obtain a cooperative characteristic vector; the responsiveness estimation module is used for calculating the responsiveness estimation of the product tracking feature vector relative to the cooperative feature vector to obtain a classification feature vector; and the equipment management result generation module is used for enabling the classification characteristic vectors to pass through a classifier to obtain a classification result, and the classification result is used for indicating whether the equipment performance of the rectifying tower and the equipment performance of the absorption tower are normal or not.
FIG. 1 illustrates an application scenario of a production management control system for electronic grade hydrofluoric acid production according to an embodiment of the present application. As shown in fig. 1, in this application scenario, first, a plurality of first parameters of the rectification column and a plurality of second parameters of the absorption column at a plurality of predetermined time points are acquired by a plurality of sensors (e.g., T1-Tn as illustrated in fig. 1) disposed at the rectification column (e.g., R as illustrated in fig. 1) and the absorption column (e.g., a as illustrated in fig. 1), and a plurality of liquid chromatograms of the hydrogen fluoride liquid (e.g., H as illustrated in fig. 1) before first filtration at the plurality of predetermined time points are acquired by a liquid chromatograph (e.g., L as illustrated in fig. 1). Here, the plurality of first parameters include a column bottom temperature of the rectifying column, a column body temperature of the rectifying column, a column top temperature of the rectifying column, a column bottom pressure of the rectifying column, a column top pressure of the rectifying column, a column mid-reflux temperature and a reflux ratio of the rectifying column, and the plurality of second parameters include a temperature of the absorbing column and a pressure of the absorbing column. Then, the plurality of first parameters of the rectifying tower and the plurality of second parameters of the absorption tower at the plurality of predetermined time points and the plurality of liquid chromatogram maps of the hydrogen fluoride liquid before the first filtration at the plurality of predetermined time points are input into a server (for example, a cloud server S as illustrated in fig. 1) deployed with a production management control algorithm for electronic-grade hydrofluoric acid preparation, wherein the server can process the plurality of first parameters of the rectifying tower and the plurality of second parameters of the absorption tower at the plurality of predetermined time points and the plurality of liquid chromatogram maps of the hydrogen fluoride liquid before the first filtration at the plurality of predetermined time points with the production management control algorithm for electronic-grade hydrofluoric acid preparation to generate a classification result for indicating whether the equipment performance of the rectifying tower and the absorption tower is normal or not.
Having described the basic principles of the present application, various non-limiting embodiments of the present application will now be described with reference to the accompanying drawings.
Exemplary System
FIG. 2 illustrates a block diagram of a production management control system for electronic grade hydrofluoric acid production according to an embodiment of the present application. As shown in fig. 2, a production management control system 200 for electronic grade hydrofluoric acid preparation according to the embodiment of the present application comprises: an equipment parameter acquiring module 210, configured to acquire multiple first parameters of a rectifying tower and multiple second parameters of an absorption tower at multiple predetermined time points, where the multiple first parameters include a tower still temperature of the rectifying tower, a body temperature of the rectifying tower, a tower top temperature of the rectifying tower, a tower still pressure of the rectifying tower, a tower top pressure of the rectifying tower, a tower internal reflux temperature and a reflux ratio of the rectifying tower, and the multiple second parameters include a temperature of the absorption tower and a pressure of the absorption tower; a product state acquisition module 220, configured to acquire, by using a liquid chromatograph, a plurality of liquid chromatograms of the hydrogen fluoride liquid at the plurality of predetermined time points before the first filtering; a product tracking encoding module 230, configured to pass the plurality of liquid chromatograms of the hydrogen fluoride liquid before the first filtering at the plurality of predetermined time points through a first convolution neural network using a three-dimensional convolution kernel to obtain a product tracking feature vector; the rectifying tower parameter encoding module 240 is configured to arrange the plurality of first parameters of the rectifying tower at the plurality of predetermined time points into a rectifying tower parameter input matrix according to the time dimension and the parameter sample dimension, and then obtain a rectifying tower feature vector through a second convolutional neural network serving as a filter; the absorption tower parameter coding module 250 is configured to arrange multiple second parameters of the absorption tower at multiple predetermined time points into an absorption tower parameter input matrix according to a time dimension and a parameter sample dimension, and then obtain an absorption tower feature vector through a third convolutional neural network serving as a filter; a parameter feature fusion module 260, configured to fuse the rectifying tower feature vector and the absorption tower feature vector to obtain a collaborative feature vector; a responsiveness estimation module 270, configured to calculate a responsiveness estimation of the product tracking feature vector with respect to the collaborative feature vector to obtain a classification feature vector; and an equipment management result generating module 280, configured to pass the classification feature vector through a classifier to obtain a classification result, where the classification result is used to indicate whether the equipment performance of the rectifying tower and the absorption tower is normal.
Specifically, in this embodiment, the apparatus parameter acquiring module 210, the product state acquiring module 220 and the product tracking encoding module 230 are configured to acquire a plurality of first parameters of the rectifying tower at a plurality of predetermined time points and a plurality of second parameters of the absorption tower, wherein the plurality of first parameters include a tower still temperature of the rectifying tower, a body temperature of the rectifying tower, a tower top temperature of the rectifying tower, a tower still pressure of the rectifying tower, a tower top pressure of the rectifying tower, a tower internal reflux temperature of the rectifying tower and a reflux ratio, the plurality of second parameters include a temperature of the absorption tower and a pressure of the absorption tower, and acquire a plurality of liquid chromatogram of the hydrogen fluoride liquid before first filtering at the plurality of predetermined time points by a liquid chromatograph, and further acquire the plurality of liquid chromatogram of the hydrogen fluoride liquid before first filtering at the plurality of predetermined time points by using a first convolution neural network of three-dimensional kernels to obtain the product tracking feature vector. As mentioned above, in order to ensure the production yield of the final electronic-grade hydrofluoric acid product, the performance of each equipment for producing electronic-grade hydrofluoric acid on the production line needs to be monitored on line. In the method for preparing the electronic grade hydrofluoric acid, the main part for monitoring the performance of each device is to monitor the rectifying tower and the absorption tower. In addition, when the rectification column and the absorption column are monitored, the liquid chromatogram of the hydrogen fluoride liquid before the first filtration is used as a response reference of the product, so that the accuracy of the judgment of the equipment performance of the rectification column and the absorption column can be further improved.
That is, specifically, in the technical solution of the present application, first, a plurality of first parameters of the rectifying column and a plurality of second parameters of the absorbing column at a plurality of predetermined time points are acquired by respective sensors disposed at the rectifying column and the absorbing column. Here, the portions of the equipment whose main need to be monitored are the temperature of the bottom of the rectifying tower, the temperature of the body of the tower, the temperature of the top of the tower, the pressure of the bottom of the tower, the pressure of the top of the tower, the reflux temperature and reflux ratio in the rectifying tower, and the temperature of the absorbing tower and the pressure of the absorbing tower, and therefore, the plurality of first parameters include the bottom temperature of the rectifying tower, the temperature of the body of the rectifying tower, the temperature of the top of the rectifying tower, the pressure of the bottom of the rectifying tower, the reflux temperature and reflux ratio in the rectifying tower, and the plurality of second parameters include the temperature of the absorbing tower and the pressure of the absorbing tower.
Meanwhile, a plurality of liquid chromatograms of the hydrogen fluoride liquid before the first filtration at the plurality of predetermined time points are obtained by a liquid chromatograph as the dynamic change responsiveness characteristics of the product. Then, a convolution neural network model with excellent performance in local implicit feature extraction of the image is used for processing the liquid chromatogram of the hydrogen fluoride liquid before the first filtration. It should be understood that, since the liquid chromatogram of the hydrogen fluoride liquid before the first filtration is subjected to feature mining, the liquid chromatogram has dynamic feature information in the time dimension, in order to extract such dynamic change feature information, in the technical solution of the present application, a first convolution neural network of a three-dimensional convolution kernel is further used to process a plurality of liquid chromatograms of the hydrogen fluoride liquid before the first filtration at the plurality of predetermined time points to obtain a product tracking feature vector.
More specifically, in the embodiment of the present application, the artifact tracking coding module includes: firstly, convolution processing, pooling processing and nonlinear activation processing based on the three-dimensional convolution kernel are respectively carried out on input data in the forward transmission process of the layer by using each layer of the first convolution neural network so as to output a tracking characteristic map from the last layer of the first convolution neural network, wherein the input of the first convolution neural network is a plurality of liquid chromatogram maps of the hydrogen fluoride liquid before first filtration of a plurality of preset time points. Then, global mean pooling based on a feature matrix is performed on the tracking feature map to obtain the tracking feature vector. It should be appreciated that pooling can be used for feature dimension reduction, mitigation of the risk of over-fitting, and reduction of the over-sensitivity of convolutional layers to detection information, while the global mean pooling can preserve important information of each of the tracking feature maps for highlighting the most important part of the response in the tracking feature maps.
FIG. 3 illustrates a block diagram of a product tracking code module in a production management control system for electronic grade hydrofluoric acid production according to an embodiment of the present application. As shown in fig. 3, the artifact tracking coding module 230 includes: a feature dynamic capturing unit 231, configured to perform convolution processing, pooling processing and nonlinear activation processing based on the three-dimensional convolution kernel on input data respectively during forward transmission of layers by using each layer of the first convolutional neural network to output a tracking feature map from a last layer of the first convolutional neural network, where an input of the first convolutional neural network is a plurality of liquid chromatogram of the hydrogen fluoride liquid before first filtering at the plurality of predetermined time points; and a global pooling unit 232, configured to perform global mean pooling based on a feature matrix on the tracking feature map to obtain the tracking feature vector.
Specifically, in this embodiment of the present application, the rectifying tower parameter encoding module 240 and the absorption tower parameter encoding module 250 are configured to arrange multiple items of first parameters of the rectifying towers at multiple predetermined time points into a rectifying tower parameter input matrix according to a time dimension and a parameter sample dimension, and then pass through a second convolutional neural network serving as a filter to obtain a rectifying tower feature vector, and arrange multiple items of second parameters of the absorption towers at multiple predetermined time points into an absorption tower parameter input matrix according to a time dimension and a parameter sample dimension, and then pass through a third convolutional neural network serving as a filter to obtain an absorption tower feature vector. That is, in the technical solution of the present application, for the plurality of first parameters and the plurality of second parameters, after the plurality of first parameters and the plurality of second parameters are respectively arranged as a rectifying tower parameter input matrix and an absorption tower parameter input matrix according to a time dimension and a parameter sample dimension, a rectifying tower eigenvector and an absorption tower eigenvector are obtained by a convolutional neural network serving as a filter, so that high-dimensional implicit associated feature information of the plurality of first parameters and the plurality of second parameters in the rectifying tower parameter matrix and the absorption tower parameter matrix, which integrate the time dimension and the sample dimension, can be respectively mined.
More specifically, in this application embodiment, the rectifying tower parameter encoding module includes: firstly, arranging a plurality of items of first parameters of the rectifying tower at a plurality of preset time points into a plurality of items of first parameter row vectors corresponding to the rectifying tower according to a time dimension, and arranging the plurality of items of first parameter row vectors into a parameter input matrix of the rectifying tower according to a parameter sample dimension; and then, carrying out convolution processing, pooling processing along a characteristic matrix and activation processing on input data in forward transmission of layers by using each layer of the second convolutional neural network to generate the rectifying tower characteristic vector from the last layer of the second convolutional neural network, wherein the input of the first layer of the second convolutional neural network is the rectifying tower parameter input matrix.
More specifically, in this embodiment of the present application, the absorption tower parameter coding module includes: firstly, arranging a plurality of items of second parameters of the absorption towers at a plurality of preset time points into a plurality of items of second parameter row vectors corresponding to the absorption towers according to a time dimension, and arranging the plurality of items of second parameter row vectors into a parameter input matrix of the absorption towers according to a parameter sample dimension; then, the layers of the third convolutional neural network are used for carrying out convolution processing, pooling processing along a feature matrix and activation processing on input data in forward transmission of the layers so as to generate the absorption tower feature vector from the last layer of the third convolutional neural network, wherein the input of the first layer of the third convolutional neural network is the absorption tower parameter input matrix.
FIG. 4 illustrates a block diagram of a rectifier parameter encoding module in a production management control system for electronic grade hydrofluoric acid production according to an embodiment of the present application. As shown in fig. 4, the rectifying tower parameter encoding module 240 includes: the first matrix constructing unit 241 is configured to arrange a plurality of items of first parameters of the rectifying tower at the plurality of predetermined time points into a plurality of items of first parameter row vectors corresponding to the rectifying tower according to a time dimension, and arrange the plurality of items of first parameter row vectors into the rectifying tower parameter input matrix according to a parameter sample dimension; a first filter unit 242 configured to perform convolution processing, pooling processing along a feature matrix, and activation processing on input data in forward direction transfer of layers using layers of the second convolutional neural network to generate the rectifying tower feature vector from a last layer of the second convolutional neural network, wherein an input of the first layer of the second convolutional neural network is the rectifying tower parameter input matrix.
Specifically, in the embodiment of the present application, the parameter feature fusion module 260 is configured to fuse the rectifying tower feature vector and the absorbing tower feature vector to obtain a synergistic feature vector. It should be understood that the data matrix in time and data sample dimensions is characterized using the convolutional neural network as a filter to obtain the rectification column feature vectorAnd the absorption tower eigenvectorIn the process of integrating feature vectors, the filter of the convolutional neural network is based on the feature vector of the rectifying tower, but the filter of the convolutional neural network is based on the feature extraction of local feature dimensions when the feature vectors are integratedAnd the absorption tower eigenvectorThe global feature distribution is fused, therefore, in the technical scheme of the application, the feature vector of the rectifying tower needs to be fusedAnd the absorption tower eigenvectorThe respective coefficients are calculated at the time of fusion to perform weighted correction on the respective coefficients. Then, the synergistic characteristic vector can be obtained by calculating the weighted sum of the rectifying tower characteristic vector and the absorption tower characteristic vector according to the position.
More specifically, in this embodiment of the present application, the parameter feature fusion module is further configured to: the weighting weight of the characteristic vector of the rectifying tower is calculated by the following formula, wherein the formula is as follows:
wherein,characteristic values representing respective positions of the rectifying tower characteristic vector,and the weighting coefficient represents the characteristic vector of the rectifying tower when the characteristic vector of the rectifying tower and the characteristic vector of the absorption tower are fused. The weighting weight of the characteristic vector of the absorption tower is calculated by the following formula, wherein the formula is as follows:
wherein,an eigenvalue representing each position of the absorption tower eigenvector,and the weighting coefficient represents the characteristic vector of the absorption tower when the characteristic vector of the rectification tower and the characteristic vector of the absorption tower are fused. It should be understood that the coefficient performs explicit generalization of semantic reasoning information by performing explicit generalization from bottom to top on semantic concepts corresponding to the feature values, so as to obtain information plasticity of spatial complexity of high-dimensional manifold corresponding to the feature in a high-dimensional semantic space by performing information reasoning from bottom to top on the feature semantics, thereby promoting the feature vector of the rectifying towerAnd the absorption tower eigenvectorAnd the collaborative feature vector is obtained by feature fusion based on global feature distribution, so that the accuracy of subsequent classification can be improved.
Specifically, in this embodiment of the present application, the responsiveness estimation module 270 and the equipment management result generation module 280 are configured to calculate a responsiveness estimation of the product tracking feature vector with respect to the cooperative feature vector to obtain a classification feature vector, and pass the classification feature vector through a classifier to obtain a classification result, where the classification result is used to indicate whether the equipment performance of the rectifying tower and the absorption tower is normal. It should be understood that, in order to monitor and control the rectification column and the absorption column more accurately, it is considered that the dynamic variation characteristics of the products with time during the production of the electronic-grade hydrofluoric acid are response characteristics with respect to the control of the rectification column and the absorption columnCalculating a responsiveness estimate of the product tracking feature vector relative to the collaborative feature vector to obtain a classification feature vector. And classifying the classified characteristic vectors through a classifier to obtain a classification result for indicating whether the equipment performance of the rectifying tower and the absorption tower is normal or not. Accordingly, in one specific example, the classification feature vector is processed using the classifier to obtain the classification result with the following formula:wherein, in the process,toIn order to be a weight matrix, the weight matrix,toIn order to be a vector of the offset,the classified feature vector is obtained.
More specifically, in this embodiment of the present application, the responsiveness estimation module is further configured to calculate a responsiveness estimation of the product tracking feature vector with respect to the collaborative feature vector to obtain the classification feature vector according to the following formula;
whereinRepresenting the product tracking feature vector(s),-representing the co-occurrence feature vector(s),a feature vector representing the classification of the feature vector,representing a dot-product of a vector,the expression takes the reciprocal of the value for each position of the vector.
In summary, the production management control system 200 for electronic-grade hydrofluoric acid production based on the embodiment of the present application is illustrated, which employs an artificial intelligence-based monitoring technology to ensure the production yield of the final electronic-grade hydrofluoric acid product by monitoring various parameters of the rectifying tower and the absorption tower in the electronic-grade hydrofluoric acid production process, and also uses the dynamic variation characteristics of the product to perform responsiveness estimation in the process, so as to more accurately monitor whether the equipment performance of the rectifying tower and the absorption tower is normal.
As described above, the production management control system 200 for electronic-grade hydrofluoric acid production according to the embodiment of the present application may be implemented in various terminal devices, such as a server of a production management control algorithm for electronic-grade hydrofluoric acid production, and the like. In one example, the production management control system 200 for electronic grade hydrofluoric acid production according to the embodiment of the present application may be integrated into a terminal device as one software module and/or hardware module. For example, the production management control system 200 for electronic grade hydrofluoric acid production may be a software module in the operating system of the terminal device, or may be an application developed for the terminal device; of course, the production management control system 200 for electronic grade hydrofluoric acid production may also be one of many hardware modules of the terminal equipment.
Alternatively, in another example, the production management control system 200 for electronic-grade hydrofluoric acid preparation and the terminal equipment may be separate devices, and the production management control system 200 for electronic-grade hydrofluoric acid preparation may be connected to the terminal equipment through a wired and/or wireless network and transmit the mutual information according to an agreed data format.
Exemplary method
FIG. 5 illustrates a flow chart of a method of control of a production management control system for electronic grade hydrofluoric acid production. As shown in fig. 5, the control method of the production management control system for electronic grade hydrofluoric acid preparation according to the embodiment of the present application comprises the steps of: s110, acquiring multiple first parameters of a rectifying tower and multiple second parameters of an absorption tower at multiple preset time points, wherein the multiple first parameters comprise the tower kettle temperature of the rectifying tower, the body temperature of the rectifying tower, the tower top temperature of the rectifying tower, the tower kettle pressure of the rectifying tower, the tower top pressure of the rectifying tower, the tower internal reflux temperature and reflux ratio of the rectifying tower, and the multiple second parameters comprise the temperature of the absorption tower and the pressure of the absorption tower; s120, acquiring a plurality of liquid chromatogram maps of the hydrogen fluoride liquid before the first filtration at the plurality of preset time points through a liquid chromatograph; s130, passing the plurality of liquid chromatogram maps of the hydrogen fluoride liquid before the first filtration of the plurality of preset time points through a first convolution neural network using a three-dimensional convolution kernel to obtain a product tracking feature vector; s140, arranging a plurality of first parameters of the rectifying tower at a plurality of preset time points into a rectifying tower parameter input matrix according to time dimension and parameter sample dimension, and then passing through a second convolution neural network serving as a filter to obtain a rectifying tower characteristic vector; s150, arranging a plurality of second parameters of the absorption tower at the plurality of preset time points into an absorption tower parameter input matrix according to time dimensions and parameter sample dimensions, and then passing through a third convolutional neural network serving as a filter to obtain an absorption tower characteristic vector; s160, fusing the rectifying tower characteristic vector and the absorption tower characteristic vector to obtain a synergistic characteristic vector; s170, calculating the responsiveness estimation of the product tracking feature vector relative to the cooperative feature vector to obtain a classification feature vector; and S180, passing the classification characteristic vector through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the equipment performance of the rectifying tower and the absorption tower is normal or not.
FIG. 6 illustrates an architectural schematic of a control methodology for a production management control system for electronic grade hydrofluoric acid production according to an embodiment of the present application. As shown in fig. 6, in the network architecture of the control method for the production management control system for electronic-grade hydrofluoric acid preparation, first, a plurality of liquid chromatogram maps (e.g., P as illustrated in fig. 6) of the hydrogen fluoride liquid before the first filtration obtained at the plurality of predetermined time points are passed through a first convolution neural network (e.g., CNN1 as illustrated in fig. 6) using a three-dimensional convolution kernel to obtain a product tracking feature vector (e.g., VF as illustrated in fig. 6); then, arranging the obtained plurality of first parameters (for example, Q1 as illustrated in fig. 6) of the rectifying tower at the plurality of predetermined time points into a rectifying tower parameter input matrix (for example, M1 as illustrated in fig. 6) according to a time dimension and a parameter sample dimension, and then passing through a second convolutional neural network (for example, CNN2 as illustrated in fig. 6) serving as a filter to obtain a rectifying tower feature vector (for example, VF1 as illustrated in fig. 6); then, after arranging the obtained plurality of second parameters (for example, Q2 as illustrated in fig. 6) of the absorption tower at the plurality of predetermined time points into an absorption tower parameter input matrix (for example, M2 as illustrated in fig. 6) according to the time dimension and the parameter sample dimension, passing through a third convolutional neural network (for example, CNN3 as illustrated in fig. 6) as a filter to obtain an absorption tower feature vector (for example, VF2 as illustrated in fig. 6); then, fusing the rectification column eigenvector and the absorption column eigenvector to obtain a synergistic eigenvector (e.g., VF3 as illustrated in fig. 6); then, calculating a responsiveness estimate of the product tracking feature vector with respect to the collaborative feature vector to obtain a classification feature vector (e.g., V as illustrated in fig. 6); and finally, passing the classification feature vector through a classifier (for example, a circle S as illustrated in fig. 6) to obtain a classification result, which is used to indicate whether the equipment performance of the rectifying tower and the absorption tower is normal.
More specifically, in step S110, step S120 and step S130, a plurality of first parameters of the rectifying column and a plurality of second parameters of the absorbing column at a plurality of predetermined time points are obtained, wherein the plurality of first parameters include a still temperature of the rectifying column, a body temperature of the rectifying column, an overhead temperature of the rectifying column, a still pressure of the rectifying column, an overhead pressure of the rectifying column, an in-column reflux temperature and a reflux ratio of the rectifying column, and the plurality of second parameters include a temperature of the absorbing column and a pressure of the absorbing column, and a plurality of liquid phases of the hydrogen fluoride liquid before first filtration at the plurality of predetermined time points are obtained by a liquid chromatograph. It should be understood that the performance of each apparatus for producing electronic grade hydrofluoric acid on the production line needs to be monitored on-line in order to ensure the production yield of the final electronic grade hydrofluoric acid product. In the method for preparing the electronic grade hydrofluoric acid, the main part for monitoring the performance of each device is to monitor the rectifying tower and the absorption tower. In addition, when the rectification column and the absorption column are monitored, the liquid chromatogram of the hydrogen fluoride liquid before the first filtration is used as a response reference of the product, so that the accuracy of the judgment of the equipment performance of the rectification column and the absorption column can be further improved.
That is, specifically, in the technical solution of the present application, first, a plurality of first parameters of the rectifying column and a plurality of second parameters of the absorbing column at a plurality of predetermined time points are acquired by respective sensors disposed at the rectifying column and the absorbing column. Here, the portions of the equipment whose main need to be monitored are the temperature of the column bottom of the rectifying column, the temperature of the column body, the temperature of the column top, the column bottom pressure, the column top pressure, the reflux temperature and reflux ratio in the rectifying column, and the temperature of the absorbing column and the pressure of the absorbing column, and therefore, the plurality of first parameters include the column bottom temperature of the rectifying column, the column body temperature of the rectifying column, the column top temperature of the rectifying column, the column bottom pressure of the rectifying column, the column top pressure of the rectifying column, the reflux temperature and reflux ratio in the rectifying column, and the plurality of second parameters include the temperature of the absorbing column and the pressure of the absorbing column.
Meanwhile, a plurality of liquid chromatograms of the hydrogen fluoride liquid before the first filtration at the plurality of predetermined time points are obtained by a liquid chromatograph as the dynamic change responsiveness characteristics of the product. Then, a convolution neural network model with excellent performance in local implicit feature extraction of the image is used for processing the liquid chromatogram of the hydrogen fluoride liquid before the first filtration. It should be understood that, since the liquid chromatogram of the hydrogen fluoride liquid before the first filtration is subjected to feature mining, the liquid chromatogram has dynamic feature information in the time dimension, in order to extract such dynamic change feature information, in the technical solution of the present application, a first convolution neural network of a three-dimensional convolution kernel is further used to process a plurality of liquid chromatograms of the hydrogen fluoride liquid before the first filtration at the plurality of predetermined time points to obtain a product tracking feature vector.
More specifically, in step S140 and step S150, the plurality of items of first parameters of the rectifying tower at the plurality of predetermined time points are arranged into the rectifying tower parameter input matrix according to the time dimension and the parameter sample dimension, and then pass through the second convolutional neural network serving as the filter to obtain the rectifying tower feature vector, and the plurality of items of second parameters of the absorbing tower at the plurality of predetermined time points are arranged into the absorbing tower parameter input matrix according to the time dimension and the parameter sample dimension, and then pass through the third convolutional neural network serving as the filter to obtain the absorbing tower feature vector. That is, in the technical solution of the present application, for the plurality of first parameters and the plurality of second parameters, after the plurality of first parameters and the plurality of second parameters are respectively arranged as a rectifying tower parameter input matrix and an absorption tower parameter input matrix according to a time dimension and a parameter sample dimension, a rectifying tower eigenvector and an absorption tower eigenvector are obtained by a convolutional neural network serving as a filter, so that high-dimensional implicit associated feature information of the plurality of first parameters and the plurality of second parameters in the rectifying tower parameter matrix and the absorption tower parameter matrix, which integrate the time dimension and the sample dimension, can be respectively mined.
More specifically, in step S160, the rectification column feature vector and the absorption column feature vector are fused to obtain a synergistic feature vector. It should be understood that feature extraction is performed on the data matrix in time and data sample dimensions using the convolutional neural network as a filter to obtain the rectifying tower feature vectorAnd the absorption tower eigenvectorIn the process of integrating feature vectors, the filter of the convolutional neural network is based on the feature vector of the rectifying tower, but the filter of the convolutional neural network is based on the feature extraction of local feature dimensions when the feature vectors are integratedAnd the absorption tower eigenvectorThe global feature distribution is fused, therefore, in the technical scheme of the application, the feature vector of the rectifying tower needs to be fusedAnd the absorption tower eigenvectorThe respective coefficients are calculated at the time of fusion to perform weighted correction on the respective coefficients. Then, the synergistic characteristic vector can be obtained by calculating the weighted sum of the rectifying tower characteristic vector and the absorption tower characteristic vector according to the position.
More specifically, in step S170 and step S180, a responsiveness estimation of the product tracking feature vector with respect to the synergistic feature vector is calculated to obtain a classification feature vector, and the classification feature vector is passed through a classifier to obtain a classification result, which is used to indicate the equipment performance of the rectifying tower and the absorption towerWhether it is normal or not. It should be understood that, considering that the dynamic variation characteristics of the products over time during the preparation of the electronic grade hydrofluoric acid are the responsiveness characteristics with respect to the control of the rectifying tower and the absorption tower, in order to more accurately monitor and control the rectifying tower and the absorption tower, the responsiveness estimation of the product tracking eigenvector with respect to the cooperative eigenvector is calculated to obtain the classification eigenvector. And classifying the classified characteristic vectors through a classifier to obtain a classification result for indicating whether the equipment performance of the rectifying tower and the absorption tower is normal or not. Accordingly, in one specific example, the classification feature vector is processed using the classifier to obtain the classification result with the following formula:whereintoIn order to be a weight matrix, the weight matrix,toIn order to be a vector of the offset,the classified feature vector is obtained.
In summary, the control method of the production management control system for electronic grade hydrofluoric acid preparation based on the embodiments of the present application is illustrated, which adopts an artificial intelligence-based monitoring technology to ensure the preparation yield of the final electronic grade hydrofluoric acid product by monitoring various parameters of the rectifying tower and the absorption tower in the electronic grade hydrofluoric acid preparation process, and in the process, also uses the dynamic variation characteristics of the product to perform responsiveness estimation so as to more accurately monitor whether the equipment performance of the rectifying tower and the absorption tower is normal or not.
Exemplary computer program product and computer-readable storage Medium
In addition to the above-described methods and apparatus, embodiments herein may also be a computer program product comprising computer program instructions that, when executed by a processor, cause the processor to perform the steps in the functions in the control method for a production management control system for electronic grade hydrofluoric acid preparation according to the various embodiments of the present application described in the "exemplary methods" section above in this specification.
The computer program product may be written with program code for performing the operations of embodiments of the present application in any combination of one or more programming languages, including an object oriented programming language such as Java, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device and partly on a remote computing device, or entirely on the remote computing device or server.
Furthermore, embodiments of the present application may also be a computer readable storage medium having stored thereon computer program instructions which, when executed by a processor, cause the processor to perform the steps in the control method of a production management control system for electronic grade hydrofluoric acid production described in the "exemplary methods" section above in this specification.
The computer readable storage medium may take any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may include, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, or device, or any combination of the foregoing. More specific examples (a non-exhaustive list) of the readable storage medium include: an electrical connection having one or more wires, a portable disk, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
The foregoing describes the general principles of the present application in conjunction with specific embodiments, however, it is noted that the advantages, effects, etc. mentioned in the present application are merely examples and are not limiting, and they should not be considered essential to the various embodiments of the present application. Furthermore, the foregoing disclosure of specific details is provided for purposes of illustration and understanding only, and is not intended to limit the application to the details which are set forth in order to provide a thorough understanding of the present application.
The block diagrams of devices, apparatuses, systems referred to in this application are only given as illustrative examples and are not intended to require or imply that the connections, arrangements, configurations, etc. must be made in the manner shown in the block diagrams. These devices, apparatuses, devices, systems may be connected, arranged, configured in any manner, as will be appreciated by those skilled in the art. Words such as "including," "comprising," "having," and the like are open-ended words that mean "including, but not limited to," and are used interchangeably therewith. The words "or" and "as used herein mean, and are used interchangeably with, the word" and/or, "unless the context clearly dictates otherwise. The word "such as" is used herein to mean, and is used interchangeably with, the phrase "such as but not limited to".
It should also be noted that in the devices, apparatuses, and methods of the present application, each component or step can be decomposed and/or re-combined. These decompositions and/or recombinations should be considered as equivalents of the present application.
The previous description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other aspects without departing from the scope of the application. Thus, the present application is not intended to be limited to the aspects shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
The foregoing description has been presented for purposes of illustration and description. Furthermore, the description is not intended to limit embodiments of the application to the form disclosed herein. While a number of example aspects and embodiments have been discussed above, those of skill in the art will recognize certain variations, modifications, alterations, additions and sub-combinations thereof.
Claims (10)
1. A production management control system for electronic grade hydrofluoric acid production comprising: the device comprises an equipment parameter acquisition module, a data processing module and a data processing module, wherein the equipment parameter acquisition module is used for acquiring a plurality of first parameters of a rectifying tower and a plurality of second parameters of an absorption tower at a plurality of preset time points, the plurality of first parameters comprise the tower kettle temperature of the rectifying tower, the tower body temperature of the rectifying tower, the tower top temperature of the rectifying tower, the tower kettle pressure of the rectifying tower, the tower top pressure of the rectifying tower, the tower in-tower reflux temperature and reflux ratio of the rectifying tower, and the plurality of second parameters comprise the temperature of the absorption tower and the pressure of the absorption tower; the product state acquisition module is used for acquiring a plurality of liquid chromatograms of the hydrogen fluoride liquid at the plurality of preset time points before first filtration through a liquid chromatograph; a product tracking encoding module, configured to obtain a product tracking feature vector by passing through a first convolution neural network using a three-dimensional convolution kernel a plurality of liquid chromatograms of the hydrogen fluoride liquid at the plurality of predetermined time points before first filtering; the rectifying tower parameter coding module is used for arranging a plurality of items of first parameters of the rectifying tower at a plurality of preset time points into a rectifying tower parameter input matrix according to time dimension and parameter sample dimension and then obtaining a rectifying tower characteristic vector through a second convolutional neural network serving as a filter; the absorption tower parameter coding module is used for arranging a plurality of second parameters of the absorption tower at the plurality of preset time points into an absorption tower parameter input matrix according to time dimensions and parameter sample dimensions and then obtaining an absorption tower characteristic vector through a third convolutional neural network serving as a filter; the parameter characteristic fusion module is used for fusing the rectifying tower characteristic vector and the absorption tower characteristic vector to obtain a cooperative characteristic vector; the responsiveness estimation module is used for calculating the responsiveness estimation of the product tracking feature vector relative to the cooperative feature vector to obtain a classification feature vector; and the equipment management result generation module is used for enabling the classification characteristic vectors to pass through a classifier to obtain a classification result, and the classification result is used for indicating whether the equipment performance of the rectifying tower and the equipment performance of the absorption tower are normal or not.
2. The production management control system for electronic grade hydrofluoric acid production of claim 1, wherein the product tracking code module comprises: a feature dynamic capturing unit, configured to perform convolution processing, pooling processing, and nonlinear activation processing based on the three-dimensional convolution kernel on input data respectively during forward layer transfer using each layer of the first convolutional neural network to output a tracking feature map from a last layer of the first convolutional neural network, where the input of the first convolutional neural network is a plurality of liquid chromatogram of the hydrogen fluoride liquid before first filtering at the plurality of predetermined time points; and the global pooling unit is used for performing global mean pooling based on a feature matrix on the tracking feature map to obtain the tracking feature vector.
3. The production management control system for electronic grade hydrofluoric acid production of claim 2, wherein the rectification column parameter encoding module comprises: the first matrix construction unit is used for arranging a plurality of items of first parameters of the rectifying tower at a plurality of preset time points into a plurality of items of first parameter row vectors corresponding to the rectifying tower according to the time dimension, and arranging the plurality of items of first parameter row vectors into a parameter input matrix of the rectifying tower according to the parameter sample dimension; a first filter unit for performing convolution processing, pooling processing along a feature matrix, and activation processing on input data in forward pass of layers using layers of the second convolutional neural network to generate the rectifying column feature vector from a last layer of the second convolutional neural network, wherein an input of a first layer of the second convolutional neural network is the rectifying column parameter input matrix.
4. The production management control system for electronic grade hydrofluoric acid production of claim 3, wherein the absorption tower parameter encoding module comprises: the second matrix construction unit is used for arranging a plurality of second parameters of the absorption tower at the plurality of preset time points into a plurality of second parameter row vectors corresponding to the absorption tower according to time dimension, and arranging the plurality of second parameter row vectors into the parameter input matrix of the absorption tower according to parameter sample dimension; a second filter unit for performing convolution processing, pooling processing along a feature matrix, and activation processing on input data in forward pass of layers using layers of the third convolutional neural network to generate the absorption tower feature vector from a last layer of the third convolutional neural network, wherein an input of a first layer of the third convolutional neural network is the absorption tower parameter input matrix.
5. The production management control system for electronic grade hydrofluoric acid production of claim 4, wherein the parameter feature fusion module is configured to calculate a position-weighted sum of the rectification column feature vector and the absorption column feature vector to obtain the synergy feature vector.
6. The production management and control system for electronic grade hydrofluoric acid production of claim 5, wherein the weighted weight of the distillation column eigenvector is calculated from the following equation:
wherein,represents the aboveThe characteristic value of each position of the characteristic vector of the rectifying tower,and the weighting coefficient represents the characteristic vector of the rectifying tower when the characteristic vector of the rectifying tower and the characteristic vector of the absorption tower are fused.
7. The production management control system for electronic grade hydrofluoric acid preparation according to claim 6, wherein the weighted weight of the absorption tower eigenvector is calculated from the following equation:
8. The production management control system for electronic grade hydrofluoric acid production of claim 7, wherein the responsiveness estimation module is further to calculate a responsiveness estimation of the product tracking feature vector relative to the synergy feature vector to obtain the classification feature vector with the formula; wherein the formula is:
whereinRepresenting the product tracking feature vector and the product tracking feature vector,representing the co-occurrence feature vector in the image,representing the classified feature vector in a manner that the classified feature vector,which represents a point-by-point multiplication of a vector,the expression takes the reciprocal of the value for each position of the vector.
9. The production management control system for electronic grade hydrofluoric acid production of claim 8, wherein the equipment management result generation module is further configured to: processing the classification feature vector using the classifier to obtain the classification result with a formula:whereintoIn order to be a weight matrix, the weight matrix,toIn order to be a vector of the offset,the classified feature vector is obtained.
10. A control method of a production management control system for preparing electronic-grade hydrofluoric acid is characterized by comprising the following steps: acquiring a plurality of first parameters of a rectifying tower and a plurality of second parameters of an absorption tower at a plurality of preset time points, wherein the plurality of first parameters comprise the tower kettle temperature of the rectifying tower, the body temperature of the rectifying tower, the tower top temperature of the rectifying tower, the tower kettle pressure of the rectifying tower, the tower top pressure of the rectifying tower, the tower internal reflux temperature and reflux ratio of the rectifying tower, and the plurality of second parameters comprise the temperature of the absorption tower and the pressure of the absorption tower; obtaining a plurality of liquid chromatograms of the hydrogen fluoride liquid at the plurality of predetermined time points before the first filtration by a liquid chromatograph; passing a plurality of liquid chromatograms of the hydrogen fluoride liquid at the plurality of predetermined time points before the first filtering through a first convolution neural network using a three-dimensional convolution kernel to obtain a product tracking feature vector; arranging a plurality of items of first parameters of the rectifying towers at a plurality of preset time points into a rectifying tower parameter input matrix according to the time dimension and the parameter sample dimension, and then passing through a second convolution neural network serving as a filter to obtain a rectifying tower characteristic vector; arranging a plurality of second parameters of the absorption towers at a plurality of preset time points into an absorption tower parameter input matrix according to a time dimension and a parameter sample dimension, and then obtaining an absorption tower characteristic vector through a third convolutional neural network serving as a filter; fusing the rectifying tower characteristic vector and the absorption tower characteristic vector to obtain a synergistic characteristic vector; calculating a responsiveness estimate of the product tracking feature vector relative to the collaborative feature vector to obtain a classification feature vector; and passing the classification characteristic vector through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the equipment performance of the rectifying tower and the absorption tower is normal or not.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210767053.0A CN115259089B (en) | 2022-07-01 | 2022-07-01 | Production management control system for preparing electronic grade hydrofluoric acid and control method thereof |
PCT/CN2022/116539 WO2024000798A1 (en) | 2022-07-01 | 2022-09-01 | Production management control system for electronic-grade hydrofluoric acid preparation and control method thereof |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210767053.0A CN115259089B (en) | 2022-07-01 | 2022-07-01 | Production management control system for preparing electronic grade hydrofluoric acid and control method thereof |
Publications (2)
Publication Number | Publication Date |
---|---|
CN115259089A CN115259089A (en) | 2022-11-01 |
CN115259089B true CN115259089B (en) | 2023-02-10 |
Family
ID=83763234
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210767053.0A Active CN115259089B (en) | 2022-07-01 | 2022-07-01 | Production management control system for preparing electronic grade hydrofluoric acid and control method thereof |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN115259089B (en) |
WO (1) | WO2024000798A1 (en) |
Families Citing this family (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115688592B (en) * | 2022-11-09 | 2023-05-09 | 福建德尔科技股份有限公司 | Rectification control system and method for preparing electronic-grade carbon tetrafluoride |
CN115903705B (en) * | 2022-11-30 | 2023-07-14 | 福建省杭氟电子材料有限公司 | Production management control system for electronic grade hexafluorobutadiene preparation |
CN117046140B (en) * | 2023-02-28 | 2024-03-08 | 福建天甫电子材料有限公司 | Device for rectifying electronic grade hydrofluoric acid |
CN116819957B (en) * | 2023-03-29 | 2024-03-08 | 福建省龙德新能源有限公司 | Tail gas treatment system and method for electronic grade lithium hexafluorophosphate |
CN117208852B (en) * | 2023-10-07 | 2024-02-13 | 福建天甫电子材料有限公司 | Purification control system and method for preparing electronic grade hydrofluoric acid |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106215447A (en) * | 2016-08-29 | 2016-12-14 | 成都灵感元素科技有限公司 | A kind of electronic-stage hydrofluoric acid production rectifying column |
CN111704109A (en) * | 2020-07-13 | 2020-09-25 | 南京德源环保科技发展有限公司 | Method and system for preparing electronic-grade hydrofluoric acid by continuous method |
CN212503988U (en) * | 2020-07-13 | 2021-02-09 | 南京德源环保科技发展有限公司 | System for preparing electronic-grade hydrofluoric acid by continuous method |
CN113651294A (en) * | 2021-07-02 | 2021-11-16 | 北京化工大学 | Continuous preparation system and method of electronic-grade hydrofluoric acid |
Family Cites Families (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH11149302A (en) * | 1997-11-17 | 1999-06-02 | Mitsubishi Chemical Corp | Operation control method for producing or processing process |
CN102183892B (en) * | 2011-05-10 | 2012-11-28 | 上海交通大学 | Load change energy consumption optimizing control method of three-column methanol distillation system |
CN103991847B (en) * | 2013-02-18 | 2015-12-23 | 福建省邵武市永晶化工有限公司 | Electronic-stage hydrofluoric acid preparation method |
CN103901130B (en) * | 2014-04-10 | 2015-04-15 | 中国科学院福建物质结构研究所 | On-line evaluation device and evaluation method for catalyst for producing dimethyl carbonate, and production method |
CN207210295U (en) * | 2017-08-29 | 2018-04-10 | 湖北工程学院 | A kind of ethanol purification system and chromatographically pure alcohol production system |
JP2021163207A (en) * | 2020-03-31 | 2021-10-11 | 月島機械株式会社 | Control device, control method, and program |
CN113703314B (en) * | 2020-05-20 | 2024-08-02 | 中国石油化工股份有限公司 | Online prediction automatic control method and system for carbon three-liquid-phase hydrogenation reactor |
-
2022
- 2022-07-01 CN CN202210767053.0A patent/CN115259089B/en active Active
- 2022-09-01 WO PCT/CN2022/116539 patent/WO2024000798A1/en unknown
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106215447A (en) * | 2016-08-29 | 2016-12-14 | 成都灵感元素科技有限公司 | A kind of electronic-stage hydrofluoric acid production rectifying column |
CN111704109A (en) * | 2020-07-13 | 2020-09-25 | 南京德源环保科技发展有限公司 | Method and system for preparing electronic-grade hydrofluoric acid by continuous method |
CN212503988U (en) * | 2020-07-13 | 2021-02-09 | 南京德源环保科技发展有限公司 | System for preparing electronic-grade hydrofluoric acid by continuous method |
CN113651294A (en) * | 2021-07-02 | 2021-11-16 | 北京化工大学 | Continuous preparation system and method of electronic-grade hydrofluoric acid |
Also Published As
Publication number | Publication date |
---|---|
WO2024000798A1 (en) | 2024-01-04 |
CN115259089A (en) | 2022-11-01 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN115259089B (en) | Production management control system for preparing electronic grade hydrofluoric acid and control method thereof | |
CN115097884B (en) | Energy management control system for preparing electronic grade hydrofluoric acid and control method thereof | |
CN115453990B (en) | Production management control system for ammonium fluoride production and control method thereof | |
CN115356434B (en) | Gas monitoring system and method for hexafluorobutadiene storage place | |
WO2024045243A1 (en) | Automatic batching system for preparation of lithium hexafluorophosphate and batching method of automatic batching system | |
Qiu et al. | Spatio-temporal tuples transformer for skeleton-based action recognition | |
WO2024000800A1 (en) | Energy management control system for preparation of lithium hexafluorophosphate and control method therefor | |
CN114870416A (en) | Rectification control system and rectification control method for preparing electronic-grade monofluoromethane | |
CN115309215B (en) | Automatic batching control system for preparing ammonium fluoride and control method thereof | |
WO2024021254A1 (en) | Intelligent separation and purification system for electronic-grade chlorine trifluoride | |
CN115202265A (en) | Control system and control method of intelligent production line of electronic-grade potassium hydroxide | |
WO2023206724A1 (en) | Rectification control system and control method for preparation of electronic-grade difluoromethane | |
WO2024045244A1 (en) | Energy management control system for anhydrous hydrogen fluoride production and control method therefor | |
WO2024021259A1 (en) | Automatic batching system for buffered oxide etch production and batching method thereof | |
CN115375691A (en) | Image-based semiconductor diffusion paper source defect detection system and method thereof | |
CN115090200B (en) | Automatic batching system for preparing electronic grade hydrofluoric acid and batching method thereof | |
CN114355240A (en) | Power distribution network ground fault diagnosis method and device | |
CN117523498B (en) | Dangerous waste treatment and recovery system and method | |
JPH03223933A (en) | Information processing system, alarm information processing system, and character recognizing system | |
CN115099684A (en) | Enterprise safety production management system and management method thereof | |
CN114610488A (en) | Data acquisition method and system based on load balancing and electronic equipment | |
CN116768206A (en) | Graphite purifying process and system | |
CN115908112A (en) | Point cloud registration method for low overlapping rate | |
CN115392320A (en) | Transformer with anti-theft function and method thereof | |
CN113092899A (en) | Transformer electrical fault identification method, system, terminal and readable storage medium based on multi-source information fusion |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
CB02 | Change of applicant information | ||
CB02 | Change of applicant information |
Address after: 364200 No. 9, Gongye Road, Pingpu village, Jiaoyang Town, Shanghang County, Longyan City, Fujian Province Applicant after: Fujian Longfu New Material Co.,Ltd. Address before: 364200 No. 9, Gongye Road, Pingpu village, Jiaoyang Town, Shanghang County, Longyan City, Fujian Province Applicant before: FUJIAN LONGFU CHEMICAL Co.,Ltd. |
|
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |